in the Nutrition Monitor (United States ing in An Update Report on Nutrition Monitoring U.S. DEPARTMENT OF AGRICULTURE Food and Consumer Services U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Public Health Service Ee ———, Poe PUBLIC HEALTH LIBRARY This report, Nutrition Monitoring in the United States: An Update Report ~~ on Nutrition Monitoring, is the product of an extramural independent effort RELEY ON performed under a contract with the Life Sciences Research Office of the .YARY | Federation of American Societies for Experimental Biology (LSRO) at the © “iv oF | request of the (U.S. Department of Agriculture and the (U.S. Department of ~LFIENIA Health and Human Services. The report was prepared by LSRO staff members in consultation with experts from nutrition, health, and related disciplines. The two departments jointly developed the scope of the contract and provided contract funds. They also provided background documents and data tabulations as requested by LSRO and the expert consultants. Although printed and distributed as part of a series of reports from the National Nutrition Monitoring System, the interpretations contained in this report do not necessarily express the views or policies of the United States Government and its constituent agencies. Copyright Information All material appearing in this report is in the public domain and may be reproduced or copied without permission; citation as to source, however, is appreciated. Suggested Citation Life Sciences Research Office, Federation of American Societies for Experimental Biology: Nutrition Monitoring in the United States — An Update Report on Nutrition Monitoring. Prepared for the U.S. Department of Agriculture and the U.S. Department of Health and Human Services. DHHS Publication No. (PHS) 89-1255. Public Health Service. Washington. U.S. Government Printing Office. September 1989. Library of Congress Catalog Card Number 89-14529 For Sale by the Superintendent of Documents U.S. Government Printing Office Washington. D.C. 20402 Nutrition Monitoring in the United States An Update Report on Nutrition Monitoring Prepared by Life Sciences Research Office Federation of American Societies for Experimental Biology (U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Public Health Service U.S. DEPARTMENT OF AGRICULTURE Food and Consumer Services Hyattsville, Maryland September 1989 DHHS Publication No. (PHS) 89-1255 C P AT. FOR TX 360 Foreword Ue N&él 1989 PuB il. This report reviews the dietary and nutritional status of the U.S. population, as well as the factors that determine status, based on the data and information available through the nutrition monitoring activities conducted by the U.S. Departments of Agriculture (USDA) and Health and Human Services (DHHS). This report is the second report on the National Nutrition Monitoring System (NNMS) and builds on the framework of the first report, Nutrition Monitoring in the United States——A Progress Report from the Joint Nutrition Monitoring Evaluation Committee, submitted in 1986. Its release follows closely the publication of The Surgeon General's Report on Diet and Health and the report of the National Academy of Sciences, Diet and Health: Implications for Reducing Chronic Disease Risk, but its focus is on the monitoring of dietary and nutritional status related to health, rather than the broader perspectives on diet, nutrition, and health encompassed in the other reports. This report was developed at the request of DHHS and USDA in accordance with the provisions of a joint contract, No. DHHS 282-87-0051, with the Federation of American Societies for Experimental Biology (FASEB). The report was prepared by the Federation's Life Sciences Research Office (LSRO). It was edited by Susan M. Pilch, Ph.D., Senior Staff Scientist, LSRO, FASEB, and was based on discussion of, and materials drafted by, the ad hoc Expert Panel on Nutrition Monitoring (EPONM). Scientists selected by FASEB as members of the Panel were chosen for their qualifications, experience, and judgment, with due consideration for balance and breadth in appropriate disciplines. Members of the EPONM and the LSRO staff and consultants who participated in the project are listed below: EXPERT PANEL ON NATIONAL NUTRITION MONITORING Milton Z. Nichaman, M.D., D.Sc. Department of Nutrition and Epidemiology University of Texas School of Public Health Houston, Texas 77225 Chairperson C. Wayne Callaway, M.D. Ronald N. Forthofer, Ph.D. 2112 F Street, N.W. Department of Biometry Washington, D.C. 20037 University of Texas School of Public Health Houston, Texas 77225 Oral Capps, Jr., Ph.D. Mildred Kaufman, M.S. Department of Agricultural Economics Department of Nutrition Texas A&M University School of Public Health College Station, Texas 77843-2124 University of North Carolina Chapel Hill, North Carolina 27599-7405 Catherine Cowell, Ph.D. A. Catharine Ross, Ph.D. Bureau of Nutrition Division of Nutrition New York City Department of Health Medical College of Pennsylvania New York, New York 10013 Philadelphia, Pennsylvania 19129 Peter R. Dallman, M.D. Howard G. Schutz, Ph.D. Department of Pediatrics Consumer Science Department University of California University of California San Francisco, California 94143 Davis, California 95616 iii CONSULTANTS Clement A. Finch, M.D. Consultant Seattle, Washington Helen A. Guthrie, Ph.D. Pennsylvania State University University Park, Pennsylvania Jean-Pierre Habicht, M.D., Ph.D., M.P.H. Cornell University Ithaca, New York Gail Harrison, Ph.D. University of Arizona Tucson, Arizona Stanley Johnson, Ph.D. Iowa State University Ames, Iowa Frederic R. Senti, Ph.D. Consultant Arlington, Virginia Theodore B. Van Itallie, M.D. Columbia University St. Luke's-Roosevelt Hospital Center New York, New York LIFE SCIENCES RESEARCH OFFICE STAFF Sue Ann Anderson, Ph.D., R.D. Senior Staff Scientist Barbara L. Durant Grants and Contracts Technical Assistant Kenneth D. Fisher, Ph.D. Director Betty Fraley Secretary Lisa Gasiewicz Technical Literature Clerk E. Lee Guttman Technical Editor J. Elaine Huey Technical Literature Specialist Judith F. Miller Administrative Assistant Susan M. Pilch, Ph.D. Senior Staff Scientist Rosemarie V. Soulen Administrative Secretary John M. Talbot, M.D. Senior Medical Consultant Martha A. Watt Word Processor/System Manager The contractual activities were overseen and assistance was provided to the EPONM and LSRO by the Joint Project Steering Committee (JPSC), which consisted of representatives from each of the Agencies within the departments concerned with nutrition monitoring (see list that follows). The cooperation and the careful, conscientious reviews provided by the JPSC were essential to the successful completion of this project. JOINT PROJECT STEERING COMMITTEE Catherine E. Woteki, Ph.D., R.D. Division of Health Examination Statistics National Center for Health Statistics Hyattsville, Maryland 20782 Project Officer and Chairperson iv Frances Cronin, Ph.D., R.D. Human Nutrition Information Service U.S. Department of Agriculture Hyattsville, Maryland 20782 USDA Liaison Melody Bacha, M.P.H., R.D.! Food and Nutrition Service U.S. Department of Agriculture Alexandria, Virginia 22032 Gerald F. Combs, Ph.D. Agricultural Research Service U.S. Department of Agriculture Beltsville, Maryland 20705 Darla Danford, M.P.H., D.Sc., R.D. Division of Nutrition Research Coordination National Institutes of Health Bethesda, Maryland 20892 Patricia M. Guenther, Ph.D., R.D. Human Nutrition Information Service U.S. Department of Agriculture Hyattsville, Maryland 20782 Gerry Hendershot, Ph.D. Division of Health Interview Statistics National Center for Health Statistics Hyattsville, Maryland 20782 Phillip Kott, Ph.D.2 National Agricultural Statistics Service U.S. Department of Agriculture Washington, D.C. 20250-2000 Linda D. Meyers, Ph.D. Office of Disease Prevention and Health Promotion Department of Health and Human Services Washington, D.C. 20201 Betty B. Peterkin, B.S.3 Human Nutrition Information Service U.S. Department of Agriculture Hyattsville, Maryland 20872 Frederick Trowbridge, M.D., M.Sc. Centers for Health Promotion and Education Centers for Disease Control Atlanta, Georgia 30333 Susan O. Welsh, Ph.D., R.D. Human Nutrition Information Service U.S. Department of Agriculture Hyattsville, Maryland 20872 Elizabeth A. Yetley, Ph.D., R.D. Center for Food Safety and Applied Nutrition Food and Drug Administration, HFF-265 Washington, D.C. 20204 Special appreciation is extended to Marie T. Fanelli-Kuczmarski, Ph.D., R.D., National Center for Health Statistics, and Susan M. Krebs-Smith, Ph.D., R.D., U.S. Department of Agriculture, for their dedicated work in coordinating data analyses and other supporting activities in the Agencies. The EPONM and LSRO would also like to thank the following Agency scientists who provided invaluable assistance in supplying information on NNMS surveys, performing data analyses, generating the data tables included in the appendices, and reviewing sections of draft reports: P. Peter Basiotis, Ph.D. U.S. Department of Agriculture Katherine M. Flegal, Ph.D. National Center for Health Statistics Margaret D. Carroll, M.S.P.H. National Center for Health Statistics Peter J. Gergen, M.D., M.P.H. National Center for Health Statistics Olivia Carter—Pokras National Center for Health Statistics Joseph Goldman, M.A. U.S. Department of Agriculture Joseph E. Ciardi, Ph.D. National Institutes of Health Joe Fred Gonzalez, Jr. National Center for Health Statistics Nancy D. Ernst, M.S., R.D. National Institutes of Health Brucy C. Gray, M.S. U.S. Department of Agriculture 3 Served October 1987 to November 1988 2 Appointment began November 1988 3 Retired October 1987 v James T. Heimbach, Ph.D. U.S. Department of Agriculture Stephen P. Heyse, M.D., M.P.H. National Institutes of Health Van S. Hubbard, M.D., Ph.D. National Institutes of Health Clifford L. Johnson, M.S.P.H. National Center for Health Statistics Robert J. Kuczmarski, Dr.P.H., R.D. National Center for Health Statistics Christine J. Lewis, Ph.D., R.D. Food and Drug Administration Ephraim Y. Levin, M.D. National Institutes of Health Anne C. Looker, Ph.D., R.D. National Center for Health Statistics Jennifer Madans, Ph.D. National Center for Health Statistics Ruth H. Matthews, M.A. U.S. Department of Agriculture Margaret McDowell, M.P.H., R.D. National Center for Health Statistics Alanna J. Moshfegh, M.S., R.D. U.S. Department of Agriculture Robert S. Murphy, M.S.P.H. National Center for Health Statistics Matthew Najjar National Center for Health Statistics Gregory Pappas, M.D., Ph.D. National Center for Health Statistics Betty P. Perloff U.S. Department of Agriculture Nancy Raper, M.S., R.D. U.S. Department of Agriculture Robert L. Rizek, Ph.D. U.S. Department of Agriculture Laurie Roidt, M.S., R.D. U.S. Department of Agriculture Michael L. Rowland National Center for Health Statistics Christopher Sempos, Ph.D. National Center for Health Statistics Zekin A. Shakhashiri, M.S., M.D., M.P.H. National Institutes of Health Ann W. Sorenson, Ph.D. National Institutes of Health Carol Tuszynski, M.S. U.S. Department of Agriculture Barbara A. Underwood, Ph.D. National Institutes of Health Faye Wong, M.P.H., R.D. Centers for Disease Control Ray Yip, M.D. Centers for Disease Control The Expert Panel met eight times between November 1987 and February 1989 to obtain background infor- mation on the NNMS, to review analyses of NNMS data published in the literature and prepared especially for this report, and to review drafts of the report. The LSRO and the Expert Panel held an open meeting on January 20, 1988 to hear oral presentations of data, information, and views on the topics related to the NNMS. At that time, one presentation was made by Mary Enig, Ph.D., Research Associate, University of Maryland, College Park, MD. A transcript of the open meeting is available from Ace Federal Reporters, 444 N. Capital St., Washington, D.C. 20001. Written comments were also invited in the Federal Register announcement of November 30, 1987 (52 FR 45504) and December 18, 1987 (52 FR 48160), and a tentative outline was made publicly available for comment. Written comments were received from the following persons and organizations: Mary Enig, Ph.D., Joseph Sampugna, Ph.D., Frank E. McLaughlin, J.D., and Mark Keeney, Ph.D. (University of Maryland); Elaine Blume (Center for Science in the Public Interest); George L. Blackburn, M.D., Ph.D. (Harvard Medical School); J. Edward Hunter, Ph.D., and Thomas H. Applewhite, Ph.D. (Institute of Shortening and Edible Oils, Inc); Lawrence J. Machlin, Ph.D. (Roche Vitamin and Fine Chemicals); and Patricia A. O'Malley (Hunger Services Network). The data, information, and views presented at the open meeting and received in writing were considered by the Panel in reaching its final conclusions. vi The EPONM reviewed each draft and the final report and provided additional documentation of conclusions and viewpoints for incorporation into the report. Members of the JPSC and the Deputy Assistant Secretaries for Food and Consumer Services, U.S. Department of Agriculture, and for Disease Prevention and Health Promotion, Department of Health and Human Services, also reviewed final drafts of the report for technical accuracy and satisfaction of the scope of work. The EPONM and the LSRO accept responsibility for the study conclusions and the accuracy of the report; however, the listing of these individuals does not imply that individual Panel members specifically endorse all statements in the report. The final report was reviewed and approved by the LSRO Advisory Committee (which consists of representa— tives of each constituent Society of FASEB) under authority delegated by the Executive Committee of the Federation Board. Upon completion of these review procedures, the report was approved and transmitted to DHHS and USDA by the Executive Director, FASEB. Although this is a report of the Federation of American Societies for Experimental Biology, it does not necessarily reflect the opinion of each individual member of the FASEB constituent Societies. uty 37 155% ILD fn \/ (date) ’ penneth D. Fisher, Ph.D. Life Sciences Research Office vii Table of Contents Page FOREWORD .......oooitiiititeietetetetese tessa saa ese esse se ste se tes ese eset essa seats e seats sates sass eases ssa h sess h esas snes esas sass ss assanansesnasannas iii LIST OF TEXT TABLES AND FIGURES .........ccccocriiiiiiiiinniienninentesesesesesssssse sess essssssssssssesssssssasens Xv EXECUTIVE SUMMARY .....c.cconiiinmimnimmimmitsts bo i £5 bess se se see ree seme nae ss sors mss sons sos sespamsaonsssnns xxii CHAPTER 1. INTRODUCTION .......ccooiiitiitrieieieieietssetessssesssissssstossssessssssssssssesssssssssssssssssssssssassssessnsssssanes 1 Charge to Expert Panel on Nutrition Monitoring .......nnnnnnnnnrrsrssenstnes sss 1 BACKFTOUNM oor sss sss sss setae ss se ses STS SS SS ES se se 2 Brief history OF te NINE .. meus srmmissmsss sisi om srs osomm sors oasis sss assassin ss re 2 Goals and purposes of the NIMS ......cmaermmmimmmmmmmssse ms sessssmss sess ress aes sass (410010394 2 Components of the NIMS ...ouunumnnanummmummmmnrrmatm moms sss assis sss spams 3 USO OF NINMB GALE ovrissusrminsarmmssrssmsrsssoamspngh 53s 150 SS EE PRE AEE FARR HERE HIST 3 Major conclusions and uses of the JNMEC 1aPOTE ...cccnmmmmmmmmmmssmmssmssssssssssssssssrsssrsrssorspsvsrsrs 6 Principles and DofInftions USE uum sisi sss ss sais sess sass ra sss sss a aassssess 6 General Conceptual MOE ..........ccoueiieiiiiiieiieieieieieieicieeceesesissssessssssssssasas assassins ss as ss nenanes 8 Organization Of the BRDOTE .cunuunnmmmmmnin tmnt mons sss sss assesses sss ss sms sass ss sass a sass raves 10 Roforencos CBA .....cimummmmmmsmmimimsmsrssssessrsnsrsrssesses srt sos shah sansansase sts beast estrone nisin mea iensi nia mds as SAREE 10 CHAPTER 2. APPROPRIATE USES OF SURVEY DATA IN THE ASSESSMENT OF DIETARY AND NUTRITION-RELATED HEALTH STATUS ...........mmmmiiine 13 Assessment of Dietary StAtUS ........ccoceeiiiiiriiiiiieec cece beens snes 13 AvAIaDIC TORIINOBOIOZIOR ocyen: ivissumpimsergsunses gins svsruss 634044 Eo4 oa hea a EE SRE RE ERE HERE EERE ATA SELES ASAI 13 Criteria for assessment of dietary intake of nutrients (and other food components) ..........ccccccueueee. 16 Contributions of supplement use and other sources of food components to QIBLATY IIRAKD uorrcrcsmmmmmmmmmmmonmnmumsm mms m—————- = AAS 17 Consistency in data sources to assess changes over time 18 Assessment of Nutrition-Related Health Status . cman smn sori mst iio seams: 18 Available MethOdOIOZIES .......cc.cuiuiriiriiiiirieieteee cies sete sb sass eases sass as nesses essen eaneanenes 18 Criteria for assessment of biochemical, hematological, and clinical measurements ........................... 19 Consistency in data sources to assess changes OVET HIME ..veveeieiieierieieeete sects e eee eseee se seaeseeseesaeaas 20 Other Considerations in Interpretation of Data... 20 Linking dietary intake data with nutritional status and health outcome ...........cccccoevrrrrirrenrrrnennnnne. 20 Noncoverage of certain population groups ..............nnns ess 21 Changing characteristics of the population ...............cci es 21 Restricted age groups in SOME SUTVEYS .........ccoriiuriiiiiieiiininiiiii ses 23 Sample size restrictions for subgroup analyses ............ciiiiiiiiniinicie ee 23 Nonresponse and adjustments fOr NONTESPONSE ..........cccvuiuiuiiiiniiniiii ae 23 Sample weights, variances, and design oftocts ee reteetete eee en ete heehee beta b ete t ete b ene t ete nesas 23 Statistical Criteria and Data Reporting ........cccoevevreninenrnntnncnnnnnnne snes esesae sees 24 5 OF ALA PLESENLEU. orien sims res sass feasens teers sept aen pens apes se seme mess ESSE TSR TOA I Er 24 Criteria for reporting and evaluating data 24 Imputation in specific surveys ................ . Age-adjustment procedures in specific SUIVeYS ............cccoevveecururecrcnrernnnes NoNresponse in SPECIFIC SUTVEYS .........cc.cvureerrreuerreresessesessesessesesssssssssssssssssssssessssssesssssssssssssssssssssssssssssssssens Variance calculation and design effects in specific surveys Use of Systeme International (SI) units References Cited... sess sess sss sessesse sss ssssssassssssssssassssnssssnssssssssssssssssssessesans 27 CHAPTER 3. UPDATE OF DIETARY AND NUTRITIONAL STATUS: INDIVIDUAL FOODS AND FOOD COMPONENTS ..........ccoecviuerrerrerneeneasnessenesessessesssessesssssssssssssssssssssssnns 31 Introduction .........cccoceevevenrnreevecnerennnnn, eee eee ee ee ee eee ee eee eee eee eee eee eee 31 Trends in Food Availability and CONSUMPLION ..........cceeuerreruecreeesrserssenesseenesssneesesesessssssesssssssssssssssssssssesssssns 31 Food availability ................... Food consumption Update on Individual FOOd COMPONENLS .........c.ceueereereererenressessessesssessesssssssssssssssssssssssssssssssssssssssssssssssssssessess 38 Quanstiy and qUALity Of AAA .........ccccocuiieiceiriieenit eee sess eens senses seen 38 lassification of food components by MONItOTINg PLIOTILY .......ccceeveererriereruerrrsreesreesiensieseeeseeseeseseneseens 42 Approach to assessing which food components represent public health issues In national RUBLION THOMIOTING wm sem ss ss sss rss 44 Discussion of individual food COMPONENLS ...........c.ceuuerueruerrrurrecisessrsessssssssessessesesesssessessssssssssssessssssssens 45 FOO ENETRY ....oveeerr renner sss snes sss ss sssss sss sstsssssssssssessssssessessssssesssssssssssss esses snsnsssens 48 PPOLBII crvvspreparisincsnrsmmasssstsmpmastgprssmssmmesssmesersomsenssoms eres seemeemesi sss sms srs ses saan sia esis vib es esomismeomemenss 50 Fat, Fatty Acids, and CROIEStETOL ...........cc.eceeuereeuererrereircrineniseiscaeseessesecsesssessssessssssssssssssssesssessssessssssons 51 OATDORYOATALES «vcuccnsnsssesismnsisnismmmmmommminmnsmesmssismsimmism omer mers ses sam a OOo STHS 54 DISLALY FIDBE suvvencrcrerciisismmmmummmsmmsmisiisismssio srs seme seas sa as a sa s8 S190 SABIE ESSE 54 AUBONOL consescinriviecnermsisctasistasiasissismssmmamsessmmmmmesem———————s A AA OAT SEAS 55 Vitamins Vitamin A an CarOloion .....cuwmsmmsemssiismsrsismreomimiessmesmssesmass mss at sess ats rsssasmessmssassmeamemen 56 VHAMIN B cicnommmmmmmmmmmmmmmmsim msn espe ssa mess ss et seams sees amomsssessm meer 57 TREATINN wonsossmncrmcrencersuss stan amos sis ismmsmsmiomets moomoo see EASA SA SR A SE SATE STROSS 58 BIDOIAVANL avconimnescassmusisstississisamssmessessossmmsmmsesmmcosonssnsemmsaesmeemesme esac senss assesses sissiamessesonsinsminnts 59 EL 60 VItAMIN BO... sa sass sass esas sees e sess seers essen 60 VHLAIN BID. ccnrmmmmimimmnammmimsimsmssiomomes sesso iesieiese nt sesmsras es ass seats seas senses msssomsemetisenss 61 VHAMIN © sovcrmnmunmecseeseescemssmssssmmmisssitis ios sspears seme ems ses sms ss ss ss set sees s——» 61 TPOLOCTIL uusmsnporirersercererssassmssiisssmsstsomsossesssosmssesmsamemmemmssmeemsss memes aS Ss ATR SH SSF SHS RS Ha TR TIAA 62 Other VILAMINS ........ocouiiuiiciriireieeiesissesie sess ses ss sses ses sssssssssssssssssesessssessessssssessesssesessssssessssssessses sens 63 Minerals TPON oussnercorscssinmsserusinsisrasiasisiasisem ons vasmssrmsssmsmsmesmesrs Hemme REE AR SH L 4 A HTH SRTRA SIA SESARIH TATIONS 64 CALCIUM ....ooeeveiieiitiiiiiciciseeis ets ss sss b essa sae sss as sass sss sss sass setae se sassssesssessessssasssesssssassaees 65 PROSPROTUS .......oceiiiiiiisesss sss estessssssessa sass ses sess sss taste assesses sass ses eases esses sas sass sera 66 MBINGBIUM orvercrncsrarnosms smmmmmemsmem sms meer ss RSA SATs ORS SSS SCRE ES 67 JUIN cccertienntnnrin nese ses assesses ssenss sess sssstssessesssesssesstsstsssass esses sesssessenssesseesessssessesesssessessssssessssssessenssaenee 68 POLBBBIII wivisiicscsrsinsisstssinssstsemsomsmssmmesmessmmemmseresosmeemes es sssenmemensessenss ss sss intsmessssesmtanississsssmomiomoernes 69 COPPET ....ooceveirritriniiiseisiseieisseissssss assesses sass bes sass ssa sass sss sss saab ass sae sasessesassanessesssssessssse ss sasssssseesassaees 70 ZINC ...oovereriisicvsiiieia sess siesastassassb sss senate basa b esses essa assesses esas tesa ee tee aee esses sess esses rane ee rans 71 Fluoride and OLher MINOTRIS uu ammmnsssmmmismmmscmmesms ramps mots messsessrssssesmsmonon 71 COTICIUBIONE cvsusncrivnunesrerseshassiashus sss tosses ssns smmssmsssamephssmesmmesmens eae SALAS RASA A HSH SATA HRSA SRS SS BBO 72 References Cited... sess sass sass sss sss sass sas sss s ses ee ses sss sess esse sess sass sssans 73 CHAPTER 4. UPDATE OF SELECTED HEALTH CONDITIONS AND BEHAVIORS: RELATIONSHIP TO NUTRITIONAL STATUS .......oorriinnniinniinniisnisensessssssssssssssssssssssssssssssassssenes 77 TIELOQUCLION ..eeveieviereenretiereeteeereeeecteeee ee sse sae ssesssessaesasssesssssessstsssessessessesssessessasssesssessessasstessesssessassesssasssssssssnssnssss 1 Nutrition, Morbidity, and MOTtality .........ccccmmmmmmmmmmmmmmiim mm msm ns 77 Health Conditions Influenced by Nutritional Status ODOTILY 1vssmissemsnsmsmmmmorsnsnsiasns iassussencnoneessrresmsmemsmensoress sess sass ssos8 14544454413 3 DIBbotes ....covmninsnsmsmemsmammens . w: CRIICRT 110s ie sn viireniminirimmanismm ranean ese SA OA HF BH EH EEA EE ERO EERE EERE EPPA IE AAAS 22 (OSLOOPOTOBIE .ererrerrerecsrrrresersrssesserssrssarsussessesansasssssss ss 4884404 KIARA LH HEISEI ERI AIE FELIPE PEI IESE ES LIES SASS HS SASSER IAT TATA S424 DIENEAL CATIES ...oovevverierirererrereiereeese reese estese tesa sessssesnsstsssstssstsbesest bess sas ass sass eases arses ars asantasastasastasastasastasassasasns Low birth weight .........cceceneuuene Growth retardation .........ccceeeurennene. os PIOgNBNCY «oii vier isimmenivoins Dietary Behaviors That May Affect Health... 87 BRASLIGORING 1. ivsssssssnpvsvmtronststrsrs os os masse ER EE AREER SSSA ANSI AISI ES EIA EIA SHIA IA 4 0014 87 VHamin/mineral SUPPIBINENt USE w...merrimistsimmisrs temas sess messes mss ssss ists ss sess assesses ses 88 CRanging GIolary PPACLIOES .....orrssmisisinsmsisssnsusnsnrnonsmsmsmssomssssssessssssessssssssssss ss ss oo oro 1960616008010 Res 12 89 Nondietary Behaviors That May Affect Nutritional Status .........cooeinnnnniinns 91 SUDBLATIOB BDUSE .ovrirrirrivmimmmmsivimssmsinisimsinissiosons sss ss SHES b Fk Ba 54 hea AR HH LEE KE SSPE ROAST SIERO 1927 91 SINOKING .ovvrvrererrrer reser sess sss sass ss eases rE EEE ST a a A RR A 0 91 EEROICISE ..ivircirurcnmrsrssrerresmansursumsaresdorsumsarasssmass sss srssss sass 48448844444 AAS LI SH HLRELH HERE KAPITI LISSA ENSERS SHH e 0303 91 Use of oral contraceptives 91 Use of medications .......cccccceveeueueee 92 SUBINATY «over irirnrersnssrs ssss s—— T T SSSSA E0 0 92 References Cited .........coceeeeiiiererieiiieeeieereitsesesseste see see ee sesssssestestssesssss est sssebasss sss ssassastessasaasssssasassessassassassassessenes 92 CHAPTER 5. NUTRITIONAL AND DIETARY FACTORS IN CARDIOVASCULAR DISEASE oor simian stam ars 546419424 £15142 42 4 HATES SEES EIS EAI A HH 608 95 PUEPIOBE: vavunvsrsssnsarsnsssnssnssns snsssmss sass sastssesesesersssnsnaresssemsaemsssssssensssssssasatass assess £4894 480414448 6 ehsk i iin HHI HEIFERS HIS 95 DEfINILIONS .verrerrcrrirvimmmssrssrsorsrrarsrsssssssssestassssesses ses sas sates 604 541A SHLAA TERE FEAR AHE EIA REL HE HEES EHS SERIES IESE SIARIETSARIARASS 95 BACKEIOUNA corsessmossmrnsisrarsosssnsnssnsconsss os sonsssstia st otareemsesessaemresesememsamssssssst sass 4458004984900 0944 5413 EAVES RTARTA IER ITI 96 COrODATY NOATE BISONS wisinmmmmmmssmisimmmmnns sens bere rere Se STIS ESET IAI IA 01 96 HYPOTENSION o.com sss sts ss re ss A A SA SA RSA 0 96 CerebrovASCUIAT AISEASE ......ccceveeverirreeiireinreestesese ee saesests estes etter tebe tebe sbeebs ass b sbeebs asa ea b essa esate sa sbansesanes 97 Conceptual MOE] ........cerrrrrr rr ———————————————————— 97 Provalence OF DISEame ......cvmmisimmisimrmiios assis ise sham orm sams se sss aes sass ses sss sss s ts sar sss ass tH 949 840060402 99 RISK FACLOTS ..veveverereierereiinieeeetesestesesae se sessese sae sessesesasssssssesssssstssestsssssssssessesessessssessssesssssassessssesessasessesestasessesesns TIYPRICHIONBBLETOIBINIA .osvsivismmssissimnnsissmsvmsns oss soso EAS PE EEE PEA AEE PEEP PESTA ASSASSINS Blood pressure ..................... Diabetes ......cccoceveerereniicneeeeeeee ODEBILY ..veveerrreererenrerenerrernnsiresnnisssessseenssesessssesenes Smoking, exercise, others DIBATY TPUCLOTS 1uiuisininisisssnsisnsnsnossvmvmossannsssn sess manta enn hs mews m ee SETI SSIS AO AAAI EH ESHA SAAB II SAA S194 POO BIOTLY: 1 vsseasnsnpsorgs savas sss vty stants so ins ss satus css mus REELS HERDER SAHIIII SA IATA TATE ATASEAARS 110 Fat, fatty acids, and Cholostor] ......cummrmirmmmrormmmsmmmmimms sass 1614004806001 61 1140051000140 110 xi Sodium ............. RRs oor pierre Calcium, potassium, magnesium, fiber ALCORNOL cuties east sass ess b ese s essa snes sense ses essa bebe se bese s Rese s ese seas eaeae nen neaene ASKICIOLIONS ssnsmenrsmmmonmsimss Ecological associations of changes in cardiovascular mortality, diet factors, and risk factors over time ............cccceverererererieeci eens 115 Cross-sectional associations of dietary intake and other risk factors ............cccceeoeueuerveeereereeecceeernenn. 117 Cross-sectional associations among nondietary risk fACtOTS ............ce.eeuecuerereruereurserreeressessesesessensenn. 118 Risk factors and subsequent morbidity and/or mortality experience (cohort studies) ........................ 120 Knowledge, Attitudes, and Behaviors Related to Cardiovascular Disease Prevention ..........cccoceeeeveveunne. 120 Major Limits to Interpretation of Data and Gaps in the Database .............ceceeeverererererenrereneceenenseeeesenns 121 OONCIUBIONE. ocovrincrnerniinnsionssnsnssssssss stessessssessssessmmmmsssssssmenss sero smsss sm RAE HEE HEE HERS SRE STRAY 123 POPUISLIONE AL TISR womens sree ves Hise Roe HR FEE Rie edi dis mana sa savas a A901 123 TTTONIAE ,.cvccnvsvsreesmsismun sass ss BRE TTS REAR ROTA RTA ime EEL SA HERE ESAS 123 Determining fACLOTS ..........coerueerererrriniresinineensinenee sees eee essere esses sees b ese sas sess sa sess asses sensesenens 124 Future Concerns and RecOmMmMENdAtiOnS .............cceeeueueurureerererenesesseseseesesessssesssesesesssssssssssssscsssssesesessensssanes 124 ROTS18n088 CHB unusumnmmnimmmmmmmmsms mma is re ime ees ee siocas sana obs ss sa 9040 0s 00800081 S 124 CHAPTER 6. ASSESSMENT OF IRON NUTRITURE ..........ococeoeueiireirereeereeenesesessssssesseesessssssesesasssens 129 PUTPOBE cuvettes seas sess sass sess essa s assess sates et atest testes ett etetetenetenennns 129 Background and DefINILIONS .onammmnmmmsumsmmmsimmismesmsmmssisimss mss esse se os sam sas se 129 Functions of iron in the DOAY ...........ccccerieiicieineeieeeeeeess sess ssse sess ss ses ssssessssesennn 129 IPOM DALATICE .....oovuieieiiiiiiiiciceciitcci teeta bess bess e sbeebs bes sa esses s eset eben assent tes sesssenesaen 129 Iron deficiency: stages of development and assesSMENt .............c.ceueveeeueuerereeernrereseiicceesesesseesessaenenne 130 Iron overload and hemoOChIOMALOSIS ...........ccceuiveieieeuereirieeiereieee tesserae sess sesae sess sasnaens 135 CONBPLUAL FIOAB 1iiiiisisnmmmmmmmmmsisismmom————————————— TS LS 135 Estimates Of PPEVAIENCE ...........ccoveuiiiiiniiieiiiciesisseess esses esesseses essa ses assesses sess sesasssssssessssssassesasassns 137 IPO BElICIANUY «ccimimmmmmunmmmmms isms rm ——————————————— 1 1 A SS A SR 137 BNOINIR oy crcriprnrvimmmmismmstisssim ovitism sss m———————— ER RAAT SISA IAS 138 POM OVRTIOB corcivrcrmmsucicmssssissmsssssrsrossssrsssssssmussssssmsumsssssseresssm sens REE SRBC SER HS A AER RHEE 139 Dietary Factors and Supplement INEAKe ...........cccoeceurieeuereeiernenieieseseiessesesesessssesessssessssesssssessssssssessssssssssees 139 Iron available in the food Supply Eieernt tre ntnse etnssgt 0010041810000 EOIN TOR ATTRA TASS TTR AR ORE SEER ERA SHORES? 139 Food sources of iron available in the food SUPPLY .........cc.cceeueiueeueiueieieiseresrieiseessessseseesesessesesssssssssseenns 140 Iron content of typical U.S. diets ........cceccemreuererreerrereecieseseseesscssesess rene sessesssesessssessssssssssesssssssssssenes 140 Mean dietary iron intakes among age and SeX GrOUPS ...........cccccuuriumurimsrimunessssncnssnesssnessssesssssesssessenns 141 Relationship of iron deficiency and iron intake .............cccceeeeueeerecereeeenereeseeeseeesesssesessesssesesessessssssseseens 141 F0Od COMDBINALIONS .......cuuiiiriiiciieririeinieinieisieesss esses sess sess sesss sess se sess sese sass ss ssssesassssassossssosssassessnn 142 Use of supplements: iron and sCOTDALE .............c.eeeuereeuerrereresireesisnssesessesesessesesssssssesessssessssssssssssssssesssnn 143 Tron fOTUfICAtion OF OOS . uu ummm rR Ae eis ana sneasas rans sa sssapss ness 143 Nondietary Factors POVETLY ouvir sees esses sess sbeebs assesses assesses tastes assess senna sane Education Eig Grow, in children . regnancy Blood loss xii Major Limits to Interpretation of Data and Gaps in the Database . we ve 146 CONCLUSIONS .coccrcerirrcrrcsrcrresrerersrserssrssersersessestss ss tss ess esses ses 4 08 HILAIRE HI SIEIEIELIEHPEEPEAIIAPEAP EES IAT EES SSES IER SERE ERAT ITAA ATS 146 Populations AL FISK ......covinrrnrnsnnrrnssismsssss is mss sss sss ————————————. 146 Trends .....cccceveeeerrucneinnennecnnnne : treereeereeneetesaeesaesateteasesserter teat sate aeerae bees 147 Determining factors .........cccccceevuenne . PASSA SSE AE STA SHS SR A RA a ep 147 Future Concerns and Recommendations ...........ccceeeeeerenrnrinunienininisninsneessiesssssssesssssssssssssssssssssssessssssssssns 147 RALOPONI008 CIOA ..ocnrcrririmmnmrivsiiormrisssmsvsimsesnesn ess sess sh S54 S44 EOE SEES HARARE NRTA ON ESSA AT SAS 147 TEOME TABLES .ueeeeeeereriieretecieeseesesseae sae seesesesse esses se sassassssssssessssssstestostsssrsessessessessessessesessessessessessessasessessessessese 150 CHAPTER 7. RECOMMENDATIONS .........cooiirinninnnininsisnssisissesesssssssssssssssssssssssssessssssssssssssssesssssssssasss 153 Comparability and Compatibility Among Components of the NNMS .........cccoovinininninnneneninencenenninns 153 Needs for Data Collection, Analysis and Dissemination ..........eiennenennnnenne. 154 Coverage of groups currently excluded... ans 154 Improved coverage of ups currently included... ne 154 Dietary data in the HANDS .....commmmummmmunimmimsmmmmsommmmssosmmsissis ssa 5 4439459194149 14041698014 155 ENOWIEAZ0 aN BUIGIAEE «cuisines ss sansa ssssssssiss sss ss ssa 939 SHIA II 9424 155 Vitamin and mineral SUPPLEMENLS ........c.cccerrieriririniniriniiiinniiinniiniessesssssssesssssssssssssssssssssassessssesssneses 155 AlCODO! CONSUMPLIOI ...vcuviiriiiiiiieerieeeeetecetsestesestststssetsssnssssesss ses sssesaess sas ssess snare saasessssassssassasassanassanasns 155 Nonresponse analyses... rsersenisabe eh TES EIT EAT R 155 Education Of data USETS .........ccccceeveerierrerenreerrenreseesesssssssssessesseisississssssessssssssssssssssssssessasssssssssesssssssssssssans 156 Responsiveness to needs of State and local data USers ............ooviiininineiinennnicccne 156 Research NEads .........cvvv remiss mas sss ss IS LS SS SS SS RSS SST TE 156 Future Reports on the NINMS ............ ccm sss ss ss ss mms sass 156 Content of reports ........ 156 Frequency of reports ; 157 The NNMS Of the FULUTE .....ceciieieieeeeereecnetetecetseesestestsnsisnestsssssssssssssessssssssessssssssssssssssssssssssssssasaens 157 References Cited ...cnsnnsinmmmmnasis i. teereeeereereaeeesestestte se teneteseaesanenestenartentns 157 APPENDICES 1: Description Of WINMS SUTLVEYE cusmmmmmmiomsm nna stsssseeseseseneseneresecsemmmesmmsssossesmsmssssmsssosssssetesssssssns I-1 II. Summary Data on Food COmMPONENLS ........ccccereriirinuiiinniininerinnienniennseesesssesssesssessssssssssssssssssssssssssns II-1 III. Glossary of Terms and ACTONYMS ........c.ccceuvururieresniriisnsnsserssenenes A ET III-1 IV. Recommended Dietary ALIOWAICES .........ccccecerirrerreerecirrisinsinsisiisiisnisniisessessessessesssssssssssssssssssssssssssssssenns Iv-1 xiii EE Tv List of Text Tables and Figures Tables Table 1-1. Sources of data from the National Nutrition Monitoring System considered in the EPONM TEPOTL .......ccccceveeruernesunurisisisissesssssisssessssssssssssssssssssssssssssssssssssssssssssssssssssssssssssasasssssasasssasas 4 Table 2-1. Effect of prevalence on predictive value when sensitivity and specificity equal 95 percent ..... trrerttrestr_tr_——_— SII IIIS E SEES E SEARS ESSAYS A HS A HS Se 20 Table 2-2. Recent estimates of population groups in the United States .........ccooveeenieviniiniiiinninnnniinininnnnn 21 Table 3-1. Percentage of persons using selected foods and mean intakes for men and women, aged 19-50 years, in 1 day in 1985 and percentage change in mean intakes from 1977: Continuing Survey of Food Intakes by Individuals 1985 and Nationwide Food Consumption Survey 1977—T8 .........ccooreireninininnnieninisnnisniiineiiiensnisnnsnssssssssssns 37 Table 3-2. Summary description of food components assessed in the EPONM PEDO. sunnmcrnunummunurismnrreenmsnommemnsepssetstsnsssbiiss rR SAAS SE SAA AR ERR rer S 39 Table 3-3. Major sources and types of data related to individual food components available from the National Nutrition Monitoring System and used to update the 1986 JINIMEC TEPOTL ...cucueueuereririniitereteteseisasisssssissse sss sss ss sssssssssssssssssstsssesssssssss ttt t asst sash she AeA SEs Ashe Es Rea eases a sts a sass essa nets 41 Table 3-4. EPONM and JNMEC classifications for monitoring priority status of OO COMPONENLS .....cvverercririrreresicscrcssisisiissassssss assesses sess sas eases sess sss ERS SS aE sabes aes 42 Table 3-5. EPONM classification of food components by monitoring priority ..........eeeenennnnnncncccnees 46 Table 3-6. Comparison of JNMEC and EPONM conclusions on classification of food components assessed in the NINMS .........coiiiniiieiiiiiiis sss sss sss snes 47 Table 3-7. Mean intake of food energy in kilocalories for women aged 20-49 years, 4 nonconsecutive days, by household income (percent of poverty): Continuing Survey of Food Intakes by Individuals 1985-86 ...........cccccovuniniiinininininmimiiieceene 49 Table 3-8. Mean intake of food energy in kilocalories for women aged 20-49 years and children aged 1-5 years, 4 nonconsecutive days, by household composition categories: Continuing Survey of Food Intakes by Individuals 1985-86 .........cccccoeuvuimniimiinmininnnea 49 Table 3-9. Age-adjusted percent of overweight and severely overweight persons aged 20-74 years, by sex and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination SUIvey, 1976-80 ...........cccocernrnneresssinisisisssssssssssssssssssssssssssiss sss ssssssssssssssssssssssssssssssssssssssssssssssssnses 49 Table 3-10. Age-adjusted percent of overweight Mexican-American and non- Hispanic persons aged 20-74 years, by sex and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1076=80 ...........ccceceverririrrenirinisinsienssessssssssssssssssssessesssssssessssesssssssassssssssssssssssassssns 50 Table 3-11. Per capita amounts of ethanol, in gallons of ethanol, based on the U.S. population aged 14 years and older, 1977-84 ..........cccouveirininennininnniniiiiii sss 56 Table 3-12. Percent of Hispanic women aged 20-44 years with low blood folate levels: Hispanic Health and Nutrition Examination Survey, 1982-84 ..........cccccoevniiinne 63 Table 3-13. Percent of women aged 16-49 years with iron deficiency determined by the MCV model: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976=80 ..........cccconiiinnnninninnnciiiiniinns 65 Xv Page Table 3-14. Age-adjusted prevalence of hypertension in persons aged 20-74 years, by ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, JOTE-80 .....co etre srrrs srs srs rst sts sss sss ses rr ————_—_—_—_—_—_—_—_—_—_—_s rors rer TTT ROSSER tn 69 Table 4-1. Total prevalence of diabetes (sum of previously diagnosed diabetes and undiagnosed diabetes), by age group, survey, and race or ethnic group: second National Health and Nutrition Examination Survey, 1976-80, and Hispanic Health and Nutrition Examination Survey, 1982-84 ao eerste sess a eases sae sa seer a enanens 81 Table 4-2. Prevalence of low birth weight (<2,500 grams), by race or ethnic group and smoking status: CDC Pregnancy Nutrition Surveillance System, 1987 ..........ooovveveereererresresseesesseosessssoon, 84 Table 4-3. Prevalence of low birth weight (<2,500 grams), by maternal age and smoking status: CDC Pregnancy Nutrition Surveillance System, 1987 ........ooeeeevveeeereerreeresseesee sesso 84 Table 4-4. Prevalence of low birth weight (<2,500 grams), by pregravid weight and smoking status: CDC Pregnancy Nutrition Surveillance System, 1987 ..........cooovevvemreereeresresseseseseesssessoon, 84 Table 4-5. Prevalence of low hematocrit at initial visit, by race or ethnic group and trimester of pregnancy: CDC Pregnancy Nutrition Surveillance System, 1987 .........ccoovvoveveeeeveeerresresrrnn. 87 Table 4-6. Prevalence of low hematocrit at initial visit, by maternal age and trimester of pregnancy: CDC Pregnancy Nutrition Surveillance System, 1987 .........oo.cooovvveeverveeveereoeeosesssooon, 87 Table 4-7. Percent of children who were ever breastfed, by year of birth and ethnic group: National Survey of Family Growth, 1982, and Hispanic Health and Nutrition Examination SUIvey, 1982-84 ..............cc..cccouruuiuiueiueeeseeeseseeseesssesesssessesssssesssss esses ses ssessssssesese sessed 88 Table 4-8. Percent of children who were breastfed for 6 months or more, by year of birth and ethnic group: National Survey of Family Growth, 1982, and Hispanic Health and Nutrition Examination Survey, 1982-1984 ...........ocoocevvevvereerveeeeeeeoeeoeeoeoeoeooon, 88 Table 5-1. Percent of persons aged 25-74 years with definite elevated blood pressure, by race, sex, and age: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976=80 ............ccooooveeemeeeeesereeereesreesreessees sees esse 108 Table 5-2. Overweight persons aged 25-74 years, by race, sex, and age: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination SUIvey, 1976-80 ..........ccccccuumummueruriuesresisiseisncsssasssssssseessessessssssssssesssssesssesssssessssses snes sssssssses essen 109 Table 5-3. Percentage contribution of selected food groups to total intake of selected food components for women aged 19-50 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals 1985 ..............ceeeeeeeneeeeseerseeesreesseessesssesssess sesso 114 Table 5-4. Trends in age-adjusted mean intakes of energy, fats, and cholesterol for persons aged 20-74 years, by sex and age: first National Health and Nutrition Examination Survey, 1971-74, and second National Health and Nutrition Examination SUrvey, 197680 ..............ccc.ecuevurueeeeenieereeeesesessessseesessssssssssssssss esses esses sessses esses 116 Table 5-5. Change in age-adjusted mean serum cholesterol levels (milligrams/dl) for persons aged 20-74 years, by race and sex: first National Health and Nutrition Examination Survey, 1971-74, and second National Health and Nutrition Examination SUrvey, 1976-80 ..........cccccouumruierrrieriesiesieessesasisssseseessesssssessssssssssessssssssssssssessessss esses esses esses 117 xvi Page Table 5-6. Age-adjusted percent distribution among groups with zero, one, and two or more risk factors for cardiovascular disease for persons aged 25-74 years, by race and sex: first National Health and Nutrition Examination Survey, 1971-74, and second National Health and Nutrition Examination Survey, 1976-80 ..........ccccccecevurerurerurrennnne. 119 Table 5-7. Percentage decrease in age-adjusted rates for observed and expected coronary heart disease mortality (ICD-8, 410-413) between 1974 and 1978 for adults aged 35-74 years, by race and sex: first National Health and Nutrition Examination Survey, 1971-74, and second National Health and Nutrition Examination SUTVEY, 197680 .........ccccceeurerierireererereereressetssestssesessssessesessssessssesessesesssssesessssessesessasessssessssessesessesssesenans 120 Table 5-8. Perceived relationships between diet and disease: Health and Diet SUIVEYS, 1982-86 .........coverieerirrererncrerentnnsssesennseesset sts sss esas ss sess ee se sees ee sees ee ees bbs 121 Table 5-9. Perceived major dietary factors related to high blood pressure: Health and Diet Surveys, 1982-86 ...........cccecerrererrerrervenns AAS AY SSeS 121 Table 5-10. Reported food control behavior: Health and Diet Surveys, 1978-86 .........c..ccccevvererereverrecrerereenene 122 Table 6-1. Stages of iron depletion and appropriate laboratory assessments ............cccccceueverereninverineneneennennnne 131 Table 6-2. Percent of persons with low hemoglobin concentrations in iron status groups as defined by the MCV model: second National Health and Nutri- tion Examination Survey, 1976-80, and Hispanic Health and Nutrition Examina- HOT SUTVEY, JOB 8 nvm osmimsismmsssss esses oss sass sass assesses aso a os SEAT ER so BEAST 132 Table 6-3. Laboratory measurements used in three different models for assessing TON AEFICIEIICY ....cuveuiiiiiiiciiiieteieterte esters testes st east sae tsb eases esse tess ease sess esas esata sa sna st eset asart eset asantesasaanansasansenerseranes 134 Table 6-4. Percent of females with iron deficiency anemia, by age and race or ethnic group: second National Health and Nutrition Examination Survey, 1976-80, and Hispanic Health and Nutrition Examination Survey, 1982-84 ...........cccocuvrrinrrinirinenineenieenreeneeresseseesenens 138 Table 6-5. Iron levels of adult male, infant, and toddler diets: Food and Drug Administration Total Diet Study, 1074-81182 ....c.cuimemimmmiims is iss ise ost aa sass 0 bosos ass sos sessossos 140 Table 6-6. Dietary iron intake (milligrams per day) for non-Hispanic children aged 1-5 years, by iron status determined by the MCV model: second National Health and Nutrition Examination Survey, 1076-80 .....csmninnismmiimimiismisimimmsmismsmmiirios 142 Table 6-7. Total, nonheme, and heme iron intakes of persons, by sex and age, 1 day: Nationwide Food Consumption Survey 1977="T8 .........ccccevrerirereereeeneeeenenssssesisesesssssisesesssssssssssssssssssssses 142 Table 6-8. Available iron intakes for persons, by sex and age, 1 day: Nationwide Food Consumption SUrvey 1977 =T8 .........cccccereerereririnrerererenssseseesssesesssssesessssssssesessssssssssssssesessssnes 143 Table 6-9. The prevalence of low hemoglobin and/or low hematocrit in low- income pregnant women at initial visits: Pregnancy Nutrition Surveillance System, TOBT ciciiiiiiinitiamnninmamisean sisson sssssrcssirss sass sosses sions sesensn sans ses sasses sess ss esses sesnntessesess us ns sasssssssssntses sess ss esses sisssmssssmssssamssmaseSS 145 Iron Table 1. Prevalence of iron deficiency determined by the MCV model in children, aged 1-4 years, by age, race, and poverty status: second National Health and Nutrition Examination Survey, 1976—80 ...........ccccecevrereniirerenreninennessesessesessessssesssssssssessssessssesasssssenes 150 Iron Table 2. Percent of persons with anemia (hemoglobin below age-specific cutoffs), by age, sex, and race or ethnic group: second National Health and Nutrition Examination Survey, 1976-80, and Hispanic Health and Nutrition Examination SUrvey, 1982-84 ...........cccocviiiriririniiieeeieietssessessse ese sessssssese ese stssesesssssssssssssesesessssssesessnsesssssensssesenssses 150 xvii Figures Figure 1-1. General conceptual model for food choice, food and nutrient intake, and nutritional ANd health SLALUS ..........c.cocerueiueeiiceecicetcieececeeeeee ses esseeessess ses ses esos s esses sees sss esse sees sees ese Figure 2-1. Effect of applying a cutoff value for an indicator of nutritional status to the distributions of values for individuals with normal status and individuals With 8DNOTIMAL SEALUS ........oeceiieiececiiieeieee esses esse ssessesess sees sess s essa sees esses esses esse sees esse senses Figure 3-1. Conceptual model for the update of dietary and nutritional status in Lhe UNIEd SAEs ............ccceeresrsrrmsrmrsenmsemssnssssssses ss ses nsssssssssssrmsssmsssmsssssssessesssnsssns senses ns ssssemss ems ese ees sss esse Figure 3-2. Per capita availability (pounds/year) of (A) meat, poultry, fish and eggs, (B) dairy products, (C) fats and oils, (D) grain products and sugars and sweeteners, (E) fruits and vegetables, and (F) legumes, nuts, and soy; coffee, tea, and cocoa; and spices in the U.S. food supply: U.S. Food Supply Series, 1909-13, 1925-29, 1935-39, 1947-49, 1957-59, 1967-69, 1977-79, ANA 1985 ........oeereeeeerereeereereeresseesseeseeseeeseee esses Figure 3-3. Annual per capita availability (pounds/year) of (A) meat, poultry, fish and eggs, (B) dairy products, (C) fats and oils, (D) grain products and sugars and sweeteners, (E) fruits and vegetables, and (F) legumes, nuts, and soy; coffee, tea, and cocoa; and spices in the U.S. food supply: U.S. Food Supply Series, 1970-85 ........ooveeoveeoooeoooooeoooooooooooonn, Figure 3-4. Decisionmaking process used by the EPONM in categorizing food components by MONItOTing PriOTIty SLALUS ...........cc.uervueevuecisriseieeesesseseeseesessssssssesssssssessses sass sees sssesss esses eee sesesen. Figure 3-5. Prevalence of "high-risk cholesterol" levels in serum in males, by age and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 .......coceeveeeereeeeerereererrennn, Figure 3-6. Prevalence of "high-risk cholesterol" levels in serum in females, by age, and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, O76-B0 conc snsssrmsvsimsnussmmismssommetivncitsssisns sss suse sss ome mmsmses mes 488 HA SHR HRS HHA SHS HRA P OR SERS SAI S88 1 eet Figure 4-1. Conceptual model for update of selected health conditions and behaviors related to NULTILIONA SEALUS ..........c.vuervuereeeieeeceeeeeeeeeeee esse seesessesessess sess sess sees sees eeeee sees eeesseeeseeeenn. Figure 4-2. Age-adjusted death rates for selected causes of death: U.S. Vital Statistics, 1950, 1983, 1985 ..........ccccrmeruummrrirrnnssssssnssssssnssssssssesssssesssssesssessssssssesssssessssessssssesssssessseseseeessesssesnseesesn Figure 4-3. Age-adjusted percent of overweight Hispanic and non-Hispanic persons, 20-74 years: Hispanic Health and Nutrition Examination Survey, 1982- 84, and second National Health and Nutrition Examination Survey, 1976-80 ..........covueeeeeeeeeeeeeeeereeerererersns Figure 4-4. Age-adjusted percent of severely overweight Hispanic and non- Hispanic persons, 20-74 years: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, EL Figure 4-5. Age-adjusted death rates for malignant neoplasms, by race and sex: U.S. Vital Statistics, 1950, 1083, 1985 ............ccooceuuueeeeereemneessessssesesssesessssssssssssssesssssssssesssssessesesssses esses eeseesesn Figure 4-6. Percent distribution of children aged 5-17 years, according to the number of decayed, missing, and filled teeth (DMFT): National Dental Caries Prevalence Study, 1979-80 .............ccoow.ueervuuereeeurieeneeeseeessesssssssesssssssssssssssessssess sees sess esses esses esse eeeee sees xviii Page Figure 4-7. Percent distribution of Mexican-American children from the Southwest United States, aged 5-17 years, according to the number of decayed, missing, and filled teeth (DMFT): Hispanic Health and Nutrition Examination SUIVEY, 1082-84 ......cccc errr Are 83 Figure 4-8. Percent of infants with low birth weight (2,500 grams or less): U.S. Vital Statistics, 1073-78, 1078-80, 10BB~BB ....cuumrivrssnsnsorssssnssnsesreseismeercreresmsmsesensmsemsrsse bem sess assesses sasssorsms somos 83 Figure 4-9. Percent of Hispanic children below the NCHS growth chart 5th percentile of height for age, by sex, age, and ethnic group: Hispanic Health and Nutrition Examination Survey, 1982-84 ........ccccvviinniiiiniintienteineteeeete eee 85 Figure 4-10. Percent of Hispanic children below the NCHS growth chart 5th percentile of weight for height, by sex, age, and ethnic group: Hispanic Health and Nutrition Examination Survey, 1982-84 .........ccccocvnivrnirnininnnnnnininiieeneeieessessessssssssssessssenens 85 Figure 4-11. Percent of low-income children below the NCHS growth chart 5th percentile of height for age (shortness) or below the NCHS growth chart 5th percentile of weight for height (thinness): CDC Pediatric Nutrition Surveil—- 190100 ByStomm, JOT =BT ..cnnuimrmssmssmstros tts tts ames seers sere esses evs es A A ASA AA TIANA TATA 86 Figure 5-1. Conceptual model for nutritional and dietary factors in cardio— VASCUIAT QISEASE ...vcvevrerierierierierieieeeeesessestesesessesae sae sessesaessessestesassessestesesesaest este btest esse as abe e seb a barb eb seb s esas aba ssassarnesnesatans 98 Figure 5-2. Age-adjusted death rates for selected causes of death, United States, selected years: National Vital Statistics System, 1950-85 ..........ccccoeurrrrrrnrinrneriiiennn ens 99 Figure 5-3. Age-adjusted death rates for diseases of the heart, by sex and race, United States, selected years: National Vital Statistics System, 1950-85 ...........cccooeunnierinniniiennnnnieninieininnns 99 Figure 5-4. Age-adjusted death rates for cerebrovascular disease, by sex and race, United States, selected years: National Vital Statistics System, 1950-85 .........cccoovueinriiniiiiiniiniiieiniens 100 Figure 5-5. Trends in coronary heart disease mortality, for all ages, United States: National Vital Statistics System, 1968—85 .........ccccecerrrerrirrerririnirinniniiniiiiniieesesesesese senses sss stents sae saesassssssssees 100 Figure 5-6. Death rates for heart disease among men aged 45-54 years, by race, United States, selected birth cohorts 1891-1940: National Vital Statistics System ..........cccecevurrrririniriicncnnnen. 101 Figure 5-7. Death rates for heart disease among women aged 45-54 years, by race, United States, selected birth cohorts 1891-1940: National Vital Statistics SIWBLEIIN woven svsnosrsnvsnsnns ress fenhnin sebsmheins fe a bri ss hei a ie nse ss Se RE Ree SER IPERS E OPA SPER SES HATER SOOT EII ALAA 0H 101 Figure 5-8. Death rates for stroke among men aged 45-54 years, by race, United States, selected birth cohorts 1891-1940: National Vital Statistics System .........ccccevvniiiniiiinniennicninnn 101 Figure 5-9. Death rates for stroke among women aged 45-54 years, by race, United States, selected birth cohorts 1891-1940: National Vital Statistics System .........ccccevreirniinnnnnnen. 101 Figure 5-10. Age-adjusted mean serum cholesterol levels, by sex and ethnic group or race, for persons aged 20-74 years: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1OTB 80 uence rerrerersresrssesrss esses vsses reser reser sees reser ssa sas A SAS SH EL AEE OAL HEAL EELS ESAS EARS SARS SAS ERS SARS SASS eRe eRe 102 Figure 5-11. Mean serum cholesterol levels in males aged 20-74 years: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Ezamination Survey, 1076-80 .cosummmmmmmnnnsmsmmnimmicnmsssermsinosstommmemsimsmssmormmssssosssisss sons 103 xix Figure 5-12. Mean serum cholesterol levels in females aged 20-74 years: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976=80 ............cccececevererirrererenrirrueeeesssesesesessssssesesssssssssssesssessssssesessssssssssessssssssssssens 103 Figure 5-13. Guidelines for treatment of elevated blood cholesterol levels in adults from the National Cholesterol Education PrOGram ..............cceeeuieereerernseneneeenssnsssssssssssssssssssssssesssses 104 Figure 5-14. Percent of persons aged 20-74 years with specified serum cholesterol levels: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 ............ccccceverererererirererensrereneerssseesssssessssssssssssssssessssssssssssssssssssssssssssssssssses 105 Figure 5-15. Percent of persons aged 20-74 years with specified serum cholesterol levels, by sex: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 ...........ccccecerevererereeenenee 4 0 HR PL SOA STEERS TA TSAR ER 105 Figure 5-16. Percent of males aged 20-74 years with specified serum cholesterol levels, by race: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 ..........cccccceeviririrerererererereieereeeesessnssesssssssssssssssssssssssssssssssssssssssessses 105 Figure 5-17. Percent of females aged 20-74 years with specified serum cholesterol levels, by race: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 ............ccccceeurririririreninrneieeeeeeeeessssssssesssssssssesssesesesssssesssssssssssssenes 105 Figure 5-18. Prevalence of hypertension (average systolic blood pressure > 140 mm mercury, and/or average diastolic blood pressure > 90 mm mercury, or history of use of antihypertensive medication) for persons aged 20-74 years, by sex and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 .............ccceceuerrrerrrrvennne. 106 Figure 5-19. Age-adjusted mean systolic and diastolic blood pressures for persons aged 20-74 years, by sex and race: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976=80 ............ccceceerrurererreererneeresesesnssesssesessessssssssssssssnns 107 Figure 5-20. Per capita amounts of saturated, monounsaturated, and polyunsaturated fats in the U.S. food supply: U.S. Food Supply Series, 1909-85 ............ccceovrerrrerrrrrrerenrererrnnnns 112 Figure 5-21. Food sources of fat in the U.S. food supply: U.S. Food Supply Series, 104740 ANA JOBE .....ccorcvircrnrismrcnerereresns ens rsssns 14190000001 900 3 04098094 EHS A SHS SASSI AEST EAHA HA ERS 112 Figure 5-22. Contributions of meat and fats and oils to per capita fat in the U.S. food supply: U.S. Food Supply Series, 1947-49 and 1985 .........ccceceeverrrerreerererirssnnssesesesssssssessssssssssssssssssssssssens 113 Figure 6-1. Changes in body iron compartments and laboratory assessments of iron status during the stages of 1001 EPIBLION cnnnnrsnnnnmmmmiw ERR eRe 131 Figure 6-2. Prevalence of iron deficiency assessed by three models in non- Hispanic children aged 4-11 years: second National Health and Nutrition Ezamination Survey, TOT8=80 unum mss sss ss soa ene iaia ie seein 133 Figure 6-3. Prevalence of iron deficiency assessed by three models in non— Hispanic males aged 12-74 years: second National Health and Nutrition Examination Survey, 1978-80 .u.cuunmummmmrmmmmmimnrsmiss sin sissies sass isso saan ss manna me sins isnsnsis 133 “ Figure 6-4. Prevalence of iron deficiency assessed by three models in non- Hispanic females aged 12-74 years: second National Health and Nutrition Examination SUrvey, 1976-80 ...........ccccccorrrrinineenininenensssssssssesssssesssesssssssesssessssssesssssssssssessssssssssssssssesssssssssaes 133 Figure 6-5. Conceptual model for the assessment of iron NULTILUTE ...........cceveverieeeeeereiieieeeeeeeeeeeerere essere 136 \ Figure 6-6. Prevalence of iron deficiency assessed by the MCV model in children and adolescents aged 4-19 years, by ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination SutVoy, 1076-80 .mnmmmmsssssmmnssmessmrnsorismstssmsmbemineseamsnite memos stsesssssssssnsrare 137 Figure 6-7. Prevalence of iron deficiency assessed by the MCV model in males aged 20-74 years, by ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, J076-80 ..cummanmmmumssmremissvississm msm iiss ssier is cisn mans ss srs 40 6060 st0apasssasnsens 137 Figure 6-8. Prevalence of iron deficiency assessed by the MCV model in females aged 20-74 years, by ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination SUTVEY, 1976-80 ...........cccccovriiirinineeernneee eres sesessssesssssssesssssssssssssssssssssesssssssssnsssssssssssssssssssssssasans 137 Figure 6-9. Prevalence of anemia (hemoglobin or hematocrit <5th percentile) in children 0-59 months: Pediatric Nutrition Surveillance System, 1973-88 .........ccceeuvuvieeeeeeeeererersresesssssenenns 139 Figure 6-10. Prevalence of anemia in children 6-60 months for each birth-year cohort, six selected States: Pediatric Nutrition Surveillance System, 1973-84 ..........ccoceoeevereeeeveveeeeerererereenn, 139 Figure 6-11. Major food sources of iron in the food supply: U.S. Food Supply Series, 1909-13, 1970, ANd 1985 .........c.eeerenrreeere eer sssssssssesssessssssssssssessssersssessssssssssssssesssns sens 140 Figure 6-12. Percent of Mexican-American and non-Hispanic women aged 16-49 years with iron deficiency assessed by the MCV model, by poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976=80 ..............ccceceeeureerreeureeeneresesesesssesssssesessesessssssssssssesesesessssssssssssss 144 Figure 6-13. Percent of women aged 20-44 years and men and women aged 45-74 years with iron deficiency assessed by the MCV model, by level of education: second National Health and Nutrition Examination Survey, 1976-80 ..............c.ceevuevuerererreieireeeeseeereeeseeeeeenenn. 144 Figure 6-14. Percent of women aged 20-44 years with iron deficiency assessed by the MCV model, by parity and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination SUIVey, 1976-80 ............cccceverrrrrererneereeesessesesessssssesessssssssssssssssssstsssssssssssssssssssssssssssssssssssssssssssesssens 145 EP EEE Executive Summary The National Nutrition Monitoring System (NNMS) includes all data collection and analysis activities of the Federal government associated with health and nutrition status measurements, food consumption measurements, food composition measurements, dietary knowledge, attitude assessment, and surveil- lance of the food supply. The ad hoc Expert Panel on Nutrition Monitoring (EPONM) was established by the Life Sciences Research Office (LSRO) of the Federation of American Societies for Experimental Biology (FASEB) to review the dietary and nutritional status of the U.S. population, as well as the factors that determine status based on the NNMS data and information available through the activities of the U.S. Departments of Agriculture (USDA) and Health and Human Services (DHHS). This report requested by USDA and DHHS is the second report on the National Nutrition Monitoring System. It builds on the framework of the report developed by the Joint Nutrition Monitoring Evaluation = Committee (JNMEC) in 1986. Charge to Panel The charge given to the EPONM for this report directed that two themes dominate: e An updating of the dietary and nutritional status as presented in the 1986 JNMEC report, and eo An in-depth analysis of the contributions of the NNMS to assessment of the status of the popula- tion as determined from two types of data—-data on diet and chronic diseases and data on dietary and nutritional problems. With respect to the first theme, the EPONM was charged to update the first report by comparing data in that report with data from NNMS surveys that be— came available since it was written. The EPONM was also charged to address methodological issues in com- paring data from different sources or times, and to identify the types of analyses necessary to make com— parisons meaningful. Trend and baseline data presented in the first report were to be updated. The EPONM was to reevaluate the categorization of food components as to the completeness of relevant data and the level of monitoring status that should be accorded each food component. With respect to the second theme, the EPONM was to include in the report in-depth analyses of two topics selected as examples of NNMS data: the first, repre— sented by data on the relationship of diet to a speci- fied chronic disease, was to emphasize dietary and nutritional factors in cardiovascular disease; the second, represented by data on a dietary and nutri- tional problem, was the assessment of iron nutriture. The objective of this part of the report was to demon- strate how NNMS data could contribute to under- standing these public health concerns, as well as to identify the strengths and weaknesses of data and information available primarily from components of the NNMS. The ability to identify the nature and magnitude of nutrition-related problems in the U.S. population was to be addressed, with focus especially on the NNMS capabilities for identifying 1) pop— ulations at risk, 2) limits to interpretations of data, 3) gaps in the database, 4) trends, and 5) determining factors. Response to Charge The update portion of this report is a followup to the 1986 JNMEC report and is intended to identify new data available from the NNMS and to examine changes and trends in dietary intake, nutritional status, and nutrition—related health conditions. The JNMEC report emphasized a coordinated review of dietary data from the Nationwide Food Consumption Survey (NFCS) 1977-78 and nutritional and health status data from the second National Health and Nutrition Examination Survey (NHANES II). New national survey data on dietary intake and nutritional status of the entire U.S. population have not become available since the JNMEC report; however, data for certain subgroups of the population have become available. Most of the available data are for the three Hispanic groups surveyed in the Hispanic Health and Nutrition Examination Survey (HHANES) and the women and children surveyed in the Continuing Sur- vey of Food Intakes by Individuals (CSFII) 1985-86. xxiii Data analyses included in the report are intended to be descriptive of dietary and/or nutritional status, trends, and associations rather than to establish causal relationships. The report is not intended to serve the purpose of program evaluation. Summary data from relevant surveys are included, by topic, in appendix II; detailed analyses are provided in the body of the report when a public health issue is identified for some of the food components included in the update (criteria are described in chapter 3) as well as in the chapters on cardiovascular disease and diet and assessment of iron nutriture. Many of the data analyses presented were prepared by the Agencies specifically for inclusion in this report; others were obtained from Agency publications or the peer-reviewed literature. Chapter 2 (on the appropriate uses of survey data for assessing dietary and nutrition-related health status) is responsive to the charge to address methodological issues in comparing data from different sources or times and to identify the types of analyses necessary to make comparisons meaningful. The discussions of data that comprise the update portion of this report are divided into two chapters, with the first (chapter 3) providing new data on dietary intake, food avail- ability, major food sources of various food compo- nents, and nutritional status with respect to various nutrients, and the second (chapter 4) providing new estimates of the prevalence of nutrition-related health conditions and behaviors. Reference is made to baseline data when appropriate. With respect to the second charge to the EPONM, the two selected topics are discussed separately. The chapter on nutritional and dietary factors in cardio- vascular disease (chapter 5) concentrates on the utility of NNMS data in assessing risk factors for cardiovascular disease and dietary and other factors related to these risk factors. The chapter on assess— ment of iron nutriture (chapter 6) concentrates on the utility of NNMS data in assessing the prevalence of iron deficiency and identifying groups at risk and the factors contributing to iron nutriture. Finally, chapter 7 contains the EPONM's specific recommen— dations for improvements in the NNMS, based on their experiences in evaluating the data analyses in this report. Update of Dietary and Nutritional Status Major Conclusions The EPONM drew the following conclusions based on their review of new NNMS data: e In the United States today, the amounts of food available in the food supply and the nutrient con- tent on a per capita basis are generally adequate to prevent undernutrition and deficiency-related diseases. Although some Americans may not have sufficient food for a variety of reasons, the supply of food that is available is abundant. eo The NNMS does not provide sufficient population— based data to permit a full assessment of nutri- tional status in some groups for whom there are special concerns about nutritional status, such as young infants and pregnant and lactating women. In addition, some other groups whose nutritional status may reasonably be suspected to differ from that of the general population, such as the home- less, institutionalized persons, migrant workers, and Native Americans living on reservations, are not included in most of the current household- based surveys of the NNMS. Finally, very little information on the dietary and nutritional status of the elderly (a group for which standards for nutri— ent adequacy and normal physiologic status have been questioned) was available in the most recent NNMS data that were the focus of this evaluation. eo Evidence from recent analyses of the U.S. food supply and from surveys of individual food con- sumption suggests that some changes are occurring in eating patterns consistent with recommended dietary guidelines for Americans to avoid too much fat, saturated fat, and cholesterol and to consume adequate amounts of starch and dietary fiber. Recent data indicate that consumers are increas- ingly choosing some lower—fat alternatives within the meat and dairy product food groups and are increasing their consumption of grain products. e Evidence available on dietary and nutritional status with respect to individual food components does not indicate substantial changes since the JNMEC report was completed in 1986. Consequently, the EPONM and JNMEC classifications of food compo- nents by public health monitoring priority are very similar. eo The principal nutrition-related health problems experienced by Americans continue to be related to the overconsumption of some nutrients and food components, particularly food energy, fat, sat— urated fatty acids, cholesterol, sodium, and alcohol. — The high prevalence of overweight among adults in the United States is evidence that energy intakes exceed energy expenditures (probably because of low energy expenditures, although this possibility cannot be assessed currently in the NNMS); however, reported intakes of food XX1v energy do not exceed standards (Recommended Energy Intakes). The JNMEC noted that more than one-quarter of the adult U.S. population was overweight, based on data collected in NHANES II. Data collected since then in the HHANES (1982-84) also indicate a high preva- lence of overweight in three Hispanic groups not previously studied (26-42 percent), especially in Mexican-American and Puerto Rican women (40 and 42 percent, respectively). Overweight is a controllable risk factor for cardiovascular disease, high blood pressure, and diabetes. — Intakes of total fat and saturated fat continue to be higher than the levels recommended by many authoritative groups; cholesterol intakes are high for adult men. These high intakes are reflected in the high prevalence (11-22 percent) of elevat- ed levels of total serum cholesterol, as defined by the 1984 NIH Consensus Development Confer— ence, found in nearly all adult groups aged 20-74 years in the United States. Elevated serum cho- lesterol levels constitute an important con- trollable risk factor for coronary heart disease. — Sodium intakes also exceed recommended levels in almost every group in the United States. Such intakes are of concern because of the sensitivity of blood pressure in some persons to sodium intake. Hypertension is prevalent (14-44 per- cent) in adult groups aged 20-74 years in the U.S. population. Hypertension is a controllable risk factor for cardiovascular disease and stroke. — Although consumption of excessive alcohol does not appear to be prevalent in a large proportion of the population, reported intakes are high in a large number of Americans and the serious nature of the health and social consequences of such intakes justifies public health concern. e In spite of the general adequacy of the supply of nutrients, there is evidence of inadequate indi- vidual dietary intake and/or impaired nutritional status in some subgroups in the population with respect to a few vitamins and minerals. — Iron deficiency continues to be the most common single nutrient deficiency, even though some recent hematological and biochemical evidence from the NNMS suggests that its prevalence has declined in children aged 1-5 years. Among groups that are assessed adequately in the NNMS, women of childbearing years and young children are at greatest risk for iron deficiency. = Although less evidence is available, the calcium status of women is a concern. The high prevalence of osteoporosis in later life is sug- gestive that the calcium intake of many women may be inadequate to permit the accretion of maximal bone mass in early adulthood and/or to maintain bone mass later in life. - Limited evidence from biochemical assessments suggests that the vitamin A, vitamin C, and folacin nutritional status of some subgroups of the population might be improved. — Intakes of zinc and vitamin B6 are also low, and poor status has been reported in some population groups in the clinical literature, but further study is needed to assess the health consequences of the reported intakes in U.S. population groups. The risk of nutrition-related disorders is gen- erally greater in low-income groups than in groups with higher incomes. — The prevalences of both overweight and iron deficiency are greater in women below poverty than in women above poverty. — The intakes of several vitamins and minerals are lower in persons below poverty than in persons above poverty. This finding is also highlighted in the low-income component of the CSFII 1985- 86. Women in the low-income survey had lower intakes of food energy than women in the all- income survey. Intakes of vitamin E, vitamin B6, folacin, calcium, magnesium, iron, and zinc were low in women in both surveys, but lower in the low-income survey than in the all-income survey. Low-income women and children who lived in households that participated in the Food Stamp Program had nutrient intakes that were generally the same or higher than those of low- income women and children living in households that did not participate in the program. The ability of the EPONM to examine excessive intakes of vitamins and minerals, and possibly to assess consequences of nutrient toxicity, was limited because none of the available NNMS sur- veys that assess nutrient intake from food included quantitative estimates of nutrient intake from vitamin/mineral supplements. Although the data available to the EPONM for their update on dietary and nutritional status of the U.S. population were not equivalent to the data reviewed by the JNMEC, in terms of the popula- tions surveyed, the conclusions of the EPONM are very consistent with those of the JNMEC. The results of recently completed and ongoing national surveys of dietary and nutritional status by USDA and DHHS will provide a more extensive database for further evaluation of the nutritional status of the U.S. population and various subgroups in future reports on the NNMS. Monitoring Priority Status for Individual Food Components The availability of relevant update data on dietary, nutritional, and health status from the various sur— veys of the NNMS for each food component varies. The data elements from the NNMS common to most of the food components considered are per capita amounts in the food supply and individual dietary intakes. The quality and quantity of data, as well as the availability of appropriate assessment criteria, differ for different components and influence the confidence with which evaluations of status may be made. The JNMEC classification and the classification used in this update report by the EPONM are similar phil- osophically (see table below). In the JNMEC report, nutrients and other food components were prioritized in three categories to contrast those components having high and moderate priority status for con-— tinued monitoring with the third group identified as those requiring further investigation. In this report, the EPONM labeled the categories somewhat differ— ently to place emphasis on their evaluation in regard to public health significance. The category of food components considered to be current public health issues by the EPONM can be equated to the JNMEC category of food components warranting public health monitoring priority status. The category of food components considered by the EPONM to be potential public health issues and requiring further study is most similar to the JNMEC category of components requiring further investigation. The type of additional study required for each component differs; basic research on the health consequences of high or low intake, additional data on food composition and dietary intake, and/or the development of methods for assessing status together with interpretative criteria may be needed. The EPONM category of food components that are not considered current public health issues is most similar to the JNMEC category of components warranting continued monitoring consideration. Assigning food components to this category does not necessarily indicate that there are no known health problems associated with these components, but that the prevalence of such problems on a national basis is known or expected to be so low that a lower level of monitoring effort than for food components in the other categories is appropriate. A schematic diagram that illustrates the decision— making process used by the EPONM for categorizing food components is shown in the figure on the oppo- site page. This process differs from the one used by the JNMEC in that the evaluation of each food com- ponent begins with the dietary intake data. This choice to begin the evaluation of each food component with consideration of the dietary intake data was made recognizing that such data are available for most of the components included; the same is not true of related health data. However, as illustrated in the figure, the resulting evaluations by both the EPONM and JNMEC are similar in that the evidence for adverse health consequences ultimately determines the categorization of food components. EPONM classification of food components JNMEC classification of food components Current public health issues Potential public health issues for which further study is required Not current public health issues Warranting public health monitoring priority Requiring further investigation Warranting continued public health monitoring consideration XxXvi NUTRIENT INTAKE Low or High Evidence of adverse Limited evidence of Little or no evidence of health conditions adverse health conditions adverse health conditions | CURRENT PUBLIC POTENTIAL PUBLIC NO CURRENT PUBLIC HEALTH ISSUE HEALTH ISSUE HEALTH ISSUE In the discussions of each of the categories that follow, the criteria for assigning food components to the category are described, update data available related to the components are tabulated, and brief conclusions about each food component are pre- sented. A "plus" in the table indicates data were available; a "minus" indicates they were not. The no- tation "limited" indicates that food composition data were available for less than 75 percent of important sources of the nutrient. eo Food components were considered to be current public health issues - if dietary intakes were low or high for a substantial proportion of the population, and if evidence from NNMS surveys of health and nutritional status indicated related health problems in the population or in subgroups of the population, or - if dietary intakes were low or high for a substantial proportion of the population, and if evidence from epidemiological or clinical studies in the literature indicated related health problems in the population or in subgroups of the population. Food components in this category are recom- mended for high priority monitoring status; that is, multiple assessments, when possible, should continue to be employed. A high priority should be given to development of assessment tools when these are lacking. The food components listed in the following table were considered to be current public health issues. Food composition Food component ta Dietary data Health data Food Energy + Food supply Overweight and associated Individual intake conditions Fat + Food supply Serum cholesterol level Individual intake Saturated fat + Food supply Serum cholesterol level Individual intake Cholesterol 4 Food supply Serum cholesterol level Individual intake Alcohol + Disappearance/sales - Individual intake (limited) Iron + Food supply Mean corpuscular volume, Individual intake rafistertih saturations, eryt Mote rotoporphyrin, hemoglobin/hematocrit Calcium + Food suppl ~ Individual Tntake Sodium + Individual intake (limited) Blood pressure xxvii Reported dietary intakes of food energy by adults are lower than Recommended Energy Intakes, but the data available from the NNMS on the high prevalence of overweight (approximately one-fourth of adults) in many groups in the United States suggest a continu- ing public health problem in regard to energy balance. Food energy should be accorded high prior— ity for monitoring status. Additional information on both energy intake and energy expenditure (physical activity) is required to evaluate the relative impact of these factors on the occurrence of obesity. The intakes of total fat, saturated fat, and cholesterol by many persons in the U.S. population exceed levels recommended by many authoritative groups. Serum cholesterol levels are affected by dietary intake of these components (and other factors); elevated levels of serum cholesterol are prevalent in the United States (11-22 percent) in men and women of all racial and ethnic groups examined and represent a risk factor for the development of coronary heart disease. Continued priority for the monitoring of serum cholesterol levels and the dietary intake of fat, fatty acids, and cholesterol is warranted. Self-reported alcohol intakes are high (an average of 1 ounce or more of ethanol per day) in a large number of persons (9 percent of adults). The public health and social consequences of excessive alcohol intake are sufficiently grave that continued efforts to improve monitoring of alcohol intake are warranted. Iron intakes are low for many in the population. Although the prevalence of iron deficiency has appar— ently declined in recent years, it is still relatively high in vulnerable groups (up to 14 percent) such as women of childbearing years. Continued monitoring is warranted and is likely to yield useful information on iron nutritional status because of the wealth of indicators available for inclusion in the NNMS. The low intakes of calcium in vulnerable groups, especially women, suggest a reason for concern. The high prevalence and severity of osteoporosis, which is possibly related, in part, to calcium intake of adoles- cents and young women, provide sufficient evidence for a public health concern. Calcium should be con- sidered a nutrient about which there is public health concern even if there is some question about its exact role in health disorders. Monitoring the intake of calcium and including assessments of bone status in NNMS surveys is warranted, as is investigating the possible overuse of calcium supplements by adults. Reported dietary intakes of sodium are high in many persons relative to estimates of safe and adequate levels of intake; reported intakes do not account for all sources of sodium. The prevalence of hyperten— sion, which is related in some persons to sodium intake as well as other factors, is high in all adult groups examined (14-44 percent). Because of the serious, and largely preventable, deleterious effects of elevated blood pressure, a high level of monitoring effort is warranted. Blood pressure measurements should continue to be included in surveys and efforts to improve and validate the assessment of total sodium intake should be pursued. eo Food components were considered to be potential public health issues, for which further study is needed, —- if dietary intakes were low or high for a substantial proportion of the population, and if limited evidence from either NNMS nutrition and health surveys or studies in the literature suggested related health problems in at least some subgroups in the population, or — if dietary intakes were adequate for the majority of the population, but limited evidence from either NNMS nutrition and health surveys or studies in the literature suggested related health problems in at least some subgroups in the population, or - if dietary intakes were low or high for a substantial proportion of the population, and if evidence was not available from either NNMS nutrition and health surveys or studies in the literature that permitted evaluation of the public health significance of the observed dietary intakes. Food components in this category are recommended for moderate monitoring priority status, with continued assessment at the least in subgroups suspected to be at risk, and moderate priority for the development of improved assessment techniques. xxviii The food components listed in the following table were considered potential public health issues, for which further study is needed. Food composition Food component ta Dietary data Health data Dietary fiber + Individual intake = (limited)* Vitamin A + Food supply Serum retinol level Individual intake Carotenes + Food supply = Individual intake Folacin + Food supply Serum and red blood cell Individual intake folate levels Vitamin B6 + Food supply ~ (limited)* Individual intake Vitamin C + Food supply - Individual intake Potassium + Food supply - Individual intake Zinc + Food suppl = Individual Yriake Fluoride - = Dental caries * Less than 75 percent analytical data for important sources of the food component. Intakes of dietary fiber are low in relation to sug- gested levels of intake, but the impact of these low intakes cannot be judged on the basis of available data. More information is required on the health effects of dietary fiber, the content in foods of various components of fiber (which have different physiolog- ical effects) as well as total dietary fiber, and recom— mendations for intake. Monitoring is recommended as this information is developed. : The content of vitamin A in the food supply and indi- vidual intakes suggest general adequacy. Intake and status may, however, warrant continued monitoring efforts in certain groups. HHANES data on low serum vitamin A levels suggest that poor young children, particularly Mexican Americans, may be such a group. Greater attention needs to be given to studying the relationships of biochemical assessments of status to functional impairments. Carotenes are also considered a potential public health issue for which further study is required. Data on intake of carotenes are available from the CSFII 1985-86 and will be available from HHANES to provide a baseline for assessing future changes in intake. Future sur- veys should continue to collect and report intake separately for carotenes and total vitamin A. Additional research is needed on the health effects of consumption of specified levels of total carotenes, as well as individual carotenes. Vitamin B6 intakes are lower than recommended levels for a substantial number of persons, especially women. In order to interpret the consequences of these intakes, further study is needed on the content and bioavailability of vitamin B6 in foods, vitamin B6 requirements, and biochemical or other techniques for assessing vitamin B6 nutritional status. Increased monitoring activity may be warranted as progress is made in these areas. Recent dietary intakes of vitamin C appear to be ade- quate in most of the population, even without consid- eration of the substantial contribution of vitamin C supplements. Older data for serum vitamin C (from NHANES II) indicate that the prevalence of low serum vitamin C levels is generally higher in groups with low socioeconomic status, especially older men, but do not provide strong evidence for vitamin C deficiency. Although these data suggest the need for some continued surveillance, changes in vitamin C XXixX fortification practices may affect intake among many segments of the population. Continued monitoring is warranted to assess the impact of these changes, but the apparently adequate intakes do not provide support for priority monitoring status. Folacin intakes are much lower than recommended in some groups, especially women. Biochemical and other evidence for deficiency is limited, but suggests a risk of deficiency in women. Further study is required to evaluate folacin requirements, to develop methods and interpretative criteria for folacin nutri- tional status, and to examine the status of groups at risk. Potassium intakes are lower than recommended levels in a substantial number of persons in the pop- ulation. Further research on the role of potassium intake in the regulation of blood pressure and on the assessment of potassium status is needed to elucidate the public health significance of the low intakes observed. Zinc intakes are lower than recommended levels in some groups, particularly women. The possibility of impaired zinc status is not supported by available biochemical or clinical data from the NNMS. How- ever, findings from the clinical literature suggest zinc deficiency in some groups in the United States. The significance of the observed low dietary intakes of zinc cannot be evaluated until additional research to determine zinc requirements and to develop better measures of zinc status is conducted. Further moni- toring is warranted. The EPONM agrees with the JNMEC's concern that fluoride intake may be too low in some groups to provide maximal benefit, but NNMS data are not currently available that permit evaluation of this possibility. Assessments of fluoride intake that take all sources into account are warranted. eo Food components were not considered to be current public health issues - if dietary intakes were adequate for the majority of the population, and evidence from either NNMS nutrition and health surveys or studies in the literature did not suggest related health problems in the population, or - if dietary intakes were low or high for a substantial proportion of the population, but evidence from either NNMS nutrition and health surveys or studies in the literature did not suggest related health problems in the population. Food components in this category are recommended for lower monitoring priority status; continued assessment should include, at a mini- mum, estimation of dietary intake. The food components listed in the following table were not considered to be current public health issues: Food composition Food component ata Dietary data Health data Protein + Food su - Toren Dake Carbohydrate + Food suppl - Individual Tntake Vitamin E + Food supply Serum a-tocopherol level (limited)* Individual intake Thiamin + Food supp = Individual Lian Riboflavin + Food suppl - Individual intake * Less than 75 percent analytical data for important sources of the food component. Food composition Food component ta Dietary data Health data Niacin + Food suppl = Individual Lsiake Vitamin B12 + Food supply - (limited)* Individual intake Phosphorus + Food suppl ~ Individual Laie Magnesium + Food suppl = (limited)* Individual Tntake Copper + Food suppl - Individual Lois * Less than 75 percent analytical data for important sources of the food component. Protein intakes appear to be adequate for almost all persons and there is no evidence of health problems associated with deficiency or excess. Monitoring should continue at a low level, especially for the elderly. Carbohydrate intakes are lower than may be desir- able, based on the dietary pattern recommended in the U.S. dietary guidelines (USDA/DHHS, 1985), but evidence does not suggest that current intakes pose a specific public health problem. Monitoring of intake should continue; if recommended decreases in the percent of energy from fats occur, concomitant increases in the proportion of energy from carbo- hydrates are expected. Although some vitamin E intakes are lower than recommended levels (especially in women), data on serum a-tocopherol levels and clinical data on the rarity of vitamin E deficiency suggest little reason for a public health focus. Interpretation of serum a-tocopherol levels is confounded by other factors such as serum lipid concentrations, and clear inter— pretative guidelines to assess marginal vitamin E status do not yet exist. Intakes of thiamin appear to be adequate, and no other evidence suggests a public health problem with respect to thiamin status. Intakes of riboflavin appear to be adequate, and no other evidence suggests a public health problem with regard to riboflavin status. Individual intakes of preformed niacin appear to be adequate (and additional niacin may be obtained from the conversion of dietary tryptophan in the body). No other evidence suggests a public health problem in relation to niacin status. Intakes of vitamin B12 appear to be adequate for the majority of the population. Clinical or biochemical evidence for a public health problem with respect to vitamin B12 deficiency is not available. Further monitoring, at a low level, is warranted. Phosphorus intakes appear to be adequate, and no other evidence exists to suggest a public health problem. Monitoring should continue at a relatively low level. Magnesium intakes appear to be low, but there are no other data on magnesium status available and magnesium deficiency is very unlikely to result from low dietary intake alone. Further research on magnesium requirements and assessment of magne- sium status would be desirable. Current information supports continued monitoring at a low level. Copper intakes appear to be low in a large number of persons in the population. Despite some unanswered questions about the estimation of intake and the assessment of status, the likelihood of a public health problem associated with copper is very low. Monitoring should continue at a low level, unless further research suggests more compelling reasons for concern. Xxxi Use of NNMS Data to Examine Dietary ° and Nutritional Factors in Cardiovascular Disease The EPONM identified the following limits to interpretation of the data and gaps in the database, based on their review of the data available from the NNMS with respect to dietary and nutrition factors in cardiovascular diseases: eo The most recent dietary survey, the CSFII 1985- 86, provides estimates of dietary intake of several food components associated with cardiovascular diseases: food energy, fat, saturated fat, mono- unsaturated fat, polyunsaturated fat, cholesterol, dietary fiber, calcium, and potassium. Estimates of sodium intake from food are also available, but these exclude sodium from salt added at the table; thus, total sodium intake is underestimated. Esti- mates of alcohol intake from self reports are also less certain, because of methodological difficulties such as underreporting. Because the properties of individual fatty acids differ, estimates of their intake would also be desirable; the nutrient com- position databases available during the EPONM's review do not contain composition information e with respect to individual fatty acids. eo With respect to implications for cardiovascular disease, the most recent data on the distribution of usual dietary intakes are limited because they are for women and children, groups not thought to be at high risk of cardiovascular disease. The earlier eo NFCS 1977-78 obtained dietary intake data for multiple days for both sexes and all ages, but the nutrient composition database was more limited with respect to information on fatty acids and Cross-sectional associations of dietary influences and risk factors for cardiovascular diseases may be examined using data from the HANES, but the power of such analyses is severely restricted because of the large day-to-day differences in the food and nutrient intake of any individual. The results of analyses of 1-day food intakes of indi- viduals do not represent the average usual intake of any individual over a longer period of time. In studying diet and disease relationships, it is generally recognized that an estimate of average, or "usual" nutrient intakes is needed. With respect to studying relationships between diet and cardiovascular disease risk factors in the NNMS, measurements of "usual" dietary intake are not obtained for the same individuals in whom measurements of risk factors are performed. Other limitations in interpretation of such associ— ations exist because of the nature of cross- sectional data; an "association" between a postu-— lated risk factor and a disease may be identified, not because it is causally related to the disease, but because it is related to another factor which is really one of the causes of the disease. Because a single 24-hour recall was used to obtain dietary intake information, there are also limitations in the interpretation of relationships between dietary intake and subsequent morbidity and mortality in the longitudinal NHANES I Epidemiologic Followup Study. Surveys have not assessed directly the impact of knowledge and attitudes about cardiovascular disease risk factors and diet on patterns of food consumption or nutrient intake. cholesterol at the time. The EPONM drew the following conclusions, based on the analyses reviewed, about the major contri- eo The ability to assess the distribution of usual butions of the NNMS to the understanding of dietary dietary intakes from the data obtained in the and nutritional factors as they relate to cardio— Health and Nutrition Examination Surveys vascular diseases: (HANES) is more limited because only 1 day of dietary data is collected. Populations at Risk e Trends in dietary intake of individuals with e respect to fat and cholesterol, as assessed by the dietary and health surveys of the NNMS, must be interpreted with caution because of differences in survey methodology and improvements in the nutrient composition databases over time. The ability to interpret changes in intake over time could be improved by the conduct of methodologi- cal studies designed to assess the consequences of changes in survey procedures on the estimates generated. xxxil Measurements of body weight, blood pressure, serum lipids, and glucose tolerance and question- naires in the HANES permit the assessment of the prevalence of several major diet— and nutrition- related risk factors for cardiovascular diseases—— obesity (overweight), hypertension, elevated serum cholesterol level, and diabetes——in nationally representative samples. By comparing the preva- lence estimates of different population groups, some characteristics of groups most affected by each risk factor can be identified. Characteristics of the groups at risk differ depending on the risk factor considered. For example, blacks are at greater risk of hypertension than whites; women of low socioeconomic status are at greater risk of obesity than women of high socioeconomic status, and persons above poverty are at greater risk of hypercholesterolemia than persons below poverty. The characteristics of individuals with multiple risk factors can also be examined: black males have a higher prevalence of multiple risk factors than white males and black or white females. However, a model that quantitates the relative contribution of all risk factors, including genetic predisposition, has not been developed that can be applied to NNMS data to assess overall risk of cardiovascular diseases. High intakes of fat, especially saturated fat, and cholesterol constitute risk factors for hyper- cholesterolemia, and characteristics of populations with high intakes can be assessed in the NNMS. For example, in women, high fat intake is associ- ated with being white, having more than a high school education, and smoking. Trends e Data from the U.S. Food Supply Series provide information on trends in the foods and amounts of food components in the food supply over time. Although the inferences about food consumption and food component intake that may be drawn from these data are limited, they are nonetheless useful to demonstrate shifts over time within food supply sources of various food components related to cardiovascular diseases (notably, total fat and fatty acid groups) in the national diet. These data indicate recent shifts from animal sources of fat to vegetable sources of fat, consistent with dietary guidance to avoid too much saturated fat. Data from the surveys that collect information on food consumption indicate a decline in the con- sumption of some high-fat foods and foods containing saturated fat. Some changes observed in the past 20 years include a shift from whole milk to low-fat milks, an increased consumption of leaner types and cuts of meat, and an increase in the use of margarine with a concomitant decrease in the use of butter. However, as noted earlier, interpretation of trends in nutrient intake is problematic because of changes in survey methods over time. Biochemical and clinical measurements that permit assessment of the prevalence of overweight, hypertension, and elevated serum cholesterol xxxiil levels have been made repeatedly over time (with limited changes in methodology). Thus, trends in the prevalence of these risk factors can be assessed. Data from the NNMS indicate a recent decline in the prevalence of hypertension and elevated serum cholesterol levels, but no decline in the prevalence of overweight. Associations between risk factors (for example, body weight and serum cholesterol level) and the occurrence of multiple risk factors in population groups can also be examined. The NNMS trend data are useful for examining concurrent changes in group intakes or status over time. For example, changes in food availability in the food supply contributing to a decrease in the content of saturated fat have been observed to precede the decline in coronary heart disease mortality. Changes in mean serum cholesterol levels consistent with changes in mean dietary intake of fats and cholesterol can also be detected between the first and second NHANES. Determining Factors e Sex and age are important determining factors for the risk of coronary heart disease and cerebro- vascular diseases. Men are at higher risk than women for coronary heart disease and hyper- tension. Although serum cholesterol levels do not vary dramatically with sex, elevated levels consti— tute a greater risk for men than for pre- menopausal women. The dietary intakes of fat, saturated fat, and cholesterol are higher in males than in females. Race also has an important impact on relative risk of cardiovascular disease. NNMS and related data indicate that blacks are at greater risk than whites for coronary heart disease, cerebrovascular disease, hypertension, and hypercholesterolemia. More black women are significantly overweight than white women or men of either race. The effects of socioeconomic factors such as pov— erty status and education do not seem consistent for all risk factors related to cardiovascular dis- eases. Indicators of high socioecomic status tend to be associated with hypercholesterolemia and higher intakes of fat, but with lower prevalences of hypertension and, for women, overweight. Several surveys of the NNMS permit the assess- ment of knowledge and attitudes about diet— and nutrition-related risk factors for cardiovascular disease. Such surveys have been repeated over time and show a trend for increasing knowledge and changing diet-related practices. Use of NNMS Data for the Assessment eo of Iron Nutriture The EPONM identified the following limits to interpretation of the data and gaps in the database, based on their review of the data available from the NNMS with respect to the assessment of iron nutri- More detailed information is needed on the type and amount of supplement intake. Total iron intake could not be determined because quan- titative estimates of iron intake from supplements were not available from any of the surveys in which estimates of intake from food were made. ture: The EPONM drew the following conclusions, based on the analyses reviewed, about the assessment of e Information about selected groups at risk of iron iron nutriture in the United States using NNMS data: deficiency, pregnant women and infants under age 1 year, is inadequate. In pregnant women, anemia Populations at Risk is associated with increased neonatal mortality and a higher prevalence of low birthweight. Iron defi- eo ciency during the brief period of infancy is believed to lead to long-term harmful consequences in regard to subsequent development. In both groups, dietary practices differ from other age and sex groups. In both of these groups, dietary intake and use of supplements over a period of 6 to 9° months determine the risk of iron deficiency. eo These groups require longitudinal assessment over at least 6 months for an adequate assessment of their nutritional status because iron status changes rapidly over a period of a few months. eo The combination of foods eaten at each meal is the most important determinant of iron absorption. e Such information is even more important than the amount of iron consumed and has only been ana- lyzed on a very limited scale. Improvements in the ability to provide such analyses should be incor- porated into the NNMS. ° e Distinguishing iron deficiency from mild inflam— matory conditions is difficult because laboratory measurements in mild inflammatory conditions or following infections may mimic iron deficiency, thereby suggesting a higher than actual preva- lence. This problem, which is greatest among the eo elderly, may be alleviated by using a recently developed four-variable model to assess iron nutriture and laboratory tests that reflect the presence of inflammation. e Criteria for anemia in blacks are uncertain. Blacks have lower hemoglobin concentrations than whites irrespective of iron status. These lower concentra- eo tions lead to misleadingly higher prevalences of anemia among blacks if uniform criteria for low hemoglobin values are used for all races. This problem can be circumvented by using the three- or four—variable models for iron deficiency (that do The variety of biochemical and hematological measures of iron nutritional status collected in the NNMS permits an estimation of the prevalence of iron deficiency and iron deficiency anemia in the U.S. population and some characterization of population groups at risk of iron deficiency. The prevalence of iron deficiency anemia (based on findings of low hemoglobin levels plus evidence of iron deficiency) in NHANES II and HHANES is low (less than 5 percent); however, the prevalence of iron deficiency without anemia is still appre- ciable (up to 14 percent) in several groups. Groups at greatest risk of iron deficiency, as indi- cated by the biochemical data from the NNMS, are young children, adolescents, and women of child- bearing age. Pregnant women and infants under 1 year of age are risk groups not well covered in current nation- ally representative surveys. Surveillance data for the Centers for Disease Control indicate that low- income pregnant women are at high risk of anemia. Dietary iron intake, assessed in the CSFII 1985-86, is very low in women of childbearing years relative to recommended levels. Available iron intake, determined by using data from the NFCS 1977-78, is also low for women relative to apparent require— ments. The intake estimates do not include the contribution from iron supplements. In contrast to iron deficiency, iron overload cannot be assessed adequately with current NNMS data to identify groups at risk. The prevalence of hemo- chromatosis is too low to be estimated reliably by available NNMS surveys. not include hemoglobin as a variable) and by using Trends laboratory tests that reflect the presence of inflammatory disease (C-reactive protein). ° eo No information on blood donation has been col- lected in the NNMS. XXXIV The best trend data available are those on the nutrient content of the U.S. food supply, which indicate an increase in the level of iron in recent years. e Assessing trends in individual intakes of iron by various population groups is more difficult because of methodological differences in the surveys over time, including revised nutrient composition data. The available NNMS data indicate little change during the last 20 years. e Assessing trends in iron nutritional status is also difficult, because the measures used to estimate the prevalence of iron deficiency have not been used in many surveys. Limited data from the Pediatric Nutrition Surveillance System suggest recent improvements in iron status among the low- income children monitored. Determining Factors e Sex and age are powerful determining factors relative to iron nutritional status. Evidence of iron deficiency is rare in males, in the elderly of both sexes, and in school children. e Univariate analyses of NNMS data indicate that the prevalence of iron deficiency is influenced by race and socioeconomic factors such as poverty status and education. Iron intake also differs with these variables, but not as consistently as iron status, suggesting differences in bioavailable iron. e Parity is also observed in NNMS data to be an influence on iron status in women during child- bearing years; women who have given birth to many children are at greater risk of deficiency. eo Other determining factors, such as iron supple— ment use, blood donation, use of medications, and growth, could not be assessed with current data from the NNMS. Recommendations The EPONM offered recommendations, based on their experiences in analyzing NNMS data for this report, for improvements in the NNMS in several areas: eo Improved comparability of nutrient composition data and coding used in different dietary surveys. eo Testing the impact of methodological differences on survey results. e Use of a common core of sociodemographic descriptors in all NNMS surveys. eo Greater similarities in NNMS data reporting. e Investigation of methods for assessing population groups currently excluded from the NNMS. e Improved coverage of some groups at nutritional risk: infants, pregnant women, lactating women, the elderly, preschool children, and adolescents. e Improved measures of usual dietary intake in the HANES. eo Collection of information for assessing the impact of knowledge and attitudes on patterns of food consumption and nutrient intake. eo Obtaining quantitative information on vitamin and mineral supplement use to better estimate total nutrient intake. eo Improving estimates of alcohol consumption. eo Improving response rates and analyzing non- response. eo Educating data users in the proper use of data from complex surveys. e Being responsive to the needs of State and local data users. The EPONM also suggested that, in future reports on the NNMS, the updates of nutritional status of the U.S. population be prepared separately from reports of detailed analyses of special topics. Updates might take the form of comprehensive reviews such as this report or might consist of tabulations of new data with more limited interpretation. XXXV Chapter 1 Introduction Charge to Expert Panel on Nutrition Monitoring The ad hoc Expert Panel on Nutrition Monitoring (EPONM) was established to review the dietary and nutritional status of the U.S. population, as well as the factors that determine status, based on the data and information available through the nutrition monitoring activities conducted by the U.S. Departments of Agriculture (USDA) and Health and Human Services (DHHS). This report requested by USDA and DHHS is the second report on the National Nutrition Monitoring System (NNMS). It builds on the framework of the first report submitted in 1986. The first report was an overview of the dietary and nutritional status of the U.S. population. It was developed by the Joint Nutrition Monitoring Evalua- tion Committee (JNMEC) to serve as a reference or baseline report for subsequent reports. The Joint Committee, as part of its analysis of data and infor— mation, partitioned food components into categories according to whether the component warranted public health monitoring priority status, public health monitoring consideration, or further investigation. Findings for the entire range of food components were presented for a variety of age and sex groupings. The JNMEC reported conclusions about the nutri- tional status of the U.S. population and made recom- mendations for improvements of the monitoring system. It also recommended that the focus of the second report be on factors that influence nutritional status. The charge given to the EPONM for this report directed that two themes dominate: e An updating of the dietary and nutritional status of the population as presented in the 1986 report. eo An in-depth analysis of the contributions of the NNMS to assessment of the status of the population as determined from two types of data—— data on diet and chronic diseases and data on dietary and nutritional problems. With respect to the first theme, the EPONM was charged to update the first report by comparing data in it with NNMS data on food and nutrient intake and nutritional status produced or released since publication of the first report. The EPONM was also charged to address methodological issues in com- paring data from different sources or times, and to identify the types of analyses necessary to make com- parisons meaningful. Trend and baseline data presented in the first report were to be updated. The EPONM was to reevaluate the categorization of food components as to the completeness of relevant data and the level of monitoring status that should be accorded each food component. Food components that were not included in the first report were to be considered if appropriate. The rationale for recategorization of previously reported food com- ponents and for the inclusion of new food components was to be stated as well as the rationale for any new assessment criteria. With respect to the second theme, the EPONM was to include in the report in-depth integrated analyses of two topics selected as examples of NNMS health and dietary data: the first, represented by data on the relationship of diet to a specified chronic disease, was to emphasize dietary and nutritional factors in cardiovascular disease; the second, represented by data on a dietary and nutritional problem, was the assessment of iron nutriture. These two topics were selected because of their public health significance and because of the breadth of data regarding them that was available from the NNMS. The JNMEC classified iron and several dietary components related to cardiovascular disease among the food components warranting public health monitoring priority status. The objective of this part of the report was to demon- strate how NNMS data could contribute to under- standing these public health concerns as well as to identify the strengths and weaknesses of data and information available primarily from components of the NNMS. The ability to identify the nature and magnitude of nutrition-related problems in the U.S. population was to be addressed, with focus especially on the NNMS capabilities for identifying (1) populations at risk, (2) limits to interpretations of data, (3) gaps in the database, (4) trends, and (5) determining factors. These factors include the following: eo Personal factors, such as age, race, ethnicity, education, height, and weight. eo Demographic and other factors, such as region, urbanization, and season. eo Household factors, such as income, work status, food assistance program participation, household size, tenancy status, usual food cost, perceived sufficiency of household food supply, and house- hold composition. eo Health-related factors, such as health status, smoking, activity level, and use of medications. e Vitamin and mineral supplement use. eo Eating pattern factors, such as food choices, source of food, food avoidance, and meal patterns. eo Knowledge and attitudes. Finally, based on experiences involved in reviewing data analyses for this report, the EPONM was to recommend ways to strengthen the NNMS. Background Brief History of the NNMS As noted in the JNMEC report (DHHS/USDA, 1986), the Food and Agriculture Act of 1977 (Public Law 95- 113) instructed the Secretary of Agriculture and the Secretary of Health, Education, and Welfare (now Health and Human Services) to submit to Congress a proposal for a comprehensive nutritional status monitoring system to integrate the ongoing nutrition survey activities of both Departments. The Depart- ments' proposal was submitted to Congress in May 1978 and, at the request of the Committee on Science and Technology, was reviewed by the General Accounting Office, which recommended the develop— ment of a comprehensive implementation plan. This plan, the Joint Implementation Plan for a Compre- hensive National Nutrition Monitoring System, was submitted to Congress in September 1981 and described ongoing nutrition monitoring activities as well as specifying goals and implementation activities. The NNMS includes all existing and proposed Federal survey and research activities with the purpose of monitoring nutritional status in the United States. The five component parts of the system are as follows: Nutritional and health status measurements. Food consumption measurements. Food composition measurements. Assessments of dietary knowledge and attitudes. Food supply determinations. The Joint Implementation Plan for a Comprehensive National Nutrition Monitoring System set two major objectives: e Achievement of the best possible coordination of the two largest components of the system, the National Health and Nutrition Examination Survey (of the DHHS) and the Nationwide Food Consumption Survey (of the USDA), and eo Development of a reporting system to translate findings from the two national surveys and other monitoring activities into periodic reports to Congress on the nutritional status of the American population. The first report of the JNMEC in 1986 and the current report of the EPONM represent activities designed to fulfill the second objective. Goals and Purposes of the NNMS The overall goals of the NNMS are as follows (DHHS/USDA, 1986): eo To provide the scientific foundation for the maintenance and improvement of the nutritional status of the U.S. population and the nutritional quality and healthfulness of the national food supply. eo To collect, analyze, and disseminate timely data on the nutritional and dietary status of the U.S. population, the nutritional quality of the food supply, food consumption patterns, and consumer knowledge and attitudes concerning nutrition. e To identify high-risk groups and geographic areas, as well as nutrition-related problems and trends, in order to facilitate prompt implementation of nutritional intervention activities. eo To establish national baseline data and to develop and improve uniform standards, methods, criteria, policies, and procedures for nutrition monitoring. eo To provide data for evaluating the implications of changes in agricultural policy related to food production, processing, and distribution which may affect the nutritional quality and healthfulness of the U.S. food supply. Specific goals for the current operational phase of NNMS activities (1987-96) are stated in the Operational Plan for the National Nutrition Monitor- ing System (DHHS/USDA, 1987) as follows: e Achieve a comprehensive system through coordi- nation among NNMS components by - including all appropriate nutrition monitoring activities; - improving coverage of major population groups at risk, including low-income groups, nursing home residents, native Americans living on reservations, and the homeless; — improving temporal coverage; and — improving comparability among surveys by increasing similarities of dietary methodologies, increasing uniformity of sociodemographic descriptors, sharing a common nutrient data— base, and using compatible survey designs. e Improve information dissemination and exchange by — continuing the reporting system to Congress (with this report); - continuing timely reporting and interpretation of data; — developing comparable reports and tape docu- mentation; — increasing policy relevance of data collected; and - improving information exchange between data users and generators. e Improve the research base for nutrition monitor- ing including - methodological research specific to the conduct of surveys and surveillance activities; and - research conducted by the broader nutrition community for its own purposes which also has application to nutrition monitoring. Components of the NNMS The components of the NNMS which represent sources of data included in the present report are summarized in table 1-1. This table contains the following information: survey or study name, sponsoring Agency, date, population studied, and data collected. For detailed descriptions of the surveys to be discussed, see appendix I. New cycles of several of the surveys listed have been undertaken; data collec- tion for the Nationwide Food Consumption Survey (NFCS) 1987-88 was recently completed and data collections for the third National Health and Nutri- tion Examination Survey (NHANES III) and the Continuing Survey of Food Intakes by Individuals (CSFII) 1989 were begun. In addition, the USDA 1988 Bridging Survey has been undertaken to com- pare the dietary methodologies used in the NFCS 1977-78 and the NFCS 1987-88. Data from these surveys were not available for inclusion in the current report. These components of the NNMS are augmented and supplemented by other Federal research and data collection activities. For example, the Agricultural Research Service of USDA and the National Institutes of Health, the Centers for Disease Control, and the Food and Drug Administration of DHHS provide much of the research base underlying the efforts of the NNMS in determining human nutritional re- quirements, methods for assessing nutritional status, and methods to measure food composition. The U.S. Vital Statistics Series of the National Center for Health Statistics and the Alcohol Epidemiologic Data System also supply data useful for national nutrition monitoring. Data from such additional sources of information are included, as appropriate, in this report. Uses of NNMS Data The many types of data from the components of the NNMS have been put to a variety of uses (National Research Council, 1984; Yetley and Johnson, 1987). A partial list of these uses includes the following: eo Determining food consumption patterns and nutrient intake of populations and subgroups. eo Demonstrating historical and secular trends in food consumption and nutritional status. eo Assessing the nutritional quality of diets of the population. eo Examining impacts of food programs and food guidance. eo Assessing the prevalence of nutrition-related health conditions. eo Developing educational materials for dietary guidance based on food intake patterns. Table 1-1. Sources of data from the National Nutrition Monitoring System considered in the EPONM! report Sponsoring Survey or study agency Date Population Data collected Nationwide Food USDA 1977-78 Private households in the Household characteristics; foods used from home Consumption 48 conterminous States and the supplies (7 days); household income; name and Survey (NFCS) individuals in those households description, quantity, form, source, and eating (all income and low income) occasion for all food and beve consumed by individuals (1-day recall, 2-days food records, 3 consecutive days); information on diet and health Continuing Survey USDA 1985 Women 19-50 years, children Household and individual characteristics, of Food Intakes 1-5 years, men 19-50 years individual food intake (one to six 24-hour Individuals (all income and low income) recalls, nonconsecutive days; 1 day only (CSFII) for men) 1986 Women 19-50 years, Household and individual characteristics, children 1-5 years (all individual food intake (one to six 24-hour income and low income) recalls, nonconsecutive days) U.S. Food Supply USDA Each year U.S. civilian population Per capita disap; ce of foods (levels of Series since 1909 nutrients in pa calculated) National Nutrient USDA Continuous NA Nutrient content of foods; basis of nutrient Data Bank composition databases for other surveys First National NCHS 1971-75 Civilian, noninstitutionalized Dietary intake (one 24-hour recall), socioeconomic Health and Nutrition pation of the United States; and demographic information, biochemical analyses Examination -'74 years of blood and urine, physical examination, body Survey (NHANES I) measurements Second National NCHS 1976-80 Civilian, noninstitutionalized Dietary intake (one 24-hour recall), socioeconomic Health and Nutrition Population of the United States; and demographic information, biochemical analyses Examination months-74 years of blood and urine, physical examination, body Survey (NHANES II) measurements Hispanic Health NCHS 1982-84 Civilian, noninstitutionalized Dietary intake (one 24-hour recall), socioeconomic and Nutrition Mexican Americans in five and demographic information, biochemical analyses Examination Southwestern States, Cuban of blood and urine, physical examination, body Survey (HHANES) Americans in Dade County, measurements Florida, and Puerto Ricans in metropolitan New York City; 6 months-74 years NHANES I NCHS 1982-84, Persons examined in NHANES I; Interviews of survivors and proxies for Epidemiological 1986 25-74 years old at baseline decedents; death certificate, hospitalization olor tudy history, health status, food frequency See footnote at end of table for definitions of acronyms. Table 1-1. Sources of data from the National Nutrition Monitoring System considered in the EPONM! report-—continued Sponsoring Survey or study agency Date Population Data collected National Health NCHS 1984 Civilian noninstitutionalized Health and living conditions of elderly aw Survey population; 55 years and over 1985 Civilian, noninstitutionalized Health promotion and disease prevention population of the United States; habits and knowledge 18 years and over 1986 Civilian, noninstitutionalized Use of vitamin and mineral supplements children (2-6 years) and adults (18 years and over) in the United States Food Label and FDA 1978, 1980 NA Prevalence of nutrition labeling and Package Survey 1982, 1983, declarations of selected nutrients and (FLAPS) 1984, 1986 ingredients Total Diet Study FDA Annually NA Mineral and contaminant content of (TDS) since 1961 representative diets for various age—sex groups Vitamin/Mineral FDA 1980 Civilian, noninstitutionalized Supplement intake, attitudes, and behaviors Supplement adults; 16 years and over (telephone interview) Intake Survey leah and Noa 1982, 1984 ii , nore tutions ed Afereness Silica, kiowisie: and iet Survey population; 18 years over viors regarding nutrition; health status and history (telephone interview) oA 1986 Civilian, goritie jh tiongliaed we attitudes knowledge. 2 NHLB pulation; 18 years over viors regarding food nutrition; NCI P health status and history (telephone interview) Pediatric Nutrition CDC Continuous Low-income, high-risk children Anthropometry, birthweight, hematological Surveillance System isemecially 1-5 years) in measures (PedNSS) 36 States Pregnancy Nutrition CDC Continuous Low-income, high-risk, pregnant Weight, health status, behavioral risk Surveillance System women in 12 States factors (PNSS) Behavioral Risk CDC Continuous Adults (18 years and over) Height, weg diet practices; salt, Factor Surveillance in 35 States in households alcohol, and tobacco use; cholesterol System (BRFSS) with telephones screening practices, awareness, and treatment 1 EPONM = Expert Panel on Nutrition Monitoring; NA = not applicable; USDA = U.S. Department of Agriculture; NCHS = National Center for Health Statistics; FDA = Food and Drug Administration; NHLBI = National Heart, Lung and Blood Institute; NCI = National Cancer Institute; CDC = Centers for Disease Control. e Examining the relationship of food consumption patterns and nutrient intakes to physical and physiological indicators of health status. eo Assessing the prevalence of specific knowledge of nutrition and certain health practices. ® Determining the economics of food consumption. e Establishing the distribution of values for indi- cators of health and nutritional status in the population. e Identifying food safety considerations. Major Conclusions and Uses of the JNMEC Report The major conclusions of the JNMEC report (DHHS/USDA, 1986) were stated as follows: e In the United States today, the food supply is safe and adequate, indeed, abundant. e The principal nutrition-related health problems experienced by Americans arise from the over— consumption of certain food components: fat, saturated fatty acids, cholesterol, and sodium. ® Twenty-eight percent of the American population ages 25-74 years, approximately 32 million people, are overweight. ® Available monitoring data suggest that, overall, Americans maintain low levels of physical activity. e Dietary and biochemical data indicate that intakes of iron and vitamin C are low in certain subgroups of the population. ® Because calcium deficiency has been implicated as a contributor to the prevalence of osteoporosis among postmenopausal white women, the rela— tively low intake of calcium among women is a cause of concern. ® Prevalences of health conditions directly or indirectly related to poor nutritional status are generally highest among the low-income popula- tion. As the first comprehensive review of data from the NNMS, the 1986 JNMEC report has been used in various ways. The report has been widely disseminated to information offices and professionals in the Government, academic scientists, the general public, and the international community. It has served as a reference for many other reports including The Surgeon General's Report on Nutrition and Health (DHHS, 1988) and activities such as evaluating progress in achieving the 1990 Nutrition Objectives (DHHS, 1986). Conclusions and recommendations of the report have been used for the following purposes: ® Guiding policy decisions of various Agencies. e Justifying additional efforts to measure the dietary and nutritional status of low-income populations. o Establishing priorities for Agency research and education activities including research on food composition and factors affecting dietary status. Principles and Definitions Used The update portion of this report is a followup to the 1986 JNMEC report and is intended to identify new data available from the NNMS and to examine changes and trends in dietary intake, nutritional status, and nutrition-related health conditions. The JNMEC report emphasized a coordinated review of dietary data from the NFCS 1977-78 and nutritional and health status data from the second National Health and Nutrition Examination Survey (NHANES II). New national survey data on dietary intake and nutritional status of the entire U.S. population have not become available since the JNMEC report: however, data for certain subgroups of the population have become available. Most of the available data are for the three Hispanic groups surveyed in the His- panic Health and Nutrition Examination Survey (HHANES) and the women and children surveyed in the CSFII 1985-86. The current report will describe the information available since the 1986 report with respect to the following three types of data: ® Cross-sectional data that may stand alone or serve as a new baseline against which data collected in the future may be compared. e Trend data (measurements repeated over time using similar techniques). e Longitudinal data on cohorts of individuals identi- fied at a specific point in time (re-examination of same individuals). Data analyses included in the report are intended to be descriptive of dietary and/or nutritional status, trends, and associations rather than to establish causal relationships. The report is not intended to serve the purpose of program evaluation. Summary data from relevant surveys are included in appendix II, by topic; detailed analyses are provided in the body of the report when a public health issue is identified for some of the food components included in the update (criteria are described in chapter 3) as well as in the chapters on cardiovascular disease and diet and assessment of iron nutriture. Many of the data analyses presented were prepared by the Agencies specifically for inclusion in this report; others were obtained from Agency publications or the peer-— reviewed literature. The chapters on dietary and nutritional aspects of cardiovascular disease and on iron nutriture are intended to illustrate the utility of NNMS data in examining these public health problems, not to provide a comprehensive review of the literature concerning these topics. The EPONM relied upon sources such as The Surgeon General's Report on Nutrition and Health (DHHS, 1988) and Diet and Health: Implications for Reducing Chronic Disease Risk (National Research Council, 1989) for summa- ries of the scientific evidence concerning nutritional status, health, and diseases. The EPONM determined that it would be useful to highlight the definition of some terms used in this report as well as in the JNMEC report (see also Glossary, appendix III). Discussions of these terms, used to describe nutrition monitoring activities, status assessment, and related factors, are provided in the following sections. Nutrition assessment refers to the measurement of indicators of dietary status and nutrition-related health status to identify the possible occurrence, nature, and extent of impaired nutritional status (ranging from deficiency to toxicity). Nutrition monitoring refers to the assessment of dietary or nutritional status at intermittent times with the aim of detecting changes in the dietary or nutritional status of a population. Nutrition surveillance refers to a continuous assessment of nutritional status for the purpose of detecting changes in trend or distribution in order to initiate corrective measures. Dietary status is defined as the condition of a pop— ulation's or an individual's intake of foods and food components, especially nutrients. Food components discussed in this report include nutrients (macro- nutrients, vitamins, and minerals) and non—nutrients that may affect health (such as dietary fiber), ob— tained from food or other sources (such as vitamin/ mineral supplements). Nutritional status is defined as the condition of a population's or an individual's health as influenced by the intake and utilization of nutrients and non—-nutrients. Measures of nutritional status reflect, directly or inferentially, the processes of food ingestion and digestion; absorption, transport, and metabolism of food components; and excretion of food components and their metabolic products. As noted in the JNMEC report (DHHS/USDA, 1986), indicators of nutritional status include (1) levels of specific food components in diets; (2) clinical, anthropometric, hematological, and biochemical measurements related to specific food components; and (3) health conditions or diseases that may be associated with inadequate or excessive intakes of several food components. Health status, as used in this report, refers to a population's or an individual's status with respect to physical state or disease condition. Various terms are used to describe nutritional status. Overnutrition refers to the condition resulting from the excessive intake of foods in general or particular food components; undernutrition refers to the con- verse situation resulting from the inadequate intake of foods in general or particular food components. Nutrient deficiency is defined as a condition associated with adverse health consequences arising from inadequate intake or utilization of a nutrient. Marginal nutritional status is defined as a condition in which nutrient stores may be low, but impairment of performance, health, or survival may not be evident. Persons with marginal nutritional status are considered at risk of nutritional deficiency, especially when subjected to stress. Nutritional imbalance is defined as a condition associated with adverse health consequences arising from insufficient or excessive intake of one nutrient or food component relative to another. Nutrient excess and/or toxicity is defined as a condition associated with adverse health consequences arising from excessive intake or utilization of a nutrient. Terms used to describe the occurrence of conditions or diseases include prevalence and incidence. Prevalence is the number of instances of a given disease or other condition in a given population at a designated time. Incidence refers generally to the number of new events (for example, new cases of disease) in a defined population, within a specified time period. Because of the JNMEC observation that health conditions related to poor nutritional status were most prevalent in the low-income population, poverty status is an important variable to examine in assess ing dietary and nutritional status. The statistical measurement of poverty was developed by the Social Security Administration in 1964. The poverty index consists of a set of cash income cutoffs that vary by the size and number of children in a family (Presi- dent's Task Force on Food Assistance, 1984). Since 1964, various cutoffs for the definition of poverty index have been used for a variety of analytical and policy applications. For example, 130 percent of the poverty index has been used to establish eligibility for several Federal nutrition and food assistance programs. (See appendices I and II for details on the calculations of poverty indices in specific NNMS surveys.) The JNMEC (DHHS/USDA, 1986) noted that poverty may result in the inability to obtain a sufficient quantity or variety of food to prevent hunger or to maintain good nutritional status. Hunger (unas— suaged) is considered the perceived need to eat. It may be the result of poverty and may or may not be associated with poor nutritional status, inadequate dietary intake, or measurable nutrient deficiencies. Hunger represents a social problem regardless of its nutritional implications. The JNMEC (DHHS/USDA, 1986) also noted the paucity of data from the NNMS available at that time to assess the occurrence and impact of hunger. Although information on perceived household food sufficiency was collected in the NFCS 1977-78 and CSFII 1985-86, the more recent data were not available for consideration in the current report. Dietary guidelines considered in this report are the qualitative recommendations from Nutrition and Your Health: Dietary Guidelines for Americans (USDA/ DHHS, 1985) which are also included in The Surgeon General's Report on Nutrition and Health (DHHS, 1988). Discussion of the specific dietary guidelines and recommendations promulgated by other groups is included for illustrative purposes and should not be construed as endorsement by the EPONM. General Conceptual Model A general conceptual model representing the relationships among food choice, food and nutrient intake, and nutritional and health status developed by the EPONM in formulating discussions of the nutritional status of the U.S. population is presented in figure 1-1. The EPONM developed the model based on their collective knowledge of the principles of nutrition, economics, and behavior and on a review of several similar models (Hautvast and Klaver, 1982; Mason et al., 1984; Mayer and Dwyer, 1979; McLaren, 1981; National Research Council, 1984; Sanjur, 1982; World Health Organization, 1976). The model identifies the major stages at which the effects of food and nutrient intake on nutrition-related health status may be assessed as well as the factors that influence each stage. The model represents a starting point rather than an exhaustive description of all possible factors and interrelationships; it is designed to allow for expansion or elaboration of detail (see chapters 3, 4, 5, and 6 for expanded models appro- priate for the discussions contained in these chapters). The components of the model shown in solid-line boxes represent the primary steps in the sequence from the food supply to health outcome(s). Components shown in broken-line boxes represent those factors that may influence each primary stage. The model does not distinguish between direct and indirect influencing factors. The relationships illustrated in the model have been interpreted from independent experimental or obser- vational studies. The NNMS measures the prevalence and distribution of various outcomes and influencing factors in the U.S. population, thus permitting the evaluation of the nature and extent of public health issues. Data from the NNMS are available for many, but not all, components of the model; when data are available, however, they do not necessarily address all relevant questions adequately. The following paragraphs describe the components of the general conceptual model and sources of NNMS data related to these components. Cross-sectional and trend data on the quantities of foods that enter the domestic food supply are pro- vided by the U.S. Food Supply Series. The division of foods into two categories, "away-from-home" and "household" food, is often desirable for characterizing acquisition and consumption patterns, but the separation between the two categories is not always clear in the current distribution system. Away-from- home foods may be obtained from restaurants, fast— food establishments, cafeterias, vending machines, and other foodservices as well as from food programs such as school lunch and breakfast and congregate meal programs for the elderly. Foods consumed away from home may have been prepared at home. Household food includes food products purchased in the market, food donations (food as gift or in-kind payment), and food grown or produced at home as well as prepared foods obtained from foodservice establishments. Factors that influence the choices to acquire and consume foods (measured in many NNMS surveys) include household income; the price of food; personal factors such as age, sex, ethnic group, education, and physiological status (such as preg- nancy); environmental factors such as advertising; and characteristics of food such as label information or convenience. These factors determine food prefer- ences, cognitions, and attitudes. In turn, food prefer— ences, cognitions, and attitudes may be affected by exposure to foods in the marketplace. Consumer demand for various products also influences the availability of foods in the food supply. Household food consumption (or money value of food consumed or used) is measured in the household NATIONAL FOOD SUPPLY —————> FOOD DISTRIBUTION ——> CONSUMPTION —— NUTRIENT UTILIZATION ————> HEALTH OUTCOME Smoking Sanitation Medication, drug use Housing Alcohol use Occupation Genetic factors Other factors Nutrient interactions Physiological status Activity, exercise Age — Sex — Race Environmental factors Agricultural factors Economic factors Policy considerations Away—from—home Away—from—home Away-from—home 7 — food available food acquired food consumed kre essen oT" weal Infection ! IT. J ! Disease | Nutrient : | requirement pTTmTTTsmmeomsseoooes : deme meen : Nutritional status Health/disease bssrmmsepenesnnnn ! Income — Price TTR. -— outcome J emma y L, @ biochemical ! Sociocultural factors WN © hematological ' Demographic factors - | > Food preferences, | Food consumed Nutrient Nutrient e anthropometric i Educational factors ‘ 1 cognitions, attitudes | by individual intake utilization eo clinical : Environmental factors | A CRT Food supply ! Physiological factors | | T | immersion supper un : Exports : i Imports | | Storage Primary representative action or consequence Household food Household food Household food gor" ! Influencing or mitigating factor available acquired consumed ' ! Figure 1-1. General conceptual model for food choice, food and nutrient intake, and nutritional and health status (see text for explanation) component of the NFCS. Individual food consump- tion is measured in the individual component of the NFCS, the CSFII, and the Health and Nutrition Examination Surveys (HANES). Combining informa- tion in the nutrient composition databases developed by the USDA with food supply or individual food con- sumption data permits the estimation of per capita nutrient content of the food supply and individual nutrient intake, respectively. The mineral content of typical diets in the United States is determined in the Total Diet Study. The use of supplements also con— tributes to nutrient intake. Supplement use has been assessed, but not quantified, in the NFCS 1977-78, CSFII 1985-86, and HANES; use was measured in the 1980 Vitamin/Mineral Supplement Intake Survey and in the 1986 National Health Interview Survey (NHIS). Monitoring of the information on nutrition labels is conducted in the Food Label and Package Survey (FLAPS). Nutritional status indicators are measured in the HANES and the Pediatric and Pregnancy Nutrition Surveillance Systems (PedNSS and PNSS, respec- tively). Although nutrient utilization is not measured directly, factors that may influence nutrient utiliza- tion and nutritional status are assessed in a variety of NNMS surveys. The prevalence of various diseases and nutrition-related health conditions is estimated in the HANES, PedNSS, and PNSS. The NHANES I Epidemiological Followup Study (NHEFS) contains cohort data that permit exploration of relationships of dietary and nutritional status and subsequent mor- bidity and mortality. The U.S. Vital Statistics Series also provides mortality data for some conditions related to diet and nutrition. Finally, knowledge, attitudes, and practices that influence nutritional and health status are assessed in selected years by the NHIS, the Health and Diet Survey, and the Behav- ioral Risk Factors Surveillance System (BRFSS). Organization of the Report With respect to the requested update of the informa- tion contained in the JNMEC report, chapter 2 (on the appropriate uses of survey data for assessing dietary and nutrition-related health status) is responsive to the charge to address methodological issues in comparing data from different sources or times and to identify the types of analyses necessary to make comparisons meaningful. The discussions of data that comprise the update portion of this report are divided into two chapters, with the first (chap- ter 3) providing new data on dietary intake, food availability, major food sources of various food components, and nutritional status with respect to 10 various nutrients, and the second (chapter 4) providing new estimates of the prevalence of nutrition-related health conditions and behaviors. Reference is made to baseline data when appropriate. With respect to the second charge to the EPONM, the two selected topics are discussed separately. The chapter on nutritional and dietary factors in cardio- vascular disease (chapter 5) concentrates on the utility of NNMS data in assessing risk factors for cardiovascular disease and dietary and other factors related to these risk factors. The chapter on assess- ment of iron nutriture (chapter 6) concentrates on the utility of NNMS data in assessing the prevalence of iron deficiency and identifying groups at risk and the factors contributing to iron nutriture. Finally, chapter 7 contains recommendations from the EPONM for improvements in the NNMS, based on their experiences in evaluating the data included in this report. References Cited Hautvast, J. G. A. J., and W. Klaver. 1982. A Con- certed Action Project on Nutrition in the European Community: The Diet Factor in Epidemiological Research. EURO/NUT Report 1. Wageningen, The Netherlands: Ponsen and Looyen. Mason, J. B., J.-P. Habicht, H. Tabatabai, and V. Valverde. 1984. Nutritional Surveillance. Geneva: World Health Organization. Mayer, J., and J. T. Dwyer, eds. 1979. Food and Nutrition Policy in a Changing World. New York: Oxford University Press. McLaren, D. S., ed. 1981. Nutrition in the Community. New York: John Wiley & Sons. National Research Council, Committee on Diet and Health. 1989. Diet and Health: Implications for Reducing Chronic Disease Risk. Washington: National Academy Press. National Research Council, Coordinating Committee on Evaluation of Food Consumption Surveys. 1984. National Survey Data on Food Consumption: Uses and Recommendations. Washington: National Academy Press. President's Task Force on Food Assistance. 1984. Report of the President's Task Force on Food Assis— tance. Washington: 726 Jackson Place, N.-W. Sanjur, D. 1982. Food and Food Intake Patterns—-— Central Issues in Their Conceptualization and Mea- surement, in V. A. Beal and M. J. Laus, eds, Proceedings of the Symposium on Dietary Data Collection, Analysis, and Significance. Research Bulletin No. 675. Amherst, Mass.: Massachusetts Agricultural Experiment Station. U.S. Department of Agriculture and U.S. Department of Health and Human Services. 1985. Nutrition and Your Health: Dietary Guidelines for Americans. Home and Garden Bulletin No. 232. 2nd edition. Washington: U.S. Government Printing Office. U.S. Deparment of Health and Human Services. 1986. The 1990 Health Objectives for the Nation: A Mid- course Review. Public Health Service. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services. 1988. The Surgeon General's Report on Nutrition and Health. DHHS Pub. No. (PHS) 88-50210. Public 11 Health Service. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1986. Nutrition Monitoring in the United States——A Progress Report from the Joint Nutrition Monitoring Evaluation Committee. DHHS Pub. No. (PHS) 86-1255. Public Health Service. Washington: U.S. Government Print- ing Office. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1987. Operational Plan for the National Nutrition Monitoring System. World Health Organization, Joint FAO/UNICEF/ WHO Expert Committee. 1976. Methodology of Nutritional Surveillance. WHO Technical Report Series No. 593. Geneva. Yetley, E., and C. Johnson. 1987. Nutritional Applica- tions of the Health and Nutrition Examination Surveys (HANES). Annu. Rev. Nutr. 7:441-463. Chapter 2 Appropriate Uses of Survey Data in the Assessment of Dietary and Nutrition—Related Health Status Assessment of Dietary Status Assessment of dietary status includes consideration of the types and amounts of foods consumed as well as intake of the nutrients and other components con- tained in foods. Information on diets may be col- lected at several levels——national food supply, household food use, individual food intake--and by a variety of methods. When data on food consumption are combined with information on the composition of food, estimates may be made of the intake of particu- lar nutrients and other food components. Available Methodologies Nutrient composition The nutrient composition databases used for dietary assessment are based on pooling of data from many sources. Methods for the characterization of the nutrient composition of foods require a variety of complex steps: selection of foods to be analyzed; appropriate sampling of foods to be analyzed; homog- enization and subsampling, extraction, separation, detection, and identification of nutrients; calculation of results and report generation; use of standards, standard reference materials, and control samples; and validation of results (Beecher and Vanderslice, 1984). In view of the variety of the chemical forms of nutrients in foods and the complexity of foods, it is not surprising that a wide variety of analytical meth- ods of variable quality exists for the determination of nutrients. Nutrient content data are lacking for some nutrients because accurate, precise, and affordable analytical methodologies are not available. Beecher and Vanderslice (1984) have summarized the state of development of methods for the analysis of nutrients in foods. The quality and quantity of available nutrient content data may also vary for other reasons. 13 Providing data on all nutrients in all types of food consumed in the United States is an overwhelming task; data from NNMS surveys of food consumption aid in the determination of foods to be included in the database. Information is not as complete for less studied nutrients as for those nutrients for which a requirement or disease relationship has long been recognized. Information also is less comprehensive for the wide variety and increasing numbers of commercially prepared foods than for traditional food commodities (Beecher and Vanderslice, 1984). Efforts to improve existing nutrient databases, in terms of the number of nutrients and foods and the amount of analytical data included, are in progress. Hepburn (1987) has summarized the percentage of analytical values (determined by chemical analysis) and imputed values for nutrients included in the Primary Nutrient Data Set used for the CSFII 1985-86. For all foods in the database, the proportion of analytical values, as opposed to imputed values, equaled or exceeded 90 percent for the more familiar nutrients that have been assessed for many years, whereas the proportion of analytical data for components newly added to the survey, such as dietary fiber and a-tocopherol, was less than 30 percent. However, the proportion of analytical values from foods that represent the major sources of each food component was relatively higher (greater than 70 percent for most nutrients). The variability of nutrient content cannot practically be reflected in the nutrient databases. For example, the content of some nutrients in foods may vary with the cultivar of plant, breed of animal, geographic region, season, and growing conditions. Brand differences, changes in formulations, introduction of new products, and alterations due to processing, packaging, storage, or cooking also introduce variability into nutrient content data. In practice, the variability of nutrient composition data does not often contribute greatly to variability in the estimate of usual nutrient intake; however, the adequacy of the database for any particular nutrient or other food component of concern should be assessed. As noted by the National Research Council (1981), "the ideal nutrient database would: 1. be current, reliable, and valid; 2. be responsive to changes in the food supply; 3. contain information on all the nutrients of interest; . have complete data (unavailable data should be extrapolated until analytical values are obtained); . be expandable as new data become available; . reflect differences associated with brands; and . be in a physical form that facilitates coding and analysis." The Human Nutrition Information Service (HNIS) of the USDA continuously maintains the National Nutrient Data Bank, the ultimate source from which data are drawn for a variety of published and machine-readable databases, including the nutrient databases for assessing diets reported in national food consumption surveys. The same Nutrient Database for Individual Food Intake Surveys was used in both the CSFII 1985-86 and the HHANES. A similar database was developed and used in the NFCS 1977- 78. In NHANES II, the nutrient composition data— base was primarily compiled by DHHS based on data from USDA's Handbook 8 (sections 1-6) and Hand- book No. 456 and data from manufacturers (if avail- able) for commercial food items reported 20 or more times. Food availability and intake The estimation of food available for consumption and of dietary intake is an essential component of the evaluation of dietary and nutritional status. Data on both food availability and individual food consump- tion from national surveys are included in the current report. Food availability. Food availability data may be assessed at the national, retail, warehouse, or house- hold level. The only data on food availability con- sidered in this report are the national disappearance data (U.S. Food Supply Series) collected by the Economic Research Service (ERS) of the USDA. The U.S. Food Supply Series estimates the amounts of approximately 300-400 foods that "disappear" into 14 the U.S. food distribution system. Disappearance is estimated by subtracting utilization (exports, military use, year—end inventories, and nonfood use) from supplies (annual production, imports, and beginning- of-the-year inventories) (see appendix I for a more detailed discussion of these procedures). Estimates of per capita availability of food are derived by dividing the total food disappearance during the year by the civilian population of the 50 States and the District of Columbia. The HNIS of USDA calculates the nutri- ent content of the U.S. food supply by multiplying the weight of food consumed per capita per year by the nutritive value of the edible portion per pound, sum- ming the results for all foods, and expressing the total on a per day basis. Disappearance is measured at different points in the distribution system for different foods. No deductions are made in the estimates for loss or waste that occurs after the food is measured, such as in further processing, marketing, or home use. Some sources of nutrients, such as alcoholic beverages and vitamin/ mineral supplements in tablet, capsule, or liquid form, are excluded from the data, but vitamins and minerals added for the fortification or enrichment of food products are included. Although per capita availability is likely to overestimate actual ingestion of food, these overestimates are probably consistent over time because of efforts to keep methodology con- stant (Welsh and Marston, 1982). Because different categories of consumers (for example, children and adults) consume different amounts and types of foods, and because the per capita availability is calculated per person without adjustment for age and sex, comparisons of per capita data across time, or comparisons of U.S. per capita data with per capita data from other countries will also reflect any differences in the demographic struc— ture of the populations in addition to that of actual food availability. All these considerations must be taken into account in interpreting the U.S. Food Supply Series data, and care must be taken in linking any trends in disappearance data to trends in disease or mortality. Nonetheless, the data provide a rapid and inexpensive indicator of the overall sufficiency of the foods available to the U.S. civilian population. In addition, these data are especially valuable for use in studies of the effects of technological, economic, and social changes on the U.S. diet and future food pro- duction as well as for examining trends in food use, nutrient levels, and food sources of nutrients. Food intake. The other type of dietary data consid- ered in this report is individual intake data. Such data may be divided into two categories based upon the methods by which they are collected: quantitative daily consumption and semi-quantitative food frequency methods (LSRO, 1986). Quantitative daily consumption methods attempt to measure the nature and quantity of individual foods consumed in a defined period of time, in contrast to food frequency methods which attempt to measure patterns of food use and implied nutrient intake across longer and often less precisely defined periods of time. The instruments used most frequently to collect dietary data by quantitative daily consumption methods are recalls or records of actual food intake over a specified period of time (usually one or more days). All foods eaten in the specified time period are reported or recorded together with an estimate of the amount ingested (aided by the use of household mea- sures, standard serving sizes, food models, or weigh— ing). The reliability of records and recalls has been found to be comparable for group estimates (Fanelli and Stevenhagen, 1986; Krantzler et al., 1982; Pao, Mickle, and Burk, 1985). The use of both types of instruments is represented in the national surveys considered in this report: the HANES (NHANES I and II and HHANES) each employed a single 24- hour recall, the NFCS 1977-78 employed a 24-hour recall followed by two consecutive days of food records, and the CSFII 1985-86 employed an initial 24-hour recall conducted in person followed on non- consecutive days by up to five additional 24-hour recalls administered by telephone (or in person, when necessary) over the period of one year. Quantitative daily consumption methods can, depending on subject memory and interviewer skill, provide reasonably accurate information on actual intakes of foods or food components; moreover, if an adequate number of replicates are included, such methods can also provide an estimate of usual intake. The major difficulty in the interpretation of data derived by these methods is the large variation in day-to-day intake of food and nutrients within indi- viduals, referred to as intraindividual variation. Intraindividual variation is often greater than inter— individual variation. Ratios of the two may differ among foods and food components; among age, sex, and socioeconomic groups; and within and between dietary intake instruments (LSRO, 1986; National Research Council, 1986). When the intraindividual variation is large and the number of recalls or records is small, the ability to detect statistically significant differences in the mean dietary intakes of groups of individuals is reduced. The number of days for which dietary recalls or records are obtained also affects the appropriate use of the data, with greater restrictions placed on the interpretation of data obtained for a single day than on data obtained over multiple days (LSRO, 1986; National Research Council, 1986). Single-day intake 15 data usually result in a distribution that is flatter and wider than the true distribution of usual intakes of individuals in the population. Thus, the prevalence of high or low intakes is overestimated. The large intraindividual variation associated with single-day data may also mask associations between dietary intake and health outcomes because of misclassifica— tion of individuals by level of intake on the basis of single-day dietary intake. Although multiple-day data may be subject to bias, use of data obtained from multiple days may provide a means of reducing the effects of intraindividual variation on estimates of usual intakes, thereby increasing the accuracy of esti- mates of mean intake. In the CSFII 1985-86, a sta— tistically significant decline in reported intake has been detected after the first day (wave) of data collec— tion (Ritenbaugh et al., 1988). Additional analyses of these data indicated that nutrient densities did not vary much by wave (USDA, 1987, 1988), suggesting that the reported quantity of foods consumed decreased but that the overall nutrient composition of the foods consumed remained similar among waves of data collection. The other category of methods for assessing individu- al dietary intake, semi-quantitative food frequency methods, includes all questionnaires (general and focused food frequencies as well as diet histories) in which subjects recall their usual dietary intake during a time period in the past. These methods are used most often to estimate usual intake of foods and of food components by rank or category according to frequency of consumption rather than to provide a quantitative measure of actual intake. Data collected by food frequency methods are regarded as more representative of the usual intakes of subjects than quantitative intake data collected for a single day or for only a few days and are less affected by intraindi- vidual variation (LSRO, 1986). However, the accur— acy of estimates of actual intake obtained with these methods is not as great as the accuracy of those ob- tained with quantitative daily methods. A combina- tion of food frequency and quantitative daily con- sumption data may provide a more comprehensive assessment of food consumption and nutrient intake than either method alone (Krantzler et al., 1982). Food frequency methods have been used less often in the national surveys considered in this report, but a food frequency questionnaire was included in the Estimates of population and individual intakes of foods and food components Single-day dietary intake data collected by recall or record may provide a reasonable estimate of the mean intake for a population if the sample size is suffi- ciently large to minimize random errors (LSRO, 1986). However, single-day data are not adequate for evaluating the usual dietary intake of individuals. The number of days of data collection required depends on the purpose of making an estimate, the precision desired, the food component(s) of interest, and the amount of intra— and interindividual varia— tion. Basiotis et al. (1987) have examined this ques— tion in a group of 29 adults who kept daily food records for one year. Each individual's average intake of nutrients and associated standard deviation over the year were considered to represent the "usual" intake and variation; the number of days of records needed to estimate individual and group intake within 10 percent of usual intake was calculated. Results indicated that the number of days of records required for an individual estimate varied among individuals for the same nutrient and within individ- uals for different nutrients. On average, food energy required the fewest days for individuals (31 days) and vitamin A the most (433 days). In contrast, the num— ber of days required to estimate mean intake for the group was considerably less (3 days for food energy and 41 days for vitamin A). Estimating the mean intake of larger groups, such as those reported in NNMS surveys, would require fewer days. Freudenheim, Johnson, and Wardrop (1987) have also examined the misclassification of nutrient intake of individuals based on food records of varying length. In 106 adult women, 1-, 2—-, 3-, and 7-day records were compared with usual intake (based on 37-72 records per subject) of food energy, calcium, vitamin A, and vitamin C. For the 1-day record, 43-67 per- cent of subjects were correctly classified to the extreme quintiles of intake; for the 7-day record, 52— 76 percent were correctly classified. These results indicate the need for caution in interpreting dietary intake data collected for a single day or a few days as representative of an individual's usual intake. Criteria for Assessment of Dietary Intake of Nutrients (and Other Food Components) Recommended Dietary Allowances As defined by the Committee on Dietary Allowances of the Food and Nutrition Board, the Recommended Dietary Allowances (RDA) are the levels of essential nutrients considered, on the basis of available scien— tific knowledge, to meet the known nutritional needs of practically all healthy persons (National Research Council, 1980). (The allowance for energy is set to meet the average needs of most of the population; for 16 vitamins and minerals that are less well studied, esti- mated ranges of Safe and Adequate Daily Dietary Intakes have been recommended.) The RDA repre- sent the average daily amounts of nutrients popula— tion groups should consume over time (expressed as amount per person per day) and are not intended to represent individual requirements. However, the RDA frequently have been used as standards to eval- uate the adequacy of nutrient intake. Because the RDA include a margin of safety, intakes below the RDA are often not inadequate, but the risk of inade— quate intake increases as the mean intake of a popu- lation falls to lower percentages of the RDA. To overcome these difficulties, proportions (three- fourths, two-thirds, one-half) of the RDA have been used in analyses of food consumption data. However, no clear rationale for the selection of any particular cutoff has been advanced. In addition, because of the differences in data available for determining nutrient requirements, the RDA for different nutrients have different margins of safety, and interpretation of the meaning of levels of intake at any proportion of the RDA cannot be considered to be the same for all nutrients (see discussion of individual nutrients in chapter 3). Considering the problems and misinterpretations occasioned by the use of the RDA as a standard for dietary adequacy, the EPONM has chosen not to express dietary intake data in this report as a percent of the RDA or to apply the RDA or any proportion of the RDA as a sole criterion for assessing whether a nutrient constitutes a public health problem because of inadequacy. However, mean intakes of population groups falling well below the RDA can be taken as rough indicators that further examination of the status of that population group is needed. This approach is elaborated in chapter 3, in the consider— ation of the public health concern with individual food components. Nutrient densi Nutrient density, the unit measurement of each nutrient per 1,000 kilocalories, is another way to measure the nutritive quality of the diet. At first consideration, nutrient density seems to offer a sim- ple and straightforward means of assessing dietary quality, but the definition of appropriate nutrient- to—energy ratios has proven difficult (Beaton, 1988). Although specific standards have not been suggested to evaluate the adequacy of nutrient density in the United States, expressing nutrient intake in this fashion can help evaluate dietary quality in some sit- uations. For example, nutrient densities have been used to assess the nutrient contributions of convenience foods consumed in households (Havlicek, Axelson, and Capps, 1983). Data for intake on a kilo— calorie basis are presented for several of the nutrients discussed in chapter 3. Probability approach The use of any fixed cutoff point for the assessment of the adequacy of nutrient intake fails to take into account the variability of requirements among indi- viduals. This weakness led the Subcommittee on Cri- teria for Dietary Evaluation (National Research Council, 1986) to describe and test a "probability approach” to the assessment of nutrient adequacy that is based on the probability that a specific intake is inadequate to meet an individual's requirement. Application of this approach requires the following: (1) estimates of average requirements and variability (standard deviation) for each nutrient, (2) informa- tion on the shape of the distribution curve of require- ments, and (3) information on the distribution of usual intakes of the nutrient. The latter may be esti— mated from multiple days of dietary intake data, with appropriate statistical adjustments to account for the contribution of intraindividual variation. The proba- bility approach can yield estimates of the prevalence of inadequacy in a population group but cannot assess the occurrence of inadequate intake by any given individual. Limitations on the application of this approach include the lack of information on the mean and shape of the requirement distributions for many nutrients, although different values can be assumed. The accuracy of estimates of inadequacy derived by the probability approach may also be constrained by the same systematic errors that affect other evalua- tions of nutrient adequacy, such as the underreport- ing or overreporting of food intake. In addition, a statistical assumption of low correlation between requirement and intake or assimilation is necessary for the application of the probability approach. For example, the assumption is violated for food energy (the level of dietary intake and requirement are highly correlated) and for iron (absorption increases as requirement increases). In theory, the probability approach may also be applied to the estimation of the prevalence of exces— sive intakes of dietary components. This application would require information on the distribution of intakes judged to be detrimental to individuals in the population. The lack of such information severely constrains the current utility of the probability approach for this application. 17 The EPONM considers the probability approach to be an attractive one that overcomes many of the diffi- culties involved in assessing population dietary intake data by use of fixed cutoff points. However, because the basic information on requirement distributions is lacking for most nutrients at this time, the EPONM has chosen not to include estimates of nutrient inadequacy derived by the probability approach in this report. Another consideration in the application of this approach is that requirement must be defined. For any given nutrient, a family of requirement dis— tributions may be generated, ranging from intakes required to prevent the clinical signs of nutrient defi- ciency to intakes required to provide ample body reserves of the nutrient. When the necessary sup- porting information is generated, the probability approach can be used more extensively and its utility established. Johnson et al. (1988) have suggested several guidelines for the use of dietary assessment methods and made recommendations for improving approaches to national nutrition assessment. Other approaches to assessment In many cases, useful information may be obtained by classification methods that do not require the use of an arbitrary standard to evaluate dietary intake data. For example, subjects can be classified by some health variable and intakes of the groups identified as dif- ferent can be compared. Alternatively, a population can be stratified by percentiles of intake and health, demographic, or other variables can be examined within these groups for differences and/or similar— ities. However, the large intraindividual variations in dietary intake result in a large proportion of indi- viduals being misclassified when individuals are clas— sified into different groups by a dietary variable (LSRO, 1986). There is a continuing need to explore other approaches to assessment. Contributions of Supplement Use and Other Sources of Food Components to Dietary Intake The evaluation of the total intake of nutrients requires the consideration of sources of intake other than foods. Current data indicate that approximately 40 percent of American adults consume one or more vitamin/mineral supplements, and that substantial amounts may be consumed from these sources (Stewart et al., 1985). Although information about supplement use is collected in several of the NNMS surveys, quantitative estimates of the contribution of supplement use to the intake of nutrients cannot be made based on any of the surveys of food consump- tion included in this report. Thus, all estimates of dietary intake are derived from foods only, and the total intake of some nutrients may be underestimated for many individuals. In addition to dietary supplements, other nonfood sources of nutrients also (ideally) should be included in estimating total nutrient intake. These include drinking water (which may supply substantial amounts of minerals and electrolytes) and over-the- counter and prescription medications (such as cal- cium-containing antacids). Consistency in Data Sources to Assess Changes Over Time Comparability of information across databases should be taken into account in the interpretation of survey results. With respect to data on dietary intake of foods and food components, two areas of comparabil- ity are of importance: the nutrient composition data- bases and the methods to assess food intake. Systematic bias in nutrient composition data may result from inadequate analytical methods or imputa- tion of incorrect values in the database. If an existing bias in a nutrient database is corrected, a false impression of change in nutrient intake over time may be created (Yetley, Beloian, and Lewis, 1986). This difficulty cannot be overcome by using the same nutrient composition database for surveys conducted at different times because adjustments must be made for changes in foods and the composition of foods in the marketplace that occur over time. Another con- sideration in examining changes over time is consis- tency in the food descriptors and degree of detail (brands) included in the databases (Yetley, Beloian, and Lewis, 1986). Other sources of variability over time or across sur- veys are differences in the methods for collecting food intake information from individual subjects. Use of different degrees of probing and food models or other measurement aids for estimating portion size can elicit different responses. Similarly, differences in the completeness of reporting may occur when a subject is interviewed in the presence of other family members rather than in private, or if a proxy report is given for a subject. Different surveys also include varying coverage of seasons as well as weekdays and weekend days when food patterns are expected to differ. Dif- ferences in estimated intake leading to systematic bias may also arise if different assumptions about foods items not described in detail by respondents 18 (such as recipes for mixed dishes or the addition of salt or fats in cooking or at the table) are used in data coding and analysis. If differences over time or among surveys are ob- served, efforts should be made to assess whether the differences are the result of actual changes in foods consumed or in the composition of foods rather than the results of such methodological changes as those described above. For example, Perloff (1988a,b) has performed such an assessment of the changes in total fat and iron in the diets of women from 1977 to 1985. The objective of this study was to determine the average change in the intake of an individual report- ing food intake data in 1985 because of changes in food composition data between 1977 and 1985. Food items that accounted for 80 percent of the total intake of the respective nutrients (fat or iron) in 1985 were identified and matched to comparable items in the 1977 database. For each item within the 80 percent cutoff, the average change per individual per day was calculated. Results indicated a net change for total fat of —0.6 grams per individual per day, with an esti— mated change of -0.5 grams due to product changes (mainly changes in the fat content of meats) and =0.1 grams due to data changes. For iron, the esti— mated net change was +0.2 grams per individual per day, with an estimated change of +0.6 grams due to product changes (increased fortification of grain products) and an estimated change of ~0.4 grams due to data changes (corrections of erroneously high val- ues for some meats) (Moss et al., 1983; Wolf, 1987). This analysis addresses one source of variation in estimated dietary intake over time. Assessment of Nutrition-Related Health Status The assessment of nutritional status includes, in addition to the measurement of dietary intakes, mea— surement of indicators of nutrition-related health status such as hematological and biochemical tests, body measurements, clinical signs of nutritional defi— ciency, tests for diseases or conditions associated with diet, and assessments of nutrition knowledge and attitudes. All of these measures are included in the surveys of the NNMS. Available Methodologies In various surveys included in this report, data have been collected for the following indicators of nutrition-related health status: eo Hematological tests (hemoglobin, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, red blood cell counts, white blood cell counts). Biochemical tests (serum iron, total iron-binding capacity, erythrocyte protoporphyrin, serum ferri- tin; serum and red blood cell folate; serum vitamin A or retinol; serum vitamin C; serum a-toco- pherol; serum total and HDL cholesterol). Body measurements (height, weight, skinfold thicknesses, circumferences, breadths). Clinical signs of nutritional deficiency (physical evidence of wasting, edema, classical signs of vitamin deficiencies). eo Tests for conditions associated with diet (high blood pressure, overweight). e Assessments of nutrition knowledge and attitudes (for example, understanding of the relationship of diet and nutrition to disease). Important concerns for all the assessment tech- niques outlined above are that methods be standard- ized, have a predictable relation to nutritional status, and be applicable in large-scale surveys. Compounds that interfere with biochemical measurements (that is, compounds that react like the compound of interest) should be recognized. Factors other than nutrition or diet that may lead to changes in the indicators should be identified and measured. Ignoring these factors or failing to maintain quality control in the measurements may compromise the reliability of the assessments of nutritional status. Criteria for Assessment of Biochemical, Hematological, and Clinical Measurements Establishment of cutoff values A major concern in estimating the prevalence of impaired nutritional or health status from biochemi- cal, hematological, and clinical measurements is in defining values for the various indicators that are related to the occurrence of impaired nutritional status (either deficiency or excess). Such cutoffs for indicators of nutritional status may be derived in several ways; some of these have been reviewed by Habicht (1980) and Habicht, Meyers, and Brownie (1982). Cutoffs may be related to some consequence of malnutrition or functional outcome and the indicator may precede, be concurrent with, or follow the functional outcome. For example, cutoffs of 19 desirable weight—for—height have been selected based on subsequent mortality experience. Cutoffs may be based on nutritional determinants examined in stud- ies in which changes in the indicator are detected following increases or decreases in intake of the nutrient. An example would be the detection of a decrease in hemoglobin concentration in response to iron depletion. Finally, cutoffs may be derived by statistical means from the normal distribution of values for the indicator in an ostensibly healthy population. Frequently such cutoffs are set to the 95 percent confidence interval, with 2.5 percent of the population at each tail of the distribution described as having potentially abnormal high or low values. If only values that are too low are of interest, a one- sided confidence interval that has all 5 percent in one tail can be constructed. This procedure statistically defines some healthy persons as having abnormal status. Use of cutoff values As discussed by Habicht, Meyers, and Brownie (1982), no diagnostic test (consisting of an indicator and a cutoff for that indicator) can reflect the true underly- ing distribution of normal and abnormal status; some misclassification inevitably occurs. With any cutoff applied to survey data, some individuals with abnor- mal status are incorrectly classified as normal (false negatives) and some persons with normal status are misclassified as abnormal (false positives) (see figure 2-1). The proportion of those with truly abnormal Cutoff value Individuals with normal status v Ni Individuals with abnormal status False positives False negatives Number of individuals Values for indicator Figure 2-1. Effect of applying a cutoff value for an indicator of nutritional status to the distributions of values for individuals with normal status and individ- uals with abnormal status status who are classified as abnormal is a measure of sensitivity, and the proportion of those with truly normal status who are classified as normal is a mea— sure of specificity. Sensitivity and specificity tend to be more stable across populations than within popu- lations, but they are inversely related for any indica- tor. The predictive value of an abnormal test result (the proportion of abnormal results that are true abnormals when the test is applied to a population containing both healthy and diseased individuals) varies directly with the prevalence of the condition. As illustrated in table 2-1, prevalence becomes more important than sensitivity and specificity in deter- mining predictive value when the prevalence of a condition such as a nutrient deficiency is low (Galen and Gambino, 1975). Table 2-1. Effect of prevalence on predictive value when sensitivity and specificity equal 95 percent (modified from Galen and Gambino, 1975) Predictive value of Prevalence an abnormal test (percent) (percent) 0.1 1.9 1.0 16.1 2.0 27.9 5.0 50.0 50.0 95.0 In interpreting the prevalence data for impaired nutritional or health status generated by the applica— tion of a specific cutoff, an understanding of the source and meaning of the cutoff is needed. Care must be taken to assure that cutoffs derived for one population are applicable to other populations (for example, criteria for normal hemoglobin values in whites are not appropriately applied to blacks). Mea- surement error in the indicator values, diurnal varia— tion, and the possible influence of variables other than the nutrient of concern (such as infection) on the values obtained for the indicator also need to be considered. These issues are discussed, as appro-— priate, in the consideration of individual nutrients in this report, together with definitions of any cutoffs used for evaluation of indicators of nutritional and health status. Other approaches to assessment As was the case with dietary data, distributional analyses of biochemical, hematological, and clinical 20 values are often useful for comparing the relative status of various population groups without defining normal and abnormal status. Consistency in Data Sources to Assess Changes Over Time A chief concern relating to consistency in data sources over time is the methodology by which values for bio— chemical, hematological, and clinical indicators of nutritional and health status are obtained. Improved biochemical methods, new instrumentation, and changes in protocols or standardization procedures are introduced (appropriately) into surveys as they become available and feasible. These changes may, however, introduce systematic changes in the values obtained that might be interpreted erroneously as secular trends (Yetley and Johnson, 1987). Ideally, when new methodology is introduced, extensive com— parisons with the earlier methodology should be con- ducted. Other Considerations in Interpretation of Data Linking Dietary Intake Data With Nutritional Status and Health Outcome The EPONM noted the importance of obtaining dietary and health data on the same sample of individuals representative of the population to achieve greater reliability and more validity. Exten- sive data on health and nutritional status indicators in individuals have been collected in the HANES, but these surveys have included only a single 24-hour recall (which is inadequate to estimate usual intake for individuals) to obtain quantitative dietary data. (Tradeoffs such as these are necessary for de- creasing respondent burden while maximizing the amount and quality of information obtained from each survey participant.) Thus, limitations are imposed on examining the cross-sectional relation- ships of dietary intake and nutritional status in the same survey; for example, the use of regression tech- niques using individual data is constrained. In the past, methodological differences have limited the ability to link data from the HANES and the USDA surveys that collect more extensive dietary data. Increased coordination between the Agencies in establishing the core content of both sets of surveys, as recommended by the National Research Council (1984) and the JNMEC (DHHS/USDA, 1986), would result in the possibility of greater linkages between the surveys in the future. Noncoverage of Certain Population Groups The subjects in the national surveys discussed in this report were selected to be representative of the civilian noninstitutionalized population of the United States (or the 48 contiguous States) living in house- holds. Surveys of the NNMS do not consistently cover population groups such as persons without fixed addresses (migrants and homeless persons), military personnel (living on base in the United States and in total), persons living in institutions (long-term care facilities, college dormitories, and prisons), and Native Americans living on reservations. (Estimates of the sizes of some of these population groups are given in table 2-2.) The nutritional status of some of these groups may be hypothesized to differ from that of the general population. Limited assessments of some of these groups have been made in small-scale studies but have not been attempted on a national basis (Cohen, 1987). Plans for sampling some of these groups in established surveys or in new, special- purpose surveys of the NNMS are under development. Changing Characteristics of the Population In interpreting the results of surveys, especially the changes in dietary and nutritional status over time, and in the planning of future surveys, attention should be given to the changing characteristics of the U.S. population. The effectiveness of public policies directed toward consumers is likely to increase with improvements in the understanding of changes in the economic, social, and demographic environments. Knowledge of the changes in characteristics of the U.S. population will also help in resolving compara— bility issues across time periods. Key characteristics of the population include (1) the distribution of households of various sizes; (2) the age structure of the population; (3) women in the labor force; (4) the racial or ethnic composition of the population; (5) the distribution of income; and (6) other demographic shifts. Perhaps the most apparent demographic change is the aging of the U.S. population (U.S. Bureau of the Census, 1984). For example, in 1978, 16 percent of all Americans were over the age of 55 Table 2-2. Recent estimates of population groups in the United States (Woteki and Fanelli-Kuczmarski, 1989, in press) Population group Year Number (in thousands) Civilian noninstitutionalized’ 1986 235,661 Institutionalized’? 1980 2,492 College dormitories? 1980 1,994 Military (in United States)’ 1986 1,836 Homeless’ 1984 250-350 White! 1986 204,301 Black’ 1986 29,306 Hispanic? 1986 18,091 Asian® 1980 3,726 Native Americans’ 1980 1,418 I" From U.S. Bureau of the Census (1987). 2 Excludes dormitories, military quarters, and boarding houses. 3 From U.S. Department of Housing and Urban Development (1984). : From U.S. Bureau of the Census (1983). From Bureau of Indian Affairs (1988). 21 years; now, 20 percent are in this age group. As well, over half of all women over 18 years of age now work outside the home, a figure that is projected to reach about 70 percent by 1995. More single—person households are evident today than ever before. More than half of all households now have two wage earners. The growth of certain ethnic populations is notable, and the population center continues to move to the South and to the West. The population is reconfiguring into smaller and more varied household units. Increasingly, house- holds consist of childless couples, children living with only one parent, unmarried parents, lone individuals, and unmarried people living together. Currently, the average household has 2.7 members; the average household size is smaller than previously recorded and the trend toward decreased size is continuing. Nearly one-quarter of all households consist of persons living alone. "Nonfamily" households are a burgeoning segment. The following paragraphs elaborate on the major changes in these characteristics in recent years and expected changes in the near future (U.S. Bureau of the Census, 1977, 1986, 1987). Single-person households have increased dramatical— ly in the past thirty years. The percentage of single— person households has more than doubled from 10.9 percent in 1950 to 22.5 percent in 1980. The growth in the share of two-person households was much more modest, from 28.8 percent to 31.3 percent over the same period. During the past thirty years there has been a decline in the proportion of more-than- two-person households from 60.3 percent in 1950 to 46.2 percent in 1980. The number of single-person households and two-person households is expected to rise 8.0 percent and 1.7 percent, respectively, over the period 1980 to 2000. Since 1970, there has been an overall aging of the U.S. population, the median age increasing from 27.9 in 1970 to 30.6 in 1982. This trend is expected to continue in the future with the median age projected to reach 36.3 by the year 2000. In addition, the number of persons in specific age groups should undergo major changes in the coming years. From 1985 to 2000, the number of persons aged 18-24 years will decrease by about 4 million; persons aged 25-34 years will decrease by about 5 million; persons aged 35-44 years will increase by about 12 million; persons aged 45-64 years will increase by roughly 16 million; and persons aged 65 years and over will increase by about 6 million. Most notably, the number of Americans aged 65 and over has doubled in the last three decades, and by the turn of the century, the total of elderly Americans will be 22 approximately 35 million. In 1950, the Census Bureau counted 12.4 million elderly persons, a segment of the population that grew to 25.7 million in 1980. Over the past several decades, more and more wome have entered the labor force, as indicated by the steady increase in the female labor force as a percent of the female population. Female labor force partici- pation has risen in monotonic fashion from 31.4 percent in 1950 to 52.1 percent in 1982. This upward trend is expected to continue in the future. The momentum of this trend towards more dual-career families foreshadows expansion in the segment of "time-sensitive" working couples who favor conve- nience foods (Capps et al., 1985; Redman, 1980). Currently, Hispanic persons constitute roughly 7 percent of the U.S. population, but this proportion is expected to rise to nearly 10 percent by the year 2000 and almost 20 percent by the year 2080. By compari- son, blacks currently constitute about 12 percent of the U.S. population; this percentage is expected to rise to 13 percent by the turn of the century and to 18 percent by the year 2080. In short, over the period 1980 to 2000, growth of the Hispanic population is expected to be roughly 6 percent, compared with 1.8 percent for blacks and 0.6 percent for whites. The distribution of income has undergone recent shifts. Annual real income in the United States is expected to grow 2.5 percent between 1980 and 2000. Prior to the 1980s, the trend was toward the relative enhancement of lower income groups with a shift in income to improve their purchasing power. However, this trend reversed in the 1980s with a shift in the income distribution toward higher income groups. The primary impact of this shift in the income distribution on food purchases lies with the emphasis placed on the value of time and the willingness to pay for convenience, value added, quality, and variety. The shift in the income distribution and the increase in real income levels in part account for the decline in the average budget for prepared-at—home food and the rise in the average budget share for away-from- home food over the past decade (Capps, 1986). Demographic shifts can be expected in the near future. The number of households located in the Northeast and the Midwest is projected to decline by 2.7 and 2.3 percent, respectively, from 1980 to 2000, while the number of households located in the West is projected to increase by 3.6 percent. Further, the number of households located in central cities is expected to fall by almost 6 percent, the number of households located in metropolitan areas is expected to rise by almost 9 percent, and the number of households located in nonmetropolitan areas is expected to rise by almost 2 percent from 1980 to 2000. Restricted Age Groups in Some Surveys Some of the surveys considered in this report were limited in their coverage of some age groups. For example, the CSFII 1985-86 was designed to provide information on groups considered to be at relatively high nutritional risk between cycles of the more comprehensive NFCS. Thus, only women aged 19-50 years and their children aged 1-5 years were included in both 1985 and 1986; men aged 19-50 years were included in 1985. In NHANES II and HHANES, persons aged 6 months-74 years were included, but not all measurements were performed for all age groups. Sample Size Restrictions for Subgroup Analyses Examining data for subgroups with specified charac— teristics such as age, race, and sex is often desirable. However, considering several of these characteristics at once may yield sample sizes too small to permit reliable estimates for means, distributions, or prevalences for the subgroup. The criteria for adequate sample sizes for subgroups in the major surveys are described later in this chapter. Groups that may be considered to be at risk (for example, pregnant women, lactating women, and infants under 6 months of age) are frequently represented in num- bers too small for analysis in national sample surveys. Nonresponse and Adjustments for Nonresponse In the surveys considered, nonresponse may occur at various levels and give rise to concern about the representativeness of the data collected. For exam- ple, in the HANES, persons in households selected for inclusion in the survey may not be interviewed, interviewed subjects may not be examined (types of unit nonresponse), and subjects who are interviewed and examined may not complete all items (item non- response). Similar types of nonresponse occur in other surveys. In addition, definitional difficulties, differences in interpretation of questions, and coding or recoding errors may also contribute to non- sampling errors. There are several ways of attempting to deal with nonresponse. No method is perfect, and all are based 23 on some assumptions which may or may not be correct. An adjustment for unit nonresponse is often done by increasing the sample weights for respon- dents with similar characteristics (those living in the same neighborhood, those of the same demographic subgroup if known, etc.). For item nonresponse, if a relation between the missing item and other data col- lected is suspected, other types of adjustment to weighting are possible. For example, if the diastolic blood pressure value is missing, but weight, age, gender, race, and smoking history are known, then a regression equation of diastolic blood pressure on these other variables could be performed and used to predict the missing blood pressure value. This impu- tation would only be done if the regression equation was reasonably accurate. This approach should be better than simply adjusting the weights because it includes more information in the imputation of the missing observation. Another alternative is to use the sample mean for persons with similar characteristics. This procedure is comparable to weighting, but it allows the obser- vation to be retained in the analysis. The regression equation approach is generally preferred to this approach because it again typically uses more infor- mation in the imputation than in the formation of the subgroup with similar characteristics to the missing person. Still another type of imputation is used extensively by the Census Bureau. It is the replacement of the missing value by the value of another person with the same characteristics. One form of this procedure is the "cold deck" method in which one person's value is used to replace the missing values for all people with these same characteristics. The other form is the "hot deck” method in which the value from the last encountered person with the similar characteristics is used to replace the missing value. The hot deck method allows some variation in the estimation process which is missing in most of the other approaches. The regression and hot/cold deck approaches may require extensive computer time to perform. The types of imputation used in the surveys consid— ered in this report are described below in the discus— sion of nonresponse adjustment and imputation in specific surveys. Sample Weights, Variances, and Design Effects Much of the data used in this report came from complex surveys such as the HANES, NFCS, and CSFII. These surveys employed stratified, multistage designs that provided for the selection of samples at each stage with a known probability. Because of the need to obtain a sufficient number of members of certain subgroups (for example, blacks in NHANES II), individuals had different probabilities of being selected into the sample. Nonsampling and other considerations also led to surveyed individuals having different sample weights. For deriving national estimates, the use of statistical analyses that fail to account for variations in sample weights and for the complex sample design is inappropriate. Conclusions based on analyses that are conducted with sample weights and that take the design effect into account can be quite different than conclusions based on analyses conducted without these factors (NCHS, 1982). The analyses reported here were conducted using each person's unique sample weight, where the weight reflects the individual's probabilities of selection, as well as adjustments for nonresponse and poststratification. In these analyses, the effect of using a sampling design other than a simple random sample is reflected in the calculation of the variance or standard error of an estimate. The design effect is defined as the ratio of the estimate of variance that takes the design into account to the estimate of variance that assumes a simple random sample of the same sample size. A design effect greater than one is often encountered in complex surveys because of pragmatic concerns; for example, the cost of a simple random sample would be far greater then that of a complex sample which may involve many fewer sampling locations. However, the trade-off for the smaller cost of the complex sample is often a greater variance of the estimates. Statistical Criteria and Data Reporting Types of Data Presented The data analyses prepared for inclusion in this report and presented in appendix II are intended to provide descriptive information rather than to provide a basis for statistical tests of hypotheses. Statistical terms used in the tables include the following: eo The mean is a measure of central tendency of a distribution of values calculated by adding all individual values and dividing by the number of values. 24 Percentiles constitute divisions of a distribution of values into equal, ordered subgroups of hun- dredths. The 50th percentile or median is a measure of cen— tral tendency that divides a distribution of values into two equal parts, with 50 percent of the values above and 50 percent of the values below this point. The standard error is the standard deviation (measure of dispersion or variation) of a statistic (mean or percent). The standard deviation is equal to the square root of the sum of the squares of deviations divided by n-1. The coefficient of variation or the relative standard error (not shown in the tables) can be determined by taking the ratio of the standard error to the mean and multiplying by 100 percent. Criteria for Reporting and Evaluating Data The statistical guidelines outlined below were used in the presentation of estimated means, medians, and percentiles in this report. Reporting Criteria in the HANES Minimum sample size requirements for presentation of estimated means, percentiles, and variances (standard errors) for NHANES II, assuming an average design effect of 1.5, have been determined (Casady, 1982) and are shown below. This approach was also used in the initial analyses of the HHANES data. These criteria are used for the presentation of data in the tables presented in appendix II. Criteria for Means and Prevalences 1. If the sample size is less than 25, the mean or percent (prevalence estimate) is not presented. An asterisk is placed in the cell. If the sample size is 25-44, the mean or percent is presented but with an asterisk. . If the sample size is 45 or more, the estimated mean or percent is presented without caveat. Criteria for Percentiles Percentiles Sample size 5th and 95th 100 10th and 90th 5 15th and 85th 35 25th and 75th 20 50th 10 If the sample sizes do not meet these minimum values, an asterisk is placed in the cell. Analyses conducted to date with NHANES and HHANES data sets indicate that it is possible to have very large relative standard errors (greater than 50 percent) and also have large sample sizes. This situa- tion occurs frequently when estimating the preva- lence of a rare condition. When situations occur in which the relative standard errors for entries in a table exceed 30 percent, analysts should use judgment in the interpretation of the data. Even with very large standard errors, depending on the condition and the objective of the analysis, it may still be worth- while to present prevalence estimates and standard errors with strong caveats about the interpretation of the data. Reporting Criteria in the USDA Surveys The criteria for the reporting of means and medians employed by the USDA are based on the coefficient of variation rather than a minimum sample size. In the tables presented in appendix II of this report, all estimated means are displayed, but a caveat is sup— plied in the form of an asterisk for those with esti- mated coefficients of variation over 20 percent. Medians or 50th percentiles, by assumption, are pre- sented (without comment) whenever means are dis— played. However, rules were developed following Woodruff (1952) for the minimum sample sizes required to present reasonably accurate values at the various percentiles. Assuming an average design effect of 2, the minimum subgroup sample size rules are as follows: Criteria for percentiles Minimum subgroup Percentiles sample sizes 5th and 95th 140 10th and 90th 80 25th and 75th 40 25 Criteria for the EPONM's Evaluation of Data The EPONM believes that a descriptive report best serves the needs of this report's intended audience. Thus, formal statistical procedures, such as t-tests and analysis of variance, have not been used exten- sively in the comparison of dietary or health-related mean values from the many different subgroups. Instead, a difference between subgroups is considered to be large if it is approximately twice the value of its standard error. This rule, although an approximate procedure, should provide reasonable guidance to the reader. For a variable with several levels, the existence of a consistent pattern across its levels also provides support for the existence of a difference. If the existence of a large difference is not consistent with the literature, this finding would need to be replicated in another study before one concludes that it is important. However, if the literature supports the existence of a difference associated with a variable, for example, poverty level, then the existence of a large difference here is probably meaningful. More formal statistical methods such as the analysis of variance and multiple comparison procedures could be used to adjust for the number of comparisons that are implicitly being done when only one difference (usually the largest) is selected for discussion. However, the use of multiple comparison procedures often makes it difficult to find real differences because it emphasizes the type I error (falsely con- cluding there is a difference) at the expense of the type II error (wrongly concluding there is no differ- ence). The EPONM is more concerned about failing to find real differences than about falsely finding differences, and the procedure described in the above paragraph is consistent with this philosophy. Imputation in Specific Surveys Imputation for missing analytical values occurs in several data sets from surveys discussed in this report. Some values for nutrient composition are imputed in the nutrient composition database. Hepburn (1987) provided data on the percentage of analytical and imputed values for nutrients in the most recent Primary Nutrient Data Set used for CSFII 1985-86 and HHANES. No imputed values were included in the dietary intake data from NFCS 1977-78, CSFII 1985-86, or HHANES. For NHANES II, some dietary intake measurements were imputed for subjects who had unsatisfactory 24-hour recalls by randomly assigning a value from the same item of information for examined persons of the same age, sex, and race. In the same survey, imputations of missing body measurements were made by substitut— ing for the missing measurements those of an indi— vidual of the same age, sex, and race who had other dimensions similar to those available for the person with missing values. Age-Adjustment Procedures in Specific Surveys Identical age—adjustment procedures were used in the calculation of estimates from the NFCS 1977-78, CSFII 1985-86, NHANES II, and HHANES data prepared for inclusion in this report. As an illustration, in the CSFII 1985-86, the mean nutrient intake estimates reported in appendix II were adjusted to reflect the 1980 U.S. Census distribution of individuals in the 10-year age groups 20-29, 30- 39, and 40-49. Thus, both males and females were adjusted to the same standard. The adjustment was done to the means for all women aged 20-49 years and the means for each category of race, poverty status, education, region, and urbanization. The associated standard errors were also computed for these adjusted means. These 1980 Census counts are: Age 1980 Census Percent of persons (years) count aged 20-49 years 20-29 40,839,623 42.93 30-39 31,526,222 33.14 40-49 22,759,163 23.93 20-49 95,125,008 100.00 Nonresponse in Specific Surveys One of the concerns of the EPONM was whether the data from the sample surveys (NFCS 1977-78, CSFII 1985-86, NHANES II, and HHANES) could be generalized to the U.S. population. Because the surveys were household based, it was clear that they excluded certain components of the U.S. population by design, for example, people living on reservations and in institutions as well as the homeless. Although these groups represent a small proportion of the total U.S. population, their nutrient intake and health status may have been so different from the rest of the population as to have called into question the findings based on the households or individuals surveyed. An additional concern was whether the surveys adequately represented the population of U.S. households. One way the surveys may have failed to 26 have been representative of the target population was if there was substantial nonresponse. Ideally, there would be none or very little nonresponse; however, over time there has been a decrease in the response rates to national surveys. Based on the EPONM's experience, carefully designed and conducted surveys that require a substantial investment of the respon- dent's time will often have nonresponse rates ranging from 20 to 40 percent. As the nonresponse rate increases, the greater the possibility for bias in the results. It is appropriate to examine the nonresponse issue in all surveys, not just in those with substantial non- response. The examination for potential bias depends on the available information. The surveys used in this report varied in their rates of nonresponse from greater than 60 percent over the 4 waves of the 1985 CSFII basic group to 25 percent in the HHANES in the Mexican-American and Puerto Rican subgroups. Within these surveys, there were demographic subgroups with even higher rates of nonresponse. The EPONM appreciates the problems that the Agencies face in attempting to collect such vast amounts of data. For example, the USDA attempts to collect multiple days of data and the NCHS data collection involves a physical examination. These efforts impose unusually heavy demands on the survey participants, and a high nonresponse rate is understandable. However, the EPONM was still reluctant to use data with nonresponse rates as large as these, particularly those from the USDA, without an investigation of the nonrespondents. Because of the EPONM's concern as well as that of the Agencies themselves, additional analyses were performed to examine this issue. Analyses of nonresponse for NFCS 1977-78, CSFII 1985-86, NHANES II, and HHANES are discussed in detail in appendix I; some of the findings are summarized here. The analysis of the CSFII 1985 1-day and 4-day subgroups showed that demographic and socio— economic differences between these subgroups existed, but that these differences did not appear to translate into differences in nutrient intake. Although the analysis did not demonstrate any major differences in nutrient intake, caution must still be used in the interpretation of the data with such large nonresponse rates because of the potential for bias. The analysis of the HHANES showed that there were some differences between the Cuban respondents and nonrespondents, and as a result of these differences coupled with the 40 percent nonresponse rate, the EPONM has some concern about the use of the Cuban data from HHANES. The respondents and nonrespondents from the Mexican-American and Puerto Rican groups did not appear to differ on the variables examined. Variance Calculation and Design Effects in Specific Surveys The program SESUDAAN (SAS Institute, Inc., 1979) was used with the design effect option (DEFT) to obtain design effects for the HANES. For variables from NHANES II, the average design effect was calculated separately for non-Hispanic white males and white females. (The value for non-Hispanic whites was also used for non-Hispanic blacks and for each poverty subgroup.) The design effect repre— sented an average across the several age—sex specific cells for a selected variable. When a design effect was less than 1.0, the value of 1.0 was used. Then the design effects for men and women were averaged and the square root of this value was used in the following formula to obtain a new estimate of the standard error: Standard Error (complex) = ¥ Design Effect x Standard Error (simple). The calculation of standard errors for the NHANES II and HHANES data in this report was performed using an average design effects approach based on SESUDAAN calculations of variances (Kovar and Johnson, 1986). The HHANES (and some of the NHANES II) variances calculated directly by SESUDAAN were found to be unstable; therefore, the average design effects approach was developed. For comparability in the methods used in this report, the average design effects approach was applied to both the NHANES II and HHANES data. To check the reasonableness of the approach, the standard errors calculated by both methods were compared for a subset of the NHANES II data and found to be nearly identical. Design effects were calculated separately for each portion of HHANES: Mexican American, Cuban, and Puerto Rican. The approach was the same as de-— scribed above. The NCHS has indicated that wide variability exists in the design effects associated with HHANES. Because of the instabilities found in the design effects, NCHS has recommended that an aver— age design effect be calculated for classes of similar variables. This procedure was used for the HHANES data presented in this report. The USDA used the program SESUDAAN to calculate directly point estimates and their corresponding standard errors. These standard error estimates were used in computing design effects for each sex and age cell. The average design effect in the USDA survey data reported herein is roughly 2. Mean intake stan— dard error estimates from the CSFII 1985-86 have been determined to be reasonably stable for sex and age cells. However, the stability of standard error 27 estimates for lower levels of aggregation (for example, sex and age and region) is more variable and caution is urged in their use for determining confidence intervals. Use of Systéme International (SI) Units Systéme International (SI) units are a uniform system for reporting numerical values based on the actual amount of reactants in moles rather than on mass concentration units. Most journals and professional societies in the biomedical sciences have endorsed the reporting of clinical laboratory data in SI units. The biochemical data included in this report are expressed in SI units, with conversions given for the more common clinical units. References Cited Basiotis, P. P., S. O. Welsh, F. J. Cronin, et al. 1987. Number of Days of Food Intake Records Required to Estimate Individual and Group Nutrient Intakes with Defined Confidence. J. Nutr. 117:1638-1641. Beaton, G. H. 1988. Criteria of an Adequate Diet, in M.E. Shils and V.R. Young, eds., Modern Nutrition in Health and Disease, Tth edition. Philadelphia: Lea and Febiger. Beecher, G. R., and J. T. Vanderslice. 1984. Determi- nation of Nutrients in Foods: Factors That Must Be Considered, in K. K. Stewart and J. R. Whitaker, eds., Modern Methods of Food Analysis. Westport, Conn.: AVI Publishing Co., Inc. Bureau of Indian Affairs. 1988. American Indians Today. Washington: U.S. Government Printing Office. Capps, O., Jr. 1986. Changes in Domestic Demand for Food: Impacts on Southern Agriculture. South. J. Agric. Econ. 18:25-36. Capps, O., Jr., J. R. Tedford, and J. Havlicek, Jr. 1985. Household Demand for Convenience and Non- convenience Foods. Am. J. Agric. Econ. 67:862-869. Casady, R. J. 1982. Memorandum to C. Johnson. Minimum Sample Size Requirements for the Presen— tation of Estimated Means, Standard Errors, Stan- dard Deviations, and Percentiles, June 24. Cohen, J. 1987. Evaluation of DHHS Efforts to Mea— sure and Track the Nutritional Status of Low-Income Populations. Prepared for Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Macro Systems, Inc. Fanelli, M. T., and K. J. Stevenhagen. 1986. Consis— tency of Energy and Nutrient Intakes of Older Adults: 24-Hour Recall vs. 1-Day Food Record. J. Am. Diet. Assoc. 86:665-6617. Freudenheim, J. L., N. E. Johnson, and R. L. Wardrop. 1987. Misclassification of Nutrient Intake of Individuals and Groups Using One-, Two-, and Seven-day Food Records. Am. J. Epidemiol. 126:703- 713. Galen, R. S.,, and S. R. Gambino. 1975. Beyond Normality: The Predictive Value and Efficiency of Medical Diagnoses. New York: John Wiley & Sons. Habicht, J.-P. 1980. Some Characteristics of Indica- tors of Nutritional Status for Use in Screening and Surveillance. Am. J. Clin. Nutr. 33:531-535. Habicht, J.-P., L. D. Meyers, and C. Brownie. 1982. Indicators for Identifying and Counting the Improp- erly Nourished. Am. J. Clin. Nutr. 35:1241-1254. Havlicek, J., Jr., J. M. Axelson, and O. Capps, Jr. 1983. Nutritional and Economic Aspects of Conve- nience and Nonconvenience Foods. Paper presented at the 1983 Agricultural Outlook Conference, U.S. Department of Agriculture. Washington, DC. November 29-December 1. Hepburn, F. N. 1987. Food Consumption/Composition Interrelationships, in U.S. Department of Agriculture, Research on Survey Methodology: Proceedings of a Symposium Held at the 71st Annual Meeting of the Federation of American Societies for Experimental Biology. Washington, April 1. Administrative Report No. 382. Hyattsville, Md.: Human Nutrition Infor- mation Service. Johnson, S. R., H. H. Jensen, G. E. Battese, et al. 1988. Methodological Factors Affecting the Measure- ment of Dietary Status. Final Report for USDA Cooperative Agreement #58-3198-6-60. Kovar, M. G., and C. Johnson. 1986. Design Effects from the Mexican American Portion of the Hispanic Health and Nutrition Examination Survey. 1986 Pro- ceedings of the Section on Survey Research Methods of the American Statistical Association. Krantzler, N. J., H. G. Schutz, H. L. Meiselman, et al. 1982. Collaborative Project on Methods of Measuring Individual Food Intake. Report prepared for the Food and Drug Administration. 28 Life Sciences Research Office, S. A. Anderson, ed. 1986. Guidelines for Use of Dietary Intake Data. Pre- pared for the Food and Drug Administration under Contract No. FDA 223-84-2059. Bethesda, Md.: Federation of American Societies for Experimental Biology. Moss, M., J. M. Holden, K. Ono, et al. 1983. Nutrient Composition of Fresh Retail Pork. J. Food Sci. 48:1767-1771. National Center for Health Statistics, J. R. Landis, J. M. Lepkowski, S. A. Eklund, and S. A. Stehouwer. 1982. A statistical methodology for analyzing data from a complex survey: the first National Health and Nutrition Examination Survey, Vital and Health Statistics. Series 2, No. 92. DHHS Pub. No. 82-1366. Public Health Service. Washington: U.S. Government Printing Office. National Research Council, Committee on Dietary Allowances. 1980. Recommended Dietary Allowances. 9th revised edition. Washington: National Academy of Sciences. National Research Council, Committee on Food Con- sumption Patterns. 1981. Assessing Changing Food Patterns. Washington: National Academy Press. National Research Council, Coordinating Committee on Evaluation of Food Consumption Surveys. 1984. National Survey Data on Food Consumption: Uses and Recommendations. Washington: National Academy Press. National Research Council, Subcommittee on Criteria for Dietary Evaluation. 1986. Nutrient Adequacy: Assessment Using Food Consumption Surveys. Washington: National Academy Press. Pao, E. M,, S. J. Mickle, and M. C. Burk. 1985. One- Day and 3-Day Nutrient Intakes by Individuals-- Nationwide Food Consumption Survey Findings, Spring, 1977. J. Am. Diet. Assoc. 85:313-324. Perloff, B. 1988a. Assessment of Change in Total Fat Data 1977-1985. Prepared for Expert Panel on Nutrition Monitoring. Perloff, B. 1988b. Assessment of Change in Iron Data 1977-1985. Prepared for Expert Panel on Nutrition Monitoring. Redman, B. J. 1980. The Impact of Women's Time Allocation on Expenditure for Meals Away From Home and Prepared Foods. Am. J. Agric. Econ. 62:234-237. Ritenbaugh, C., G. Beaton, C.-S. Goodby, et al. 1988. Methodological Issues in Food Consumption Surveys. Final Report for USDA Cooperative Agreement #58- 3198-6-62. SAS Institute, Inc. 1979. SAS Users' Guide. 1979 Edition. Raleigh, N.C. Stewart, M. L., J. T. McDonald, A. S. Levy, et al. 1985. Vitamin/Mineral Supplement Use: A Telephone Survey of Adults in the United States. J. Am. Diet. Assoc. 85:1585-1590. U.S. Bureau of the Census. 1977. Projections of the Population of the United States: 1977-2050. Current Population Reports. Series P-25, No. 704. Washing- ton: U.S. Government Printing Office. U.S. Bureau of the Census. 1983. General Social and Economic Characteristics, United States Summary (PC80-1-C1). Washington: U.S. Government Print- ing Office. U.S. Bureau of the Census. 1986. Statistical Abstract of the United States:1987. 107th edition. Washington: U.S. Government Printing Office. U.S. Bureau of the Census. 1987. Statistical Abstract of the United States: 1988. 108th edition. Washington: U.S. Government Printing Office. U.S. Bureau of the Census, J. S. Siegel, and M. Davidson. 1984. Demographic and Socioeconomic Aspects of Aging in the United States. Current Popu- lation Reports, Special Studies. Series P-23, No. 138. Washington: U.S. Government Printing Office. U.S. Department of Agriculture. 1987. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Women 19-50 Years and Their Children 1-5 Years, 4 Days, 1985. NFCS, CSFII Report No. 85-4. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture. 1988. Nationwide Food Consumption Survey, Continuing Survey of Food 29 Intakes by Individuals, Women 19-50 Years and Their Children 1-5 Years, 4 Days, 1986. NFCS, CSFII Report No. 86-3. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1986. Nutrition Monitoring in the United States--A Progress Report from the Joint Nutrition Monitoring Evaluation Com— mittee. DHHS Pub. No. (PHS) 86-1255. Public Health Service. Washington: U.S. Government Printing Office. U.S. Department of Housing and Urban Development. 1984. A Report to the Secretary on the Homeless and Emergency Shelters. Washington: U.S. Government Printing Office. Welsh, S. O,, and R. M. Marston. 1982. Review of Trends in Food Use in the United States, 1909 to 1980. J. Am. Diet. Assoc. 81:120-125. Wolf, W. R. 1987. Inorganic Nutrient Composition of Retail Beef in the United States. J. Food Comp. Anal. 1:11-17. Woodruff, R. S. 1952. Confidence Intervals for Medians and Other Position Measures. J. Am. Stat. Assoc. 47:635-646. Woteki, C. E.,, and M. T. Fanelli-Kuczmarski. In press. The National Nutrition Monitoring System, in M. Brown, ed., Present Knowledge in Nutrition, 6th edition. Washington: International Life Sciences Institute-Nutrition Foundation. Yetley, E., and C. Johnson. 1987. Nutritional Applica- tions of the Health and Nutrition Examination Surveys (HANES). Annu. Rev. Nutr. 7:441-463. Yetley, E. A, A. Beloian, and C. J. Lewis. 1986. Dietary Methodologies for Food and Nutrition Moni- toring. Paper presented at NHANES III Dietary Survey Workshop. Chapter 3 Update of Dietary and Nutritional Status: Individual Foods and Food Components Introduction This chapter is the first of two intended to provide an update of information reviewed in the JNMEC report. It contains data, mainly from the NNMS, on dietary and nutritional status related to individual foods and food components that have been released since publication of the JNMEC report. A conceptual model, adapted from the general model shown in chapter 1, that illustrates the topics and data to be considered in this chapter is presented in figure 3-1. Components of the model relevant to the discussions in this chapter are highlighted by the shaded boxes; individual topics noted with an asterisk are those for which data are available for discussion. Potential data sources are represented by the numbers that appear above or below the boxes; numbers noted with an asterisk represent those surveys or studies from which data were obtained for consideration in this chapter. Information on the surveys that provided most of the data for the foods and food components discussed in this section is presented below (see appendix I for a detailed description of these and other surveys included): eo US. Food Supply Series (1909-85; especially 1985)-—per capita data for foods and nutrients. e Continuing Survey of Food Intakes by Individuals (CSFII 1985-86) (all income and low-income)-- dietary intake data for men, 19-50 years (1 day); women, 19-50 years (1-4 days); children, 1-5 years (1-4 days). e Hispanic Health and Nutrition Examination Survey (HHANES) (1982-84)--health and nutritional status data for Mexican Americans from the Southwest, Cubans from Dade County, Florida, and Puerto Ricans from the Metropolitan New York area, 6 months-74 years. Dietary data (a single 24-hour recall and food frequency) were also 31 collected in HHANES but were not available for inclusion in this report. e Centers for Disease Control (CDC) Pediatric and Pregnancy Nutrition Surveillance Systems (PedNSS and PNSS, respectively) (1987)--anthro- pometric and hematological data for low-income children and pregnant women in selected States. e Food and Drug Administration (FDA) Vitamin/ Mineral Supplement Intake Survey (1980)--data on use of vitamin and mineral supplements by adults. Data from these sources are used to examine trends in food availability and consumption, to assess the dietary and nutritional status of the U.S. population, and to reevaluate the public health monitoring priority status accorded to individual food compo- nents by the JNMEC. The ability to draw conclusions about the U.S. population by the EPONM in this update is limited by the selected population groups for which new data are available. Trends in Food Availability and Consumption Food Availability The availability of food in the United States is determined by production and market demand for various foods. Demand is primarily, although not exclusively, a reflection of economic factors such as real income and prices. The interrelationships of economic and other factors with components of the national food supply are depicted in figure 3-1. Since 1909-13, major changes in the availability of various foods and patterns of availability within several food groups have occurred in the United States (USDA, 1988a). The broad patterns of change in the availability of various foods over this time period are ze NATIONAL FOOD SUPPLY ————> FOOD DISTRIBUTION ——> i Environmental factors | 1 Agricultural factors CONSUMPTION ————> NUTRIENT UTILIZATION —————> HEALTH OUTCOME Away—from—home 3 Away-from—home food acquired 4 7] food available 1 25 6+«7«8 9 13 14 15 16 19... Sanitation Housing Occupation Other factors | Economic factors : 1 Policy considerations ! go -—¥---- ' Newser wine premise te momen ' + Infection : : Disease RN 12+ 5x67 8 9 1 Food production : LC 22 3 yr EEE Health/disease tp LL Incomes - Prices 1# 2% 45 54 6+ outcome | Sociocultural factorss Br bb {serum retinal, ALAN ' Pemographic factors« food consumed Nutrient gb 3} >t tocopherof, fe/ 1 1 Educational factors by individuals intakes TBC, cholesterol, rn 18+ ' Environmental foctorss ~ folate; RBC folate, Food supply» | Physiological factors 1 | Characteristics of foods ol a (hemoglobin, a cokes : 1725679 120 i Exports 1 i Imports a 3 i Storage 1 256+7+8 9 Household food 14« 15 16 19 acquired 3 \ Household food available Primary representative action or consequence Influencing or mitigating factor National Nutrition Monitoring System and other data sources: 1 = CSFIl 1985-86, 2 = NFCS 1977-78, 3 = U.S. Food Supply Series, 4 = National Nutrient Data Bank, 5 = NHANES |, 6 = NHANES Il, 7 = HHANES, 8 = NHEFS, 9 = NHIS, 10 = FLAPS, 11 = Total Diet Study, 12 = Vit/Min, 13 = Health and Diet Study, 14 = PedNSS, 15 = PNSS, 16 = BRFSS, 17 = U.S. Vital Statistics, 18 = AEDS, 19 = NHES. See appendix lil for definitions of acronyms. Shaded boxes highlight portions of the model discussed; an asterisk (+) indicates data and data sources considered in this chapter Figure 3-1. Conceptual model for the update of dietary and nutritional status in the United States (see text for explanation) illustrated in figure 3-2. Since 1970, noteworthy changes in food consumption patterns in the United States have spurred further shifts in availability of various foods. Data for food availability on a yearly basis for 1970 through 1985 are shown in figure 3-3. Since the early 1970s, the total per capita availability of animal products has increased slightly (approxi- mately 3 percent), while that of crop products has increased by 11 percent. In terms of percentage changes, increases occurred (in descending order) for lowfat milks, vegetable fats and oils, cheese, poultry, dark green and deep yellow vegetables, sugars and sweeteners, fish, fruits, legumes, nuts, soy, and grain products; decreases occurred for whole milk, animal fats, eggs, coffee, tea, and cocoa, and meat (mainly beef). Traditional demand determinants that have influenced these trends are changes in relative prices, changes in real income (income adjusted for infla- tion), shifts in the demographic structure of house- holds, and changes in tastes and preferences. As well, changes in nontraditional demand determinants, such as concern for nutrition and health, changes in life- styles, changes in technological forces, advertising techniques, and recommendations by public health officials on diet and nutrition have influenced food availability and consumption patterns. The following paragraphs describe these influencing factors and their impacts on foods available in the food supply. During the period 1970 through 1985, real prices of most major food items decreased rather substantially (Capps, 1986). The most notable declines were for food purchased for home consumption such as red meat (primarily pork), poultry, eggs, dairy products, fats and oils, and processed vegetables. On the other hand, real price increases were evident for sugar and sweets, fishery products, fresh fruit, and food away from home. Real price changes of cereal and bakery products and processed fruit were for the most part negligible. Relative price changes can make some foods less attractive and others more attractive. For example, over the last 15 years, the real price of poul- try has decreased 28 percent, while the real price of red meat has declined 13.5 percent. The per capita quantity of poultry increased roughly 14 pounds from 1970-74 to 1980-84, while the per capita quantity of red meat decreased roughly 8 pounds over the same time (figure 3-3A); one factor that may have contrib— uted to this decline is the relative price difference. Another example is the increase in the per capita quantities of vegetable oils in the food supply (figure 3-3C) during the time of the expansion of the U.S. soybean industry that brought about a dramatic increase in the supply of soybean oil at competitive prices. From 1970-74 to 1975-79, real income rose by 12.1 percent; from 1975-79 to 1980-84, real income rose 33 by 6.5 percent. The effect of income growth on food consumption patterns depends on the level of income and the "income elasticity" of specific foods. (Income elasticity refers to the percentage change in con- sumption associated with a 1 percent change in income.) As real income rose, demand increased for some foods (for example, beef, poultry, shellfish, fresh fruits, and vegetables) and decreased for others (for example, sugar, processed milk, potatoes, eggs, and cereal products) (Economic Research Service, 1981). Because the income elasticity of most foods is small, large increases in real income are necessary to gener- ate substantial increases in consumption. In addition, rising real incomes are associated with an increase in demand for convenience attributes of food products (Capps, Tedford, and Havlicek, 1985; Connor, 1981; Redman, 1980a) and with a rise in demand for away- from-home food (Kinsey, 1983). Raunikar, Huang, and Purcell (1985) noted that prior to the 1980s, the trend in real income increases was toward the relative enhancement of the lower income groups, but this trend reversed in the 1980s with a shift in income distribution toward the high income groups. The pri- mary impact of this shift in the income distribution is the emphasis placed on the value of time and willing- ness to pay for convenience, quality, and variety. The shift in income distribution and the increase in real income levels account, in part, for the decline in the average budget for all food and the rise in the average budget share for away-from—home food over the past two decades. Shifts in population demographics also influence patterns of food availability and food consumption. From 1970-74 to 1975-79 and from 1975-79 to 1980-84, the population grew 5.4 and 5.7 percent, respectively; the total quantity of food used domestically has changed in direct proportion to the increase in population. The composition of the population also plays a role in the changing demand for food and for particular types of foods. The proportions of persons in the age groups 18-44 years and 65 years and older are rising (U.S. Bureau of the Census, 1977, 1986). The shift in age distribution of the population may account, in part, for the decline in whole milk consumption (figure 3-3B) and the increases in demand for soft drinks, fruit drinks, and other beverages. Demographic shifts and changes in income distribution interact to affect food consumption and expenditures away from home (Sexauer, 1979). Finally, the distribution of households of various sizes influences the domestic food market (Sexauer and Mann, 1979). The percentage of single-person and two-person households has increased in the last 25 years, while the percentage of more—-than—two—person households has declined (U.S. Bureau of the Census, 1986). Single—person and two-person households use more A. U.S. Food Supply: Meat, Poultry, Fish, and Eggs Pounds Meat ee pn Poultry Fish ce Ym ree 180 160 + 140 + 120 100 + 80+ 60 40 + 20+ 0 1905 Whole milk Pounds 325 1915 1925 1935 1945 1955 1965 1975 1985 Year B. U.S. Food Supply: Dairy Products Other Lowfat milk Cheese milk products —— ———"; ——Xm 300 275 250+ 225+ 200} 175+ 150 125 100+ 751 50 + 251 0 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year C. U.S. Food Supply: Fats and Oils Salad, cooking, Lard and and other Margarine Shortening beef tallow edible oils re Ye cee Yl i sm Ys 0 Butter sie si X X NX 1 1935 1945 1965 1975 1985 1925 1955 Year D. U.S. Food Supply: Grain Products and Sugars Grain products ee Yn, Sugars and sweeteners — Yr Pounds 300 275 250 225 200 175 150 125 100 75 50 25 0 1905 4 In 1 1 1 1 1915 1925 1935 1945 1955 1965 L 1985 1975 Year E. U.S. Food Supply: Fruits and Vegetables Dark green, White deep yellow Other potatoes vegetables vegetables Noncitrus fruits Citrus fruits —_—— Pounds 220 200+ 180 160 + 140+ 120+ 100 + 80+ 60 + Pou 2 1 1 1 1 1 1905 1915 HE 1925 1965 1945 1955 1975 1985 Year 1935 F. U.S. Food Supply: Miscellaneous Foods Coffee, tea, and cocoa Legumes, nuts, and soy Sami imi nds 0 8k 6} 4} ak oF 8t 6 4 2+ Q Heme 2 Hm Won om Yee mK L L 1925 1935 1945 1955 1965 1975 1985 Year Figure 3-2. Per capita availability (pounds/year) of (A) meat, poultry, fish and eggs, (B) dairy products, (C) fats and oils, (D) grain products and sugars and sweeteners, (E) fruits and vegetables, and (F) legumes, nuts, and soy; coffee, tea, and cocoa; and spices in the U.S. food supply: U.S. Food Supply Series, 1909-13, 1925-29, 1935- 39, 1947-49, 1957-59, 1967-69, 1977-79, and 1985 34 A. U.S. Food Supply, 1970-85: Meat, Poultry, Fish, Eggs D. U.S. Food Supply, 1970-85: Grain Products and Sugar Grain Sugars and Meat Poultry Fish Eggs products sweeteners Pounds Pounds 180 180 160 IE I 160 140 140 120+ 120+ 100) 100 + 80 80} 60 60} - BUD fon ms ws em mm EE E50 HR em tt wi 40 BD En. i er 1 A] i 55 So—— i] S—: i S—— 20+ 0 1 1 1 A A A 1 0 i A 1 ail A A A 1970 1972 1974 1976 1978 1980 1982 1984 1986 1970 1972 1974 1976 1978 1980 1982 1984 1986 Year Year B. U.S. Food Supply, 1970-85: Dairy Products E. U.S. Food Supply, 1970-85: Fruits and Vegetables Dark green, Other Citrus ~~ Noncitrus White deep yellow Other Whole milk Lowfat milk Cheese milk products fruits fruits potatoes vegetables vegetables Pounds Pounds 240 180 220 60h co oe i me TT i EE 200 180 Wot 160 120+ mmm - 140 FY ia mmr ni 120 100 80F ~~ ____ — —_— _ _— 80 60 F 60 40} 40F mim mmm me mm 20 20F i) \ . \ . . 1 \ 0 \ \ ) \ L A . 1970 1972 1974 1976 1978 1980 1982 1984 1986 1970 1972 1974 1976 1978 1980 1982 1984 1986 Year Year C. U.S. Food Supply, 1970-85: Fats and Qils F. U.S. Food Supply, 1970-85: Miscellaneous Foods Butter and Lard and Salad, cooking, and Legumes, Coffee, tea, margarine Shortening beef tallow other edible oils nuts, and soy and cocoa Spices Pounds Pounds 26 20 24 18 22 18k 14 16 12 14} 121 10 10} 8t 8r L 6} 4F— _ idem, ie — Ar Sms sme oo 2+ -— 2+ 0 1 1 1 1 i i i 0 A T LC = 1 - 1 1 1970 1972 1974 1976 1978 1980 1982 1984 1986 1970 1972 1974 1976 1978 1980 1982 1984 1986 Year Year Figure 3-3. Annual per capita availability (pounds/year) of (A) meat, poultry, fish and eggs, (B) dairy products, (C) fats and oils, (D) grain products and sugars and sweeteners, (E) fruits and vegetables, and (F) legumes, nuts, and soy; coffee, tea, and cocoa; and spices in the U.S. food supply: U.S. Food Supply Series, 1970-85 35 convenience foods per person than do more-than- two-person households because time available for food preparation may be scarcer and there is less of a tendency for household members to specialize in food preparation (Capps, Tedford, and Havlicek, 1985). In addition to the traditional demand determinants discussed above, notable changes have been evident in nontraditional demand determinants such as concern for nutrition and health. Recent trends in the availability of various foods in the food supply may reflect this concern and indicate the possibility that more Americans are making some dietary changes consistent with the U.S. dietary guidelines (USDA/DHHS, 1985). For example, the availability of red meat (mainly beef) fell 6 percent over the period 1970-85 (figure 3—-3A). Beef is still the most popular meat, but available quantities of poultry and fishery products, which are often perceived as lower in fat, are rising. Similarly, as the quantities of lowfat and skim milk increased dramatically, probably in res- ponse to public health concerns with dietary fat, the availability of whole milk declined more than 44 percent since 1970 (figure 3-3B). Although the total quantities of fats and oils in the food supply have not declined, there has been a major shift towards a greater proportion of vegetable fats and oils and a decline in the proportion of fats of animal origin (figure 3-3C) which again may reflect efforts on the part of consumers to switch from saturated fats to unsaturated fats and oils. Other changes such as the decline in the quantities of eggs available (figure 3- 3A), and increases in the quantities of grain products, fruits, and vegetables available (figures 3-3D and 3- 3E) may also be influenced by the improved dissemi- nation of information about the links between diet and health. The dynamic nature of food demand is also attribut— able in part to changes in lifestyles of the U.S. population (Padberg and Westgren, 1983; Redman, 1980b) and to technological forces. For example, the frequency of both spouses working outside the home has resulted in less time for food preparation. With more women working outside the home, breakfast habits are changing; together with a concern for health, these changing habits may have contributed to the decline in the quantities of eggs in the food supply. New technology in household food prepara- tion, especially microwave ovens, and concomitant innovations in food processing continue to decrease the time needed for meal preparation. During the past few decades, a myriad of convenience foods, par— ticularly frozen items, ready-to-serve items, and mixes have been introduced into the marketplace. The total retail market for prepared foods reached approximately $8 billion in 1986, up 2.9 percent from the previous year. In addition, enormous growth has 36 occurred in the number of fast-food restaurants. The market for frozen potato products, cheese, tomatoes, and (more recently) chicken has increased because of increased consumption away from home. Improve- ments in processing and marketing have also boosted the popularity of certain foods. The food industry has been responsive to the desire by consumers for food ingredients that could have an influence on health, such as fiber and calcium, and has actively promoted foods that contain such ingredients or foods to which these ingredients have been added. In summary, changes in relative prices, the level and distribution of real income, population distribution and demographics, concern for health and nutrition, lifestyles, and technological forces have had impor- tant impacts on food consumption in the United States during the period 1970-85. Their varied and sometimes conflicting influences are reflected in the shifts in the quantities of food available in the U.S. food supply during that period. Food Consumption Recent trends in food consumption can be examined by comparing the consumption of men and women assessed in the CSFII 1985 with consumption assessed in the NFCS 1977-78 (USDA, 1985, 1986a). Selected results, based on data for a single day, are presented in table 3-1. The total intake of meat, poultry, and fish per person declined slightly in men and women (9 and 3 percent, respectively). Evaluating the changes in this food group is difficult because meats are categorized as meat mixtures (reported as a unit) and individual meats (reported separately). (Meat mixtures are defined as mixtures having one or more types of meat, poultry, or fish as a major ingredient, such as stews; casseroles; sandwiches, including hamburgers; and frozen dinners.) The quantities of beef reported separately declined dramatically (35 percent for men and 45 percent for women). Smaller declines were observed in the amounts of poultry and pork reported separately. The quantity of fish consumed increased for both men and women (50 and 18 percent, respec— tively). These changes are not entirely consistent with trends observed in the quantities of these foods in the food supply over the same time period. Changes in the total quantities of meats consumed may be obscured by the shift away from eating meat separately and toward eating meat as part of a mixture, especially for women. Another change in individual intakes between 1977 and 1985 occurred for milk. The intake of lowfat and LE Table 3-1. Percentage of persons using selected foods and mean intakes for men and women, aged 19-50 years, in 1 day’ in 1985 and percentage change in mean intakes from 1977 Continuing Survey of Food Intakes by Individuals 1985 and Nationwide Food Consumption Survey 1977-78 (USDA, 1985; USDA, 1986a) Men Women Mean intakes’ Mean intakes’ Individuals Change from Individuals Change from Food group or subgroup using 1985 1977 to 1985 using 1985 1977 to 1985 Percent Grams Percent Percent Grams Percent Total meat, poultry, and fish 93 268 -9 88 181 -3 Meat mixtures 40 110 +5 37 88 +35 Beef (reported separately) 28 52 -35 23 27 -45 Pork (reported separately) 25 26 ~7 20 14 -22 Poultry {reported separately) 16 25 -22 19 22 -8 Fish and shellfish (repo 11 21 +50 12 13 +18 separately) Total fluid milk 48 205 -5 51 141 -5 Whole milk 27 117 -25 26 64 -35 Lowfat or skim milk 21 87 +53 26 77 +60 Cheese 33 17 +6 34 18 +6 Eggs 28 26 -26 24 18 -28 Total vegetables 85 272 +3 83 173 -8 Total grain products 94 278 +8 94 209 +29 Grain mixtures 25 94 +31 26 74 +72 Total carbonated soft drinks 61 433 +74 54 287 +53 Regular soft drinks 48 332 +43 36 179 +28 Low-calorie soft drinks 16 101 +494 20 105 +123 Data for men were collected on a day in the summer; data for women were collected on a day in the spring. Comparisons of data collected in 1985 with data collected in 1977 should be made cautiously because of changes in data collection procedures and probing techniques which might affect conclusions about increases or decreases in the intake of certain foods. In some cases, further analysis is required to determine whether a change in intake between the two periods should be attributed to a change in the diets of individuals or to a change in methodology. See appendix I for information on differences between the two surveys. Mean values represent intakes of all persons. skim milks increased greatly (63 and 60 percent for men and women, respectively), while the intake of whole milk declined (25 and 35 percent for men and women, respectively). These changes correspond to changes in the availability of whole and lowfat milks in the food supply over the same time. Other notable changes include decreases in the intake of eggs reported separately; increases in the intake of total grain products and grain mixtures, with larger increases seen for women than for men; and a small increase in the intake of vegetables by men and a small decrease in the intake of vegetables by women. Because many meat mixtures and grain mixtures (such as pizza, spaghetti with sauce, and rice dishes) include vegetables, some of the vegetables consumed are reported elsewhere. Thus, several of the ob- served changes are in directions consistent with dietary guidelines that suggest avoiding too much fat, saturated fat, and cholesterol and eating foods with adequate starch and fiber (USDA/DHHS, 1985). In addition, there have been changes in beverage consumption. The consumption of carbonated soft drinks has increased substantially, with a greater increase occurring in the intake of low-calorie soft drinks than regular soft drinks, even though regular soft drinks are consumed by a larger proportion of both men and women. The proportion of men and women consuming alcoholic beverages and the mean intakes of alcoholic beverages both increased between 1977 and 1985 (USDA, 1985, 1986a), a finding that is contrary to the declining trend in the retail sales of alcoholic beverages over this time (Alcohol Epidemio- logical Data System, 1986). A 1985 survey question that probed for 'forgotten food items including alcoholic beverages may have contributed to the in- crease in the amounts reported (USDA, 1985, 19864). Update on Individual Food Components Food components for which data have become available since the completion of the JNMEC report are the following: Food Energy Vitamin A Iron Protein Carotenes Calcium Fat, Fatty Vitamin E Phosphorus Acids, and Thiamin Magnesium Cholesterol Riboflavin Sodium Carbohydrates Niacin Potassium and Dietary Vitamin B6 Copper Fiber Vitamin B12 Zinc Alcohol Vitamin C Folacin Dietary and nutritional status related to these components, as well as to fluoride, are considered by the EPONM in this report. For each food component discussed, a summary of its function in the body, identification of good sources in the diet, conse- quences of inadequate and excessive intake, and available indicators of status are presented in table 3-2. Foods are identified as good sources of the various nutrients and other components in the diet based primarily on information contained in Good Sources of Nutrients (USDA, 1989) as well as the JNMEC report (DHHS/USDA, 1986) and Recom- mended Dietary Allowances (National Research Council, 1980). For each nutrient with a U.S. RDA, the criterion of providing 10 percent of the U.S. RDA per serving was used to identify foods that are good sources of the nutrient. These foods do not neces- sarily represent the foods that are the major sources of the component in the diet. For example, grain products do not contain a large amount of iron per serving (less than 10 percent of the U.S. RDA) and, thus, are not considered good sources of iron; however, they are consumed in large amounts and represent a major source (41 percent) of iron in the food supply. Quantity and Quality of Data The availability of relevant update data on dietary, nutritional, and health status from the various surveys of the NNMS for each food component discussed in this chapter is summarized in table 3-3. The data elements from the NNMS common to most of the food components listed above are per capita amounts in the food supply and individual dietary intakes. The quality and quantity of data, as well as the availability of appropriate assessment criteria, differ for different components and influence the confidence with which evaluations of status may be made. Since the JNMEC report was prepared, new food composition data have become available for fatty acids, cholesterol, dietary fiber, carotenes, vitamin E, vitamin B6, vitamin B12, folacin, magnesium, potas- sium, copper, and zinc. However, composition data are still relatively less complete for dietary fiber, vitamin E, vitamin B6, vitamin B12, and magnesium than for other components. Only vitamin E and dietary fiber have analytical data available for fewer than half the major sources in the diet (Hepburn, 1987). Data on the intake of sodium are also some- what limited; estimates of amounts in the food supply are not available and the accuracy of estimates of individual intake are uncertain because of difficulties in quantifying salt used in cooking or at the table. Table 3-2. Summary description of food components assessed in the EPONM report Food Consequences of Consequences of Indicators of component Function in body Good sources in diet inadequate intake excessive intake status in NNMS Food ene Metabolic functions: - Underweight, semi- Overweight, obesity weight for w maintenance of starvation, growth height Ss temperature, go retardation in cator of energy and repair of bones children balance and tissue, and move- ment of muscles; body fat storage Protein Involved in most meta- Meat, poul Protein-energy Protein intake Serum albumin bolic processes; essen— fish, eggs, wk, malnutrition converted to levels tial for growth, develop- cl body fat ment, and maintenance of tissues; amino acids are structural elements of muscle, connective tissue, bone, enzymes, hormones, and antibodies Fat, Concentrated source of Saturated fatty acids: No biic health Relation to Serum levels of pug acids energy; carrier for the butter, lard 2 lem; clinical development of total cholesterol cholesterol fat-soluble vitamins; ficiencies of obesity, cardio- HDL-cholesterol, structural and function- Monounsaturated essential fatty disease, triglycerides al components of cell fatty acids: olive oil acids and fat- and some cancers membranes; precursors soluble nutrients of compounds involved Po! turated fatty have occurred in aspects of acids: corn and soybean metabolism oils Oholast grat liver, ey, brains, egg yolk, meat, poultry, fish, cheese Omega-3 fatty acids: fatty fish Carbohydrate Energy source; stored Simple carbo tes: Ketosis Contribution to = as i or converted fruits, vegetables, excess calorie to at milk, sugar, other intake caloric sweeteners Complex carbohydrates: oi roducts and vegetables Fiber Bulk in diet; promotes Whole-grain products, Constipation Possible decrease # normal elimination many fruits and vegeta- in mineral bles, dry beans, nuts absorption Alcohol Source of calories; Beer, wine, and other el Inadequate nutrient - depressant alcoholic beverages intake; possible over— weight; cirrhosis of liver, pancreatitis, cancer of the throat; social and public safety prob Vitamin A Formation and maintenance Preformed retinol: Changes in eyes Teratogenicity; Serum retinol (retinol) of skin, hair, and mucous liver and vitamin A- (keratomalacia) toxicity symptoms levels membranes; essential for fortified foods and skin, vision inclu ects vision, bone growth, tooth roblems on skin and bone development, and Carotene: dark green xerophthalmia) reproduction leafy vegetables, yellow and orange vegetables and fruits Vitamin E Antioxidant functions Vegetable fats and oils, Enhanced fragility ~ Serum tocopherol (tocopherol) some cereal products, of red blood cells levels nuts, some seafood Thiamin Component of enzymes in- Pork, beef and pork - - volved [pene Toei gholism, required for cell reproduction, fatty acid metabolism, and nervous system functions liver, whole-grain and enriched grain products Loss of appetite, shirk muscle tone, depression, neurologi (beriberi, 39 Table 3-2. Summary description of food components assessed in the EPONM report-—continued Food Consequences of Consequences of Indicators of component Function in body Good sources in diet inadequate intake excessive intake status in NNMS Riboflavin Component of enzymes in— Milk, some cheeses, Cracks at corner of a - volved in protein and energy meat, liver mouth; soreness and metabolism and inflammation of mouth, lips, and tongue Niacin Involved in energy Liver, peanuts, poultry, Dermatitis, diarrhea, Vascular dilation =< metabolism and synthesis red meat, fish (also depression (pellagra) of protein and fat dairy products and as sources of tryptop Vitamin B6 Functions in metabolism of Poultry, fish, bananas, Depression, confu- Possible damage - (pyridoxal) rotein; nervous system meat sion, convulsions to peripheral ction nervous system Vitamin B12 Required for formation of red Liver, red meat, fish, Pernicious anemia; - = (cobalamins) blood cells, building genetic eggs, milk neurological material, function of nervous system, and metabolism of protein and fat Vitamin C Formation of co) n; main-— Citrus fruits, other Hemorrhages in skin Gastrointestinal Serum ascorbic (ascorbic acid) tenance of capillaries, bones, fruits, tomatoes, and gums, weakness, symptoms acid levels and teeth; iron absorption; potatoes, dark green defects in bone antioxidant vegetables development (scurvy) Folacin Formation of hemoglobin Liver, dark n Pallor, weakness, =~ Serum and red (folate) and genetic materi Josip vegeta les, neurological blood cell dry , wheat germ changes, anemia folate levels Iron Carrier of ox gen in body; Liver, beef, dry Iron deficiency, Iron overload Hemoglobin red blood cell formation beans, spinach iron deficiency (persons with hematocrit, Mcv, , anemia netic pre— erythrocyte proto- isposition) porphyrin, trans- errin saturation, serum ferritin Calcium Formation and maintenance Milk, cheese, broccoli, Osteoporosis Renal calculi; - of bone and teeth; muscle spinach, turnip greens, (other factors possible soft contraction; blood clotting; canned fish also contribute tissue calci- integrity of cell membranes significantly) fication Phosphorus Structural element of bones Dairy products, meat, None of practical ~ - and teeth; gartirinaiee ina poultry, fish (protein concern variety of chemical reactions sources) Magnesium Component of bone; protein Whole-grain products, Rarely - muscle a = synthesis; release of energy some , some spasms, tremor, from glycogen; regulation dark green vegetables, nausea, apathy, of ms temperature and nuts convulsions, coma blood pressure Sodium Regulation of body fluid Salt (sodium chloride), - Edema, - volume and acid- bal- sodium-containi hypertension ance of blood; transmission additives and of nerve impulses; principal condiments extracellular cation Potassium Regulation with sodium of Red meat, milk, many = Hyperkalemia, - fluid volume and acid- fruits and vegetables, cardiac arrest base balance; principal seafood (acute intoxication) intracellular cation Copper Component of several pro— Shellfish, nuts, liver, Anemia, bone disease i Serum copper teins and enzymes; iron kidney, corn oil, levels utilization margarine, lentils Zinc Formation of protein; com- Shellfish, red meat, Growth retardation, Emesis (acute Serum zinc ponent of many enzymes; poultry, ricotta cheese, Boos appetite, mental intoxication) levels wound healing, blood forma- whole-grain cereals, et , skin tion, general growth and dry beans c , retarded maintenance of all tissues sexual development Fluoride Component of bones and Fluoridated water, Impaired dental Mottled teeth - tooth enamel certain fish, tea (also health from fluoridated tooth- pastes and rinses) 40 Tv Table 3-3. Major sources and types of data related to individual food components available from the National Nutrition Monitoring System and used to update the 1986 JNMEC report U.S. Food Nutrient Supply CSFII CDC CDC CDC FDA FDA NHIS Database Series 1985-86 HHANES PedNSS PNSS BRFSS Vit/Min Health & Diet HPDP Food (nutrient (per capita Goa (nutritional (nutritional (nutritional (knowledge, (supplement (knowledge, (knowledge, component composition) amounts) intake status) status) status) behavior) intake) behavior) behavior) >a ~ > ~ x 2 Xa ~ Food energy Protein Total fat 3 Fatty acids Casbonyare te Dietary fiber Alcohol (ethanol) Vitamin A Carotenes Vitamin E Thiamin Riboflavin Niacin Vitamin B6 Vitamin B12 Vitamin C Folacin Vitamin D Pantothenic acid Biotin Iron Calcium Phosphorus Magnesium Sodium Poiastiun pper Zinc Iodine Manganese Fluoride pa = MLS Xu pel ~ ~ i IADR | | DADA DEE DD IDA ERA DDI Ppa ~ nN 111 DADA | DADADADA | 1 1 DADA DADA DADA | 1 ADIDAS (IEE IAD DADAD DDI DADE IAD DA DADE PUEDE Re PUL EM ph pad 1 DEDEDE DADG | DADADADADAMADADADADADADADADADE I DA 1 LH Porro Pr ~ o VEL nM rrr ~ ~ EERE BALL ELE Body weight and height or stature (self-reported in CSFII 1985-86 and measured in HHANES, PedNSS, and PNSS); caloric intake measured in CSFII 1985-86 and HHANES. Serum cholesterol, HDL-cholesterol, and triglyceride levels. Saturated, monounsaturated, and polyunsaturated fatty acids. Analytical values in CSFII 1985-86 database for <75 percent of important sources of nutrient. Alcohol estimates from the daily recalls administered are likely to be underestimates. Analytical values in CSFII 1985-86 database for 89 percent of the important sources in International Units; 73 percent of the best sources as retinol equivalents. Serum retinol levels. Serum alpha-tocopherol levels. 10 Serum and red blood cell folate levels for a subsample. 11 Hemoglobin, hematocrit, mean corpuscular volume (MCV), transferrin saturation, erythrocyte protoporphyrin, and serum ferritin. 12 Hemoglobin or hematocrit. ONDA WN Blood pressure. CSFII 1985-86 = Continuing Survey of Food Intakes by Individuals 1985-86 BRFSS = Behavioral Risk Factors Surveillance System HHANES = Hispanic Health and Nutrition Examination Survey FDA = Food and Drug Administration CDC = Centers for Disease Control Vit/Min = Vitamin/Mineral Supplement Intake Survey PedNSS = Pediatric Nutrition Surveillance System NHIS = National Health Interview Si PNSS = Pregnancy Nutrition Surveillance System HPDP = Health Promotion and Disease Prevention Biochemical, hematological, anthropometric, and/or clinical assessments of nutritional status or health conditions related to food energy; fat, fatty acids, and cholesterol; vitamin A; folacin; iron; and sodium are available in recent data from the NNMS. Some bio- chemical measurements that were included in NHANES II (serum levels of albumin, vitamin C, vit- amin B12, copper, and zinc), however, are not avail- able in the more recent HHANES. Criteria for the interpretation of many of these values are uncertain. Details on the limitations in data and availability of interpretative criteria are provided in the discussions of individual components later in this chapter. Classification of Food Components by Monitoring Priority As part of the update of information on dietary and nutritional status and the classification by monitoring priority, the EPONM addressed the question of which food components examined in the NNMS might con- stitute public health issues with respect to deficiency or excess. Ideally, all food components shown to be required for the maintenance of good health or for which excesses may represent a health risk should be monitored; practically, greater monitoring efforts should be directed to those shown to represent public health problems in the population. The classification scheme developed by the EPONM, together with the equivalent classification scheme used by the JNMEC, is presented in table 3-4. The JNMEC classification and the classification used in this update report by the EPONM are similar philosophically. In the JNMEC report, nutrients and other food components were prioritized in three cate- gories to contrast those components having high and moderate priority status for continued monitoring with the third group identified as those requiring further investigation. In this report, the EPONM labeled the categories somewhat differently to place emphasis on their evaluation in regard to public health significance. The category of food components considered to be current public health issues by the EPONM can be equated to the JNMEC category of food components warranting public health monitoring priority status. The category of food components considered by the EPONM to be potential public health issues and requiring further study is most similar to the JNMEC category of Table 3-4. EPONM and JNMEC classifications for monitoring priority status of food components EPONM classification JNMEC classification e Food components were considered to be current public health issues — if dietary intakes were low or high for a sub— stantial proportion of the population, and if evi- dence from NNMS surveys of health and nutri- tional status indicated related health problems in the population or in subgroups of the population, or - if dietary intakes were low or high for a sub- stantial proportion of the population, and if evi— dence from epidemiological or clinical studies in the literature indicated related health problems in the population or in subgroups of the population. Food components in this category are recommended for high priority monitoring status; that is, multiple assessments, when possible, should continue to be employed. A high priority should be given to devel- opment of assessment tools when these are lacking. eo Food components warranting public health monitoring priority status — if evidence from health and nutrition surveys indicated related health problems in the population, and a substantial proportion of the population had 3-day dietary intakes consid- erably higher or lower than recommended levels, or — if evidence from epidemiological and controlled clinical studies indicated related health problems in the population, and a substantial proportion of the population had 3-day dietary intakes considerably higher or lower than recommended levels. Table 3-4. EPONM and JNMEC classifications for monitoring priority status of food components——continued EPONM classification JNMEC classification oe Food components were considered to be potential public health issues, for which fur- ther study is needed, - if dietary intakes were low or high for a sub- stantial proportion of the population, and if lim- ited evidence from either NNMS nutrition and health surveys or studies in the literature sug- gested related health problems in at least some subgroups in the population, or if dietary intakes were adequate for the majority of the population, but limited evidence from either NNMS nutrition and health surveys or studies in the literature suggested related health problems in at least some subgroups in the pop- ulation, or if dietary intakes were low or high for a sub- stantial proportion of the population, and if evi- dence was not available from either NNMS nutrition and health surveys or studies in the literature that permitted evaluation of the public health significance of observed dietary intakes. Food components in this category are recommended for moderate monitoring priority status, with contin- ued assessment at the least in subgroups suspected to be at risk, and moderate priority for the development of improved assessment techniques. eo Food components were not considered to be current public health issues — if dietary intakes were adequate for the majority of the population, and evidence from either NNMS nutrition and health surveys or studies in the literature did not suggest related health problems in the population, or — if dietary intakes were low or high for a sub- stantial proportion of the population, but evi- dence from either NNMS nutrition and health surveys or studies in the literature did not sug- gest related health problems in the population. Food components in this category are recommended for lower monitoring priority status; continued assessment should include, at a minimum, estimation of dietary intake. eo Food components requiring further investi- gation - if information from dietary and health surveys was insufficient to permit judgment about public health significance, or - if intakes deviated from recommended levels for many in the population, but related health prob- lems could not be found or methods for identi- fying health problems were not available, or - if, despite theoretical reasons for believing that the food component might have public health significance, intakes were in an acceptable range and related health problems could not be identified. oe Food components warranting continued public health monitoring consideration - if no currently available evidence from health and nutrition examination surveys indicated related health problems in the population, and most of the population had 3-day dietary intakes that met recommended levels, or - if potential health problems related to inadequate intakes were ruled out at the time. (The JNMEC assigned some components to more monitoring consideration and some to less monitoring consideration.) 43 components requiring further investigation. The type of additional study required for different components differs; basic research on the health consequences of high or low intake, additional data on food com- position and dietary intake, and/or the development of methods for assessing status together with interpretative criteria may be needed. The EPONM category of food components that are not considered current public health issues is most similar to the JNMEC category of components warranting continued monitoring consideration. Assigning food components to this category does not necessarily indicate that there are no known health problems associated with these components, but that the prevalence of such problems on a national basis is known or expected to be so low that a lower level of monitoring effort than for food components in the other categories is appropriate. All of the categorizations by the EPONM in this report should be considered provisional. As new data from the NNMS and other sources become available, changes are likely to occur in the assessment of public health significance and level of monitoring appropriate. A schematic diagram that illustrates the decision— making process used by the EPONM for categorizing food components is shown in figure 3-4. This process differs from the one used by the JNMEC in that the evaluation of each food component begins with the dietary intake data. This choice to begin the evalua- tion of each food component with consideration of the dietary intake data was made recognizing that such data are available for most of the components included; the same is not true of related health data. NUTRIENT INTAKE Neither Low nor High Lo 3 Limited evidence of Little or no evidence of odverse health conditions Evidence of adverse health conditions CURRENT PUBLIC HEALTH ISSUE adverse health conditions | POTENTIAL PUBLIC HEALTH ISSUE Vv NO CURRENT PUBLIC HEALTH ISSUE Figure 3-4. Decisionmaking process used by the EPONM in categorizing food components by moni- toring priority status (Dotted lines indicate less likely outcomes.) 44 However, as illustrated in figure 3-4, the results of both processes are similar in that the evidence for adverse health consequences ultimately determines the categorization of food components. Criteria used by the EPONM in assessing dietary intake and health consequences are described in the following section. Approach to Assessing Which Food Components Represent Public Health Issues in National Nutrition Monitoring The variety of NNMS data available for the different components suggests that no single approach to assessment can be recommended for all components. Each component must be considered independently in terms of the data available, primarily from the NNMS, but also from the general biomedical litera—- ture in some cases. When data from the literature are used, reports such as The Surgeon General's Report on Nutrition and Health (DHHS, 1988) or Diet and Health: Implications for Reducing Chronic Disease Risk (National Research Council, 1989) that represent a consensus of scientific opinion constitute the pri- mary sources of information. For assessing the potential public health issues with respect to inadequacy of various food components, the RDA (or safe and adequate intakes) may be helpful as a first step in examining dietary intakes of food components. For the reasons indicated in chapter 2, it is inadvisable to use specified levels of the RDA, or numbers or percentages of persons with intakes above or below the RDA, as a sole criterion for identifying a public health issue with respect to dietary deficiency. The RDA (except for energy) are estimated to exceed the requirements of most healthy individuals. As noted by the Committee on Dietary Allowances of the Food and Nutrition Board (National Research Council, 1980), "In assessing dietary surveys of popu- lations, if the amounts of nutrients consumed fall below the RDA for a particular age-sex group, some individuals can be assumed to be at nutritional risk. When the proportion of individuals with such low intakes is extensive, the risk of deficiency in the population is increased." The approach used by the EPONM to assign food components to categories in its classification scheme was essentially similar to that used by the JNMEC. To de-emphasize reliance on the RDA as a standard for dietary adequacy, the EPONM has not tabulated dietary intake data in this report as a percent of the RDA. However, in the evaluations of individual food components that follow, the RDA (or safe and adequate intake), if available, was used as a rough initial screen for possible deficiency. Initially, a determination was made to ascertain whether the mean intake (based mainly on multiple days of data) of the population groups considered was above or below the RDA and whether there were substantial numbers of very low intakes in the population. Then: e If the mean intake of a specified component ex- ceeded its RDA, then very few individuals would be likely to be at risk of deficiency in that population. — If the biochemical and clinical evidence available from the NNMS (or other sources) confirmed the absence of a nutritional deficiency problem, the specified component was not considered to be a current public health issue with respect to deficiency in the population surveyed. On the other hand, if additional evidence from the NNMS or other sources indicated the potential of deficiency in at least some groups in the population, the food component was classified as one considered to be a potential public health issue, for which further study is required to determine the nature and extent of the potential problem. e Alternatively, if the mean intake of a component fell substantially below the RDA or if a substantial number of persons in the population or particular subgroups had very low intakes, a possible problem of undernutrition may be indicated. - However, additional evidence of a nutritional problem based on clinical or biochemical data from the NNMS was needed to determine that the component should be considered a current public health issue. Such clinical and biochemical data were accorded more weight than the dietary intake data. If such data were not available from the NNMS, information in the general medical literature was used to assess the possibility that the specified component should be considered a current public health issue. If intakes were low, but no data from the NNMS or elsewhere indicated a problem, the food component was not considered to be a current public health issue. If intakes were low, and no data on health or nutritional status were available from the NNMS or any other source to assess the potential for deficiency, then the component was considered to be a potential public health issue, for which further study is required. 45 The EPONM used a similar approach for assessing high intakes of food components; however, the RDA is less useful in identifying potential public health problems with respect to excess intake or overnutri- tion. If the mean intake was near or above the RDA and the distribution of intakes was skewed to the side of high intakes, the possibility of deleterious health effects was evaluated. Other standards may be more applicable to components such as fat, saturated fat, cholesterol, and sodium for which excessive intake is presumed to be harmful. Evidence for possible exces— sive consumption from supplements was also consid— ered. The same types of confirmatory evidence from the NNMS or from other sources about deleterious effects were required for the categorization of high intakes as for low intakes. Evidence for health consequences is provided mainly by the biochemical and clinical data from the NNMS. The EPONM used the same criteria as the JNMEC to evaluate these data. As noted by the JNMEC (DHHS/ USDA, 1986), "Much can be inferred about the nutritional status of the population, even with imper— fect data judged by imperfect criteria, especially when a wider knowledge of nutrition is brought to bear." Thus, in addition to the criteria above, the members of the EPONM have applied their experience and judgment in categorizing food components. The classification of food components into the cate— gories by the EPONM is presented in table 3-5 and comparisons with the JNMEC categorizations are presented in table 3-6. The individual evaluations of each food component that justify the classifications follow. Discussion of Individual Food Components In the following sections on the individual food components, discussions are organized around the question, "What is the evidence from the current data (mainly from the NNMS) that a public health issue exists with respect to this component?” Comparisons of current data and earlier data are made as appro- priate. Greatest weight is given to biochemical and clinical evidence. For the discussions of each food component, the following are included: e Reasons for concern about inadequacy or excess of the component, including the JNMEC classi- fication. e Information on trends in the amount and food sources of the food component in the U.S. food supply, with emphasis on current status. Table 3-5. EPONM classification of food components by monitoring priority’ Current Potential public Not current public health health issue; public health issue further study needed issue Food energy Dietary fiber Protein Fat Vitamin A Carbohydrates Saturated fat Carotenes Vitamin E Cholesterol Folacin Thiamin Alcohol Vitamin B6 Riboflavin Iron Vitamin C Niacin Calcium Potassium Vitamin B12 Sodium Zinc Magnesium Fluoride Copper Phosphorus I a graded priority should be accorded to those food components judged to be current or potential public health issues. e Information on current individual dietary intakes and changes observed in recent surveys. e Information on biochemical/clinical assessments, if any, related to the food component. e Information on the extent of supplement use, if any. : e Evaluation of public health significance of findings based on the approach described above. e Conclusion on monitoring priority status based on public health significance. The data tables and graphs that support the discussions of the individual food components are included, by topic, in appendix II (these tables and graphs are numbered II-1, II-2, II-3, and so on). For some components identified as "current public health issues," additional analyses of the influence of sex, age, ethnic group, and socioeconomic status are shown in the text. The data (if available) presented in appendix II for each food component include the following: 46 Graphs of per capita amounts provided by the U.S. food supply (U.S. Food Supply Series, 1909-85). Graphs of food sources in the U.S. food supply (U.S. Food Supply Series, 1985). Tables of dietary intake from food only (4 days) for women aged 20-49 years and their children aged 1-5 years (CSFII 1985-86). Tables of mean dietary intake from food only (1 day) for both sexes and all ages (NHANES I, 1971-74; NFCS 1977-78; NHANES II, 1976-80; and CSFII 1985-86). Tables of biochemical and clinical data for Mexican Americans, Cubans, and Puerto Ricans (HHANES, 1982-84) and comparable data for non-Hispanic whites and blacks (NHANES II, 1976-80). Tables reporting vitamin and mineral supplement use (FDA Vitamin/Mineral Supplement Intake Survey, 1980) (see tables II-132 and 11-133). Table 3-6. Comparison of JNMEC and EPONM conclusions on classification of food components assessed in the NNMS Food component JNMEC classification EPONM classification Comments Food energy Public health monitoring Current public health issue EPONM and JNMEC in agreement priority status Protein Public health monitoring Not current public health issue EPONM and JNMEC essentially consideration (more) in agreement Fat Public health monitoring Current public health issue EPONM and JNMEC in agreement priority status Saturated fat Public health monitoring Current public health issue EPONM and JNMEC in agreement priority status Cholesterol Public health monitoring Current public health issue EPONM and JNMEC in agreement priority status Carbohydrates Public health monitoring Not current public health issue EPONM and JNMEC in agreement consideration (less) Die fiber Further study required Potential public health issue EPONM and JNMEC in agreement ay (further th needed) Alcohol Public health monitoring Current public health issue EPONM and JNMEC in agreement priority status Added caloric Further study required Not evaluated No comparison possible sweeteners Vitamin A Public health monitoring Potential public health issue EPONM gave more weight to low consideration (more) (further study needed) serum levels in some groups Carotenes Not evaluated Potential public health issue No comparison possible (further study needed) Vitamin E Not evaluated Not current public health issue No comparison possible Thiamin Public health monitoring Not current public health issue EPONM and JNMEC essentially consideration (more) in agreement Riboflavin Public health monitoring Not current public health issue EPONM and JNMEC essentially consideration (more) in agreement Niacin Public health monitoring Not current public health issue EPONM and JNMEC essentially consideration (more) in agreement Vitamin B6 Further study required Potential public health issue EPONM and JNMEC in agreement (further study needed) Vitamin B12 Public health monitoring Not current public health issue EPONM and JNMEC in agreement consideration (less) Vitamin C Public health monitoring Potential public health issue EPONM noted improved intake; priority status (further study needed) little evidence of health effects Folacin Further stu uired Potential public health issue EPONM and JNMEC in agreement dy req (further oy needed) Iron Public health monitoring Current public health issue EPONM and JNMEC in agreement priority status Calcium Public health monitoring Current public health issue EPONM and JNMEC in agreement priority status Phosphorus Public health monitoring Not current public health issue EPONM and JNMEC in agreement consideration (less) Magnesium Further study required Not current public health issue Little evidence of health effects Sodium Public health monitoring Current public health issue EPONM and JNMEC in agreement priority status Potassium Not evaluated Potential public health issue No comparison possible (further study needed) Copper Not evaluated Not current public health issue No comparison possible Zinc Further study required Potential public health issue EPONM and JNMEC in agreement (further study needed) Fluoride Public health monitoring Potential public health issue EPONM concluded data too limited priority status (further study needed) to assess extent of possible problem 47 Food Energy The diet must provide sufficient energy to meet the body's requirements for growth and development, metabolic functions, muscle activity, and repair of damage caused by injury or illness (DHHS, 1988). Food energy is provided by fat, carbohydrate, protein, and alcohol. Food energy is measured in kilocalories, commonly referred to as calories. The pure forms of fat, carbohydrate, protein, and alcohol provide approximately 9, 4, 4, and 7 kilocalories per gram, respectively (Merrill and Watt, 1973). When sufficient food is available, healthy individuals normally consume enough food to maintain body weight and support growth. Chronically low intakes of energy in relation to requirements lead to under— weight and, in extreme cases, semistarvation. In children, low energy intake may also lead to growth retardation. Most, but not all, persons in the United States have access to sufficient food, but insufficient consumption occurs among some individuals because of insufficient resources to purchase adequate food and as a clinical problem among patients with a vari— ety of physical and mental diseases (DHHS/USDA, 1986). Over time, consumption of energy in excess of needs leads to overweight and obesity. Body weight depends on complex physiological con— trols of the balance between energy intake and energy expenditure; both factors are equally important in regulating body weight (DHHS, 1988). The JNMEC accorded food energy a high priority for public health monitoring status because of evidence that energy imbalance was reflected in the high prevalence of overweight in the United States. Body mass index (BMI), defined as weight (kg)/height (m)?, was con— sidered to reflect long-term energy balance; the cri- teria for overweight and severe overweight were set, respectively, at the 85th and 95th percentiles of BMI for men and women aged 20-29 years. (Overweight, defined in this fashion, was considered an approxi- mation of excess body fatness or obesity.) The cri— teria used by the JNMEC for assessing overweight were also used in this report. These cutoffs represent a statistical approach to defining overweight and are not based on morbidity or mortality experience; if the health consequences of overweight were considered, BMI standards might differ with age. Per capita energy content of the food supply (3500 kilocalories in 1985) is substantially higher than the intakes recorded in surveys of individuals. There have been fluctuations in the energy content of the food supply over time, but no clear trends (figure II-1). The most notable changes in the food sources of energy from 1909 to 1985 have been an increase in the percentage of energy contributed by fats and oils 48 (especially shortening and salad, cooking, and other edible oils), an increase in the percentage contributed by sugars and sweeteners, and a decrease in the per— centage contributed by grain products. In 1985, the major sources of food energy in the food supply were fats and oils (20 percent); grain products (19.9 per- cent); meat, poultry, and fish (19 percent); and sugars and sweeteners (17.8 percent) (figure 11-2). Concern about low energy intakes in childhood arises from the possibility of growth retardation, based on evidence of short stature among children of low socioeconomic status (see chapter 4). However, the CSFII 1985-86 data indicate that children aged 1-5 years have mean intakes of energy that fall within the range of Recommended Energy Intakes (table 11-2). Intakes did not differ greatly for blacks and whites or for children above and below poverty. Data from CSFII 1985-86 suggest that a different situation exists for women (table II-1). Their reported mean intake falls below the recommended range, and intake is lower in older (40-49 years) than in younger (20-29 years) women, lower in blacks than in whites, and lower for women below poverty than those above poverty. Relationships of energy intake to the range of household income and to household composition are shown in tables 3-7 and 3-8. Intakes are observed to rise slightly with increasing income but plateau at higher income levels, and are substantially lower in women with children who lived in households without a male head than in women living in any other type of household. Mean (1-day) individual energy intakes measured in surveys con- ducted in the 1970s and 1980s show some fluctuations, but no clear trends, over time for most age and sex groups. Data on overweight for persons aged 20-74 years in the HHANES (table II-4) permit comparison of the three ethnic groups in that survey to non-Hispanic blacks and whites in NHANES II (table II-7) (see also figure 4-2 in chapter 4). In all three Hispanic groups, the age-adjusted prevalence of overweight was higher in females than in males (table 3-9). Mexican- American and Cuban males had a slightly higher prevalence of overweight (31 and 28 percent, respec- tively) than did Puerto Ricans (26 percent); the prev— alence in the latter group was similar to that in non- Hispanic whites and blacks (table 3-9). The preva- lence of overweight in Mexican-American and Puerto Rican females (42 and 40 percent, respectively) was higher than in Cubans (32 percent) or non-Hispanic whites (24 percent) and similar to the prevalence in non-Hispanic blacks (44 percent) (table 3-9). The prevalence of overweight in Mexican-American men was not affected by poverty status, but the prevalence was greater in Mexican-American women below Table 3-7. Mean intake of food energy in kilocalories for women aged 20-49 years, 4 nonconsecutive days, by household income (percent of poverty): Continuing Survey of Food Intakes by Individuals 1985-86 Table 3-8. Mean intake of food energy in kilocalories for women aged 20-49 years and children aged 1-5 years, 4 nonconsecutive days, by household composi- tion categories: Continuing Survey of Food Intakes by Individuals 1985-86 Intake Intake Household income Standard error Household Standard error (percent of poverty) n Mean of the mean composition n Mean of the mean All women 2,056 1,617 16 Women, 20-49 years Quintiles Male head and First 369 1,418 42 — No children 412 1,526 28 (0-113) - Child(ren) 1,151 1,546 19 Second 378 1,495 37 No male head and (114-207) — No children, no Third 405 1,543 37 other adult 124 1,532 47 (208-305) — No children, other adult 91 1,593 80 Fourth 370 1,588 28 — Child(ren) 278 1,355 47 (306-426) Children, 1-5 years Fifth 368 1,562 29 (427+) Male head 563 1,421 24 No male head 84 1,453 45 Table 3-9. Age-adjusted percent of overweight and severely overweight persons aged 20-74 years, by sex and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80’ Percent overweight Percent severely overweight Ethnic group or race Male Female Male Female Mexican American 30.9 41.6 10.8 16.9 Cuban 27.6 31.6 10.7 6.6 Puerto Rican 25.6 40.2 8.0 15.7 Non-Hispanic white 24.2 23.9 7.7 9.4 Non-Hispanic black 26.0 444 10.0 19.8 I See text for definition of overweight and severely overweight. 49 poverty than in those above poverty (tables 3-10 and II-5). The same relationship was seen for non-His— panic women below and above poverty; the preva- lence of overweight was somewhat lower in non- Hispanic men below poverty than those above poverty (tables 3-10 and II-7). Although there are some exceptions, a general trend for higher prevalence of overweight with increasing age until the oldest age groups (60-69 and 70-74 years) exists (tables 11-4 and II-7). Table 3-10. Age-adjusted percent of overweight Mexican-American and non-Hispanic persons aged 20-74 years, by sex and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examina- tion Survey, 1976-80" Sex and Percent overweight ethnic group Below poverty Above poverty Males Mexican American 31.8 31.1 Non-Hispanic 21.5 24.6 Females Mexican American 46.1 40.1 Non-Hispanic 38.5 24.6 I See text for definition of overweight. The low calorie intakes reported may be considered paradoxical in view of the high prevalence of over- weight in this country, especially in women of low socioeconomic status. The JNMEC report (DHHS/ USDA, 1986) concluded that, if reported diets repre— sent usual energy intake and such a large proportion of the population is overweight, many Americans are underactive. Methodological constraints in assessing the role of eating practices in the genesis of over— weight and obesity include possible underreporting of food intake and possible higher energy intakes at some time in the past, but other factors may also con— tribute to the development of obesity. These include heredity, decreased physical activity without an appropriate reduction in energy intake, and other reasons such as altered metabolism of adipose tissue, defective or decreased heat production in the body (thermogenesis), and the use of certain drugs (DHHS, 1988). Further research is needed to evaluate the relative contributions of these factors to the occur— rence of obesity and overweight; data available from the NNMS do not permit assessment of their impacts. However, like the JNMEC, the EPONM considers it probable that low physical activity has a significant 50 relationship to the high prevalence of overweight ob- served in the United States. The U.S. dietary guide- lines (USDA/DHHS, 1985) recommend achieving and maintaining desirable weight by an appropriate bal- ance of energy intake and expenditure. Recommen- dations indicate that energy intake can be reduced by limiting consumption of foods relatively high in cal- ories, fats, and sugars and minimizing alcohol con- sumption; energy expenditure can be increased through regular and sustained physical activity. In the opinion of the EPONM, the current low levels of reported intake suggest that further reductions in energy content of diets, without a concomitant change in dietary pattern toward greater nutrient density, may compromise overall nutritional status, especially in women. These observations lead the EPONM to conclude that greater emphasis should be given by the scientific community to the role of phys- ical activity in the etiology and treatment of obesity. Conclusion: Food energy is considered a current public health issue. Reported dietary intakes of adults are lower than Recommended Energy In- takes, but the data available from the NNMS on the high prevalence of overweight (approximately one-fourth of adults) in many groups in the United States suggest a continuing public health problem in regard to energy balance. Food energy should be accorded high priority for monitoring status. Additional information on both energy intake and energy expenditure (physical activity) is required to evaluate the relative impact of these factors on the occurrence of obesity. Protein Protein is required for growth, development, and maintenance of body tissues and is involved in almost all metabolic processes. Food protein supplies the amino acids needed to form the body proteins that comprise muscle, connective tissue, bone matrix, and other body components, as well as enzymes, anti- bodies, and hormones. Protein—energy malnutrition is a serious public health problem in some countries but occurs rarely in the United States, except in association with child neglect, food faddism, or serious underlying disease (DHHS/USDA, 1986). Dietary protein in excess of need may be converted to body fat. High intakes of protein have been hypoth- esized to contribute to the decline in kidney function with age, but there is no evidence that current in- takes by the U.S. population adversely affect the prevalence of renal disease (DHHS, 1988). Although associations between high intakes of animal protein and the incidence of coronary heart disease and some cancers have been suggested, these associations are likely confounded by the high correlation between animal protein and fat intake (National Research Council, 1989). The JNMEC report classified protein as a component warranting continued public health monitoring con- sideration, because intakes exceeded recommended levels and there was no evidence from health and nutrition examination surveys of related health prob- lems. The JNMEC evaluation, as well as information from the literature, suggested that there has been little concern about inadequacy or excess in regard to protein intake in any population subgroup. A review of the per capita protein content in the U.S. food supply shows that the 1985 level of 104 grams per day is similar to the level available during most of the century (figure II-3). In 1985, the major sources of protein in the food supply were meat, poultry, and fish (43.4 percent); dairy products (20.6 percent); and grain products (19 percent) (figure II-4). The sources of dietary protein have changed over time, with in— creases since 1909-13 in the proportion provided by meat, poultry, and fish and dairy products and a large decrease in the proportion provided by grain prod- ucts. Over the last 20 years, the percentage of dietary protein provided by meat has declined, with a con- comitant increase in the percentage supplied by poul- try and fish. During the same 20 years, the contribu- tion of total dairy products has remained the same, but the proportion of protein supplied by particular dairy products has shifted, with decreased percent- ages for whole milk and increased percentages for lowfat milk and cheese. Based on the CSFII 1985-86 data, mean intakes of dietary protein in all groups among women aged 20- 49 years, and among children aged 1-5 years, were well above the RDA. Individuals above poverty and those with higher levels of education had greater intakes of protein. However, the differences in intake were relatively small; even among those individuals below poverty and those with less than a high school education mean protein intakes were 20-30 percent above the RDA. There did not appear to be any large differences in protein intake by age, race, region, or degree of urbanization. These data suggest that the percentage of individuals with a low protein intake that would put them at risk for having adverse health conditions probably continues to be very low. Data from surveys conducted during the 1970s and 1980s indicate little change in protein intake (1-day means) over this time (table II-14). The data also indicate that mean intakes are lower in the elderly than in younger adults, but still exceed the RDA. Although biochemical assessments of protein nutri— tional status have not been included in NNMS 51 surveys completed since the publication of the JNMEC report, earlier results from NHANES I and II showed little or no evidence of protein deficiency based on serum albumin levels. Protein status may be a greater concern in the elderly, especially the institutionalized elderly, but data are not available from the NNMS to address this concern. Conclusion: Protein is not considered to be a current public health issue. Intakes appear to be adequate for almost all persons and there is no evidence of health problems associated with deficiency or excess. Monitoring should continue at a low level, especially for the elderly. Fat, Fatty Acids, and Cholesterol Excess consumption of dietary fats has been associated with heart disease, certain cancers, obesity, and gall bladder disease. The JNMEC classified total fat, saturated fatty acids, and cholesterol as com- ponents warranting public health monitoring priority on the basis of high dietary consumption. Other than the general recommendation that 3 percent of energy be obtained from essential fatty acids, there are no RDAs for fat or cholesterol. When calorie intake is adequate, public health problems have not been as- cribed to inadequate intakes of fat or cholesterol. Excessive intake is, however, a public health concern; the dietary guidelines for Americans (USDA/DHHS, 1985) recommend avoiding too much fat, saturated fat, and cholesterol. The Surgeon General's Report on Nutrition and Health (DHHS, 1988) concurs that reducing the consumption of fat (especially saturated fat) and cholesterol is an issue for most people. The Adult Treatment Panel of the National Cholesterol Education Program has recommended a reduction in the intake of total fat to less than 30 percent of calories, saturated fatty acids to less than 10 percent of calories, and cholesterol to less than 300 milli- grams per day for adults with high serum cholesterol levels and other risk factors for cardiovascular disease (NHLBI, 1987; NIH, 1987). Similar quantitative reductions have been recommended for the general population (for adults or for persons over 2 years of age) by other authoritative groups (American Cancer Society, 1984; American Heart Association, 1986; NIH Consensus Development Panel, 1985; National Research Council, 1989). The per capita amount of fat in the U.S. food supply has increased between 1909 and 1985, equaling 169 grams per day in 1985 (figure II-5). (These food supply estimates include fat that is lost or discarded as waste and, thus, are much higher than estimates of the fat intake of individuals.) The distribution of types of fat in the food supply has shown change over the last several decades. The per capita amount of saturated fat has remained nearly constant (near 60 grams per day) (figure II-7), while the amount of monounsaturated fat has gradually increased since 1909 and equaled 68 grams per day in 1985 (figure 11-9). The largest change has been in the amount of polyunsaturated fat which has more than doubled since 1909, with the greatest increase occurring since the mid-1960s (figure 11-11). In 1985, per capita polyunsaturated fatty acids equaled 33 grams per day. Cholesterol in the U.S. food supply has decreased slightly since reaching a high in 1945, equaling 500 milligrams per capita per day in 1985 (figure 11-13). Food groups contributing fat in the food supply have shifted. The proportion of total fats from meat, poultry, and fish has changed only slightly, equaling 31.4 percent in 1985 (figure 11-6). The proportion of fat from whole milk has declined steadily from a high of 10.4 percent in 1947-49 to 3 percent in 1985, while the proportion from fats and oils has increased from 38 percent to 47 percent in the same period. Other food groups make little contribution. For saturated fat, although there have been only small changes between 1909 and 1985 in the overall proportions provided by meat, fish, and poultry and dairy prod- ucts, there has been a shift within the dairy group with the proportion from whole milk decreasing and that from cheese increasing. Within the fats and oils groups, the proportion of saturated fat from animal sources has decreased while the proportion from veg- etable sources has increased. In 1985, meat, poultry, and fish; fats and oils; and dairy products contributed almost all of the saturated fat in the food supply (figure 11-8). The proportion of polyunsaturated fatty acids from meat, poultry, and fish has decreased, but the proportion from the fats and oils group has more than doubled over the last 76 years. In 1985, 68 percent of polyunsaturated fatty acids were contrib— uted by this food group (figure II-12). The three major food groups supplying cholesterol in 1985 were meat, poultry, and fish (43 percent); dairy products (13 percent); and eggs (39 percent) (figure 11-14). Based on 4-day dietary intake data from the CSFII 1985-86, estimates of the fat intakes of women aged 20-49 years and children aged 1-5 years were 37 and 35 percent of calories, respectively (tables II-17 and 11-19). Only approximately 10 percent of women had fat intakes below 30 percent of calories. For women in the CSFII 1985-86, total fat intakes were higher for whites than for blacks and higher for those of higher socioeconomic status than for those of lower socioeconomic status; however, race and socioeco- nomic factors had little effect on percent of calories from fat (tables II-16 and II-17). Saturated fatty acids comprised an estimated 13 and 14 percent of calories in the diets of women and children, respec— tively (tables II-23 and 11-25). Monounsaturated fats provided 13 percent of calories for both women and children (tables I-28 and II-30). Polyunsaturated fats provided 6 percent of calories for children and 7 percent for women (tables II-33 and II-35). Mean cholesterol intakes were 277 and 228 milligrams per day for women and children, respectively (tables 11-37 and II-38). More than 25 percent of women had mean cholesterol intakes in excess of 300 milli- grams per day (table II-37). The mean intake of dietary cholesterol of women did not vary greatly “with age, poverty status, or education. 52 One-day data (table 11-20) from the CSFII 1985-86 show that mean intakes of total fat for children and adult males are similar to the NFCS 1977-78 means, whereas for females 20-49 years old, total fat intake appears to have decreased about 10 percent between these surveys. The intakes for women from CSFII 1985-86 are similar to those from NHANES I and II. Whether the differences in intake seen in the differ- ent surveys truly reflect changes in consumption over time, or result from differences in survey methods, is uncertain. The analysis by Perloff (1988), described in chapter 2, evaluated one methodological difference between the NFCS 1977-78 and CSFII 1985-86 that may have affected the estimates of fat intake. This analysis examined changes in the nutrient composi- tion database for fat and concluded that product changes contributed far more than data changes. However, the effects of other methodological differ— ences among the surveys, such as degree of probing and rules for coding responses, have not been evalu- ated. Thus, the available data do not permit a confi- dent assessment of whether fat intake has changed over time. The percent of calories from total fat (table 11-21) has apparently decreased slightly for children (from 37 percent in 1977 to 35 percent in 1985-86), decreased for males (from 41-42 percent in 1977 to 36-37 percent in 1985), and decreased for females (from 40-42 percent in 1977 to 36-37 percent in 1985-86). Increases in consumption of carbo- hydrate during this time period may have contributed to the decline in the percent of calories from fat. Prior to 1985-86, intakes of the fatty acid groups were not reported in USDA surveys. The interpreta- tion of cholesterol intake over time (table 11-39) is complicated by the fact that data on the cholesterol content of many foods were not available at the time of NHANES I (Dresser, 1983). However, the estimated intakes of men have remained high; mean (1-day) values in CSFII 1985 (423-466 milligrams per day) are well in excess of recommended levels. Despite the limitations in the interpretation of the individual dietary intake data (especially in the assessment of trends over time), it is clear that many persons have intakes of total fat, saturated fatty acids, and cholesterol that exceed recommended levels. Data on mean serum cholesterol levels are available from HHANES (tables 11-40 and II-41) and may be compared to NHANES II data (tables II-42 and 11-43). In order to maintain comparability with the JNMEC evaluation in the update portion of this report, the criteria for "high-risk cholesterol" levels in serum were those recommended by the 1984 NIH Consensus Development Conference (NIH, 1985): >5.69 mmol/L (>220 milligrams/dl) for persons aged 20-29 years; >6.21 mmol/L (>240 milligrams/dl) for persons aged 30-39 years; and >6.72 mmol/L (>260 milligrams/dl) for persons aged 40 years and older. These criteria were chosen on the basis of observed statistical distributions of cholesterol values rather than on the basis of disease relationships. New stan- dards for defining risk based on elevated blood cho- lesterol levels and other factors were defined by the National Cholesterol Education Program during preparation of this report; these are discussed in chapter 5. The prevalences of "high-risk cholesterol" levels in Mexican Americans, Cubans, and Puerto Ricans in HHANES and non-Hispanic whites and "High—Risk Cholesterol" Levels: Males Non-Hispanic Non-Hispanic Mexican American Cuban Puerto Rican white black SN BR [1] Percent 40 35 30 25 N 20 B N ql Ri + 15 N K N K AN NK NE * N R QR NH NK = N 10F HNp N NB NK i N NE NH NE ING RI: iN NK NK NH IN N IN 5 NK N Ki NB NK iN & iN NE NS NH NK INH] AR o LENE Nk NB NE NE [IN * 20-29 30-39 40-49 50-59 60-69 70-74 Age in years Figure 3-5. Prevalence of "high-risk cholesterol" levels in serum in males, by age and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 (See text for serum cholesterol values that are considered "high- risk." The asterisk indicates an unstable statistic or a statistic not reported because of small sample size.) 53 blacks in NHANES II are shown in figures 3-5 and 3-6 (see also tables I-40 and 11-42). For adult men (20-74 years) in the three Hispanic groups, the mean serum cholesterol levels ranged from 5.26 to 5.35 mmol/L, with 13.5-15.5 percent considered to have "high-risk cholesterol." For women, mean serum cholesterol levels were almost identical to those of men (5.15 to 5.39 mmol/L), and 11.3-15.4 percent had "high-risk cholesterol." Mean serum cholesterol levels did not differ greatly among the three Hispanic ethnic groups, whereas means for males above poverty tended to be greater than those below poverty, particularly in older adults. The prevalence of "high-risk cholesterol” in men and women of all three Hispanic groups was lower than the prevalence in non-Hispanic whites and blacks. Factors contributing to high serum cholesterol in- clude obesity, dietary saturated fat and cholesterol, nondietary behaviors, and genetic factors (see chapter 5 for a detailed discussion of these factors). Conclusion: Total fat, saturated fat, and cholesterol are considered to be current public health issues. The intakes of these food components by many "High-Risk Cholesterol" Levels: Females Non-Hispanic Non-Hispanic Mexican American ~~ Cuban Puerto Rican white black _ SS BN (1 am Percent 40 35 30 25 * ; \ K N 20 NB QB N & NE N N 15 fl N NK N ’ R NE NK N Fi B v § A NE \ INS Kd ¥ \ K NH N HN KI K K NK NH N NI q bi bd NK NK N >I UNE SE LE NE NE \ o LUNE A: NK NE NE N* 20-29 30-39 40-49 50-59 60-69 70-74 Age in years Figure 3-6. Prevalence of "high-risk cholesterol" levels in serum in females, by age and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 (See text for serum cholesterol values that are considered "high- risk." The asterisk indicates an unstable statistic or a statistic not reported because of small sample size.) persons in the U.S. population exceed levels recommended by many authoritative groups. Serum cholesterol levels are affected by dietary intake of these components (and other factors); elevated levels of serum cholesterol are prevalent in the United States (11-22 percent) in men and women of all racial and ethnic groups examined and represent a risk fac— tor for the development of coronary heart disease. Continued priority for the monitoring of serum cholesterol levels and the dietary intake of fat, fatty acids, and cholesterol is warranted. More specific recommendations are included in chapter 5. Carbohydrates Carbohydrates are sources of energy for the body and serve as constituents of various cellular structures and substances. Carbohydrates are classified as monosaccharides, disaccharides, and polysaccharides. The mono- and disaccharide sugars are called simple carbohydrates and the polysaccharides (such as starch) are called complex carbohydrates. The JNMEC classified total carbohydrate as a component warranting less consideration for public health moni- toring priority. Dietary guidelines (USDA/DHHS, 1985) for the gen-— eral American public have focused on food choices designed to maintain health and reduce the risks of chronic and degenerative diseases. = Among the dietary guidelines are recommendations to eat foods with adequate starch and fiber and to avoid too much sugar. A reduction in frequency of sugar consump- tion, particularly sugar found in sticky-type foods, is recommended to reduce dental caries, especially in vulnerable groups such as children. There are no specific guidelines for a quantitative requirement for carbohydrates in the diet. The per capita amount of carbohydrates in the food supply, which had declined since 1909, has shown a fluctuating upward trend since the mid-1960s (figure 11-15). In 1985, sugars and sweeteners contributed 39.6 percent of the carbohydrates in the food supply; grain products, vegetables, and fruits provided 35.8, 9.2, and 6.6 percent, respectively (figure II-16). There has been a substantial decrease in the propor- tion of carbohydrates contributed by grain products in the food supply since 1909. During the same time, there has been an increase in the proportion of car— bohydrates supplied by sugars and sweeteners; most of the change since the 1960s has been caused by increased use of high fructose corn syrup. Based on 4-day data from the CSFII 1985-86, the carbohydrate intake of women aged 20-49 years was 175 grams per day (46 percent of calories); the intake 54 of children aged 1-5 years was 184 grams per day (52 percent of calories) (tables II-44 through 11-47). The intake of carbohydrate as percent of calories was not greatly affected by race, poverty status, or educational level. Based on 1-day estimates, there has been a general trend during the 1970s and early 1980s for an increase in the mean percent of calories from carbo- hydrate for most age and sex groups (table 11-49). There are no health assessments, other than decayed teeth, specifically thought to be associated with intake of carbohydrates. Conclusion: Carbohydrates are not considered to be a current public health issue. Carbohydrate intakes are lower than may be desirable, based on the dietary pattern recommended in the U.S. dietary guidelines (USDA/DHHS, 1985), but evidence does not suggest that current intakes pose a specific public health problem. Monitoring of intake should continue; if recommended decreases in the percent of energy from fats occur, concomitant increases in the proportion from carbohydrates are expected. Dietary Fiber Dietary fiber is a term that refers to a heterogeneous group of plant food components that are resistant to digestion by enzymes produced by the human gastro— intestinal tract. Different dietary fibers are classified as soluble or insoluble depending upon their response to various extraction techniques. In general, the soluble fibers include gums, mucilages, and some pec- tins and hemicelluloses, while the insoluble fibers include cellulose, lignin, and other pectins and hemi- celluloses. Grains, fruits, and vegetables are good sources of fiber; oat bran, beans, and dried fruit are good sources of soluble fiber components, while wheat bran is a good source of insoluble fiber components. The JNMEC report considered dietary fiber to be a component that required further investigation. The Surgeon General's Report on Nutrition and Health (DHHS, 1988) concluded that consuming foods with dietary fiber is usually beneficial in the management of constipation and diverticular disease. In addition, some evidence from clinical studies suggests that soluble fibers are associated with lower blood glucose and blood lipid levels; although inconclusive, some evidence also suggests that an overall increase in intake of foods high in fiber might decrease the risk for colon cancer. The dietary guidelines for Ameri- cans advise eating foods with adequate starch and fiber (USDA/DHHS, 1985). Although there is dis- agreement whether quantitative recommendations for dietary fiber intake are appropriate at this time (National Research Council, 1989), the National Can— cer Institute (1984) currently recommends that adults consume 20-30 grams per day with an upper limit of 35 grams for day. Estimates of mean (4-day) daily dietary fiber intake from the CSFII 1985-86 are 11 grams for women aged 20-49 years and 10 grams for children aged 1-5 years (tables II-50 and II-51). These data indicate that only 5 percent of the women surveyed had in- takes of 20 grams or more of dietary fiber per day. Intake of dietary fiber was greater among women with higher levels of education. Multivariable analy- ses conducted by Cronin (1988) showed that higher fiber intakes by women were associated with having more education, being in the "other" racial category, living in the Midwest or West, living alone, being employed part time, and not smoking. Additional analyses (Cronin, 1988) indicated that vegetables, grains, and fruits supplied 50, 30, and 12 percent, re- spectively, of the dietary fiber in the diets of women. One-day data from the CSFII 1985 indicate that on average the dietary fiber intake of men is higher than that of women (approximately 17 grams per day). Because data on the fiber content of foods have become available only recently, there are neither estimates of individual fiber intake before the CSFII 1985-86 nor estimates of fiber contained in the food supply. The nutrient composition database for the CSFII 1985-86 contained analytical values for total dietary fiber for only 40 percent of the best sources of fiber (Hepburn, 1987). Additional food composition data for total dietary fiber, as well as for different components of fiber that have different physiological effects, and more basic research on the health effects of dietary fiber are needed. Conclusion: Dietary fiber is considered a potential public health issue for which further study is needed. Dietary intakes of fiber are low in relation to suggested levels of intake and are not consistent with recommended dietary patterns, but the im— pact of these low intakes cannot be judged on the basis of available data. More information is re- quired on the health effects of dietary fiber, the content in foods of various components of fiber (which have different physiological effects) as well as total dietary fiber, and recommendations for intake. Monitoring is recommended as this infor— mation is developed. Alcohol Despite the paucity of data on alcohol available from the NNMS, the JNMEC classified alcohol as a 55 component warranting public health monitoring pri- ority status because of concerns about high intake. Reasons for concern about alcohol intake related to diet and nutritional status include the following: alcohol is high in calories and can supply a high per- centage of calories in the diets of drinkers; alcoholic beverages are low in nutrients other than calories; excessive consumption can lead to nutritional inade— quacies caused by poor diet or interference with nutrient absorption; excessive consumption can ele- vate blood pressure and can lead to diseases such as cirrhosis of the liver, pancreatitis, cancer of the throat, and fetal alcohol syndrome. In addition, con- sumption is linked to poor fetal growth and develop- ment, accidents, homicide, assault, and other social problems. The Economic Research Service of USDA compiles data on per capita alcoholic beverages on the basis of industry data. In 1985, per capita figures based on the adult population (all those aged 21 years and older) were 34.5, 3.8, and 2.5 gallons each for beer, wine, and distilled spirits (USDA, 1987). Data on long- range and recent trends in per capita amounts of alcohol are also available from the Alcohol Epidemio- logic Data System (1986), based on beverage sales and/or tax receipts from 35 States and the District of Columbia and shipments data from industry for the nonreporting States. These disappearance data, ex- pressed on an ethanol-equivalent basis, overestimate consumption because they do not account for waste or storage at various stages of distribution and use. Data for the years 1977-84 are shown in table 3-11. As indicated in this table, there was a slight decrease in the first four years since 1980. The overall changes mask changes in the types of beverages, with recent decreases in beer and spirits coupled with a slight in- crease in wine. The data for earlier periods indicated a very small increase in per capita alcohol through the 1950s, a rather rapid increase through the 1960s, and a moderate increase through the 1970s. Even though abstainers are not taken into account in the per capita figures, these values are higher than self-reported alcohol intakes from national food con— sumption surveys. Difficulties in assessment such as deliberate underreporting by individuals, under- representation of very heavy drinkers, failure to classify as drinkers persons who did not consume alcohol during the survey period, proxy reports, and failure to assess consumption on atypical days are recognized as factors that may lead to underestimates of true consumption. In the CSFII 1985, the mean (1-day) intake of alcoholic beverages by women was 84 grams, with 59 grams from beer and ale (USDA, 1985); the mean (1-day) intake by men was 304 grams with 271 grams from beer and ale (USDA, 1986a). An extensive questionnaire on alcohol Table 3-11. Per capita amounts of ethanol, in gallons of ethanol, based on the U.S. population aged 14 years and older, 1977-84] All Year Beer Wine Spirits beverages 1977 1.29 0.29 1.06 2.64 1978 1.32 0.31 1.07 271 1979 1.37 0.32 1.06 2.75 1980 1.38 0.34 1.04 2.76 1981 1.39 0.35 1.02 2.76 1982 1.38 0.36 0.98 2.72 1983 1.37 0.36 0.96 2.69 1984 1.35 0.37 2.65 0.94 I Adapted from Alcohol Epidemiologic Data System (1986). use was administered in the HHANES, but data analyses of these data have not yet been completed. Data were also collected in the NHIS on Health Promotion and Disease Prevention with respect to drinking behavior, knowledge, and attitudes (NCHS, 1988). With "heavier drinking" defined as an average of more than 1 ounce of alcohol (two drinks or more) per day, the results of this survey indicated that 13 percent of men and 3 percent of women had con-— sumed this level in the past two weeks. One-fourth of current drinkers (35 percent of men and 12 percent of women) reported they drank five or more drinks in 1 day at least five times in the past year. The national prevalence estimate of chronic heavy drink- ing was 8.7 percent based on data collected in the Behavioral Risk Factors Surveillance System, 1981- 83 (Bradstock et al., 1985; Gentry et al., 1985). In the same survey, 6.1 percent of adults gave self-reports of drinking and driving (Bradstock et al., 1987). NNMS data have also been used to examine the relationship of alcohol intake to various outcome variables. Data from NHANES II and the Behavioral Risk Factors Surveillance System have been used to examine the relationship of alcohol intake and body weight; in both surveys, alcohol consumption was associated with lower body weight (after adjustment for other factors that influence weight) in women but not in men (Williamson et al., 1987). In the NHANES I Epidemiologic Followup Study, data show a higher 56 relative risk for breast cancer in moderate and high consumers of alcohol than in nonconsumers (Schatz— kin et al, 1987). However, evidence for this association is contradictory (DHHS, 1988); the association appears to be specific to type of breast cancer and a more substantial database is required before conclusions can be drawn regarding risk. Conclusion: Alcohol is considered a current public health issue. Self-reported intakes are high (an average of 1 ounce or more of ethanol per day) in a large number of persons (9 percent of adults). The public health and social consequences of excessive alcohol intake are sufficiently grave that continued efforts to improve monitoring of alcohol intake are warranted. Vitamins Vitamin A and Carotenes Vitamin A (retinol and pro-vitamin A carotenoids) is a fat-soluble nutrient required in adequate amounts for vision, reproduction, bone growth, and mainte- nance of healthy epithelia. Clinical signs of chronic retinol deficiency include poor night vision and char- acteristic changes in the eyes and skin. Because the body can store vitamin A, acute deficiency does not produce symptoms in otherwise well nourished indi- viduals. In chronic excess, retinol has adverse effects, particularly on bone and in reproduction; it is a potential teratogen. The JNMEC classified vitamin A as a component warranting continued public health monitoring consideration based on low dietary in- takes. Earlier, the possibility of vitamin A deficiency, as suggested by low serum levels of the vitamin, was identified as a potential problem in some groups in the U.S. population by the Ten-State Nutrition Sur- vey (DHEW, 1972a, 1972b), especially in the Hispanic population, and in children and low-income groups. Beta-carotene and other pro-vitamin A carotenoids can be converted to vitamin A in the body. Interest in the carotenoids has increased in recent years because of the accumulation of a large body of evi- dence that foods high in carotenoids are protective against a variety of epithelial cancers (DHHS, 1988). High intakes of carotenes are not known to produce serious adverse effects. Vitamin A content in the food supply is reported as "retinol equivalents" to express the equivalent quan-— tities of retinol and carotenes in foodstuffs. The U.S. food supply shows an increase in per capita amount of vitamin A from 1909-85 (figure II-17) and an increase in carotenes since 1965 (figure II-19). Vitamin A and carotene levels both reached a peak in 1985 at 1,610 and 660 retinol equivalents, respec tively. The recent gains primarily reflected the development of new varieties of deep yellow vegetables with higher carotene contents. Fortification of margarine and some dairy products also contributed to the increase. Vegetables have accounted for three-fourths of the carotenes in the food supply over the years; dark green and deep yellow types were the most common source (figure 11-20). Vitamin A (in the form of retinol) is also supplied by meat, poultry, and fish (26.4 percent) as well as dairy products (15.8 percent) (figure 11-18). Individual dietary intake data from CSFII 1985-86 indicate that the mean (4-day) intake of vitamin A for all women, 832 retinol equivalents, is near the RDA (table 11-52). Nonetheless, individual variation was considerable, with the median intake (50th percentile) equaling 614 retinol equivalents. (The National Research Council [1980] estimated that 500— 600 retinol equivalents per day is needed by adults to maintain serum vitamin A levels, with greater amounts needed to provide reserves.) In the CSFII 1985-86, median intakes appear to vary little with age and poverty status, but to differ by race and education, in the adult females. The mean intake for children, as well as the intake at the 25th percentile, is above the 1980 RDA. Individual 1-day intakes, expressed in International Units, from surveys conducted during 1971-86 show an increase over time (table II-54). Caution is needed in interpreting these estimates obtained with very few days of intake data as usual intakes because of the very high day-to-day variation in vitamin A intake (see chapter 2). The intake of carotenes was also estimated in the CSFII 1985-86 (see tables II-59 and II-60 for data for women and children, respectively). Standards for evaluating the adequacy of intakes of carotenes are not available. These data should, however, provide useful baseline information on intake levels. An evaluation of the data from the various HANES concluded that serum vitamin A levels alone are inadequate to provide estimates of the prevalences of vitamin A deficiency and toxicity in the U.S. popula- tion (LSRO, 1985). Nonetheless, epidemiological and clinical data indicate that there are age-specific physiological correlates of low serum vitamin A levels related to deficiency (LSRO, 1985). The prevalences of serum vitamin A levels below 0.70 pmol/L (20 pg/dl) were relatively low (0-6.1 percent) for all age groups in all three HANES. (Comparisons among surveys should be made cautiously because total serum vitamin A levels were measured in NHANES I and II, but serum retinol levels were measured in HHANES.) The prevalences of serum vitamin A 57 levels between 0.70 and 1.01 pmol/L (20 and 24 pg/dl) were generally higher (0-23.5 percent), especially in children. The likelihood that functional impairment is associated with these low serum levels of vitamin A is greater for adults than for children. Multivariable analysis indicated that black persons had a greater prevalence of low serum vitamin A regardless of economic status, and that, generally, poor persons had a greater prevalence of low serum vitamin A regardless of race (LSRO, 1985). Data from HHANES (tables II-55 through II-58) do not indicate a significant prevalence of low serum vitamin A in any group, with the possible exception of 4-5 year-old Mexican Americans below poverty; 10.1 percent had serum retinol levels below 0.70 pmol/L (20 pg/dl). Clinical signs of overt vitamin A deficiency were not seen in this group. A 1980 telephone survey conducted by the FDA on vitamin/mineral supplement use (Stewart et al., 1985) found that 25 percent of the adult U.S. population obtained vitamin A from supplementary sources. Among users of vitamin A supplements, the median intake from supplements was 125 percent of the RDA; at the 95th percentile, vitamin A obtained from supplements equaled 430 percent of the RDA. Conclusion: Vitamin A is considered to be a potential public health issue for which further study is required. The content of vitamin A in the food supply and individual intakes suggest general ade— quacy. Intake and status may, however, warrant continued monitoring efforts in certain groups. HHANES data on low serum vitamin A levels sug- gest that poor young children, particularly Mexican Americans, may be such a group. Greater attention needs to be given to studying the relationships of biochemical assessments of status to functional impairments. Carotenes are also considered a potential public health issue for which further study is also required. Data on intake of carotenes are available from the CSFII 1985-86 and will be available from HHANES to provide a baseline for assessing future changes in intake. Future surveys should continue to collect and report intake separately for carotenes and total vitamin A. Additional research is needed on the health effects of consumption of specified levels of total carotenes, as well as individual carotenes. Vitamin E Vitamin E (tocopherol) is a fat-soluble nutrient. Its biological function is poorly understood but is most likely related to its antioxidant properties. Clinical or biochemical evidence of vitamin E deficiency has not been reported in adults (National Research Council, 1980). Decreased erythrocyte survival time has been reported in low-birth-weight infants with depleted vitamin E stores (National Research Council, 1980). Nutritional status is commonly estimated from serum or plasma concentrations. However, because most vitamin E is transported by plasma lipoproteins, there is a high correlation between total plasma lipids and vitamin E concentration (Bieri, 1976; Horwitt et al., 1972; Rubinstein, Dietz, and Srinavasan, 1969). Consequently, this must be considered when serum or plasma vitamin E levels are used to assess nutritional adequacy. The Food and Nutrition Board stated that, "it is generally accepted that a plasma level of total tocopherols below 0.5 milligrams/100 ml is undesir— able, although it has not been shown that lower con— centrations in adults, unless of a duration of a year or longer, are associated with inadequate tissue concen- trations" (National Research Council, 1980). Com- pared to other fat-soluble vitamins, vitamin E is relatively nontoxic. Isolated and inconsistent reports of adverse effects of very high doses in adults have appeared; large doses in anemic children have inhi- bited the hematological response to iron treatment (National Research Council, 1980). Vitamin E was not evaluated by the JNMEC. Neither vitamin E consumption in foods nor serum vitamin E levels have been measured previously in the NNMS. Thus, information on intake in the CSFII 1985-86 and the serum a-tocopherol concentrations deter— mined in the 1982-84 HHANES provide baseline data. Vitamin E in the food supply is reported as milligram equivalents of a-tocopherol. Fats and oils are the major sources. The vitamin E content of the food supply has increased steadily, doubling since 1909 to the current (1985) level of 16.1 milligrams per capita per day (figure II-21). Fats and oils contributed 66.4 percent of vitamin E, with 35.5 percent from salad, cooking, and other edible oils (figure II-22) in 1985. Vegetables contributed less than 8 percent and grain products just over 4 percent. In the CSFII 1985-86, mean vitamin E intake (4 days) of all women equaled 7.0 milligrams per day, with considerable range (table I-61). At the 5th percentile, intake averaged 2.5 milligrams per day, whereas at the 95th percentile, intake averaged 13.8 milligrams per day. Thus, although the average intake is close to the RDA, most individuals were below this value. The mean intakes of dietary vitamin E, determined in the CSFII 1985-86, are above or very near the RDA for children 1-5 years and for men 20-49 years. The nutrient composition database for vitamin E contains analytical values for fewer food items than for most nutrients 58 (approximately 40 percent of best sources), introducing some uncertainties in the estimates of intake (Hepburn, 1987). The mean serum a-tocopherol levels for adults fall between 22 and 25 pmol/L (equal to 1-1.1 milligrams/dl) for the three groups of Hispanic Americans (table II-64). Children have somewhat lower serum a-tocopherol levels, averaging 14-18 pmol/L (equal to 0.6-0.8 milligrams/dl) (table II-63). The lower concentrations of total serum lipids nor- mally found in children provide a likely explanation for this difference because of the correlation between serum lipids and a-tocopherol (National Research Council, 1980). Interpretation of serum a-tocopherol levels will remain problematic until clear interpreta- tive guidelines are established. It can be noted, how- ever, that all mean values exceed the level of 0.5 milligrams/dl that the Food and Nutrition Board (National Research Council, 1980) has considered to be the lowest desirable concentration. There is little relationship between socioeconomic status or sex and serum a-tocopherol concentrations. According to a 1980 FDA survey (Stewart et al., 1985), 28 percent of the adult U.S. population uses vitamin/ mineral supplements containing vitamin E. The me- dian intake among users equaled 200 percent of the RDA. At the 95th percentile, intake from supple- ments equaled 6000 percent of the RDA. Thus, a substantial portion of the U.S. adult population con- sumes large amounts of vitamin E from nonfood sources. Conclusion: Vitamin E is not considered to be a cur- rent public health issue. Although some intakes are lower than recommended levels (especially in women), data on serum a-tocopherol levels and clinical data on the rarity of vitamin E deficiency suggest little reason for a public health focus. Interpretation of serum a-tocopherol levels is con- founded by other factors such as serum lipid con- centrations, and clear interpretative guidelines to assess marginal vitamin E status do not yet exist. Thiamin Thiamin, an essential component of enzymes involved in the release of energy from carbohydrate and fat, is a water-soluble B vitamin. Formerly called vitamin B1, thiamin also plays a role in cell reproduction, fatty acid metabolism, and normal functioning of the nervous system. Clinical signs of thiamin deficiency in adults primarily involve the nervous and cardio- vascular systems. Beriberi is the classic thiamin defi- ciency disease, occurring most frequently in areas where the diet consists mainly of unenriched white rice and white flour. In the United States, thiamin deficiency is usually associated with conditions such as alcoholism. The JNMEC classified thiamin as a component warranting continued public health moni- toring consideration. Key sources of thiamin in the food supply include grain products (42.3 percent) and meat, poultry, and fish (25.7 percent) (figure I-24). Other notable food sources of thiamin are vegetables (10.9 percent) and dairy products (8 percent). Thiamin provided by the U.S. food supply in the 1980s is roughly 2.2 milligrams per capita per day (figure II-23). This figure is approximately 40 percent higher than in the pre-World War II era (1909-39, 1.6 milligrams per capita per day). The introduction of enrichment of flour was responsible for this increase. Women in the CSFII 1985-86 aged 20-49 years had mean (4-day) intake levels slightly above the RDA (1.00 to 1.05 milligrams per day) (table II-67). Only 5 percent had intakes below 50 percent of the RDA. Mean intake levels were similar among all age groups, and mean intake levels were slightly above the RDA among whites and other races (excluding blacks) and slightly below the RDA among blacks. Mean intake levels were reasonably similar regardless of poverty status, region, and urbanization; mean intake levels, however, rose with increases in educational status. The mean thiamin intakes of children aged 1-5 years (CSFII 1985-86) were above the RDA among all races and were highest among black children (table II-68). Mean intake levels were reasonably similar and above the RDA regardless of poverty status, education, region, and urbanization. One-day data from the NFCS 1977-78, and from the CSFII 1985-86 (table II-69) indicate that mean intake levels of thiamin increased 9.3 percent for children 1-5 years of age; 18.0 percent for males 20— 49 years of age; and 10.8 percent for females 20-49 years of age. All mean intake levels were above the RDA; they were the highest for males and lowest for females. Biochemical assessments of thiamin status were not included in recent NNMS surveys. No clinical evi- dence of thiamin deficiency was detected. In 1980, supplements containing thiamin were in- gested by approximately 30 percent of the population, either singly or in combination with other vitamins and minerals (Stewart et al., 1985). This percentage is second only to vitamin C (35 percent of the popu-— lation). Among users, the median intake of supple— mental thiamin was 550 percent of the RDA; intake at the 95th percentile was 6,000 percent of the RDA. 59 Conclusion: Thiamin is not considered to be a cur- rent public health issue. Intakes appear to be adequate and no other evidence suggested a public health problem with respect to thiamin status. Riboflavin Formerly called vitamin B2, riboflavin is a water—sol- uble B vitamin. Riboflavin is a component of enzymes involved in release of energy from protein, fat, and carbohydrate. Clinical signs of riboflavin deficiency include cracks at the corners of the mouth and sore- ness and inflammation of the mouth, lips, and tongue. If riboflavin deficiency occurs, thiamin and niacin deficiencies are usually associated. The JNMEC con- sidered riboflavin a component warranting continued public health monitoring consideration. Major sources of riboflavin in the food supply include dairy products (34.7 pecent); grain products (24 percent); and meat, poultry, and fish (24.3 percent) (figure 11-26). Riboflavin provided by the U.S. food supply in the 1980s is about 2.3 to 2.4 milligrams per capita per day (figure II-25). This figure is roughly 28 to 33 percent higher than in the pre-World War II era (1909-39, 1.8 milligrams per capita per day). However, riboflavin provided by the U.S. food supply in the 1980s is essentially the same as that provided in the period 1947-79. The changes observed reflect the introduction of enrichment of flour. The mean (4-day) intake of riboflavin by women aged 20-49 years (CSFII 1985-86) was 12.5 percent above the RDA, with only 5 percent of individuals having intakes below 50 percent of the RDA (table II-70). Mean intake levels were highest among women aged 20-29 years, followed by those aged 30-39 years, and were lowest among those aged 40-49 years. Mean in- take levels increased with increasing socioeconomic status and education but were reasonably similar ac- cording to region and urbanization. Mean riboflavin intakes in children aged 1-5 years (CSFII 1985-86) were at least 60 percent above the RDA, with 95 percent of children having intakes of at least 0.9 milligrams per day (table II-71). Mean intakes were above the RDA among all races. Mean intake levels were above the RDA in all regions and were reasonably similar regardless of poverty status, education status, and urbanization. One-day data from the NFCS 1977-78 and from the CFSII 1985-86 (table II-72) indicate that the mean intake levels of riboflavin increased 4.3 percent for children 1-5 years of age; 8.1 percent for males 20-49 years of age; and 8.3 percent for females 20-49 years of age. Mean intake levels were above the RDA; they were highest for males and lowest for females. Biochemical assessments of riboflavin status were not included in recent NNMS surveys. Clinical evidence of riboflavin deficiency was not detected. Supplementary riboflavin was ingested by 30 percent of the adult population, either singly or in combi- nation with other vitamins and minerals (Stewart et al., 1985). This percentage is third only to vitamin C and thiamin (35 percent and 30 percent of the popu- lation, respectively). Among users, the median intake of riboflavin from supplements was 420 percent of the RDA; intake at the 95th percentile was 5,000 percent of the RDA. Conclusion: Riboflavin is not considered to be a current public health issue. Intakes appear to be adequate and no other evidence suggests a public health problem with regard to riboflavin status. Niacin Niacin plays an essential role in the release of energy from protein; it may be ingested from food as "pre— formed niacin" or produced in the body from ingested tryptophan (one of the amino acids found in protein). The JNMEC classified niacin as a component war— ranting continued public health monitoring consid— eration. The per capita amount of preformed niacin in the food supply was 26 milligrams in 1985 (figure 11-27). The level has increased, especially since the mid- 1940s when enrichment of flour was introduced. The major sources of niacin in the food supply are meat, poultry, and fish (46 percent) and grain products (30 percent) (figure 11-28). The contribution of grain products has increased over time. Based on the results of the CSFII 1985-86, the mean (4-day) intakes of dietary preformed niacin in all age groups among women aged 20-49 years and among children aged 1-5 years were well above the RDA (tables II-73 and II-74) (these values would be even higher if the contributions of tryptophan to total nia— cin intake were included). Individuals above poverty status and those with more than a high school educa- tion had greater intakes of niacin. The differences in intake were very small, however, and among individ— uals below poverty and those with a high school edu- cation or less, mean niacin intakes were still 10-15 percent above the RDA. There did not appear to be any differences in niacin intake by age, race, or degree of urbanization. These data suggest that the percentage of individuals with low niacin intakes that would put them at risk for having adverse health conditions is probably very low. One-day mean 60 intakes of preformed niacin (table II-75) indicate slight increases over time in surveys conducted during the period 1971-86. Although biochemical assessments related to niacin nutriture have not been made in any of the component parts of the NNMS, data from the clinical examinations carried out in the NHANES I and II showed few, if any, clinical signs in relation to niacin deficiency. Supplemental niacin was ingested by 30 percent of the adult population in 1980 (Stewart et al., 1985). The median intake of users was 190 percent of the RDA; the 95th percentile of intake was 1,200 percent of the RDA. Conclusion: Niacin is not considered to be a current public health issue. Individual intakes of preformed niacin appear to be adequate (and additional niacin may be obtained from the conversion of dietary tryp- tophan in the body). No other evidence suggests a public health problem in relation to niacin status. Vitamin B6 Vitamin B6 (a collective term for pyridoxine, pyri- doxal, and pyridoxamine) is required for the forma- tion of amino acids needed for protein synthesis, the conversion of tryptophan to niacin, and the produc- tion of substances involved in nervous system function. The JNMEC classified vitamin B6 as a component requiring further investigation. The per capita amount of vitamin B6 in the food sup- ply has not changed appreciably since the early 1900s (figure 11-29), but food sources have changed: the contribution of meat, poultry, and fish has increased dramatically and the contributions of grains and veg- etables (notably potatoes) have declined. In 1985, the per capita amount of vitamin B6 in the food supply was 2.1 milligrams; the major sources of vitamin B6 in the food supply were meats, poultry, and fish (41.1 percent); vegetables (21.9 percent); dairy products (10.7 percent); and fruits (10.6 percent) (figure II-30). Dietary intakes of vitamin B6 by children aged 1-5 years in the CSFII 1985-86 exceeded the RDA and did not differ by race, poverty, or education (table II- 78). In contrast, the mean dietary intakes of women were well below (approximately half) the RDA and varied with age (table II-76). The requirement for vitamin B6 is related to protein intake and the RDA for vitamin B6 is set assuming a protein intake of 100 grams for adult women and 110 grams for adult men (National Research Council, 1980). These assumed protein levels are higher than the levels actually consumed; thus, allowances for vitamin B6 may be overly generous and the low intake levels seen in women may not be indicative of potential deficiency. Moshfegh (1988) has determined that many more women met the protein-based allowance for vitamin B6 intake than met the RDA. Another complication in the interpretation of vitamin B6 intakes is the nutrient composition database; analytical values were available for 70 percent of important sources of vita— min B6 in the CSFII 1985-86 (Hepburn, 1987). As is true for all nutrients, the composition database can- not account for differing bioavailability, a factor that is important in understanding adequacy of vitamin B6 intakes. Biochemical or clinical assessments of vitamin B6 status are not available from the NNMS. Frank vitamin B6 deficiency resulting in clinical manifestations is not suspected to be widespread in the general population. However, evidence of impaired status has been reported in some population groups, most notably the elderly and alcoholic individuals (National Research Council, 1980). In 1980, 30 percent of the adult U.S. population con— sumed supplements containing vitamin B6 (Stewart et al.,, 1985). The median intake among users was approximately 140 percent of the RDA; the 95th percentile of intake was 5,000 percent of the RDA. Excessive intake of the vitamin from high-potency supplements is of concern because of possible damage to the peripheral nervous system (Dreyfus, 1988). Conclusion: Vitamin B6 is considered to be a poten- tial public health issue for which further study is needed. Intakes are lower than recommended levels for a substantial number of persons, espe- cially women. In order to interpret the conse- quences of these intakes, further study is needed on the content and bioavailability of vitamin B6 in foods, vitamin B6 requirements, and biochemical or other techniques for assessing vitamin B6 nutri- tional status. Increased monitoring activity may be warranted as progress is made in these areas. Vitamin B12 Vitamin B12 (cobalamins) is required for formation of red blood cells, synthesis of genetic material, function of the nervous system, and metabolism of protein and fat. The JNMEC classified vitamin B12 as a compo-— nent warranting less consideration for continued public health monitoring. Severe vitamin B12 defi- ciency can result in pernicious anemia and in neuro-— logical damage if deficiency is prolonged. Deficiency 61 is caused more often by impaired absorption rather than inadequate intake (DHHS/USDA, 1986). How- ever, a dietary vitamin B12 deficiency can occur in strict vegetarians who avoid all foods derived from animals and in the breast-fed children of mothers who consume such diets (DHHS, 1988). The per capita amount of vitamin B12 in the food supply has increased slightly since the 1940s and was 10.1 micrograms in 1985 (figure II-31). Vitamin B12 is found only in animal products, so the major sources in the food supply have remained fairly stable over time; in 1985, meat, poultry, and fish contributed 75.4 percent, and dairy products contributed 17.6 percent of the vitamin B12 in the food supply (figure 11-32). Individual intake data (4 days) from the CSFII 1985- 86 indicate that almost 50 percent of women aged 20- 49 years, and over 90 percent of children aged 1-5 years, had intakes in excess of the RDA (tables 11-80 and 11-81). The early laboratory manifestations of vitamin B12 deficiency include an abnormally large number of lobes of the nuclei of neutrophils (a type of white blood cell) and a low serum concentration of vitamin B12. These findings are rare among hospital patients and indicate that health problems related to a de- ficient intake of vitamin B12 are too unusual to be detected in population surveys, although some con- cern has been raised about vitamin B12 status in the elderly. Serum vitamin B12 measurements were made in NHANES II but have not been released. In 1985, 30 percent of the adult U.S. population con- sumed supplements (singly or in combination with other vitamins and minerals) of vitamin B12 (Stewart et al., 1985). Among users, intake at the median level was 200 percent of the RDA and intake at the 95th percentile exceeded 3,300 percent of the RDA. Conclusion: Vitamin B12 is not considered to be a current public health issue. Intakes appear to be adequate for the majority of the population. Clini- cal or biochemical evidence for a public health problem with respect to vitamin B12 deficiency is not available. Further monitoring, at a low level, is warranted. Vitamin C Vitamin C functions in the formation of collagen; the maintenance of capillaries, bones, and teeth; the pro- motion of iron absorption; and protection of other vitamins and minerals from oxidation. Some evidence suggests a protective effect against certain cancers (DHHS, 1988). Vitamin C deficiency causes scurvy, characterized by weakness, hemorrhages in the skin and gums, and defects in bone development in chil- dren (DHHS/USDA, 1986). Although the average intake of vitamin C was above the RDA, the JNMEC report suggested that vitamin C was a nutrient war— ranting priority monitoring status based on low intakes in certain groups. Evidence of measurable clinical deficiency in all population groups was negligible, and the prevalence of low serum vitamin C levels was low in most groups. The per capita amount of vitamin C in the food supply has fluctuated since the early 1900s but has not changed consistently; it was 115 milligrams in 1985 (figure 11-33), an amount well in excess of the highest RDA. The major food sources have changed, with increases in the contributions of citrus fruits and decreases in the contributions of potatoes and vegetables other than dark green and deep yellow types. In 1985, the food groups that contributed the major shares of vitamin C in the food supply were vegetables (47.9 percent) and fruits (42.7 percent), especially citrus fruits (27.7 percent) (figure 11-34). Based on 4-day data from the CSFII 1985-86, mean intakes of dietary vitamin C in all age groups among women aged 20-49 years and among children aged 1- 5 years were well above the RDA (tables I-82 and 11-83). Women above poverty and those with higher levels of education had higher intakes of vitamin C. Among those individuals below poverty, mean intake of vitamin C was at or close to the RDA, while among individuals above poverty, mean intake of vitamin C was 37 percent greater than the RDA. Women with less than a high school education had mean vitamin C intakes close to the RDA, while individuals with greater than a high school education had intakes that exceeded 50 percent more than the RDA. Among children aged 1-5 years, intakes of vitamin C were much greater than the RDA in all groups, and the differences seen with the indices of socioeconomic status were similar to those seen among women. There did not appear to be any meaningful differ— ences in vitamin C intake by age and race in the children. These data suggest that the percentage of individuals with low vitamin C intake that would put them at risk for having adverse health conditions is probably very low. However, the lower intakes of vitamin C among women and children of lower socioeconomic status suggest that they may be at greater, albeit still low, risk for adverse health outcomes related to low vitamin C intake than groups of higher socioeconomic status. One-day data for vitamin C intakes assessed in surveys conducted in the period 1971-86 (table 11-84) suggest some increases over the period studied, but results are not consistent over all surveys. Greater 62 changes may be seen in the future because of the recent introduction of higher levels of vitamin C fortification in a variety of foods and beverages. Although serum vitamin C determinations were not made in the HHANES, data from the NHANES II, 1976-80, indicated that among children the prev- alence of low serum vitamin C levels was low. The prevalence of low values was higher in men than in women and higher in persons below poverty than in persons above poverty; the prevalence was highest (19.5 percent) in men below poverty aged 55-74 years. There was very little or no evidence of vitamin C deficiency in data from the clinical examination conducted in the NHANES II. In 1980, 35 percent of the adult U.S. population con— sumed vitamin C in the form of supplements (Stewart et al, 1985). Vitamin C, consumed by more than 90 percent of all supplement users, was the most widely consumed nutrient. Among users of vitamin C sup- plements, median intake was 333 percent of the RDA; the 95th percentile of intake was 2,800 percent of the RDA. Conclusion: Vitamin C is considered to be a potential public health issue for which more study is needed. Recent dietary intakes appear to be adequate in most of the population, even without consideration of the substantial contribution of vitamin C supplements. Older data for serum vitamin C (from NHANES II) indicate that the prevalence of low levels is generally higher in groups with low socioeconomic status, especially older men, but do not provide strong evidence for vitamin C deficiency. Although these data suggest the need for some continued surveillance, changes in vitamin C fortification practices may affect intake among many segments of the population. Continued monitoring is warranted to assess the impact of these changes, but the apparently adequate intakes do not provide support for priority monitoring status. Folacin Folacin, also called folate, a descriptor for compounds with properties and chemical structures similar to folic acid or pteroylglutamic acid, is required for the formation of red blood cells and genetic material. Folacin deficiency can result in anemia, weakness, forgetfulness, sleeplessness, and periods of euphoria. The JNMEC classified folacin as a component requiring further investigation. Individual dietary intake data for folacin were not available to the JNMEC. The per capita amount of folacin has not changed substantially since the early 1900s; it was nearly 300 micrograms in 1985 (figure 11-35). The contributions of meat, poultry, and fish and of fruits have increased, and the contributions of grain products have de- creased over time. In 1985, the major share of folate in the food supply was contributed by vegetables (24.8 percent); legumes, nuts, and soy (19.5 percent); grain products (12.7 percent); meat, poultry, and fish (12.6 percent); and fruits (12.4 percent) (figure II-36). Mean dietary folacin consumption (4 days) from CSFII 1985-86 is estimated to be below the 1980 RDA in over 95 percent of women aged 20-49 years and over 50 percent of children aged 3-5 years (tables 11-85 and II-87). In contrast, over 90 percent of children aged 1-2 years had folate intakes above the RDA for this age group. For women, intake was higher in persons of higher socioeconomic status (as indicated by poverty status and education level). Some uncertainties exist in the nutrient composition database for folate. Folate in food is difficult to measure, and the database contained analytical values for approximately 70 percent of best sources of the nutrient (Hepburn, 1987). Serum and red blood cell folate levels were measured in subgroups in both NHANES II and HHANES. Analyses of the NHANES II data indicated that women aged 20-44 years were at greatest risk of folate deficiency (LSRO, 1984). Results for Hispanic women aged 20-44 years from HHANES are pre- sented in table 3-12. Prevalences of low values for serum and red blood cell folate levels were lower in women of all three Hispanic groups than in the women of all races and ethnic groups in NHANES II; however, different methods were used in the two sur— veys to measure blood folate levels. In addition, interpretative criteria for blood folate levels are not certain (LSRO, 1984); thus, prevalence of low values cannot be interpreted as indicating the prevalence of folate deficiency. Hematological signs of folate deficiency have not been detected in surveys. How- ever, folate deficiency is believed to be rare, and there is little evidence that it constitutes a major public health problem. Furthermore, the discrepancy between observed intakes below estimates of recom— mended intakes of folate and the lack of evidence for folate deficiency suggests to the EPONM that allowances need to be reexamined. Groups that are known to be most vulnerable to folate deficiency include premature infants and women during the last half of pregnancy. These groups are not sampled in sufficient numbers in the NNMS to analyze the available data on the preva- lence of either low intakes or low blood values in these groups. A larger group that might be Table 3-12. Percent of Hispanic women aged 20-44 years with low blood folate levels: Hispanic Health and Nutrition Examination Survey, 1982-84 Standard error Population group n Percent of percent Low red blood cell folate level’ Mexican American 1,040 3.5 0.7 Cuban 194 10.8 2.4 Puerto Rican 434 9.2 1.8 Low serum folate level® Mexican American 1,090 11.1 1.2 Cuban 205 9.8 2.3 Puerto Rican 427 8.3 1.8 I Red blood cell folate <140 nanograms/milliliter. 2 Serum folate <3.0 nanograms/milliliter. vulnerable to folacin deficiency is women who use oral contraceptives, because these agents may depress folacin absorption from foods in the diet. Although blood folate values were, on the average, lower in oral contraceptives users than in nonusers in NHANES II, the difference was not statistically significant (LSRO, 1984). Supplemental folacin (in the form of folic acid) was consumed by 20 percent of the adult U.S. population in 1980 (Stewart et al., 1985). Intake among user: was equivalent to the RDA at the median and was 2( . percent of the RDA at the 95th percentile. Conclusion: Folacin is considered to be a potential public health issue for which further study is required. Dietary intakes are much lower than recommended in some groups, especially women. Biochemical and other evidence for deficiency are limited, but suggest a risk of deficiency in women. Further study is required to evaluate folacin requirements, to develop methods and inter- pretative criteria for folacin nutritional status, and to examine the status of groups at risk. Other Vitamins Dietary and nutritional status with regard to vitamin D, vitamin K, pantothenic acid, and biotin is not currently assessed in the NNMS. The EPONM is not aware of evidence for public health issues with respect to these vitamins that is strong enough to justify a recommendation for their inclusion in the NNMS at this time, although the vitamin D status of the elderly may be of interest. Minerals Iron Dietary iron deficiency is considered to be the most common nutritional deficiency in the United States. The JNMEC ranked iron as a component warranting public health monitoring priority status and it is one of the two conditions selected for special emphasis in this report (see chapter 6). The risk of iron deficiency varies markedly according to the iron requirements in different age/sex groups. Iron deficiency is common among infants and young children between 6 months and 3 years of age, in whom the rate of growth is high, and whose diet is often dominated by milk, which contains little iron. Low-birth-weight infants are also at high risk of iron deficiency because they have particularly high iron requirements to support their rapid growth. The growth spurt during adolescence is also associated with a high prevalence of iron deficiency. Among adults, young women have much higher iron require— ments than men, as a result of iron lost in menstrual blood or maternal iron lost to the fetus, particularly during the last half of pregnancy. Men and post- menopausal women are rarely iron deficient unless there is excessive blood loss due to chronic aspirin use, frequent blood donation (more than three times per year), or diseases such as gastrointestinal ulcer or cancer. Iron deficiency may be assessed using a combination of measurements undertaken in various NNMS surveys (see chapter 6). The amount of iron in the food supply increased during the mid-1940s (in response to the introduc- tion of iron enrichment of flour) and has also in- creased recently, equaling 17 milligrams in 1985 (figure II-37). The percentages of iron supplied by various food groups in the food supply were 41.0, 23.8, and 12.6 percent for grain products; meat, poultry, and fish; and vegetables, respectively, in 1985 (figure 11-38). Estimates of the mean iron intake (from 1-day data) in NHANES I and II, NFCS 1977-78, and CSFII 1985-86 are remarkably close, with a range of 9.2 to 10.8 milligrams per day in women of childbearing years (table II-93). The mean intakes are less than 60 percent of the RDA for this group. The distribution of intakes (based on 4-day data) in CSFII 1985-86 shows that over 95 percent of women aged 20-49 years, and over 90 percent of infants aged 1-2 years, have iron intakes below their respective RDA (tables 11-89 and 11-91). Over 50 percent of children aged 3-5 years have intakes below the RDA. The mean dietary iron intake, but not the iron density of diets, of women differed by race and socioeconomic status (tables II-89 and 11-90). These intake data would lead one to anticipate an even higher prevalence of iron deficiency than was found in the analysis of the extensive biochemical data collected in NHANES II and HHANES using the mean corpuscular volume (MCV) model (see chapter 6 for discussion of the model and tables 11-98 through 11-101 for data for persons aged 4-74 years). Esti— mates of the prevalence of iron deficiency in women of childbearing years (one of the vulnerable groups for which the NNMS provides extensive data) ranged from 2.4 to 14 percent (table 3-13). The discrepancy between the modest prevalence of iron deficiency and low intakes suggests to the EPONM that the RDA for iron are not only overly generous, they are also rarely achievable from usually consumed diets in the population groups who are at risk of developing iron deficiency. A major reason for the apparent contra- diction is that intake of iron is not related to nutri- tional status outcome in a straightforward fashion; absorption of iron from food increases dramatically when body iron stores become low. Efforts to decrease the prevalence of iron deficiency in the U.S. population have apparently been success— ful in the past decade, particularly among infants and small children. Data from the PedNSS indicate a declining prevalence of anemia in low-income chil- dren (see discussion of data in chapter 6). Factors responsible for this success include an increased prevalence of breast feeding, more use of iron—forti- fied infant formula, and avoidance of cows milk in early infancy. Furthermore, the food industry has increasingly used better absorbed forms of iron to fortify bread and other cereal products. These changes are not readily apparent in the individual intake data in the NNMS. Iron deficiency among young women may prove diffi- cult to eradicate, because it is likely to be related to unusually heavy menstrual blood loss for which ade- quate iron cannot be supplied readily by dietary means alone. Supplemental iron (alone or in combination with other minerals and vitamins) was consumed by ap- proximately 22 percent of the adult U.S. population in 1980 (Stewart et al., 1985). Iron was consumed by 56 Table 3-13. Percent of women aged 16-49 years with iron deficiency determined by the MCV model’: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 Age and ethnic Standard error group or race n Percent of percent 16-19 years Mexican American 277 7.9 1.6 Cuban 42 0.0* 4.6 Puerto Rican 117 7.9 24 Non-Hispanic white 456 3.8 1.1 Non-Hispanic black 83 13.8 44 20-29 years Mexican American 468 8.2 1.5 Cuban 60 2.8 2.5 Puerto Rican 162 7.2 2.5 Non-Hispanic white 956 2.4 0.6 Non-Hispanic black 163 3.5 1.8 30-39 years Mexican American 407 11.7 1.8 Cuban 87 7.2 3.3 Puerto Rican 155 7.9 2.7 Non-Hispanic white 741 6.5 1.2 Non-Hispanic black 104 59 2.9 40-49 years Mexican American 288 12.9 2.0 Cuban 100 10.2 3.3 Puerto Rican 155 8.3 24 Non-Hispanic white 587 6.5 14 Non-Hispanic black 90 14.0 44 The MCV model requires abnormal values for at least two of the following measurements: MCV, transferrin saturation, erythrocyte protoporphyrin. See chapter 6 for cutoffs indicative of abnormal values. * Indicates a statistic that may be unreliable because of small sample size. percent of all supplement users and was the most commonly consumed mineral. The intake of sup- plemental iron was approximately 120 percent of the RDA at the median and 500 percent of the RDA at the 95th percentile for users of iron supplements. Conclusion: Iron is considered to be a current public health issue. Intakes are low for many in the population. Although the prevalence of iron deficiency has apparently declined in recent years, it is still relatively high in vulnerable groups such as women of childbearing years. Continued monitoring is warranted and is likely to yield useful information on iron nutritional status because of the wealth of indicators available for inclusion in the NNMS. A detailed discussion of iron nutriture data is contained in chapter 6. Calcium Calcium is essential in the formation of bones and teeth and in the maintenance of bone strength; it is also required for contraction of muscle, clotting of blood, and integrity of cell membranes. Low calcium intake is one of several factors associated with osteo porosis, a loss of bone mass that increases susceptibil- ity to fractures (DHHS/USDA, 1986). Although bone mass declines with age and osteoporosis is common in older persons (more so in women than in men), de- finitive data for estimation of the prevalence of osteoporosis are lacking. Although the precise rela- tionship of dietary calcium to osteoporosis has not been elucidated, it has been suggested that higher intakes of dietary calcium could increase peak bone mass during adolescence and early adulthood and de- lay the onset of bone fractures later in life (DHHS, 1988). Thus, increased consumption of foods rich in calcium may be especially beneficial for adolescents and young women. Adequate calcium intake is also important in later life to maintain bone mass. Cal- cium has also been found in some studies to be a fac tor in the control of hypertension although this effect is not certain and results of clinical studies must be regarded as inconclusive (DHHS, 1988; National Research Council, 1989). The JNMEC classified cal- cium as a component warranting public health moni- toring priority because of low intake by some groups. The per capita amount of calcium in the U.S. food supply in 1985 was greater than 900 milligrams (figure 11-39). These data indicate that, overall, the U.S. food supply contains an adequate amount of cal- cium for the population. Dairy products supply the greatest proportion of dietary calcium in the food supply (76.8 percent in 1985) (figure 11-40). The CSFII 1985-86 data indicate that the mean (4- day) calcium intake of women aged 20-49 years (630 milligrams per day) continues to be below the RDA (table II-102). The mean intake (562 milligrams per day) for women aged 40-49 years was lower than for younger women (20-29 years). This finding is of concern because postmenopausal women (especially in the early stages of menopause) excrete more calcium than younger women and require more dietary calcium to maintain calcium balance. These data also indicate that black women in these age groups have lower mean calcium intakes than white women and that lower calcium intakes are associated with lower income and lower education levels. Despite the intake differences by race, black women have a much lower prevalence of osteoporosis, sug- gesting that factors other than calcium intake con- tribute to the prevalence of osteoporosis. In a multi- variable analysis, factors found to be associated with higher intakes of calcium by women were: being younger, being in a racial group other than black, having more education, having a higher income, being employed part-time, being a participant in the Food Stamp Program, living in a central city, living in the Midwest or West, having a child aged 1-5 years in the household, being taller, being a regular supplement user, being pregnant, and lactating (Welsh, 1988). For children aged 1-5 years, mean intake of calcium over 4 days was 804 milligrams per day (table II-104). The median intake for this group was 769 milligrams per day, indicating that over half of these children consumed less than the RDA. As with the women, mean intakes of black children were lower than those of white children. Lower mean intakes of calcium of children were also associated with lower income levels and lower education level of the mother. Trends in calcium intake since 1971-74 are available based on 1-day food consumption data (table II-106). For males aged 20-49 years, data from NHANES I, NFCS 1977-78, NHANES II, and CSFII 1985-86 indicate that mean calcium intakes have ranged from about 750 milligrams per day to about 1,100 milli- grams per day with little change over this time period, suggesting that calcium intakes for most adult men are probably adequate. For women aged 20-49 years, mean calcium intakes have ranged from 530 to 690 milligrams per day, well below the RDA. Each of the surveys shows the same trend of lower calcium intake within this age group. For children aged 1-5 years, mean calcium intakes ranged from 750 to 920 milligrams per day. Overall, levels of calcium intake seem to have stayed fairly constant over this 15-year period. 66 Clinical and biochemical assessments of calcium status have not been included in NNMS surveys because methods for direct assessment of calcium status appropriate for field surveys are not available. Measurements of bone density are planned for NHANES III. Analysis of data collected in 1980 on supplement use indicated that 13.5 percent of the U.S. population and 34.9 percent of supplement users consumed calcium supplements (Stewart et al., 1985). The median and 95th percentiles of calcium intake supplied by supple- ments were 16 and 113 percent, respectively, of the 1980 RDA for calcium, indicating that calcium supplements were consumed in relatively small doses. Promotion and use of calcium supplementation, as well as use of calcium-fortified foods, as a means of preventing osteoporosis has occurred since the 1980 survey. This promotion provides an added impetus to monitor calcium intake, particularly intake of cal- cium from supplements and calcium—fortified foods by adult women. Conclusion: Calcium is considered to be a current public health issue. The low intakes of calcium in vulnerable groups, especially women, suggest a reason for concern. The high prevalence and severity of osteoporosis, which is possibly related, in part, to low calcium intake of adolescents and young women, provide sufficient evidence for a public health concern. Calcium should be consid- ered a nutrient about which there is public health concern even if there is some question about its exact role in health disorders. Monitoring the intake of calcium and including assessments of bone status in NNMS surveys is warranted, as is investigating the possible overuse of calcium supplements by adults. Phosphorus Phosphorus is an essential mineral because, together with calcium, it forms an insoluble compound that gives rigidity and strength to bones and teeth; as a part of other biological components, phosphorus is essential for many metabolic reactions. Some studies in experimental animals have suggested that a high ratio of phosphorus to calcium in the diet may decrease calcium utilization; however, more recent studies in humans suggest that the calcium:phos- phorus ratio is less important than the adequacy of calcium intake. The JNMEC classified phosphorus as a component warranting less consideration for con- tinued public health monitoring. Phosphorus is widely distributed in foods and phos- phates are often used in food processing. Thus, dietary deficiencies of phosphorus are very unlikely (National Research Council, 1980). The per capita amount of phosphorus in the U.S. food supply in 1985 was greater than 1500 milligrams (figure II-41). These data indicate that the U.S. food supply contin— ues to provide a generous amount of phosphorus for the population. The major sources of phosphorus in the food supply in 1985 were dairy products (35.7 percent); meat, poultry, and fish (29.2 percent); and grain products (13.2 percent) (figure 11-42). The CSFII 1985-86 (4-day) data indicate that the mean phosphorus intake of women aged 20-49 years (975 milligrams per day) continues to be above the RDA (table II-107). Mean intakes were lower in black women and in women with lower economic and educational status; however, in all cases, mean intakes were at or above the RDA for phosphorus. For children aged 1-5 years, mean intake of phos- phorus over 4 days was 1010 milligrams per day (table 11-108). Mean intakes for all racial, education, eco- nomic, geographic, and urbanization groups were greater than the RDA for these children. Information on trends in phosphorus intake since 1971-74 is available based on 1-day food consump- tion data (table II-109). For males aged 20-49 years, data from NHANES I, NFCS 1977-78, NHANES II, and CSFII 1985-86 indicate that mean phosphorus intakes have varied little and have been consistently well above the RDA for adult men over this time period. For women aged 20-49 years, mean phosphorus intakes have also been above the RDA although their intakes have been lower than those of men. For children aged 1-5 years, mean phosphorus intakes have been above the RDA. The elderly have lower intakes than young adults. Overall, levels of phosphorus intake within age and sex groups seem to have stayed fairly constant over this 15-year period. Direct biochemical or clinical assessments of phos— phorus status are not available for use in field surveys. The analysis of data collected in 1980 on supplement use indicated that 8.4 percent of the U.S. population and 21.8 percent of supplement users consumed phosphorus from supplements (Stewart et al., 1985). Among users, the median and 95th percentiles of phosphorus intake supplied by supplements were 14 and 73 percent of the RDA, respectively, indicating that phosphorus supplements were consumed in relatively small doses. Conclusion: Phosphorus is not considered to be a current public health issue. Intakes appear to be 67 adequate, and no other evidence exists to suggest a public health problem. Monitoring should continue at a relatively low level. Magnesium Magnesium is an essential mineral in the diet, used in bone formation, protein synthesis, energy release from muscle glycogen, and body temperature and blood pressure regulation. The JNMEC classified magnesium as a component requiring further inves— tigation because of low dietary intakes. The magnesium content of the food supply has de- clined since the beginning of this century, but a slight increase has occurred since 1980 (figure 11-43). Cur- rently, the major sources of magnesium in the food supply are dairy products (19.8 percent); grain products (17.7 percent); vegetables (15.8 percent); meat, poultry, and fish (15.4 percent); and legumes, nuts, and soy (13 percent) (figure 11-44). There has been a large decline in the percentage of magnesium obtained from grain products since 1909-13. Data from the CSFII 1985-86 show that the mean (4- day) dietary intake of magnesium for women aged 20-49 years falls below the RDA (table II-110). This finding raises some concern about the possibility of inadequate dietary intake of magnesium in this group. However, the RDA is based primarily on balance studies of young men in which widely varying results were obtained (National Research Council, 1980). Among women in the CSFII 1985-86, mean intakes were lower for blacks than for whites, for persons below poverty than for persons above poverty, and for persons with a lower level of education than for persons with a higher level of education. Notable differences by age, region, or urbanization were not observed. The mean (1-day) intake of males of this same age group was higher than that of women and slightly lower than the RDA. The mean intake of children aged 1-5 years (table II- 112) was also close to the RDA for this group, and the effects of race, poverty status, and education level on dietary intake were not as striking as they were for women. There has been little change in the mean individual intakes of these groups (small increases for most groups) since the NFCS 1977-78. There are no clinical or biochemical indicators of magnesium status available in surveys of the NNMS (indeed, there are no good indicators of status available to measure). Magnesium deficiency has not been reported to occur in response to low dietary intake alone (National Research Council, 1980). Generally, deficiency occurs when conditions such as alcoholism or prolonged vomiting or diarrhea inter— fere with absorption or when magnesium—free parenteral solutions are administered. Supplemental magnesium was consumed by approxi- mately 14 percent of the adult U.S. population in 1980 (Stewart et al., 1985). The median intake of magnesium from supplements was 22 percent of the RDA and the 95th percentile of intake was nearly equal to the RDA among users. Conclusion: Magnesium is not considered a current public health issue. Dietary intakes appear to be low, but there are no other data on magnesium status available and magnesium deficiency is very unlikely to result from low dietary intake alone. Further research on magnesium requirements and assessment of magnesium status would be desir- able. Current information supports continued monitoring at a low level. Sodium Sodium is an essential mineral in the diet. The JNMEC classified sodium as a component warranting public health monitoring priority because of concerns about high intakes. The JNMEC report (DHHS/ USDA, 1986) noted that when sodium intake exceeds excretion, there may be an increase in the sodium content of the extracellular fluid where most of this mineral is found in the body. The result of this imbalance between sodium intake and excretion is edema, which is manifest by swelling of the hands, feet, and legs. Epidemiological studies indicate a relationship between a high sodium intake and the occurrence of high blood pressure and stroke (DHHS, 1988). Clinical studies have generally shown that sodium restriction reduces elevated blood pressure (DHHS, 1988). However, the response to sodium is variable; while some individuals maintain normal blood pressure over a wide range of sodium intakes, others appear to be "salt-sensitive" and display increased blood pressure in response to high sodium intakes (DHHS, 1988). The U.S. Dietary Guidelines recommend that Ameri— cans avoid too much sodium (USDA/DHHS, 1985). The text accompanying the Guidelines and The Surgeon General's Report on Nutrition and Health (DHHS, 1988) cite the major reasons for recommen— dations that most Americans reduce their sodium intakes, even though all individuals are not equally susceptible to the effects of sodium. These include the lack of a practical indicator for individual sodium sensitivity, the potential benefit to persons whose blood pressures are affected by sodium, and the lack 68 of harm from moderate sodium restriction. The Food and Nutrition Board (National Research Council, 1980) considered 1,100 to 3,300 milligrams of sodium per day as a safe and adequate intake for healthy adults, with 3,300 milligrams as the upper limit for adults. The Surgeon General's Report on Nutrition and Health (DHHS, 1988) estimated, on the basis of survey results and industry information, the current average total sodium intake for adults to be in the range of 4,000 to 6,000 milligrams per day, well above the advised "safe and adequate" level. Dietary sources of sodium include foods in which it occurs naturally, salt and sodium-containing com- pounds added to foods during processing, salt added in cooking or at the table, and drinking water. The JNMEC report (DHHS/USDA, 1986) estimated the sodium content of municipal water supplies to aver- age 28 milligrams per liter, and noted that many water softeners increase the sodium content of drink- ing water. In addition, many over-the-counter med- ications such as antacids contain sodium. This multi- plicity of sources makes it difficult to obtain accurate estimates of the total intake of sodium. The quantity supplied by salt used in cooking and at the table is particularly difficult to assess. The recent introduc- tion of many low-sodium foods by food manufac- turers has added an additional complication. Many foods in the marketplace are labeled with terms such as "unsalted," "no salt added," or "without added salt." The FDA defines other terms to be used in nutrition labeling, such as "sodium free" (<5 milligrams per serving), "very low sodium" (35 milligrams or less per serving), "low sodium" (140 milligrams or less per ser— ving), or "reduced sodium" (a 75 percent reduction in sodium content). Subjects must be able to identify such products correctly in dietary reports, and food composition data for such products must be included in databases to produce accurate estimates of intake for persons consuming these foods. Data on sodium in the U.S food supply are not available. The CSFII 1985-86 data provide estimates of indi- vidual sodium intake including naturally occurring sodium, sodium contributed by compounds used in food processing, and an assumed amount used in food preparation. These estimates exclude sodium from salt added at the table. The mean (4-day) intake of women aged 20-49 years was 2,372 milligrams per day and many intakes exceeded the upper limit of the estimated range of safe and adequate intakes (table 11-114). Intakes were slightly higher in whites than in blacks, slightly higher in those above poverty than in those below poverty, and higher among those with higher levels of education. The mean sodium intake of children aged 1-5 years was 2,036 milligrams per day; race and socioeconomic factors seemed to have less influence on the sodium intakes of children than on the intakes of women (table 11-115). Estimates of the mean (1-day) intakes of sodium for various age and sex groups are available from NNMS surveys conducted during the period 1971-86 (table 11-116). These data must be interpreted with caution because different assumptions about the salting of foods were used in different surveys. Esti- mates from the NHANES II and CSFII 1985-86, however, were similar. Mean intakes in excess of 3,300 milligrams per day were reported for males aged 12-49 years. The only health condition assessed in the NNMS related to sodium intake is hypertension. (Hyperten-— sion is also influenced by a variety of other factors in addition to sodium.) New data on the age-adjusted prevalence of hypertension in adults are available for the three Hispanic ethnic groups in HHANES (tables 11-119 and 11-120) and these are compared to esti- mates for non-Hispanic whites and blacks from NHANES II (tables II-121 and II-122) in table 3-14. Hypertension is defined herein (as in the JNMEC report) as a condition in which an individual had an average systolic blood pressure greater than 140 mm mercury, or had an average diastolic blood pressure greater than 90 mm mercury, or was taking anti- hypertension medication. Males and females of all three Hispanic ethnic groups had lower prevalences of hypertension than did non-Hispanic whites and blacks; prevalence estimates were highest in blacks. Among Mexican Americans and non-Hispanic persons, hypertension was slightly more prevalent in those below poverty than in those above poverty (tables 11-120 and 11-122). See chapter 5 for a more detailed discussion of hypertension. Table 3-14. Age-adjusted prevalence of hyper- tension’ in persons aged 20-74 years, by ethnic group or race: Hispanic Health and Nutrition Examina- tion Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 Male Female Mexican American 23.9 20.3 Cuban 20.7 144 Puerto Rican 214 19.2 Non-Hispanic white 33.8 25.1 Non-Hispanic black 41.6 43.8 1 See text for definition of hypertension. Conclusion: Sodium is considered a current public health issue. Reported dietary intakes are high in many persons relative to estimates of safe and ade- quate levels of intake; reported intakes do not account for all sources of sodium. The prevalence of hypertension, which is related in some persons to sodium intake as well as other factors, is high in all adult groups examined (14-44 percent). Because of the serious, and largely preventable, deleterious effects of elevated blood pressure, a high level of monitoring effort is warranted. Blood pressure measurements should continue to be included in surveys and efforts to improve and validate the as— sessment of total sodium intake should be pursued. Potassium Potassium is an essential mineral and is a cation like sodium; however, potassium is concentrated in the cell rather than in the extracellular fluid like sodium. Population studies have shown that low potassium intake is associated with high blood pressure (DHHS, 1988). Although sodium intake may be the most important dietary determinant of blood pressure, variation in the ratio of sodium to potassium in the diet may also affect blood pressure under certain conditions (National Research Council, 1980). The Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure (1988) has noted that high potassium intake has a modest blood pressure-lowering effect, but that evidence in this regard is still developing. The Surgeon General's Report on Nutrition and Health (DHHS, 1988) also noted that the relationship between sodium, potas— sium, and blood pressure in normotensive adults may be dependent on the family history of hypertension and that further research is needed to evaluate the effects on blood pressure of both increased potassium and reduced sodium intake. The National Research Council (1980) has set Estimated Safe and Adequate Daily Dietary Intake (ESADDI) ranges for potassium. The JNMEC did not evaluate potassium in its report. The per capita amount of potassium in the U.S. food supply had been declining since 1909 but has in- creased slightly after reaching a nadir in 1980-81 (3,300 milligrams) to 3,460 milligrams in 1985 (figure 11-45). In 1985, the food groups contributing the major shares of potassium in the food supply were vegetables (27.1 percent); dairy products (21.1 per- cent); meat, poultry, and fish (19.3 percent); and fruits (11.3 percent) (figure 11-46). Based on 4-day data from the CSFII 1985-86, the mean potassium intake of women aged 20-49 years was 2,073 milligrams per day (table 11-123). At least 25 percent of mean intakes of women in this age group fell below the lower limit of the range suggested to be safe and adequate. Intakes were higher for white women than for black women, for women above poverty than those below poverty, and for women with more education. For children aged 1-5 years (table II-124), mean intakes were almost all above the lower limit of the safe and adequate range, and some intakes exceeded the upper limit. Few consistent changes in potassium intakes (based on 1-day data) have been detected in surveys of individual intake over the period 1971-86 (table II- 125); however, the improvements in the food composition database for potassium over the same time period may have obscured changes in intake if any occurred. There are no biochemical or clinical measurements of potassium nutritional status available; serum potassi— um levels may be measured but wide variations in intake are not reflected in changing levels of circulat- ing potassium because potassium homeostasis is reg- ulated tightly by excretion through the kidney. In 1980, only 4.5 percent of the adult U.S. population obtained potassium in the form of supplements (Stewart et al., 1985). Among users, the median and 95th percentile levels of intake barely exceeded the upper limit of the ESADDI. Conclusion: Potassium is considered a potential pub- lic health issue for which further study is needed. Intakes are lower than recommended levels in a substantial number of persons in the population. Further research on the role of potassium intake in the regulation of blood pressure and on the assess— ment of potassium status is needed to elucidate the public health significance of the low intakes observed. Copper Copper is an essential mineral that functions in a variety of enzymes and other proteins. It was not evaluated by the JNMEC. Copper deficiency in humans is rare; it is not known to occur among adults under normal circumstances, but has been diagnosed in malnourished children in Peru, in premature infants fed exclusively a cow milk formula, and in infants on prolonged parenteral nutrition (National Research Council, 1980). More recent studies by the Agricultural Research Service of USDA have sug- gested a role for copper in heart function, but further 70 study is needed in this area. The National Research Council (1980) has set a range of ESADDI for copper. The per capita amount of copper in the food supply has declined 19 percent since 1909 to a level of 1.7 milligrams in 1985 (figure 11-47). The contributions of vegetables and grain products to copper in the food supply have declined, while the contributions of legumes, nuts, and soy have increased; in 1985, the food groups contributing the major share of copper were meat, poultry, and fish (20.6 percent); vegetables (20.3 percent); grain products (18 percent); and legumes, nuts, and soy (17.7 percent) (figure 11-48). Estimates of dietary copper consumption from 4-day data collected in the CSFII 1985-86 indicate that over 90 percent of women aged 20-49 years and children aged 1-5 years had intakes below the range of ESADDI (tables II-126 and II-127). Several questions may be raised about these apparently low intakes in view of the rarity of copper deficiency. First, intake estimates are uncertain because analytical data on the copper content of foods are lacking in the nutrient composition database for approximately 30 percent of food sources. Second, the National Research Council (1980) has noted that the remarkably steady tissue concentrations of copper in adults in the United States are probably an indication of a sufficient dietary intake and strong homeostatic control. Early laboratory manifestations of copper deficiency include a low serum copper and decreased number of neutrophils in the blood. Serum copper levels were measured in NHANES II but not in HHANES; inter- pretative criteria to assess deficiency are uncertain. In the opinion of the EPONM, it does not appear that collection of such data should have a high priority at present. In 1980, approximately 12 percent of the adult U.S. population consumed copper in the form of supple- ments (Stewart et al., 1985). The median intake of users was 67 percent of the upper limit of the ESADDI and intake at the 95th percentile was equal to the upper limit of the ESADDI. Conclusion: Copper is not considered to be a current public health issue. Intakes appear to be low in a large number of persons in the population. Despite some unanswered questions about the estimation of intake and the assessment of status, the likelihood of a public health problem associated with copper is very low. Monitoring should continue at a low level, unless further research suggests more com- pelling reasons for concern. Zinc Zinc is an essential mineral in the diet and is a component of many enzymes. As such, it is involved in many metabolic processes including protein syn- thesis, wound healing, immune function, and growth and maintenance of tissues. The JNMEC classified zinc as a component requiring further investigation. Severe zinc deficiency characterized by hypogonadism and dwarfism has been observed in the Middle East, and evidence of milder forms of zinc deficiency (detected by biochemical and clinical measurements and responsiveness to increased zinc intake) has been found in several population groups in the United States (National Research Council, 1980). The per capita amount of zinc in the U.S. food supply was essentially the same in 1985 as in 1909, although fluctuations have occurred in the intervening years (figure 11-49). In 1985, the major sources of zinc were meats, poultry, and fish (48.7 percent); dairy products (19 percent); and grain products (12.6 per- cent) (figure II-50). There has been a large decline in the percentage of zinc obtained from grain prod- ucts since 1909-13. Data from the CSFII 1985-86 show that the mean (4- day) dietary intake of zinc for women aged 20-49 years is approximately half of the RDA and that a large percentage of intakes falls well below the RDA. This finding raises concern about the possibility of inadequate dietary intake of zinc and, consequently, zinc deficiency in this group. Among women in the CSFII 1985-86, mean intakes of zinc were lower for blacks than for whites, for persons below poverty than for persons above poverty, and for persons with a lower level of education than for persons with a higher level of education. There were no notable differences in zinc intake due to age, region, or urbanization. The mean intake (based on 1-day data) of males of this same age group was higher and closer to the RDA. The mean intake of children aged 1-5 years was also close to the RDA, and there was no notable effect of race, poverty status, region, or urbanization on dietary intake of zinc in children. The effects of education level on dietary zinc intake were not as striking for the children as they were for the women. Dietary zinc was not assessed in the NFCS 1977-78 or earlier surveys because adequate data on the zinc content of foods were not available at that time. Serum zinc levels were not measured in HHANES. Very few low values for serum zinc were detected in NHANES II. However, serum zinc is not a reliable indicator of zinc status because factors other than zinc deficiency (such as infection, inflammation, or acute inflammatory response) can influence its level. 71 Low zinc status has been reported to be associated with depressed growth, delayed sexual maturation, and impaired taste function in small groups studied in the United States. These associations have not been detected or (in most cases) measured in NNMS surveys. An estimated 13.5 percent of the adult population used supplements containing zinc in 1980 (Stewart et al, 1985). The median intake of users from supple- ments was 50 percent of the RDA. A substantial por— tion of these users consumed doses in excess of the RDA (approximately 5 percent consumed levels three times the RDA). Excessive consumption of zinc can interfere with copper metabolism. Conclusion: Zinc is considered a potential public health issue, for which further study is needed. Dietary intakes are lower than recommended levels in some groups, particularly women. The possibility of impaired zinc status is not supported by available biochemical or clinical data from the NNMS. However, findings from the clinical litera- ture suggest evidence of zinc deficiency in some groups in the United States. The significance of the observed low dietary intakes of zinc cannot be evaluated until additional research to determine zinc requirements and to develop better measures of zinc status is conducted. Further monitoring is warranted. Fluoride and Other Minerals Fluoride is a preventive factor in dental caries and may have some benefit in increasing bone mass under certain conditions (DHHS, 1988). The efficacy, safety, and cost—effectiveness of fluoride in the prevention of tooth decay have been well established (DHHS, 1988). Even though fluoride may occur naturally in water or be added to municipal water supplies, the JNMEC expressed concern that fluoride intake might be too low for many Americans to benefit. They classified fluoride as a food component warranting public health monitoring priority status. Data on intake were not available from the NNMS then, and are not available now. The Surgeon General's Report on Nutrition and Health (DHHS, 1988) also noted the uncertainty regarding current fluoride intakes. When the optimal level for fluoride addition to drinking water was set, the water supply was the major source of the fluoride. Now, sources include toothpaste, mouth rinses, topical applications, and beverages and foods prepared with fluoridated water. The availabil- ity of fluoride has increased, perhaps to levels that may induce mild dental fluorosis (mottling in developing teeth) (DHHS, 1988). The incidence of caries in children has declined as much as 30-50 percent in the past two decades; this decline has been attributed to increased fluoride from drinking water, food, toothpaste, mouth rinses, and topical applica- tions, but decreased intake of cariogenic foods and improved dental hygiene and care may also be con— tributing factors (DHHS, 1988). Based on the fore— going information, the EPONM considers fluoride to be a potential public health issue for which further study is needed. The EPONM agrees with the JNMEC concern that fluoride intake may be too low in some groups to provide maximal benefit, but data are not currently available that permit evaluation of this possibility. Assessments of fluoride intake that take all sources into account are warranted. Other minerals for which RDA or ESADDI have been set are iodine, manganese, chromium, selenium, and molybdenum; these are not monitored in the NNMS at present. Despite current research interest in these minerals and others (for example, boron), the EPONM does not consider the evidence with respect to public health issues regarding these minerals compelling enough to justify a recommendation for their inclusion in the NNMS at this time. Conclusions e In the United States today, the amounts of food available in the food supply and the nutrient content on a per capita basis are generally adequate to prevent undernutrition and deficiency-related diseases. Although some Americans may not have sufficient food for a variety of reasons, the supply of food that is available is abundant. e The NNMS does not provide sufficient population— based data to permit a full assessment of nutri- tional status in some groups for whom there are special concerns about nutritional status, such as young infants and pregnant and lactating women. In addition, some other groups whose nutritional status may reasonably be suspected to differ from that of the general population, such as the home- less, institutionalized persons, migrant workers, and Native Americans living on reservations, are not included in most of the current household- based surveys of the NNMS. Finally, very little information on the dietary and nutritional status of the elderly (a group for which standards for nutri— ent adequacy and normal physiologic status have been questioned) was available in the most recent NNMS data that were the focus of this evaluation. e Evidence from recent analyses of the U.S. food supply and from surveys of individual food 72 consumption suggests that some changes are occurring in eating patterns consistent with recom- mended dietary guidelines for Americans (USDA/ DHHS, 1985) to avoid too much fat, saturated fat, and cholesterol and to consume adequate amounts of starch and dietary fiber. Recent data indicate that consumers are increasingly choosing some lower—-fat alternatives within the meat and dairy product food groups and are increasing their consumption of grain products. Evidence available on dietary and nutritional status with respect to individual food components does not indicate substantial changes since the JNMEC report was completed in 1986. Consequently, the EPONM and JNMEC classifications of food compo- nents by public health monitoring priority are very similar (see table 3-6). The principal nutrition-related health problems experienced by Americans continue to be related to the overconsumption of some nutrients and food components, particularly food energy, fat, saturated fatty acids, cholesterol, sodium, and alcohol. — The high prevalence of overweight among adults in the United States is evidence that energy intakes exceed energy expenditures (probably because of low energy expenditures, although this possibility cannot be assessed currently in the NNMS); however, reported intakes of food energy do not exceed standards (Recommended Energy Intakes). The JNMEC noted that more than one-quarter of the adult U.S. population was overweight, based on data collected in NHANES II (1976-80). Data collected since then in the HHANES (1982-84) also indicate a high prevalence of overweight in three Hispanic groups not previously studied (26-42 percent), especially in Mexican-American and Puerto Rican women (40 and 42 percent, respectively). Overweight is a controllable risk factor for car— diovascular disease, high blood pressure, and diabetes. — Intakes of total fat and saturated fat continue to be higher than the levels recommended by many authoritative groups; cholesterol intakes are high for adult men. These high intakes are reflected in the high prevalence (11-22 percent) of elevated levels of total serum cholesterol, as defined by the 1984 NIH Consensus Development Conference (NIH, 1985), found in nearly all adult groups aged 20-74 years in the United States. Elevated serum cholesterol levels constitute an important controllable risk factor for coronary heart disease. - Sodium intakes also exceed recommended levels in almost every group in the United States. Such intakes are of concern because of the sensitivity of blood pressure in some persons to sodium intake. Hypertension is prevalent (14-44 per-— cent) in adult groups aged 20-74 years in the U.S. population. Hypertension is a controllable risk factor for cardiovascular disease and stroke. - Although consumption of excessive alcohol does not appear to be prevalent in a large proportion of the population, reported intakes are high in a large number of Americans and the serious nature of the health and social consequences of such intakes justifies public health concern. e In spite of the general adequacy of the supply of nutrients, there is evidence of inadequate indi- vidual dietary intake and/or impaired nutritional status in some subgroups in the population with respect to a few vitamins and minerals. — Iron deficiency continues to be the most common single nutrient deficiency, even though some recent hematological and biochemical evidence from the NNMS suggests that its prevalence has declined in children aged 1-5 years. Among groups that are assessed adequately in the NNMS, women of childbearing years and young children are at greatest risk for iron deficiency. — Although less evidence is available, the calcium status of women is a concern. The high preva- lence of osteoporosis in later life is suggestive that calcium intake of many women may be inadequate to permit the accretion of maximal bone mass in early adulthood and/or to maintain bone mass later in life. - Limited evidence from biochemical assessments suggests that the vitamin A, vitamin C, and folacin nutritional status of some subgroups of the population might be improved. - Intakes of zinc and vitamin B6 are also low, and poor status has been reported in some population groups in the clinical literature, but further study is needed to assess the health consequences of the reported intakes in U.S. population groups. The risk of nutrition-related disorders is gen- erally greater in low-income groups than in groups with higher incomes. —- The prevalences of both overweight and iron deficiency are greater in women below poverty than in women above poverty. 73 - The intakes of several vitamins and minerals are lower in persons below poverty than in persons above poverty. This finding is also highlighted in the low-income component of the CSFII 1985-86 (USDA, 1986b, 1987, 1988b). Women in the low- income survey had lower intakes of food energy than women in the all-income survey. Intakes of vitamin E, vitamin B6, folacin, calcium, mag- nesium, iron, and zinc were low in women in both surveys, but lower in the low-income survey than in the all-income survey. Low- income women and children who lived in households that participated in the Food Stamp Program had nutrient intakes that were gener- ally the same or higher than those of low-income women and children living in households that did not participate in the program. eo The ability of the EPONM to examine excessive intakes of vitamins and minerals, and possibly to assess consequences of nutrient toxicity, was lim- ited because none of the available NNMS surveys that assess nutrient intake from food included quantitative estimates of nutrient intake from vitamin/mineral supplements. e Although the data available to the EPONM for their update on dietary and nutritional status of the U.S. population were not equivalent to the data reviewed by the JNMEC, in terms of the popula- tions surveyed, the conclusions of the EPONM are very consistent with those of the JNMEC. The results of recently completed and ongoing national surveys of dietary and nutritional status by USDA and DHHS will provide a more extensive database for further evaluation of the nutritional status of the U.S. population and various subgroups in future reports on the NNMS. References Cited Alcohol Epidemiologic Data System, D. Doernberg, F. S. Stinson, and G. D. Williams. 1986. Surveillance Report #2. Apparent Per Capita Alcohol Consumption: National, State and Regional Trends, 1977-1984. National Institute on Alcohol Abuse and Alcoholism. Washington: CSR, Inc. American Cancer Society. 1984. Nutrition and Can- cer, Cause and Prevention. An American Cancer Society Special Report. Ca--A Cancer Journal for Clinicians. 34:121-126. American Heart Association. 1986. Dietary Guidelines for Healthy Adult Americans. Circulation 74:1465A. Bieri, J. G. 1976. Vitamin E, in Present Knowledge in Nutrition. 4th edition. New York: The Nutrition Foundation, Inc. Bradstock, M. K.,, J. S. Marks, M. R. Forman, et al. M985. The Behavioral Risk Factor Surveys: III. Chronic Heavy Alcohol Use in the United States. Am. J. Prev. Med. 1:15-20. \Bradstock, M. K.,, J. S. Marks, M. R. Forman, et al. 1987. Drinking-Driving and Health Lifestyle in the United States: Behavioral Risk Factors Surveys. J. Stud. Alcohol. 48:147-152. Capps, O., Jr., 1986. Changes in Domestic Demand for Food: Impacts on Southern Agriculture. South. J. Agric. Econ. 18:25-36. Capps, O., Jr., J. R. Tedford, and J. Havlicek, Jr. 1985. Household Demand for Convenience and Non- convenience Foods. Am. J. Agric. Econ. 67:862-869. Connor, J. M. 1981. Food Product Proliferation: A Market Structure Analysis. Am. J. Agric. Econ. 63:607-617. Cronin, F. J. 1988. Dietary Fiber in the Diets of Women. Paper presented at the American Dietetic Association Conference for Advanced Practice and Research, Chicago, Ill. March 12. Dresser, C. M. 1983. From Nutrient Data to a Data Base for a Health and Nutrition Examination Survey: Organization, Coding, and Values—-Real or Imputed. Paper presented at the Eighth National Nutrient Data Base Conference. MinneapolisS, Minn. July 25. Dreyfus, P. M. 1988. Diet and Nutrition in Neuro- logic Disorders, in M. E. Shils and V. R. Young, eds. Modern Nutrition in Health and Disease. Tth edition. Philadelphia: Lea & Febiger. Economic Research Service, D. Smallwood and J. Blaylock. 1981. Impact of Household Size and Income on Food Spending Patterns. Technical Bulletin No. 1650. Washington: U.S. Department of Agriculture. Gentry, E. M.,, W. D. Kalsbeek, G. C. Hogelin, et al. 1985. The Behavioral Risk Factor Surveys: II. Design, Methods, and Estimates from Combined State Data. Am. J. Prev. Med. 1:9-14. Hepburn, F. N. 1987. Food Consumption/Composition Interrelationships, in U.S. Department of Agriculture, Research on Survey Methodology. Proceedings of a Symposium Held at the 71st Annual Meeting of the Federation of American Societies for Experimental Biology. Washington, April 1. Administrative Report 74 No. 382. Hyattsville, Md.: Information Service. Human Nutrition Horwitt, MK, C. C. Harvey, C. H. Dahm, Jr., and M. T. Searcy. 1972. Relationship between tocopherol and serum lipid levels for determination of nutritional adequacy. Ann. N.Y. Acad. Sci. 203:223-236. Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure. 1988. The 1988 Report of the Joint National Committee on Detec— tion, Evaluation, and Treatment of High Blood Pressure. Public Health Service. NIH Pub. No. 88- 1088. Bethesda, Md.: National Institutes of Health. Kinsey, J. 1983. Working Wives and the Marginal Propensity to Consume Food Away From Home. Am. J. Agric. Econ. 65:10-19. Life Sciences Research Office, S. M. Pilch, ed. 1985. Assessment of the Vitamin A Nutritional Status of the U.S. Population Based on Data Collected in the Health and Nutrition Examination Surveys. Contract No. FDA 223-84-2059. Bethesda, Md.: Federation of American Societies for Experimental Biology. Life Sciences Research Office, F. R. Senti and S. M. Pilch, eds. 1984. Assessment of the Folate Nutritional Status of the U.S. Population Based on Data Collected in the Second National Health and Nutrition Exam- ination Survey, 1976-1980. Contract No. FDA 223- 83-2384. Bethesda, Md.: Federation of American Societies for Experimental Biology. Merrill, A. L., and B. K. Watt. 1973. Energy Values of Foods . . . Basis and Derivation. U.S. Department of Agriculture. Agriculture Handbook No. 74. Washington: U.S. Government Printing Office. Moshfegh, A. 1988. Vitamin B-6: Levels of Intake and Associated Factors in the Diets of Women 19-50 Years of Age. Paper presented at the American Dietetic Association Conference for Advanced Practice and Research, Chicago, Ill. March 12. National Cancer Institute. 1984. Cancer Prevention. DHHS Pub. No. 84-2671. Public Health Service. Bethesda, Md.: National Institutes of Health. National Center for Health Statistics, C. A. Schoenborn. 1988. Health Promotion and Disease Prevention: United States, 1985. Vital and Health Statistics. Series 10, No. 163. DHHS Pub. No.(PHS) 88-1591. Public Health Service. Washington: U.S. Government Printing Office. National Heart, Lung, and Blood Institute. 1987. Facts About . . . Blood Cholesterol. NIH Pub. No. 88-2696. Public Health Service. Washington: U.S. “Schatzkin, A., D. Y. Jones, R. N. Hoover, et al. 1987. Government Printing Office. National Institutes of Health Consensus Development Panel. 1985. Lowering Blood Cholesterol to Prevent Heart Disease. JAMA. 253:2080-2086. National Institutes of Health, Expert Panel on Detec- tion, Evaluation, and Treatment of High Blood Cho- lesterol in Adults. 1987. National Cholesterol Education Program. Bethesda, Md.: National Heart, Lung, and Blood Institute. National Research Council, Committee on Diet and Health. 1989. Diet and Health: Implications for Reducing Chronic Disease Risk. Washington: National Academy Press. National Research Council, Committee on Dietary Allowances. 1980. Recommended Dietary Allowances. 9th revised edition. Washington: National Academy of Sciences. Padberg, D. I, and R. E. Westgren. 1983. Adaptability of Consumers and Manufacturers to Changes in Cultural Patterns and Socioeconomic Values, in P. L. Farris, ed. Future Frontiers in Agricultural Marketing Research. Ames, Iowa: Iowa State University Press. Perloff, B. 1988. Assessment of Change in Total Fat Data 1977-1985. Prepared for Expert Panel on Nutrition Monitoring. Raunikar, R., C. L. Huang, and J. C. Purcell. 1985. The Changing United States Food Market. J. Agri— business. 3:42—-45. Redman, B. J. 1980a. The Impact of Women's Time Allocation on Expenditure for Meals Away From Home and Prepared Foods. Am. J. Agric. Econ. 62:234-2317. Redman, B. J. 1980b. Household Expenditure on Meat Versus Nonmeat Sources of Protein in the United States. South. J. Agric. Econ. 12:51-54. Rubinstein, H. M., A. A. Dietz, and R. Srinavasan. 1969. Relation of Vitamin E and Serum Lipids. Clin. Chim. Acta. 23:1-6. Sexauer, B. 1979. The Effect of Demographic Shifts and Changes in the Income Distribution on Food- Away-From-Home Expenditure. Am. J. Agric. Econ. 61:1046-1057. Sexauer, B. H., and J. S. Mann. 1979. Food Spending in Single-Person Households. Natl. Food Rev. (Summer):32-33. 75 Alcohol Consumption and Breast Cancer in the Epidemiologic Follow-up Study of the First National Health and Nutrition Examination Survey. N. Engl. J. Med. 316:1169-1173. Stewart, M. L., J. T. McDonald, A. S. Levy, et al. 1985. Vitamin/Mineral Supplement Use: A Telephone Survey of Adults in the United States. J. Am. Diet. Assoc. 85:1585-1590. U.S. Bureau of the Census. 1977. Projections of the Population of the United States: 1977-2050. Current Population Reports. Series P-25. No. 704. Washing- ton: U.S. Government Printing Office. U.S. Bureau of the Census. 1986. Statistical Abstract of the United States: 1987. 107th edition. Washington: U.S. Government Printing Office. U.S. Department of Agriculture. 1985. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII Report No. 85-1. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture. 1986a. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Men 19-50 Years, 1 Day, 1985. NFCS, CSFII Report No. 85-3. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture. 1986b. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII Report No. 85-2. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture. 1987. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1986. NFCS, CSFII Report No. 86-2. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture. 1988b. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 4 Days, 1985. NFCS, CSFII Report No. 85-5. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture, K. L. Bunch. 1987. Food Consumption, Prices, and Expenditures 1985. Statistical Bulletin No. 749. Washington: U.S. Government Printing Office. U.S. Department of Agriculture, M. Hogbin, and L. Fulton. 1989. Good Sources of Nutrients. Administrative Report 371. Hyattsville, Md.: Human Nutrition Information Service. U.S” Department of Agriculture, N. Raper, and R. Marston. 1988a. Nutrient Content of the U.S. Food Supply. Tables of Nutrients and Foods Provided by the US. Food Supply. HNIS Adm. Rep. No. 299-21. Hyattsville, Md.: Human Nutrition Information Service, USDA. U.S. Department of Agriculture and U.S. Department of Health and Human Services. 1985. Nutrition and Your Health: Dietary Guidelines for Americans. Home and Garden Bulletin No. 232. 2nd edition. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services. 1988. The Surgeon General's Report on Nutrition and Health. DHHS Pub. No. (PHS) 88-50210. Public Health Service. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1986. Nutrition 76 Monitoring in the United States--A Progress Report from the Joint Nutrition Monitoring Evaluation Committee. DHHS Pub. No. (PHS) 86-1255. Public Health Service. = Washington: U.S. Government Printing Office. U.S. Department of Health, Education, and Welfare. 1972a. Ten-State Nutrition Survey, 1968-1970. Vol. I-II. DHEW Pub. No. (HSM) 72-8130. Washington: U.S. Government Printing Office. U.S. Department of Health, Education, and Welfare. 1972b. Ten-State Nutrition Survey, 1968-1970. Vol. IV. DHEW Pub. No. (HSM) 72-8132. Washington: U.S. Government Printing Office. Welsh, S. 1988. Calcium: Levels of Intake and Associated Factors in the Diets of Women. Paper presented at the American Dietetic Association Conference for Advanced Practice and Research, Chicago, Ill. March 12. Williamson, D. F., M.R. Forman, N. J. Binkin, et al. 1987. Alcohol and Body Weight in United States Adults. Am. J. Public Health. 77:1324-1330. Chapter 4 Update of Selected Health Conditions and Behaviors: Relationship to Nutritional Status Introduction This chapter provides a brief summary that high- lights recent information from the NNMS and related data sources associated with prevention, development, or prevalence of a number of health conditions that are influenced by nutritional status; dietary behaviors that may affect health; and other behaviors that may affect nutritional status. This chapter is intended to serve as an update rather than an exhaustive review of these topics because most of the health conditions and behaviors included were discussed in the JNMEC report. Cardiovascular diseases and anemia are dis- cussed in detail in chapters 5 and 6, respectively; therefore, they are not included in this section. In addition, disorders such as cancer, kidney disease, AIDS, and other chronic diseases may have effects on nutritional status and may require special nutrition management. The nutritional consequences of such diseases are not appropriately included in the NNMS and are not considered in this report. A conceptual model, figure 4-1, adapted from the general conceptual model shown in chapter 1, illus- trates the topics and data considered. Components of the model that are most relevant to the discussions in this chapter are highlighted by the shaded boxes; individual topics noted with an asterisk are those for which some data are available. Potential NNMS and related data sources are represented by the numbers that appear above or below the boxes; numbers noted with an asterisk represent those surveys or studies from which data were obtained for consideration in this chapter. The information from the NNMS in this chapter bears on the model in numerous ways, from consideration of factors that influence food choice to health outcomes. Nutrition, Morbidity, and Mortality Several of the leading causes of death in the United States are associated with dietary and nutritional 77 status. These include cardiovascular disease, cere— brovascular disease, some cancers, chronic liver disease (especially cirrhosis), and diabetes mellitus. The extent to which changes in diet might alter mor- bidity and mortality caused by these diseases is un- certain. The JNMEC report compared age-adjusted death rates from these and several other causes for the years 1950 and 1983. Similar information includ- ing the most recent data from 1985 is presented in figure 4-2. No striking changes in death rates have occurred between 1983 and 1985. The JNMEC (DHHS/USDA, 1986) noted that the downward trend in mortality rates for cardiovascular disease, diabetes, and digestive system cancers occurred simultaneously with improvements in the diagnosis and treatment of these diseases, improved health and nutrition education, and growing public awareness of the contribution of lifestyle factors to health. Mortality experience and cause of death distribution for the NHEFS approximated closely the Vital Statis— tics data (Madans et al., 1986), suggesting the utility of this data set for examining associations between indicators of nutritional and dietary status and subsequent mortality. Health Conditions Influenced by Nutritional Status Obesity Obesity has adverse effects on health and longevity; it is associated with the prevalence and incidence of hypertension; hypercholesterolemia; non-insulin- dependent diabetes mellitus; certain cancers; osteoarthritis of weight bearing joints, especially the knees; and psychological problems (DHHS, 1988; National Research Council, 1989). In the JNMEC report, obesity was defined as the accumulation of excess body fat. Overweight was defined as excess body weight for height; the criterion for overweight was body mass index (BMI) at or above the 85th 8L NATIONAL FOOD SUPPLY ————> FOOD DISTRIBUTION ——> CONSUMPTION ——> NUTRIENT UTILIZATION A. HEALTH OUTCOME 1¢2 546+ 7+ 8+ 9+ Sanitation ; Housing $ Occupation + Other factors : 1 Environmental factors 1 Agricultural factors , Economic factors ' . . . 1 Policy considerations Away—from-home 3 Away—-from—home ( > food available food acquired 1*2 56789 10.11.13 16 : - 18 Food supply 12 5% 6% 7+ 829s | 1 14% 15+ 16+ 19+ | Exports ! ! Imports | 5% 6% 7+ 849 13« ) pimmnnd HR ! 14% 15% 16% 17« 19+ 3 Nr 3, ; Storage ; Primary representative action or consequence 3 \ Household food Household food Household food Fos : Influencing or mitigating factor available acquired consumed : ! te TTT National Nutrition Monitoring System and other data sources: 1 = CSFIl 1985-86, 2 = NFCS 1977-78, 3 = U.S. Food Supply Series, 4 = National Nutrient Data Bank, 5 = NHANES |, 6 = NHANES II, 7 = HHANES, 8 = NHEFS, 9 = NHIS, 10 = FLAPS, 11 = Total Diet Study, 12 = Vit/Min, 13 = Health and Diet Study, 14 = PedNSS, 15=PNSS, 16 = BRFSS, 17 = U.S. Vital Statistics, 18 = AEDS, 19 = NHES. See appendix lll for definitions of acronyms. Shaded boxes highlight portions of the model discussed; an asterisk (+) indicates data and data sources considered in this chapter. Figure 4-1. Conceptual model for update of selected health conditions and behaviors related to nutritional status (see text for explanation) Age—adjusted Death Rates 1983 1985 — Deaths per 100,000 320 280 240 200 160 120 80 40 Nex Diabetes mellitus 0 \! 2 Accidents and adverse effects Chronic liver neoplasms and influenza diseasa Diseases of heart vascular Figure 4-2. Age-adjusted death rates for selected causes of death: U.S. Vital Statistics, 1950, 1983, 1985 percentile of the NHANES II reference population and the criterion for severe overweight was body BMI at or above the 95th percentile of the same population. Although BMI is highly correlated with body fat, increases in BMI can also result from increases in lean body mass or large body frame size (DHHS, 1988). Therefore, HHANES data on the prevalence of overweight and severe overweight, rather than obesity, are presented and compared to NHANES II data for non-Hispanics in figures 4-3 and 4-4 (see also tables II-4 through II-9 in appendix II). The prevalence of overweight was generally higher in women than in men. Among the different ethnic groups compared, non-Hispanic white women showed the lowest prevalences of overweight, followed by Cuban women. Non-Hispanic black women had the highest prevalence, and the estimates for Mexican- American and Puerto Rican women fell between the two extremes. In general, socioeconomic status below poverty was associated with higher prevalences of overweight in females. Among the males, non-His- panic whites had the lowest and Mexican Americans had the highest prevalences of overweight. The prevalence of overweight and obesity is high in the United States and higher, at most ages, than the prevalences in Britain or Canada (Millar and Stephens, 1987). The exact prevalence estimated depends on the criteria used to assess weight status (BMI, desirable weight standards) and the population to which the criteria are applied. Nevertheless, BMI and the prevalence of obesity are associated with socioeconomic status. Flegal, Harlan, and Landis (1988a, 1988b) have examined secular trends in BMI in men and women aged 18-34 years with data from the National Health Examination Survey (NHES) 79 Prevalence of Overweight Mexican Non—Hisponic Non-Hispanic American Puerto Rican Cuban white black _.. SS RRR OC] a Percent 50 40 0% 20 oo 3 BSS 1 KS CC 0 2 Female Figure 4-3. Age-adjusted percent of overweight Hispanic and non-Hispanic persons, 20-74 years: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 Prevalence of Severe Overweight Mexican American Percent 50 Non—Hispanic Non-Hispanic Cuban white black RRR [1 a. Puerto Rican RN 40 30 Female Figure 4-4. Age-adjusted percent of severely overweight Hispanic and non-Hispanic persons, 20— 74 years: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 (1960-62), NHANES I (1971-74), and NHANES II (1976-80). Mean BMI for black and white men was similar and changed little over the time period studied. In both black and white men, mean BMI was higher with higher levels of education, except that mean BMI at the lowest education level increased over time. In both black and white men, higher income was associated with higher BMI. For young women, the mean BMI increased over this time period (1960-80) for both blacks and whites at all levels of education and income. However, higher mean BMI was observed in black and white women with lower educational levels; differences in BMI by education level increased over time. In addition, Flegal, Harlan, and Landis (1988a) observed that mean BMI of women tended to be lower in higher income categories but the differences at the various income levels became smaller over time. Additional analyses of these data by Gillum (1987) showed that rural and southern black women aged 25-74 years were more overweight than their urban, northern, and western counterparts. Overall, age— adjusted prevalences for overweight were 47.1, 46.8, and 48.1 percent for 1960-62, 1971-74, and 1976-80, respectively. Gillum (1987) also confirmed that overweight in black women was inversely related to family income and education. Self-reports of body weight and self-assessments of overweight are collected in other NNMS activities. Data from the BRFSS collected from 28 States and the District of Columbia in 1981-83 indicated that 23 percent of respondents considered themselves over— weight (Forman et al., 1986). Based on comparison with desirable weight—for—height tables, more blacks and Hispanics were classified as overweight than whites. More overweight adults than normal-weight adults had uncontrolled hypertension, were binge drinkers, and had a sedentary lifestyle (Forman et al., 1986). Data from this survey also suggest the prevalence of obesity differs for men and women among the states participating; however, no clear trend in the prevalence of overweight was noted in data from states participating for the full study period (1984-86) (CDC, 1988). Among women aged 19- 50 years surveyed in the CSFII 1985-86, those classified as overweight on the basis of BMI derived from self-reported body weight and height were more likely to be older (ages 35-50 years), to be black, and to have an income less than 130 percent of poverty. Moreover, these women less likely to report rigorous leisure activity and excellent health status than those classified as normal weight or underweight (Moshfegh, 1987). Somewhat similar observations were reported by Dawson (1988) based on analyses of data from the 1985 NHIS. Williamson, Kahn, and Remington (1988) have used the NHEFS data to estimate the 10-year incidence of overweight in a national cohort of adults aged 20-74 years. The incidence of overweight was defined for those who were not overweight at baseline as an increase in BMI to 227.8 and 227.3 for men and women, respectively, and was highest in those aged 35-44 years (23.4 and 18.3 percent for men and women, respectively) and lowest for those aged 65-74 years (7.0 and 7.6 percent for men and women, respectively). The incidence of major weight gain 80 (defined as an increase of =5.0 BMI units regardless of weight at baseline) was twice as high in women as in men, with the peak incidence occurring in persons aged 25-34 years (6.0 and 12 percent for men and women, respectively). Men who were underweight and women who were overweight at baseline had the highest incidence of major weight gain. Diabetes Diabetes mellitus includes a variety of conditions associated with the abnormal secretion or action of insulin from the pancreas and the resulting abnormal metabolism of glucose. Insulin—dependent (Type I) diabetes is distinguished clinically from non-insulin- dependent (Type II) diabetes by usual age of onset, pathology, and treatment. Most national statistics on the prevalence of diabetes do not distinguish between the two types. The development of diabetes is strongly influenced by genetic factors; obesity greatly increases the risk for developing Type II diabetes. The JNMEC report presented data from NHANES II on the prevalence of diabetes in the general U.S. population. Persons were considered to have diabetes if results of an oral glucose tolerance test indicated a diabetic condition or if diabetes was reported during the medical history. Update data available from the NNMS are those for the prevalence of diabetes in the three Hispanic groups in HHANES (table 4-1). (The response rate for participation in the oral glucose tolerance test in HHANES was generally low and was especially low for the Cubans and Puerto Ricans as compared to the Mexican Americans.) The oral glu- cose tolerance test was administered to a subsample in each survey according to the recommendations of the National Diabetes Data Group: subjects fasted overnight for 10-16 hours (9-17 hours for HHANES); the test was performed in the morning; a fasting blood sample was taken; subjects drank a solution containing 75 grams of glucose; and additional blood samples were taken after one hour and two hours. Diabetes was considered to be indicated by the results of the glucose tolerance test if the fasting plasma glucose was 7.8 mmol/L (140 milligrams/deciliter) or more, or if the fasting value was less than 7.8 mmol/L (140 milligrams/deciliter) and the two-hour value was 11.1 mmol/L (200 milligrams/deciliter) or more. Estimates of the prevalence of diabetes were, as might be expected, much higher in older (aged 45- 74 years) than younger persons (aged 20-44 years). Prevalence estimates were lowest for non-Hispanic whites and highest for Mexican Americans and Puerto Ricans. Recent analyses of the NHANES II data confirmed the strong association between the Table 4-1. Total prevalence of diabetes (sum of previously diagnosed diabetes and undiagnosed diabetes), by age group, survey, and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 (Flegal et al., 1988c) Age Prevalence Approximate 95 percent group Survey Ethnic group or race (percent) confidence interval 20-44 HHANES Mexican American 3.8 2.0-5.5 years Cuban 24 0.0-5.0 Puerto Rican 4.1 2.8-5.3 NHANES II Non-Hispanic white 1.6 11-22 Non-Hispanic black 3.3 0.0-5.7 45-74 HHANES Mexican American 23.9 20.8-27.1 years Cuban 15.8 10.5-21.1 Puerto Rican 26.1 22.2-30.1 NHANES II Non-Hispanic white 12.0 10.7-13.2 Non-Hispanic black 19.3 15.1-23.6 prevalence of diabetes and body weight (NCHS, 1987a). Persons who were 50 percent or more above ideal body weight had diabetes at five times the rate of persons who were at ideal weight or lighter. Kovar, Harris, and Hadden (1987) have pointed out that comparisons of data collected over the past 40 years with results of the oral glucose tolerance tests in NHANES II suggest strongly that only about one- half of persons with diabetes are aware they have the disease. That is, while 3.4 percent of the adult popu- lation have been diagnosed as diabetic, an additional 3.2 to 3.4 percent meet the criteria for the disease but have not been diagnosed as diabetic. The analyses of Kovar, Harris, and Hadden (1987) suggest that prev- alence of diabetes in the U.S. adult population may be twice the rate estimated from medical histories alone. Examination of data from the NHEFS showed that the age-adjusted death rates for white men and women aged 40-77 years with diabetes were twice the rates for persons without diabetes, with little difference by sex or age (Kleinman et al., 1987). The percentage of the excess mortality attributable to coronary heart disease was approximately 75 percent and 57 percent for men and women, respectively. The relative risk of death and ischemic heart disease mortality remained higher in persons with diabetes even after adjusting for systolic blood pressure, serum cholesterol, body mass index, and smoking. Cancer Associations of certain types of cancer with body weight and calories; intake of dietary fat and fiber; and consumption of fruits and vegetables and smoked, salted, and pickled foods have been suggested. With 81 respect to diet and cancer, The Surgeon General's Report on Nutrition and Health (DHHS, 1988) concluded the following: e A decrease in fat consumption by the general public might reduce risk for certain cancers. Maintenance of desirable weight was recom- mended. Intake of foods high in dietary fiber might decrease the risk for colorectal cancer. An increase in consumption of fruits and vegetables containing carotenoids for persons who consume low amounts of these foods might be beneficial. A decrease in alcohol intake among heavy drinkers would help to reduce the prevalence of cancers of the mouth, esophagus, pharynx, and perhaps other sites. Selenium intake should not increase above levels now in the average diet. Evidence does not justify a recommendation to the general public to decrease protein intake on the basis of its relationship to cancer. The public should continue to limit its intake of salt—pickled, salt-cured, and smoked foods to current low levels of consumption. The JNMEC report compared age-adjusted cancer death rates for 1950 and 1982. Similar information including the most recent data from 1985 is presented in figure 4-5. No striking changes between 1983 and Cancer Death Rates 1950 1983 1985 NNN Deaths per 100,000 240 200+ 160 120 80 40t 0m White males White females Black males Black females Figure 4-5. Age-adjusted death rates for malignant neoplasms, by race and sex: U.S. Vital Statistics, 1950, 1983, 1985 1985 are evident. The NHIS included a series of questions on cancer epidemiology and control in 1987. Nutrition-related questions addressed frequency of eating and portion sizes for 62 food items as well as use of vitamin and mineral supplements and changes in diet and cooking practices due to health reasons. Analyses of the data were not available to the EPONM. Data collected by components of the NNMS must be interpreted with considerable caution in examining associations of diet and cancer. Even the dietary data collected in the more comprehensive components of NNMS is incomplete when viewed in relation to the time lag between possible dietary factors and occur- rence of cancers. Similarly, Jabine (NCHS, 1987b) has noted that sensitive information such as occur— rence of cancer is more likely to be underreported by subjects in telephone and face-to-face interviews than less sensitive personal medical information such as occurrence of asthma or hay fever. Osteoporosis Osteoporosis is a skeletal disorder characterized by a decrease in the amount of bone so severe that frac— tures may occur even after minimal trauma. There is moderate evidence that low dietary calcium is a positive risk factor for osteoporosis (DHHS, 1988). Other positive risk factors include age, postmeno- pausal status, corticosteroid use, extreme immobility, alcohol consumption, low body weight, and cigarette smoking. Protective factors include black race, estrogen use, and heavy exercise. 82 The Surgeon General's Report on Nutrition and Health (DHHS, 1988) noted that defining the relationships between diet and osteoporosis is difficult because of the many dietary factors associated with bone mass (calcium, phosphate, vitamin D, protein, sodium, fluoride, calories, and alcohol), the universality of bone loss with age, the interaction of diet and lifestyle with genetic factors, and the difficulties in measuring bone mass in populations. Data on the prevalence of osteoporosis have not been collected in the NNMS, but some information is available to suggest the scope of the problem. Data from the 1986 National Hospital Discharge Survey indicate that 816,000 of a total of 10,716,000 patients aged 65 years and over were discharged from U.S. hospitals with a first— listed diagnosis of fracture of the neck of the femur. Data from the 1985 National Nursing Home Survey show that 66,300 patients (4.4 percent) had a diagnosis of the same fracture upon admission to a nursing home. Data on bone density are being collected in the NHANES III and will be available. Age-specific data on dietary intake patterns collected in the NNMS, particularly the HANES, may be useful for future analyses of associations of diet and osteoporosis. Dental Caries Dental caries, or tooth decay, are caused by a pro- gressive dissolution of mineral from tooth surfaces by acid produced by oral bacteria. Advanced disease can result in tooth loss. A causative role of dietary sugar (especially in sticky foods) in caries production and a protective role of fluoride are relatively well estab— lished. Ismail et al. (1987a, 1987b, 1988) have analyzed the dental exam data from the Mexican- American subjects in HHANES. The numbers of decayed, missing, and filled teeth for children from a national sample and Mexican-American children are shown in figures 4-6 and 4-7, respectively. With respect to findings in adults, Mexican Americans had lower scores for decayed, missing, and filled teeth overall, but higher numbers of untreated decayed teeth, than did participants of NHANES 1 (1971-74) of the same region. Low Birth Weight Low birth weight may result from inadequate fetal growth or premature birth, or both. The term low birth weight describes infants weighing less than 2,500 grams (about 5.5 pounds) and the term de- scribes very low birth weight as infants weighing less than 1,500 grams (about 3.3 pounds). The lower the 1—4 DMFT 39.8% Zero DMFT 36.6% Figure 4-6. Percent distribution of children aged 5-17 years, according to the number of decayed, missing, and filled teeth (DMFT): National Dental Caries Prevalence Study, 1979-80 Zero DMFT 46.2% Figure 4-7. Percent distribution of Mexican—-Amer- ican children from the Southwest United States, aged 5-17 years, according to the number of decayed, missing, and filled teeth (DMFT): Hispanic Health and Nutrition Examination Survey, 1982-84 (Ismail et al., 1988) birth weight the greater the risk of neonatal death (DHHS, 1988). Birth weight is a strong determinant of the chances of survival and postnatal growth, development, and health of the infant. Low birth weight is also associated with increased morbidity, congenital abnormalities, and increased susceptibility to infection. Rates of low birth weight are linked to differences in mortality rates among subpopulation groups. Vital Statistics data (1983-85) are available to update information presented in the JNMEC report (figure 4-8). Black women are twice as likely as white women to have a low birth weight baby. The average annual rate of low birth weight from 1983-85 was 12.5 percent for nonwhite women 83 and 5.6 percent for white women. The ratio of approximately two to one of low birth weight between black and white women has remained fairly constant over the past two decades. Such racial differences (much higher prevalences of low birth weight infants in blacks) are detected even in mothers at low risk (Kleinman and Kessel, 1987). V 7, Vi / Per 100 live births 14 12 10 Vii oO N + OO T T Tv T 1973-75 1978-8 Oo 1983-85 Figure 4-8. Percent of infants with low birth weight (2,500 grams or less): U.S. Vital Statistics, 1973-75, 1978-80, 1983-85 Several risk factors are associated with low birth weight. These include, but are not limited to, lack of or late entry into the prenatal care system, high par- ity, unintended pregnancy, teenage and older mater— nal age pregnancy, being unmarried, previous low birth weight, poor obstetrical health history, anemia, chronic diseases, low socioeconomic status, low maternal weight gain, smoking, and alcohol and drug abuse. Any one or more of these factors are useful in defining high-risk pregnancy and in targeting appro- priate interventions. Several studies have shown that nutrition has a posi- tive influence on birth weight; however, the extent of its effect is difficult to evaluate because of the inter- relationship of prepregnancy weight, weight gain during pregnancy, and socioeconomic status. Data from the 1987 PNSS are presented in tables 4-2 through 4-4. These data are not from a nationally representative sample but rather represent a low- income, high-risk population of pregnant women from participating states. Nonetheless, they clearly demonstrate the association of smoking with low birth weight. In this population, the prevalence of low birth weight was lowest in whites and Hispanics and highest in blacks and Native Americans. The Table 4-2. Prevalence of low birth weight (<2,500 grams), by race or ethnic group and smoking status: CDC Pregnancy Nutrition Surveillance System, 1987 Total Smokers Nonsmokers Race or ethnic group n’ Ferent n Percent n Percent Total 69,346 6.7 20,393 9.3 48,953 5.6 White 37,207 5.6 15,335 8.3 21,872 3.7 Black 24,290 9.0 4,328 13.8 19,962 8.0 Hispanic 7,024 4.5 645 5.6 6,379 44 Native American 267 71 57 7.0 210 7.1 Asian 558 54 28 0.0 530 5.7 I n is the number of persons in the sample; for n < 100, interpret data with caution. Table 4-3. Prevalence of low birth weight (<2,500 grams), by maternal age and smoking status: CDC Pregnancy Nutrition Surveillance System, 1987 Total Smokers Nonsmokers Age (years) n! Pefoent n Peroent n Peroont Total 69,530 6.7 20,410 9.3 49,120 5.6 <20 20,803 7.0 5,000 8.3 15,803 6.5 20-24 25,453 6.3 8,076 8.5 17,377 52 25-29 14,603 6.5 4,886 9.7 9,717 4.9 30-34 6,220 7.5 1,825 12.7 4,395 5.3 > 34 2,451 7.9 623 16.1 1,828 5.1 I 1 is the number of persons in the sample; for n < 100, interpret data with caution. Table 4-4. Prevalence of low birth weight (<2,500 grams), by pregravid weight and smoking status: CDC Pregnancy Nutrition Surveillance System, 1987 Total Smokers Nonsmokers Pregravid weight 1 Percent Percent Percent (percent of desirable weight) n low n low n low Total 68,323 6.7 20,097 9.3 48,226 5.6 < 90 percent 13,882 9.9 5,026 12.8 8,856 8.3 90-120 percent 35,768 6.4 10,300 8.8 25,468 5.5 > 120 percent 18,673 4.9 4,771 6.9 13,902 4.1 Tn is the number of persons in the sample; for n < 100, interpret data with caution. 84 prevalence was not greatly affected by maternal age, but was related to maternal weight before pregnancy. Growth Retardation Slowed growth may be one of the first clinically measurable indicators of inadequate dietary intake in children. Anthropometric data such as height and weight provide valuable information for identifying poor nutritional status in children. In cases in which intake of an inadequate diet is chronic and mild, the child's linear growth is often slowed, and height is low Low Height for Age: Males Mexican American Cuban Puerto Rican bercont RNY BA er: 14 12} 3 [3 0 10+ KR PRY fe 81 KX] CX 4] KX] RXR 6 20 pd 1 RY 0.0.0, POC] 900 oo RX RX RY 4r eX pd bd (XX DO] KX PAN RXXI 13 B PX 2+ KX] PH KS PSY P D J 1X BX PRX 0 KX RXR N RX 6-11 12-17 Age in years Low Height for Age: Females Mexican American Cuban Puerto Rican NN BX Percent 14 12} 10} (RX xX RI RX 8l NN ES NY ER NE 6 N RR N BRK RXR N BX oe 4r N 3 RRR N Bd 2+ N I] RY oe 0 2-5 6-11 Age in years Figure 4-9. Percent of Hispanic children below the NCHS growth chart 5th percentile of height for age, by sex, age, and ethnic group: Hispanic Health and Nutrition Examination Survey, 1982-84 (The aster— isk indicates an unstable statistic because of small sample size.) 85 for age. This condition is termed "stunting" or "shortness." If energy intake is severely inadequate, the child loses weight and often has a low weight—to- height ratio. This condition is termed "wasting," sometimes called "thinness." Information on the growth of children available since the 1986 JNMEC report consists of analyses of data from NHANES I and NHANES II, HHANES, and the PedNSS. Height for age and weight for height of children aged 2-17 years in the HHANES were com— pared to NCHS growth charts (see figures 4-9 and 4-10). The expected prevalence of both shortness Low Weight for Height: Males Mexican American Cuban Puerto Rican EEA == BRA Percent 10 8st 6 pe: 4} 2 - x Lo BE Bl 2-5 Age in years Low Weight for Height: Females Mexican American Cuban Puerto Rican EEE = RR Percent 10 8} 6} 4 ¥ N 2F 0.0.0.0.4 x i ee [XXX 0 i 0 RR IXRXN] 2-5 6-9 Age in years Figure 4-10. Percent of Hispanic children below the NCHS growth chart 5th percentile of weight for height, by sex, age, and ethnic group: Hispanic Health and Nutrition Examination Survey, 1982-84 (The asterisk represents an unstable statistic because of small sample size; the zero represents a prevalence estimate of 0.0 percent.) and thinness in the reference population is 5 percent. Mexican-American, Cuban, and Puerto Rican males aged 2-5 years had a higher prevalence of shortness than the reference population; however, a higher prevalence of shortness was evident only for Mexican American and Puerto Rican males by the ages of 12- 17. Cuban and Puerto Rican females aged 2-5 years tended to be shorter than the reference population; however, by ages 12-17 years, females in all three Hispanic groups exhibited a higher prevalence of shortness than the reference population. For Hispanic males and females aged 2-9 years, incidence of underweight was less than in the reference population of children of these ages. Although the data on height—for—age could be considered indicative of mild dietary inadequacy, other factors, both genetic and environmental, need to be evaluated before making this conclusion. Also, height for age is regarded as a more sensitive indicator of dietary adequacy in prepubertal children than in adolescents. The PedNSS continuously monitors the nutritional status of high risk groups of low-income infants and children in various states. The database is obtained from selected health service delivery programs such as Maternal and Child Health; Early and Periodic Screening, Diagnosis, and Treatment; the Supplemental Food Program for Women, Infants and Children; and Head Start. Data compiled between 1973 and 1987 have consistently shown 9-11 percent Shortness and Thinness in Low—Income Children Shortness Thinness rere irre Percent 14 12 10+ ST Te 8} 6} 4r B=, Oc ~O. 08mg. gO Om 6. —.g—0--0—0 2+ 0 L L L . 1 L L 1972 1974 1976 1978 1980 1982 1984 1986 1988 Year Figure 4-11. Percent of low-income children below the NCHS growth chart 5th percentile of height for age (shortness) or below the NCHS growth chart 5th percentile of weight for height (thinness): CDC Pediatric Nutrition Surveillance System, 1973-87 of children below the NCHS growth chart 5th percentiles for height for age (shortness or stunting) and 3-4 percent below the 5th percentile for weight for height (thinness) (figure 4-11). These CDC data are based on children who are in low-income families participating in government- supported service programs; consequently, this group is less representative of the total U.S. population with respect to poverty status and related characteristics. This fact helps to explain the higher risk of stunting that exists in the CDC surveillance population than in the reference population that was used to establish the 5th percentile criterion. The EPONM is in agreement with the JNMEC that the CDC surveil lance system seems to be sensitive enough to detect trends in nutritional status among persons seeking health services. Growth parameters in black and white children aged 1-17 years in the NHANES I (1971-75) and NHANES II (1976-80) were analyzed in relation to poverty status (Jones et al., 1985). Mean values for height and weight were generally lower for children living below the designated poverty threshold than for children living above poverty. Growth factors for fatness (skinfolds) were less affected by poverty status. The magnitude of these poverty—associated differences decreased between the times of NHANES I and NHANES II. The differences in growth were not consistently associated with differences in energy intakes between poverty status groups or between surveys, suggesting that environmental factors other than nutrition (for example, infections, parasitic diseases, or family instability) may be responsible for part of the differences in growth observed. Pregnancy There is increased demand for nutrients during pregnancy. The CSFII 1985-86 data indicate that pregnant women make specific and appropriate alterations in their diets which include avoidance of alcohol, increased dairy product consumption, and use of vitamin/mineral supplements; these changes would be expected to lead to a higher nutrient density for several nutrients in pregnant than in nonpregnant women (Harrison et al., 1988; Krebs- Smith, 1988). The number of pregnant women included in most national sample surveys of the general population is usually too small for analyses that take into account the nutrition-related physiological changes that occur over the nine months of pregnancy. However, data for the prevalence of low hematocrit, by trimester, from the 1987 PNSS are presented in tables 4-5 and 4-6. (Criteria for low hematocrit are given in chapter 6.) Prevalence of low hematocrit was highest in the third trimester, was higher in blacks than in other ethnic groups, and was higher in younger women than in older women. Dietary Behaviors That May Affect Health The NNMS provides some data on dietary behaviors that may affect health. Most of these behaviors are motivated by food preferences, attitudes, and cognitions (see conceptual model in figure 4-1). Some may be related to perceptions about health (breastfeeding, vitamin/mineral supplement use, vegetarianism, weight reducing diets); some may be related to body image (weight reducing diets); some may be motivated by religious or ethical beliefs (vegetarianism); and some may be responsive to advertising (food away from home) and education (breastfeeding). Breastfeeding A number of benefits of breastfeeding have been reported; human milk contains factors that provide Table 4-5. Prevalence of low hematocrit at initial visit, by race or ethnic group and trimester of pregnancy: CDC Pregnancy Nutrition Surveillance System, 1987 Total First trimester Second trimester Third trimester Race or ethnic group n’ Peoent n Pepoont n Fefoont n Percent Total 77,771 14.7 15,261 5.6 36,367 9.9 26,143 26.8 White 43,908 9.4 10,575 3.2 19,885 5.8 13,448 19.4 Black 25,702 24.2 3,609 12.2 12,584 16.8 9,509 38.6 Hispanic 7,130 13.6 902 7.5 3,450 7.9 2,778 22.8 Native American 263 9.1 48 0 111 4.5 104 18.2 Asian 592 15.2 99 9.1 265 121 228 21.5 1 n is the number of persons in the sample; for n < 100, interpret data with caution. Table 4-6. Prevalence of low hematocrit at initial visit, by maternal age and trimester of pregnancy: CDC Pregnancy Nutrition Surveillance System, 1987 Total First trimester Second trimester Third trimester Age (years) of Toe A Tet a ThE a Total 77,771 14.7 15,261 5.6 36,367 9.9 26,143 26.8 <20 23,064 17.3 4,044 6.1 10,819 11.3 8,201 30.6 20-24 28,882 14.1 5,861 5.3 13,332 9.5 9,692 25.6 25-29 16,375 13.2 3,439 5.6 7,726 9.1 5,210 24.4 30-34 6,829 13.5 1,418 5.8 3,203 9.1 2,208 24.7 > 34 2,615 12.5 499 6.0 1,286 73 830 24.0 1 87 n is the number of persons in the sample; for n < 100, interpret data with caution. some degree of immunity against infections, supplies protection from iron deficiency, and rarely causes allergic reactions (DHHS, 1988). Data on breastfeed— ing are presented from an analysis of HHANES by Carter et al. (1988) in tables 4-7 and 4-8. The nationwide increases over time in the percent of chil- dren ever breastfed and the percent breastfed for 6 months or more that have been observed in the non- Hispanic population were also observed in the Mexican-American population. Dietary intake data for breastfeeding women from the CSFII 1985-86 indicated that intakes of energy and energy-yielding nutrients (protein, fat, and carbohydrate) appeared to be higher than intakes of nonpregnant, nonlactating women. Intakes of all vitamins and minerals analyzed in CSFII 1985-86 also appeared to be higher among women who were breastfeeding. Values for nutrient intakes represent nutrient content of all foods and beverages (except water); they do not include nutrient intakes from vitamin and mineral supplements (Krebs—Smith, 1988). Eighty—-two percent of breastfeeding women reported using nutrient supplements and thereby had higher intakes for some vitamins and minerals than intakes from foods alone would indicate. Vitamin/Mineral Supplement Use There are concerns about nutrient toxicity or imbal- ances due to intake of high doses of vitamin/mineral supplements. For data from the FDA Vitamin/ Mineral Supplement Intake Survey, see tables 11-132 and II-133 in appendix II. Data from this survey indicated that, excluding pregnant and lactating women, nearly 40 percent of the adult population used one or more supplements (Stewart et al., 1985). Of those users, more than half consumed only one supplement; vitamin C, either alone or in combina- tion with other nutrients, was most widely consumed (more than 90 percent of users). Supplement use was more prevalent among women than among men, and more prevalent in the West than in other parts of the country. A wide range of intakes (up to 500 times the RDA for individual nutrients) was found. Heavy supplement use was more common in older adults, Table 4-7. Percent of children who were ever breastfed, by year of birth and ethnic group: National Survey of Family Growth, 1982, and Hispanic Health and Nutrition Examination Survey, 1982-84 (Carter et al., 1988) Year of birth Ethnic group 1971-73 1974-76 1977-79 1980-82 Non-Hispanic 27.9 37.0 44.8 53.2 Mexican American 33.7 33.6 39.7 50.6 Cuban 39.8 27.2 44.5 32.9 Puerto Rican 17.8 16.0 15.9 26.1 Table 4-8. Percent of children who were breastfed for 6 months or more, by year of birth and ethnic group: National Survey of Family Growth, 1982, and Hispanic Health and Nutrition Examination Survey, 1982-84 (Carter et al., 1988) Year of birth Ethnic group 1971-73 1974-76 1977-79 1980-82 Non-Hispanic 6.7 14.9 20.7 18.8 Mexican American 14.0 12.9 16.6 18.8 Cuban 14.9 4.1 5.7 5.0 Puerto Rican 6.8 3.6 4.7 7.1 88 and such users were typically taking two or more specialized vitamin and mineral products at a time as part of a personalized supplement regimen (Levy and Schucker, 1987). Heavy supplement use was also found to be associated with more frequent visits to health food stores, greater interest in and concern with influencing their own nutritional status, and less physician involvement in prescribing nutrient supplements. Data from the NFCS 1977-78 indicated that 35 per- cent of individuals one year of age and older used supplements, regularly or irregularly, and confirmed higher prevalences of use for women and persons living in the West (Moshfegh, 1985). Education was also positively associated with supplement use. The diets of supplement users provided a larger amount of many nutrients from food alone than did the diets of nonusers. Comparison of supplement use by partici—- pants in the NFCS 1977-78 and participants in the CSFII 1985 showed greater use of vitamin and mineral supplements by men aged 19-50 years (50 percent increase), women aged 19-50 years (48 per— cent increase), and children aged 1-5 years (26 per— cent increase) (USDA, 1985, 1986). In all cases supplement use was greater than the average of 40 percent reported in 1980 from the FDA Vitamin/ Mineral Supplement Intake Survey: 45 percent for men, 58 percent for women, and 60 percent for children in 1985. Changing Dietary Practices Vegetarianism In the CSFII 1985-86, very few individuals (no more than about 2 percent of men and women and no more than 1.7 percent of children) identified themselves as vegetarians (USDA, 1985, 1986). Vegetarianism was found to be associated with lower intakes of protein, vitamin B12, iron, zinc, and thiamin (Harrison et al., 1988). Vegetarianism was not found to be signifi- cantly associated with body weight status. Weight Reducing Diets Eight percent of the women in the CSFII 1985-86 reported they were on a special diet, which they described as a weight-loss diet. The percent was higher (13 percent) in women classified as overweight as determined by body mass index (Moshfegh, 1987). Results of the NHIS on Health Promotion and Disease Prevention (NCHS, 1988a) indicated that 56 89 percent of persons who were classified as overweight (24 percent of the total population) were trying to lose weight. Food Away From Home Since 1970, the percentage of disposable income going to food at home has declined steadily; the percentage going to food away from home has remained relatively constant (The Food Institute, 1988). Americans spent an estimated $454 billion for food, both at home and away from home, in 1987. Consumer food expenditures accounted for 15.3 percent of the total spent for all goods and services in the United States during the year (The Food Institute, 1988). Of the total spent for food in 1987, $305.8 billion, or 67 percent, was spent for food consumed at home. About one-third of personal consumption expenditures for food was for food away from home, which totaled $148.7 billion in 1987 (The Food Institute, 1988). The $454.4 billion spent for food in 1987 represented 14.3 percent of disposable personal income, continu- ing the historic decline in the proportion of income Americans allot to food. The proportion allocated to food at home has dropped from 17 percent in 1950-54 to 13.6 percent in 1970-74, to 11.6 percent in the three most recently detailed years. Away-from- home spending experienced a decline from the early 1950s to the mid-1960s; however, the percentage of expenditures allotted to food outside the home has increased slightly since 1970-74 from 4.5 percent to 4.9 percent (The Food Institute, 1988). Based on the Bureau of Labor Statistics Consumer Expenditure Survey for 1984, expenditures for food at home took 19.1 percent of the pre-tax income of the $5,000-$9,999 group and 4.9 percent of the income of the $40,000-and-over group (The Food Institute, 1988). The percentage of income going to food away from home was 6.6 percent for the $5,000-$9,999 group and 3.7 percent for the $40,000—and-over group (The Food Institute, 1988). Analysis of the proportion of the food dollar for at-home versus away-from—-home consumption by income shows that as income increases, the proportion of the food dollar for away—from-home consumption increases and at— home consumption decreases in monotonic fashion (The Food Institute, 1988). Based on the results of a recent survey of the take- out food market, eight of ten households buy take-out food in any month (purchased outside but consumed inside the home) (Food Marketing Institute, 1987). Put another way, 71 million households buy take-out food on a regular basis. According to this study, approximately 15 percent of total food dollars go to take-out purchases; 19 percent of total food dollars is spent on food eaten in restaurants; and the remaining 66 percent is spent on food prepared at home. In es— sence, then, currently one of every three dollars spent on food is going to away—from—home food outlets. The Census Bureau divides eating and drinking place sales by broad format type. One group includes res— taurants, lunchrooms, and cafeterias, called full- menu restaurants. The other includes what the Cen- sus Bureau terms "refreshment places,” primarily fast-food operations, called limited-menu restau-— rants. In 1987, full-menu operations accounted for 54.1 percent of eating and drinking place sales, while limited—menu operations accounted for 36.9 percent. The total volume of eating and drinking place sales was almost $153 billion (The Food Institute, 1988). Since 1982, sales in eating and drinking places have continued to rise. Sales by various types of establish— ments have remained relatively constant over the five-year period with eating place sales at 92 percent and drinking place sales at 8 percent of total food- based retail trade. Within eating place sales alone, full-menu units command about 60 percent and lim- ited-menu units 40 percent of total sales, although the percentage of sales by limited-menu units has increased from 38.7 percent in 1982 to 40.1 percent in 1986. The evidence substantiates the claim that nominal sales from away-from-home outlets, either from take-out purchases or from full-menu or limited- menu restaurants, are growing over time. Away-— from-home outlets are key components of the retail food distribution sector. Adults interviewed in the CSFII 1985 reported that they consumed 25 to 33 percent of their intakes of food energy and nutrients away from home (USDA, 1985, 1986). According to Haines et al. (1988), be— tween 1977-78 and 1985, significant shifts were evi— dent in the numbers of women aged 19-50 years classified into various eating patterns. The eating patterns were described multidimensionally in terms of percent of calories consumed at eight eating locations. The numbers of women classified into restaurant groups increased by nearly 60 percent; those classified into fast-food groups increased approximately 120 percent; and those classified into the cafeteria group increased by 38 percent. In short, over this period, the rise in the number of women who frequent restaurants, fast-food places, and cafeterias is noteworthy. Unequivocally, eating patterns of women shifted between 1977-78 and 1985. Changes in employment status, income, age, and education of the female head of household con- tributed to these trends. Frequently, it is assumed that away-from-home food is high in energy, fat, and sodium relative to food consumed at home. Although differences existed among the diets of women classified into the various away—-from—home groups, overall, these women had higher intakes of energy, fat, and saturated fat than women who reported eating all food at home. Some, but not all, away-from-home groups had cholesterol intakes that were lower than those of women con- suming all food at home. Intakes of fat, as percent of energy, tended to be higher for women categorized as consuming food away from home, whereas intakes of carbohydrate and protein, as percent of energy, tended to be lower. For the most part, intakes of sodium, potassium, copper, and zinc tended to be higher in women categorized in the away—from-home groups. Patterns of intake of other food components, including calcium, dietary fiber, folacin, vitamin C, vitamin A, and carotenes differed among categories of away-from-home eaters and were not consistent in direction of change from those of women consuming all foods at home (Haines et al., 1988). Guenther and Ricart (1988), using data from the CSFII 1985, explored the relationship between eating at food service establishments and the nutritional quality of women's diets (intakes per 1,000 kilocal- ories of 13 dietary components). Food service estab— lishments included restaurants, fast-food places, and cafeterias. Small but significant correlations were found between extent of food service eating and lower densities of carbohydrate, vitamin C, fiber, calcium, and iron. On the other hand, small, but significant, correlations were found between extent of food ser- vice eating and higher densities of alcohol, polyun- saturated fatty acids, saturated fatty acids, and total fat. No statistically significant relationships were evident for protein, cholesterol, vitamin A, or zinc. In summary, national survey data indicate that nutri- ent intakes are affected by the increase in consump- tion of food away from home. Overall, intakes of total energy and fat appear to be higher while intake of carbohydrate appears to be lower away from home. Less consistent changes have been observed for intakes of other nutrients. These changes may be more related to the type of away-from-home eating location than to eating away from home, per se. School Lunches Akin et al. (1983a,b) evaluated the impact of the National School Lunch Program on nutrient intakes of children and adolescents aged 6-18 years. Data from the NFCS 1977-78 were used for these analyses. Participants in the school lunch program had greater daily nutrient intakes than children and adolescents who consumed other types of lunches or who skipped lunch. The increased nutrient intakes were generally greater for children aged 6-11 years than for adolescents. Nondietary Behaviors That May Affect Nutritional Status The NNMS provides some limited data on the preva- lence and impact of several nondietary behaviors that may affect nutritional status (substance abuse, smoking, exercise, and use of oral contraceptives and medications). Many of these behaviors influence nutritional status by affecting nutrient requirements and/or utilization, as well as having independent effects on health outcome (see conceptual model in figure 4-1). Substance Abuse Excessive use of alcohol may precipitate nutrient deficiencies, lead to cirrhosis of the liver, and increase blood pressure; addiction to illicit drugs may also contribute to poor nutritional status. See data on alcohol in chapter 3. An extensive questionnaire on illicit drug use was administered in HHANES, but the data analysis was incomplete, and therefore the data were not available to the EPONM. Smoking Smoking is a risk factor for cancer and cardiovascular disease. The data from the NHIS on Health Promo- tion and Disease Prevention (NCHS, 1988a) indicated that 30 percent of persons aged 18 years and older smoked in 1985. The prevalence of smoking was equal for men and women under 30 years of age. Of all age groups, the highest prevalence of smoking was found in the men aged 30-44 years. The food consumption patterns and dietary intakes of smokers and nonsmokers differ. For example, 18 percent of nonsmokers habitually skip breakfast but 38 percent of smokers do not eat breakfast. However, approximately similar percentages (38 percent) of both groups report eating snacks on a daily basis (NCHS, 1988b). Data from the CSFII 1985-86 91 indicate that women smokers consumed less of fruits and vegetables and more of eggs, sugars, coffee, and alcoholic beverages than nonsmokers (Larkin et al., 1989 in press). Their intakes of carbohydrate, fiber, vitamin C, and thiamin per 1,000 kilocalories were also lower than the intakes of nonsmokers. In this sample, smoking was associated with lower income, less education, youth, and lack of employment outside the home (Harrison et al, 1988). Smoking was a marker for a poor diet, in terms of both nutrient intake and adherence to the major recommendations of the Dietary Guidelines for Americans (USDA/ DHHS, 1985). Analysis of NHANES II data showed that median vitamin C intakes were lower in current cigarette smokers than in nonsmokers, and the reported frequency of consumption of fruits and vegetables high in vitamin C was also lower in cur- rent smokers than in nonsmokers (Woteki, Johnson, and Murphy, 1986). Exercise Forty percent of the respondents to the NHIS on Health Promotion and Disease Prevention (NCHS, 1988a) reported that they exercised or played sports regularly in 1985, but only 28 percent were consid- ered very physically active as defined by energy expenditure of 3 or more kilocalories/kilogram body weight/day. (Examples of activities that achieve this level of energy expenditure are walking with a mod- erate increase in heart rate for 45 minutes every day and running or jogging with a large increase in heart rate for 15 minutes every day.) Regular exercise was more prevalent among men (43 percent) than women (38 percent); however, walking for exercise was more prevalent among women (46 percent) than men (38 percent). Fifty percent of women aged 18-29 years walked for exercise while only 37 percent of men in the same age group reported walking as exercise. Data from the CSFII 1985-86 indicate that over- weight women were less likely to be physically active than normal-weight women (Moshfegh, 1987). Physical activity levels were rated as light (golf or strolling occasionally), moderate (rigorous exercise once or twice a week or steady walking three or more times per week), or rigorous (running or playing tennis three or more times per week). Use of Oral Contraceptives Some evidence suggests that the use of oral con- traceptives may negatively affect nutritional status with respect to zinc, folacin, and vitamin B6; however, oral contraceptive use may improve iron nutritional status by decreasing menstrual blood loss. The pre- valence of use of oral contraceptives by premeno- pausal females aged 12-54 years has changed little in the time between NHANES I and NHANES II, having been 17.9 percent in 1971-74 and 16.7 percent in 1976-80 (Russell-Briefel et al., 1985). In NHANES II, women aged 20-44 years who used oral contracep— tives were found to have decreased glucose tolerance compared with women of the same age who did not use oral contraceptives (Russell-Briefel et al., 1987). Use of Medications Some medications may exacerbate nutrient defici— encies (for example, aspirin stimulates gastrointes— tinal blood loss that can lead to iron deficiency) or contribute sources of nutrients (calcium-containing antacids). The HHANES data on medication use, however, are not currently available. Special concern is justified about the relationship of medication use and nutritional status in the elderly. The elderly take more drugs than other groups and may be particu- larly vulnerable to adverse interactions. Summary The availability of the data discussed in this chapter is a strength of the NNMS, contributing to a more comprehensive view of nutrition-related health out— comes and practices that affect dietary intake and nutritional status as well as their associations. Much of this information adds support to the identification of public health issues associated with nutrition in chapter 3. References Cited Akin, J. S., D. K. Guilkey, P. S. Haines, and B. M. Popkin. 1983a. Impact of the School Lunch Program on Nutrient Intakes of School Children. School Food Service Res. Rev. 7:13-18. Akin, J. S., D. K. Guilkey, and B. M. Popkin. 1983b. The School Lunch Program and Nutrient Intake: A Switching Regression Analysis. Am. J. Agric. Econ. 65:477-485. Carter, O., T. Ezzati, and S. Gray. 1988. Trends in Breastfeeding Among Mexican-Americans, Cubans and Puerto Ricans. (Preliminary draft.) 92 Centers for Disease Control. 1988. Prevalence of Overweight in Selected States——Behavioral Risk Factor Surveillance, 1986. JAMA. 259:797-798. Dawson, D. A. 1988. Ethnic Differences in Female Overweight: Data from the 1985 National Health Interview Survey. Am. J. Public Health. 78:1326- 1329. Flegal, K. M., T. M. Ezzati, M. I. Harris, et al. 1988c. Prevalence of Diabetes in Mexican Americans, Cubans and Puerto Ricans from the Hispanic Health and Nutrition Examination Survey, 1982-84. (Preliminary draft) Flegal, K. M.,, W. R. Harlan, and J. R. Landis. 1988a. Secular Trends in Body Mass Index and Skinfold Thickness with Socioeconomic Factors in Young Adult Women. Am. J. Clin. Nutr. 48:535-543. Flegal, K. M., W. R. Harlan, and J. R. Landis. 1988b. Secular Trends in Body Mass Index and Skinfold Thickness with Socioeconomic Factors in Young Adult Men. Am. J. Clin. Nutr. 48:544-551. The Food Institute. 1988. Food Retailing Review - 1988 Edition. Food Marketing Institute. 1987. Shopping a la Carte. The Changing Environment of the Take-out Food Market. Conducted for the Food Marketing Institute and the Campbell Soup Company by Lieberman Research, Inc. Washington: Food Marketing Institute. Forman, M. R,, F. L. Trowbridge, E. M. Gentry, et al. 1986. Overweight Adults in the United States: The Behavioral Risk Factor Surveys. Am. J. Clin. Nutr. 44:410-416. Gillum, R. F. 1987. Overweight and Obesity in Black Women: A Review of Published Data from the National Center for Health Statistics. JAMA. 79:865-871. Guenther, P. M,, and G. Ricart. 1988. Effect of Eating at Foodservices on the Nutritional Quality of Women's Diets, in M. F. Moyal, ed. Dietetics in the 90s. Role of the Dietitian/Nutritionist. London: John Libbey Eurotext Ltd. Haines, P. S., B. M. Popkin, and D. K. Guilkey. 1988. Dietary Status and Eating Patterns. Prepared for the U.S. Department of Agriculture under Cooperative Agreement No. 53-3198-6-58. Harrison, G. G., T. Moon, E. J. Graver, et al. 1988. Health-Related Factors Affecting Dietary Intake. Final report for USDA Cooperative Agreement #58-319R-7-010. Ismail, A. I., B. A. Burt, and J. A. Brunelle. 1987a. Prevalence of Dental Caries and Periodontal Disease in Mexican American Children Aged 5 to 17 Years: Results from Southwestern HHANES, 1982-83. Am. J. Public Health. 77:967-970. Ismail, A. I., B. A. Burt, and J. A. Brunelle. 1987b. Prevalence of Total Tooth Loss, Dental Caries, and Periodontal Disease in Mexican-American Adults: Results from the Southwestern HHANES. J. Dent. Res. 66:1183-1188. Ismail, A. I., B. A. Burt, J. A. Brunelle, and S.M. Szpunar. 1988. Dental Caries and Periodontal Disease Among Mexican-American Children from Five Southwestern States, 1982-83. MMWR. 37(SS-3): 33-45. Jones, D. Y., M. C. Nesheim, and J.-P. Habicht. 1985. Influences in Child Growth Associated with Poverty in the 1970's: An Examination of HANES I and HANES II, Cross-sectional U.S. National Surveys. Am. J. Clin. Nutr. 42:714-724. Kleinman, J. C., R. P. Donahue, M. I. Harris, et al. 1987. Mortality Among Diabetics in a National Sample. Am. J. Epidemiol. 128:389-401. Kleinman, J. C., and S. K. Kessel. 1987. Racial Dif- ferences in Low Birth Weight: Trends and Risk Factors. N. Engl. J. Med. 317:749-753. Kovar, M. G., M. I. Harris, W. C. Hadden. 1987. The Scope of Diabetes in the United States Population. Am. J. Public Health. 77:1549-1550. Krebs—Smith, S. M. 1988. Letter and attachments to C. Suitor, National Research Council, National Academy of Sciences, dated October 18. Larkin, F. A., P. P. Basiotis, H. A. Riddick, et al. In press. Dietary Patterns of Women Smokers and Non- smokers. J. Am. Diet. Assoc. Levy, A. S., and R. E. Schucker. 1987. Patterns of Nutrient Intake Among Dietary Supplement Users: Attitudinal and Behavioral Correlates. J. Am. Diet. Assoc. 87:754-760. Madans, J. H,, C. S. Cox, J. C. Kleinman, et al. 1986. 10 Years After NHANES I: Mortality Experience at Initial Follow-up, 1982-84. Public Health Rep. 101:474-481. 93 Millar, W. J., and T. Stephens. 1987. The Prevalence of Overweight and Obesity in Britain, Canada, and United States. Am. J. Public Health. 77:38-41. Moshfegh, A. 1985. Characteristics of Vitamin and Mineral Supplement Users and Nonusers. Paper pre- sented at the Annual Meeting of the American Dietetic Association, New Orleans, La. October 7-11. Moshfegh, A. 1987. Weight Status and Associated Characteristics of Adult Women—--CSFII. 1985. Paper presented at the Twentieth Annual Meeting of the Society for Nutrition Education, San Francisco, Calif. July 9. National Center for Health Statistics, W. C. Hadden and M.I. Harris. 1987a. Prevalence of Diagnosed Diabetes, Undiagnosed Diabetes, and Impaired Glu- cose Tolerance in Adults 20-74 Years of Age, United States, 1976-80. Vital and Health Statistics. Series 11, No. 237. DHHS Pub. No. (PHS) 87-1687. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, T. B. Jabine. 1987b. Reporting Chronic Conditions in the National Health Interview Survey: A Review of Tendencies from Evaluation Studies and Methodological Test. Vital and Health Statistics. Series 2, No. 105. DHHS Pub. No. (PHS) 87-1379. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, C. A. Schoenborn. 1988a. Health Promotion and Disease Prevention, United States, 1985. Vital and Health Statistics. Series 10, No. 163. DHHS Pub. No. (PHS) 88-1591. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, C. A. Schoenborn and V. Benson. 1988b. Relationships Between Smoking and Other Unhealthy Habits: United States, 1985. Advance Data from Vital and Health Statistics. No. 154. DHHS Pub. No. (PHS) 88-1250. Hyattsville, Md.: Public Health Service. National Research Council, Committee on Diet and Health. 1989. Diet and Health: Implications for Reducing Chronic Disease Risk. Washington: National Academy Press. Russell-Briefel, R., T. Ezzati, and J. Perlman. 1985. Prevalence and Trends in Oral Contraceptive Use in Premenopausal Females Ages 12-54 Years, United States, 1971-80. Am. J. Public Health. 75:1173-1176. Russell-Briefel, R., T. M. Ezzati, J. A. Perlman, and R. S. Murphy. 1987. Impaired Glucose Tolerance in Women Using Oral Contraceptives: United States, 1976-80. J. Chronic Dis. 40:3-11. Stewart, M. L., J. T. McDonald, A.S. Levy, et al. 1985. Vitamin/Mineral Supplement Use. A Telephone Survey of Adults in the United States. J. Am. Diet. Assoc. 85:1585-1590. U.S. Department of Agriculture and U.S. Department of Health and Human Services. 1985. Nutrition and Your Health: Dietary Guidelines for Americans. Home and Garden Bulletin No. 232. 2nd edition. Washington: U.S. Government Printing Office. U.S. Department of Agriculture. 1985. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII Report No. 85-1. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture. 1986. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Men 19-50 Years, 1 Day, 1985. 94 NFCS, CSFII Report No. 85-3. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Health and Human Services. 1988. The Surgeon General's Report on Nutrition and Health. DHHS Pub. No. (PHS) 88-50210. Public Health Service. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1986. Nutrition Monitoring in the United States--A Progress Report from the Joint Nutrition Monitoring Evaluation Committee. DHHS Pub. No. (PHS) 86-1255. Public Health Service. Washington: U.S. Government Print— ing Office. Williamson, D. F., H. S. Kahn, and P. L. Remington. 1988. The Ten-Year Incidence of Overweight and Major Weight Gain Among United States Adults. Presentation at the Annual Meeting of the American Heart Association, Washington. November 14-17. Woteki, C., C. Johnson, and R. Murphy. 1986. Nutri- tional Status of the U.S. Population: Iron, Vitamin C, and Zinc, in National Research Council, What is America Eating? Washington: National Academy Press. Chapter 5 Nutritional and Dietary Factors in Cardiovascular Disease Purpose Cardiovascular disease is a major cause of morbidity and mortality in the United States. Despite recent declines in mortality, cardiovascular diseases still account for more deaths annually than any other group of diseases. Several of the food components identified as public health issues by both the JNMEC and EPONM (food energy, total fat, saturated fat, alcohol, and sodium) were considered to be of con- cern, in part, because of their relationships to the development of cardiovascular diseases (see chapter 3). The purpose of this chapter is to demonstrate how NNMS data can contribute to the understanding of dietary and nutritional factors as they relate to cardiovascular diseases as well as to identify the strengths and weaknesses of data and information available from specific components of the NNMS. The major strength of NNMS and related data is their potential to provide information that permits moni- toring of trends in morbidity and mortality from nutrition-related diseases, nutrition—related risk fac— tors for diseases, and dietary intake of components associated with diseases. Although the NNMS data are not intended to contribute to determining the mechanism of diet and disease relationships, they do permit investigation of cross-sectional associations of disease with risk factors, and of disease or risk factors with food and nutrient intake. The focus of this chapter is on identifying and discussing NNMS data related to 1) populations at risk, 2) trends in disease prevalence and risk factor status, and 3) determining factors. Limits to interpretation of data and gaps in the database are also addressed. To accomplish the objectives described above, data from the NNMS and related sources are used to examine the prevalence of cardiovascular diseases, primarily coronary heart disease; the distribution and prevalence of nondietary risk factors for car- diovascular diseases; the distribution of and trends in dietary factors associated with cardiovascular diseases as well as factors affecting dietary intake; and 95 relationships among dietary and other risk factors and disease. Definitions Cardiovascular disease includes a variety of pathological processes pertaining to the heart and blood vessels. Terms employed in the discussion of cardiovascular diseases in this chapter are defined below. The relevant codes from the most recent re- vision of the International Classification of Diseases (ICD) (DHHS, 1980) for the disorders described below are also specified. Coronary (or ischemic) heart disease (ICD-9 410-414) is a term used to identify several cardiovascular disorders resulting from inadequate circulation of blood to local areas of the heart muscle. This impaired blood circulation to the heart is almost always the result of focal narrowing of the coronary arteries by atherosclerosis. Atherosclerosis is a progressive process that begins in childhood with the appearance of lesions in the form of fatty streaks in the lining of the coronary arteries or aorta. The fatty streaks may eventually progress to fatty and fibrous plaques or even larger, more complicated lesions. As the lesions develop, the progressive narrowing of the vessels reduces blood flow to the tissues supplied by the affected vessels, resulting in angina pectoris (chest pain), myocardial infarction (heart attack), or sudden death. These are the most common manifes- tations of coronary heart disease. Hypertension (ICD-9 401) is defined as persistently elevated arterial blood pressure. Hypertensive heart disease (ICD-9 402) includes hypertensive cardiomegaly (enlargement of the heart), hypertensive cardiop- athy (disease or disorder of the heart), hypertensive cardiovascular disease, and hypertensive heart fail- ure. Cerebrovascular disease (ICD-9 430-438) includes a group of disorders characterized by ischemic stroke (atherothrombotic stroke), a serious and sudden decrease of blood supply to the brain resulting from atherosclerosis, and by cerebral hemorrhage (hemorrhagic stroke). Cardiovascular disorders not thought to be related to diet, such as congenital heart disease and ruptured aortic aneurysm, are not considered in this report. Background An extensive literature on the relationship of dietary and nutritional factors, as well as other risk factors, to the development of cardiovascular diseases has accumulated in recent decades. The evidence linking diet and nutrition to cardiovascular diseases has been examined in The Surgeon General's Report on Nutri- tion and Health (DHHS, 1988) and in the report of the National Academy of Sciences’ Committee on Diet and Health (National Research Council, 1989). The major conclusions of these two authoritative reports are summarized below. Comparisons of populations in different countries and within countries show large differences in incidence and mortality for cardiovascular diseases, particularly coronary heart disease, and the underlying patho- logical vascular process for many of these diseases, atherosclerosis. The differences in rates between and within populations are strongly associated with differences in average levels and distributions of the blood lipoproteins (for coronary heart disease) and of blood pressure (for stroke). Coronary Heart Disease Coronary heart disease rates and the levels of risk across populations are strongly related to average plasma cholesterol levels and, more specifically, to low density lipoprotein (LDL)-cholesterol levels. Mean cholesterol levels, in turn, have been found to be related to the composition of a population's habitual diet, especially the intake of saturated fat and cholesterol. In populations with a high prevalence of coronary heart disease, the impact of dietary varia— tions on individual risk is strongly influenced by inherent (genetic) differences in blood lipoprotein levels, and by the presence of other characteristics such as high blood pressure, cigarette smoking, and diabetes. Thus, within populations, depending on the homogeneity of a number of population characteris— tics and the degree of variability in the nutrient intake among individuals, varying degrees of associa tion with blood lipid and lipoprotein levels have been found. 96 Although habitual physical activity has not been shown to be related to population risk of coronary heart disease (National Research Council, 1989), individual risk as well as fatality rates of myocardial infarction may be lowered in those who are physically active. Within populations, the major risk factors of elevated serum cholesterol levels, blood pressure, and smoking, individually and combined, are related to the risk and occurrence of coronary heart disease in a strong, continuous, and graded fashion (DHHS, 1988). In controlled experiments in both humans and animals, changes in dietary composition, particularly of fatty acids (type and amount) and cholesterol, influence individual levels of lipoproteins fairly predictably. Dietary saturated fat and cholesterol increase blood total cholesterol and LDL-cholesterol levels; polyunsaturated fat decreases blood total cholesterol and LDL-cholesterol levels, while mono- unsaturated fat may decrease blood cholesterol. In randomized clinical trials, experimental lowering of total and LDL-cholesterol levels, and possibly elevation of high density lipoprotein (HDL)-choles— terol levels, by diet and/or medication, have, in a majority of studies, been associated with a reduction in risk of coronary heart disease in proportion to the degree of lowering. Influences of diet may also be manifest through effects on arterial thrombosis, a blood clot occluding a blood vessel by affecting the stickiness of the platelets responsible for blood clotting. Evidence suggests that intake of particular polyunsaturated fats, the omega- 3 fatty acids, may offer some protection against the development of clinical manifestations of atheroscler- osis by decreasing platelet aggregation and clotting activity and preventing arterial thrombosis (National Research Council, 1989). Hypertension With respect to hypertension, the most prominent dietary influences are energy imbalance (obesity), alcohol, and, in some persons, the intake of sodium. Effects of other dietary components such as calcium, potassium, chloride, magnesium, dietary fiber, and fatty acids are currently under investigation. That increased body weight is related to increased blood pressure is well supported by studies in both the epidemiological and clinical literature (DHHS, 1988). Weight loss is an important component in the treatment of hypertension in obese persons. Epidemiological studies have shown a relationship between the consumption of large amounts of alcohol and elevated blood pressure. In many of these studies, the relationship of alcohol intake to blood pressure was independent of age, body weight, exer- cise, and smoking. The available evidence indicates the potential importance of alcohol restriction in blood pressure control (DHHS, 1988). Ecological evidence from comparisons of diet and blood pressure in populations throughout the world shows that in non-Western populations with low salt intake there is no rise in blood pressure with age and little or no hypertension. Studies among individuals within populations have shown inconsistent results, in part, because of large variation in salt intake within individuals and, in many populations, because of homogeneity in regard to the level of usual salt intake. In addition, obtaining accurate estimates of sodium intake is very difficult. In clinical studies, moderate reduction of dietary sodium intake results in reduction of elevated blood pressure, but response is variable, suggesting that only some individuals are salt-sensitive (DHHS, 1988). In regard to other minerals for which intake has been related to blood pressure level, a number of studies have suggested that potassium intake may have an independent and beneficial antihypertensive effect on blood pressure. Clinical trials using potassium salts have indicated that their lowering effects on blood pressure are moderated by the amount of sodium in the diet. That calcium intake may be associated with level of blood pressure has been suggested by experi- mental studies in animals, epidemiological studies, and clinical trials. Because of conflicting evidence, The Surgeon General's Report on Nutrition and Health (DHHS, 1988) concluded that the role of calcium in blood pressure remains uncertain and that current evidence is inconclusive. Another mineral nutrient considered is magnesium. Clinical trials have pro- vided most of the data relating magnesium to blood pressure (DHHS, 1988). The same ecological data suggesting a relationship between calcium intake and blood pressure, that is, the evidence that elevated blood pressure is more prevalent in hard-water areas than soft-water areas, also suggest a possible rela— tionship of magnesium intake and blood pressure. Increased levels of intake of dietary fiber also have been suggested as being associated with lower levels of blood pressure as well as with lower serum choles— terol levels. However, because of the association of dietary fiber with other nutritional factors known to affect blood pressure levels, the finding of an associa— tion of dietary fiber and blood pressure may not be due to an independent effect of fiber. In the limited number of studies available (epidemio— logical studies and short-term trials), increased 97 intake of polyunsaturated fats, including omega-3 fatty acids, has been shown to be associated with lower blood pressure levels. This relationship has not been established in long-term studies, including clinical trials. Cerebrovascular Disease Cerebrovascular disease in the form of atherothrom- botic stroke is largely influenced by the same risk factors as coronary heart disease and has a generally similar distribution. Dietary associations may differ and, in general, are a neglected area of research. The risk of hemorrhagic stroke, on the other hand, is more strongly influenced by average blood pressure and by level of control of hypertension, both in indi- viduals and populations, than by serum cholesterol levels and other risk factors associated with coronary heart disease. There is some evidence that high in- takes of omega—3 fatty acids increase the risk for hemorrhagic stroke (National Research Council, 1989). Conceptual Model Based on the background presented above, a con- ceptual model, modified from the general conceptual model (figure 1-1) described in chapter 1, that illus- trates the broad relationships between dietary and nutritional status and cardiovascular disease and risk factors for cardiovascular disease is provided in figure 5-1. Components of the model that are most relevant to cardiovascular disease are shown in shaded boxes; individual topics noted with an asterisk are those for which data are available. Potential NNMS and related data sources are represented by the numbers above or below the boxes; numbers noted with an asterisk indicate those surveys or studies from which data were used in this chapter. The U.S. Food Supply Series data provide information on the per capita amounts of foods and nutrients related to cardiovascular diseases and can be used to examine trends from 1909 through 1985. Individual intakes of foods and nutrients from foods are assessed in the NFCS 1977-78 and CSFII 1985-86 (which provide the best measures of usual intake) and in the HANES. Trends in individual intake may also be examined in these surveys, but care is needed in their interpretation because of methodological differences between the different surveys and differences within surveys over time. Direct measures of nutritional status related to car- diovascular diseases (such as serum lipid levels, blood pressure, and body weight) are made in the HANES; 86 NATIONAL FOOD SUPPLY —> FOOD DISTRIBUTION ——> CONSUMPTION ——— 5 NUTRIENT UTILIZATION —— > HEALTH OUTCOME | Environmental factors | | Agricultural factors ( \ Economic factors National Nutrition Monitoring System and other data sources: 12 A] Away-from-home food available Away—from—home food acquired Away—from—-home food consumed 12567809 10011 13416 1#2 5% 6% 7% 8+ 9+ Nutrient requirement 1% 2% 4% 5% 6» 79 11 (3%) Sanitation ; Housing Occupation i Other factors ' Infection i Disease Household food Household food Household food available acquired consumed Y 2 12567912 1 2 5% 6 7482 0 14 15 16 19s * «7s 0% 750 14 15 16% 17% 19% Primary representative action or consequence : Influencing or mitigating factor 1 = CSFIl 1985-86, 2 = NFCS 1977-78, 3 = U.S. Food Supply Series, 4 = National Nutrient Data Bank, 5 = NHANESI, 6 = NHANES, 7 = HHANES, 8 = NHEFS, 9 = NHIS, 10 = FLAPS, 11 = Total Diet Study, 12 = Vit/Min, 13 = Health and Diet Study, 14 = PedNSS, 15 = PNSS, 16 = BRFSS, 17 = U.S. Vital Statistics, 18 = AEDS, 19 = NHES. See appendix lll for definitions of acronyms. Shaded boxes highlight portions of the model discussed: an asterisk (x) indicates data and data sources considered in this chapter. Figure 5-1. Conceptual model for nutritional and dietary factors in cardiovascular disease (see text for explanation) self-reports about occurrence and/or knowledge about some of these risk factors are collected in several surveys (CSFII 1985-86, BRFSS, NHIS). Data on behaviors related to cardiovascular disease (such as smoking, exercise, and diet) are also collected in several surveys. Only in the HANES are data on both dietary intake and direct measures of nutrition— related risk factors for cardiovascular disease ob- tained for the same persons. Emerging information from the NHEFS permits evaluation of the relation— ship of measures of dietary and other risk factors taken at one time to subsequent development of car- diovascular disease. Finally, the Vital Statistics Sys— tem provides data on mortality rates from cardiovas— cular diseases; changes in these rates over time can be examined in relationship to changes in dietary and other risk factors over the same time. Prevalence of Disease Since 1950, Vital Statistics data (NCHS, 1988a) indicate that the age-adjusted death rate for all causes of death has declined markedly in the U.S. population (figure 5-2). Similar declines, of lesser magnitude, have been observed for death rates from diseases of the heart and cerebrovascular disease (figure 5-2). In the period 1950-85, heart disease continued to be the leading cause of death in the United States and cerebrovascular disease was the Age—adjusted Death Rates Cerebrovascular All causes Diseases of heart disease Ye ES — —— Deaths per 100,000 900 800} 700 + 600 500 400 + 300 ~—— mmm Bre mo. 20fF TTT ©-~0-6-0-¢ 1006 _ oo _ _ _ ~ 0 A 1 A 1 A - hs 1950 1955 1960 1965 1970 1975 1980 1985 Year Figure 5-2. Age-adjusted death rates for selected causes of death, United States, selected years: National Vital Statistics System, 1950-85 99 third leading cause; together, the two accounted for approximately one-half of total deaths. Thus, despite the declines in mortality from these cardiovascular diseases in recent years, they remain major public health concerns. There are striking differences by sex and race in mortality from heart disease and cerebrovascular disease (figures 5-3 and 5-4, respectively), with males experiencing greater mortality than females and blacks experiencing greater mortality than whites. All four of the major sex—race groups have experi- enced a decline in death rates from these diseases during the period 1950-85, but the rates of change have differed among the groups. Sempos et al. (1988) have examined this phenomenon in greater detail for coronary heart disease mortality during the period 1968-85 (figure 5-5). Their evaluations were based on reclassification of cause of death to maximize the comparability of cause-of-death codes from the 8th and 9th Revisions of the International Classification of Diseases (DHHS, 1980). The analysis indicated a leveling off in the rate of decline of coronary heart disease mortality since 1976. During 1968-75, the age-adjusted absolute rate of decline in coronary heart disease mortality rates was essentially the same in white and black males and black females, and was slightly lower in white females. In the 1976-85 period, however, mortality rates continued to decline at the same rate for white males, but the decline was much less for the other three sex—race groups. Death Rates for Diseases of the Heart Black males White females Black females steiner White males —_—— Deaths per 100,000 450 400} 350 300+ 250 + 200 + 150 + 100 + S50 ——— 1 1960 L L s 1970 1975 1980 Year 0 1 1 1950 1955 1965 1985 Figure 5-3. Age-adjusted death rates for diseases of the heart, by sex and race, United States, selected years: National Vital Statistics System, 1950-85 Death Rates for Cerebrovascular Disease Black males White females Black females etl ——f— White males —— Deaths per 100,000 160 —— aot FEtizin 120} | 100 f 80 60 40} 20} 1 1 1 0 A 1 1 1950 1955 1960 1965 1970 1975 1980 1985 Year Figure 5-4. Age-adjusted death rates for cerebro- vascular disease, by sex and race, United States, selected years: National Vital Statistics System, 1950-85 Coronary Heart Disease Mortality White males White females Black males Black females Rate per 100,000 350 300 ~— = rm ————m—~. 1967 1969 1971 1973 1975 1977 1979 1981 Year 1983 1985 Figure 5-5. Trends in coronary heart disease mortality, for all ages, United States: National Vital Statistics System, 1968-85. Coronary heart disease coding: 1968-78 (ICDA 8), ICDA Nos. 410-413; 1979-85 (ICD 9) ICD Nos. 410-414, 402, 429.2 (Sempos et al., 1988) Another way to examine the changes in mortality from heart disease and cerebrovascular disease over time is to examine the death rates at specified ages among persons in successive birth cohorts. Such an analysis (NCHS, 1988a) indicates that the heart disease death rate for middle-aged men (45-54 years) increased in the successive birth cohorts of 1891-1900 and 1901-10, and then declined in the next three birth cohorts, with lower deaths rates and a greater decline in death rates being observed in whites than in blacks (figure 5-6). For middle-aged women (45- 54 years), heart disease death rates declined in each successive birth cohort from that of 1891-1900 to that of 1931-40, with rates being higher in blacks than in whites (figure 5-7). In all birth cohorts, death rates were higher among males than among females. A similar analysis for stroke death shows a progressive decline in rates for males and females of all races in the successive birth cohorts from 1891- 1900 to 1931-40 (figures 5-8 and 5-9). Racial dif- ferences in stroke death rates are striking, with much higher rates in blacks than in whites of both sexes and in all birth cohorts. Data from the National Hospital Discharge Study also provide information on the occurrence of cardiovas- cular disease. There was no downward trend in the hospital discharge rate for first listed or all listed diagnoses of myocardial infarction in the period 1970-81 (Gillum, 1987a), but the hospital case fatality rates have declined. Hospital discharge rates increased for all coronary artery-related diagnoses between 1984 and 1986 (Feinleib et al., 1988). These data provide information on the magnitude of the public health problem associated with cardio- vascular disease and help to identify groups at greater or lesser risk with respect to coronary heart disease mortality. Males are at greater risk than females and, within sex, blacks are at greater risk than whites. With respect to cerebrovascular disease mortality, blacks are at greater risk than whites and, within races, men are at greater risk than women. Risk Factors The major risk factors for cardiovascular diseases are hypercholesterolemia, smoking, and hypertension. Each of these factors may be controlled or modified to reduce risk. Other risk factors such as diabetes, obesity, and physical inactivity are also controllable. Increasing age and male sex are also risk factors. The risk factors considered in detail in this section are 100 Death Rates for Heart Disease: Men Aged 45-54 Years 600 [J whee BJ All other roces mm Block 3 Q oe o o ~ Q Qa 0 5 = ? o 200 | 1891 1901- 1911-1921 1931 1900 1910 1920 1930 1940 1891- 1901— 1911-1921- 1931 1900 1910 1920 1930 1940 Birth Cohort Figure 5-6. Death rates for heart disease among men aged 45-54 years, by race, United States, selected birth cohorts 1891-1940: National Vital Statistics System Death Rates for Heart Disease: Women Aged 45-54 Years 600 3 we BX Another races Hl so 3 S 400 3 2% — CX 3 38 Q CXS a 585 2 0 3 KK 9 3 200 XX o — 2 RX XX IRL 5854 3031 XK 555054 XX (XC) GAL CX) 5 1891- 1901— 1911-1921- 1931— 1900 1910 1920 1930 1940 1891- 1901 1911-1921 1931— 1900 1910 1920 1930 1940 Birth Cohort Figure 5-7. Death rates for heart disease among women aged 45-54 years, by race, United States, National Vital selected birth cohorts 1891-1940: Statistics System Death Rates for Stroke: Men Aged 45-54 Years 300 C1] woe Ba All other races Loa] Black 3 Is} 200 oO o — J a wn = + 3 a 100 1891— 1901— 1911-1921- 1931— 1900 1910 1920 1930 1940 1891— 1901— 1911—1921— 1931— 1900 1910 1920 1930 1940 Birth Cohort Figure 5-8. Death rates for stroke among men aged 45-54 years, by race, United States, selected birth cohorts 1891-1940: National Vital Statistics System Death Rates for Stroke: Women Aged 45-54 Years 300 J White BX] Al other races Ml sock Deaths per 100,000 1891— 1801— 1911-1921 1931— 1900 1910 1920 1930 1940 1891- 1901— 1811—1821— 1931— 1900 1910 1820 1930 1940 Birth Cohort Figure 5-9. Death rates for stroke among women aged 45-54 years, by race, United States, selected birth cohorts 1891-1940: National Vital Statistics System 101 those for which a dietary or nutritional relationship has been hypothesized-—hypercholesterolemia, hypertension, diabetes, and obesity. Because the risk of cardiovascular disease also differs by age, sex, and race, these factors are considered in the discussion that follows. Data from the NNMS that permit assessment of the distribution and prevalence of these risk factors are highlighted. Hypercholesterolemia Cholesterol is the blood lipid most strongly related to cardiovascular disease. Cholesterol is produced by the body as well as being obtained from the diet; it is an essential component of cell membranes and serves as a precursor for bile acids and steroid hormones. The lipoproteins on which cholesterol is transported in the blood are: low density lipoproteins (LDL), high den- sity lipoproteins (HDL), and very low density lipopro— teins (VLDL) (DHHS, 1988). The LDL usually con- tain 60-70 percent of the total serum cholesterol and high levels are associated with risk for coronary heart disease. Most of the total cholesterol is contained in the LDL, and serum total cholesterol and LDL-cho- lesterol are highly correlated, so that high levels of serum total cholesterol are also associated with cor— onary heart disease risk. The HDL usually contain 20-30 percent of the total cholesterol, and high levels are associated with low risk for coronary heart disease. The VLDL contain approximately 10-15 percent of the total cholesterol and are composed largely of triglyceride. Whether elevated serum tri- glyceride levels pose an independent risk for coronary heart disease is still uncertain. The level of total circulating cholesterol and the par- titioning of cholesterol among its lipoprotein carriers are controlled partly by genetic factors and partly by dietary intake, particularly of saturated fats and cholesterol. Other factors such as obesity and phys- ical inactivity also play a role. Total serum cholesterol levels were measured in each of the HANES and in the earlier NHES, 1960-62, presenting the opportunity to examine time trends in this risk factor. (From this point forward in this report, the term "serum cholesterol" refers to "total serum cholesterol.") Data on the levels of HDL- cholesterol are also available from NHANES II. Data for the means and prevalences of elevated serum cholesterol from the most recent HANES (HHANES) are tabulated in appendix II (tables II-40 through II- 43) and discussed in chapter 3, together with com- parable data for non-Hispanic whites and blacks from NHANES II. 102 NHANES II and HHANES data for mean serum cho- lesterol levels by racial and ethnic groups are illus- trated in figure 5-10; there was little difference among the groups. Analyses of the most recent national data on serum cholesterol (NCHS, 1986a) show the effects of demographic and socioeconomic variables. Mean serum cholesterol levels are signifi— cantly higher in older persons with levels peaking at age 45-54 years among men and levels peaking at age 55-64 years among women. After age 54 years, mean cholesterol levels are higher in women than in men. Racial (black-white) differences in mean levels were not significant for either sex. In the four race—sex groups, age-adjusted mean serum cholesterol levels were higher among individuals with higher economic levels (based on poverty status and family income), but lower among individuals with higher levels of education. Mean Serum Cholesterol Levels by Ethnic Group/Race Non-Hispanic Non-Hispanic Mexican American Cuban American Puerto Rican white black Ny BEER [1] a mmol/L Female Figure 5-10. Age-adjusted mean serum cholesterol levels, by sex and ethnic group or race, for persons aged 20-74 years: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 Values for mean serum cholesterol levels over time for specific age groups are presented in figures 5-11 and 5-12 for men and women, respectively. These data, analyzed after standardization of the values obtained with the different analytical methodologies used in the three surveys (NCHS/NHLBI, 1987), indi- cate that age-adjusted cholesterol levels decreased 3-4 percent in the total adult population between the Mean Serum Cholesterol Levels: Males 1960-1962 1971-1974 1976-1980 CJ i mmol/L 7 | Se (NE (NH | NB NE | NE | NE | N NE NE | N \ NE | N Nl: 3b INE N N NE N NE HH N N N HH N N tH NE N N N ft N NE L N N N KN Ei N N i 2 NE N N N Ei N N HH Nf N N N EH NE NE NE N N N Ei NE N i Nf N N NE N Ei N HH 20-24 25-34 35-44 45-54 55-64 65-7 Age in years Figure 5-11. Mean serum cholesterol levels in males aged 20-74 years: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 1960-62 and 1976-80 surveys. Declines were statistically significant for both men and women and for all whites, but not for blacks. Although differences in mean levels over time are very small, even small changes in the distribution of serum total cholesterol values may significantly shift the prevalence of risk for coronary heart disease. The development of cutoff values for serum choles— terol levels that identify persons or groups with increased risk of coronary heart disease has been an evolving process. The cutoffs used to define "high risk" in the tables II-40 through II-43 in appendix II and discussed in chapter 3 were the age—specific ones recommended by the 1984 NIH Consensus Develop— ment Conference (NIH, 1985) and were reported to maintain comparability with the JNMEC report. Recently, new guidelines for the treatment of high blood cholesterol in adults have been promulgated as part of the National Cholesterol Education Program (NCEP) (NIH, 1987). These classification recom- mendations are outlined in figure 5-13. Although consideration of all components of the screening and treatment plan is beyond the scope of this project and the available data, the EPONM considered it useful to examine the distribution of serum cholesterol levels over time according to the initial serum cholesterol 103 Mean Serum Cholesterol Levels: Females 1960-1962 1971-1974 1976-1980 / _ Age in years Figure 5-12. Mean serum cholesterol levels in fe— males aged 20-74 years: National Health Examina- tion Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 screening categories delineated in these new guide- lines (see figures 5-14 through 5-17). Over the period from 1960-62 to 1976-80, informa- tion available from the NHES and NHANES I and II shows that in the total population the age-adjusted prevalence of serum cholesterol levels <5.20 mmol/L (<200 mg/dl) has increased more than 18 percent, while the prevalence of levels 26.20 mmol/L (=240 mg/dl) has decreased 16 percent (figure 5-14). These population distributional shifts have been such that the prevalence of serum cholesterol levels in the range 5.20-6.19 mmol/L. (200-239 mg/dl) has de- creased only 4.7 percent. More than 55 percent of the adult American population aged 20-74 years had high serum cholesterol levels =5.20 mmol/L, (=200 mg/dl) that may lead to increased risk of developing coronary heart disease. According to the NCEP criteria in figure 5-13, many of these persons would be referred for lipoprotein analysis; men would be more affected because being male constitutes an additional risk factor. Among males, the pattern of change in prevalence was similar to that seen in the total population, whereas among the women there was a small increase in the proportion of individuals with cholesterol levels 5.20-6.19 mmol/L (200-239 mg/dl) between 1960-62 and 1971-74 (figure 5-15). Initial classification and recommended followup based on total cholesterol Classification and treatment decisions based on LDL-cholesterol A. Classification <5.20 mmol/L (<200 mg/dl) 5.20-6.19 mmol/L Borderline-high blood cholesterol (200-239 mg/dl) >6.20 mmol/L (=240 mg/dl) B. Recommended followup A. Classification <3.35 mmol/L (<130 mg/dl) 3.35-4.14 mmol/L Borderline-high-risk LDL-cholesterol (130-159 mg/dl) 24.15 mmol/L (2160 mg/dl) Desirable blood cholesterol Desirable LDL-cholesterol High blood cholesterol ig cholestero High-risk LDL-cholesterol 40) 8 B. Dietary treatment Initiation level Minimal goal Total cholesterol <5.2 mmol/L. —. Repeat within 5 years . (<200 mg/dl) Without CHD or two other 24.15 mmol/L. <4.15 melt risk factors (=160 mg/dl) (<160 mg/dl Total cholesterol 5.20-6.19 mmol/L . (200-239 mg/dl) With CHD gr two other 23.35 mmol/L. <3.35 mmol/l, risk factors (=130 mg.dl) (<130 mg.dl) Without definite CHD Dietary information or two other CHD risk and recheck annually C. Drug treatment factors (one of which Without CHD or two other 24.90 mmol/L. <4.15 mmol/L can be male sex) risk factors (2190 mg/dl) ~~ (<130 mg/dl) With definite CHD With CHD or two other 24.15 mmol/L <3.35 mmol/L or two other CHD risk risk factors’ (=160 mgdl) (=130 mg/dl) factors (one of which IN can be male sex) Lipoprotein analysis; further action based on Patients have a lower initiation level and goal if they are at high risk because they already have definite CHD, or because they have any two of the following risk factors: male sex, family history of premature CHD, cigarette smoking, hypertension, low HDL-cholesterol, diabetes mellitus, definite cerebrovascular or peripheral vascular disease, or severe obesity. LDL-cholesterol level Total cholesterol = 6.20 mmol/L (= 240 mg/d]) 2/3 Roughly equivalent to total cholesterol <6.20 mmol/L (<240 mg/dl) (2) or <5.20 mmol/L (<200 mg/dl) (3) as goals for monitoring dietary treatment. I cHD- coronary heart disease Figure 5-13. Guidelines for treatment of elevated blood cholesterol levels in adults from the National Cholesterol Education Program (NIH, 1987) Serum Cholesterol Levels: All Races, Both Sexes 1971-74 197 NN 80 Percent 50 45} 40 35F 30 + 25+ 5. 20- 6.19 Serum cholesterol level (mmol/L) <5.20 >6.20 Figure 5-14. Percent of persons aged 20-74 years with specified serum cholesterol levels: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 Serum Cholesterol Levels: All Races 1960-62 1971-74 1976-80 Z 7 / Percent 50 45+ 40 + 35+ 301 25¢ 20 15 10+ Females 5.20 5.20- 6.19 56.20 5.20-6.19 >6. 20 Serum cholesterol level (mmol/L) Figure 5-15. Percent of persons aged 20-74 years with specified serum cholesterol levels, by sex: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 105 Serum Cholesterol Levels: White and Black Males 1960-62 1971-74 CC] 1976-80 Percent 50 4st 40} 35 30} 25} 20} 15} 10} 5F <5.20 5.20 6.19 6.20 Serum cholesterol level (mmol/L) 5.20- 6.19 56.20 Figure 5-16. Percent of males aged 20-74 years with specified serum cholesterol levels, by race: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 Serum Cholesterol Levels: White and Black Females 1960-62 CC] 1971-74 1976-80 RRR iit i Percent 50 45} 40} 35} 30} 25} fl Black females __ White females 5.20 5.20—- 6.19 56.20 Serum cholesterol level (mmol/L) Figure 5-17. Percent of females aged 20-74 years with specified serum cholesterol levels, by race: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 Among black males as compared to white males, the patterns of change were quite different (figure 5-16). The prevalence of serum cholesterol levels <5.20 mmol/L. (<200 mg/dl) increased only 4 percentage points among black males as compared to 8 percentage points among white males. In fact, the prevalence of cholesterol levels 26.20 mmol/L (=240 mg/dl) decreased only 0.2 percent age points among black males suggesting that the increase in the lowest category came from the shift of individuals from the intermediate category rather than from the high cholesterol category. Among black and white women the pattern of net increase in the lowest cholesterol category was similar to that seen among males, while the percent decrease in the proportion with the highest cholesterol levels was similar in black and white women (figure 5-17). Both white men and women showed a greater increase in the proportion with cholesterol levels <5.20 mmol/L (<200 mg/dl) than among black men and women. However, a higher proportion of blacks than whites had serum cholesterol levels <5.20 mmol/L (<200 mg/dl). Data are also available from NHANES II for serum HDL-cholesterol levels, a factor which is inversely associated with risk of coronary heart disease. Levels of HDL-cholesterol were found to be higher in women than in men, and higher in blacks than in whites (Linn et al., 1989). In a multivariable model, predictors of high serum HDL-cholesterol were female gender and black race, while higher frequency of alcohol consumption, smoking, and higher body mass index were associated with lower HDL- cholesterol levels. These data, and the age-specific data available in appendix II, illustrate the utility of the repetitious nature of the NCHS components of the NNMS in identifying changes in serum cholesterol levels, one of the major nutrition-related risk factors for coronary heart disease, and the nature and magnitude of these changes in different population subgroups. These data are clearly important in identifying those subgroups of the population that should be targeted for increased education and intervention efforts. Blood Pressure Elevated blood pressure is a risk factor for both coronary heart disease and cerebrovascular disease. The major nutrients and diet-related factors that may influence blood pressure and their possible mechanisms were identified in The Surgeon General's Report on Nutrition and Health (DHHS, 1988); they are obesity, alcohol intake, and, in some persons, sodium intake. In addition, potassium, calcium, 106 magnesium, and unsaturated fatty acids may also be related to blood pressure levels. The strength and independence of these associations remain incon- clusive and constitute an active research area. Data for the prevalence of hypertension from the most recent HANES (HHANES) are tabulated in appendix II, tables II-119 through II-122 and are presented in figure 5-18, together with comparable data for non-Hispanic whites and blacks from NHANES II. Persons were classified as hypertensive if the average of three systolic blood pressure read- ings was greater than or equal to 140 mm mercury, and/or the average of three diastolic blood pressure readings was greater than or equal to 90 mm mer- cury, or if individuals reported during the medical history interview that they were taking antihyper- tensive medication. The blood pressure criteria were those recommended by the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure (Subcommittee on Definition and Prevalence, 1984). Among Mexican Americans, Cubans, and Puerto Ricans the prevalence of hyper- tension among both males and females was lower than among non-Hispanic whites or blacks. Among both males and females, non-Hispanic blacks had the highest prevalences of hypertension. Prevalence of Hypertension Non-Hispanic black Non-Hispanic white CJ Cuban NANA Puerto Ricon [es Mexican American Percent 50 40 30 20 Female Figure 5-18. Prevalence of hypertension (average systolic blood pressure 2140 mm mercury, and/or average diastolic blood pressure 290 mm mercury, or history of use of antihypertensive medication) for persons aged 20-74 years, by sex and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 Blood pressure measurements from the NHES, NHANES I, and NHANES II provide an opportunity for examining time trends in this risk factor. For these trend data, blood pressure values are given for the first (sitting) measurement in each survey, because these were collected in an identical fashion. Values for mean systolic and diastolic blood pressure over time (Persky et al., 1986) are presented in figure 5-19, showing differential changes occurring by race with greater decreases in blacks. In all surveys, mean diastolic blood pressure levels were higher in men than in women, and were higher in blacks than in whites, while systolic blood pressure was higher in women than in men. In the most recent national data (NHANES II), mean systolic blood pressure levels were higher in older age groups and were higher among blacks than whites in most age groups (NCHS, 1986b). Dannenberg et al. (1987) have also shown that, in all the examination surveys, mean blood pressures were higher in older age groups. Trends in the prevalence of "elevated blood pressure” have also been assessed using criteria different from those described above for hypertension. Over the period 1960-62 through 1976-80, the prevalence of "elevated blood pressure” (systolic pressure of at least 160 mm mercury and/or diastolic pressure of at least 95 mm mercury, based on a single measurement of blood pressure) has increased slightly in white males, decreased in black males and females, and decreased modestly in white females (table 5-1). Changes in the prevalence figures for hypertension are a result of a combination of factors including prevention and improved methods of treatment and control. Mean Systolic Blood Pressure Males Females White Black White Black —] a 1 KC] mm Hg mm Hg 160 160 150 150 140 140 130 130 120 120 110 110 oC of NHES NHANES | NHANES if NHES NHANES | NHANES Ii Mean Diastolic Blood Pressure Males Females White Black White Black C1 BREE C1 eR mm Hg mm Hg 100 100 95 95 90 90 85 85 80 80 75 75 ol oC NHES NHANES | NHANES II NHES NHANES | NHANES Il Figure 5-19. Age-adjusted mean systolic and diastolic blood pressures for persons aged 20-74 years, by sex and race: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 (Persky et al., 1986) Table 5-1. Percent of persons aged 25-74 years with definite elevated blood pressure’, by race, sex, and age: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971- 74; and second National Health and Nutrition Examination Survey, 1976-80 (NCHS, 1988a) White Black Sex and age 1960-62 1971-74 1976-80 1960-62 1971-74 1976-80 Percent of population Male Age adjusted, 25-74 years 19.0 21.7 22.3 36.3 35.8 29.7 25-34 years 6.1 8.3 12.2 21.8 16.1 134 35-44 years 14.9 17.2 15.2 28.0 36.8 33.2 45-54 years 19.6 25.8 28.6 34.6 .37.0 29.3 55-64 years 27.5 31.2 29.7 49.7 49.5 45.7 65-74 years 38.6 35.1 32.7 63.3% 50.3 32.1 Female Age adjusted, 25-74 years ~~ 19.2 18.5 16.3 37.7 37.4 26.2 25-34 years 2.3 3.8 3.2 8.8 10.7 5.8 35-44 years 8.2 9.9 9.9 29.2 28.2 17.4 45-54 years 18.8 18.8 20.1 44.3 49.4 42.9 55-64 years 32.5 32.0 24.4 50.5 54.2 34.2 65-74 years 53.8 42.9 35.0 79.0° 59.8 40.0 I Definite elevated blood pressure is defined as either systolic pressure of at least 160 mm mercury or diastolic pressure of at least 95 mm mercury or both based on a single measurement. Based on fewer than 45 persons. Diabetes The Diabetes Data Group (NIH, 1985) noted that approximately twice as many persons with diabetes have a medical history of heart disease or cardiac abnormalities as do persons without diabetes. About half of all diabetic individuals have a history of hypertension. The occurrence of stroke in diabetic persons is about 2 to 6 times greater than in nondiabetic persons. Data on the prevalence of diabetes in recent NNMS surveys are presented in chapter 4. These prevalence estimates were based on identifying persons with ab- normal results for an oral glucose tolerance test plus persons who gave a history of diabetes. In NHANES II, only half of the persons classified as diabetic knew of their condition (NCHS, 1987). The prevalence of diabetes, both diagnosed and undiagnosed, was greater among older individuals and was higher in blacks than in whites. At younger ages, women were 108 more likely to be diabetic than men. Prevalence increased with higher levels of percent desirable weight. Kovar et al. (1987) noted that the prevalence of self-reported diabetes increased over the period 1958 to 1985. The major reason for including dia- betes in this section is its relationship to obesity and overweight. These NNMS data, although not adding directly to our knowledge of the association between diabetes and cardiovascular diseases, indicate that diabetes is a highly prevalent health condition which is known to be related to these diseases. Obesity Obesity and/or overweight is positively associated with the prevalence of hypertension and with risk of cardiovascular disease. Obesity is also clearly and strongly associated with diabetes, a previously noted risk factor for cardiovascular disease (DHHS, 1988). Rates of both diabetes and hypertension are nearly tripled in persons 20 percent or more overweight (DHHS, 1988). Data on overweight from the most recent NNMS surveys are presented in chapters 3 and 4 and tables II-4 through II-9 in appendix II Changes in the prevalence of overweight during the period 1960-62 through 1976-80 are shown in table 5-2, and indicate little change over time. Although estimates of the prevalence of obesity and/or overweight vary with the criteria used, most estimates indicate that at least 25 percent of the adult American population is either overweight or obese. Clearly, components of the NNMS provide some of the best available data on body weight and changes in body weight over time. The interpretation of these data in terms of risk of cardiovascular disease is com— plicated by the use of different standards to classify persons as overweight and by uncertainty about the degree of overweight that increases risk. Further information on the relationships of obesity and other risk factors for cardiovascular disease can be found later in this chapter. Smoking, Exercise, Others Smoking has been identified in numerous studies as a major risk factor for the development of coronary heart disease. Overall, the weight of evidence sug- gests that increased physical activity is associated with reduced risk of cardiovascular disease. The 1985 NHIS on Health Promotion and Disease Prevention (NCHS, 1988b,c) showed that smokers, who com- prised 30 percent of the adult population, perceived themselves to be less physically active and more likely to be sedentary in terms of leisure time sports Table 5-2. Overweight’ persons aged 25-74 years, by race, sex, and age: National Health Examination Survey, 1960-62; first National Health and Nutrition Examination Survey, 1971-74; and second National Health and Nutrition Examination Survey, 1976-80 (NCHS, 1988a) White Black Sex and age 1960-62 1971-74 1976-80 1960-62 1971-74 1976-80 Percent of population Male Age adjusted, 25-74 years 25.1 26.0 26.7 24.1 27.6 30.9 25-34 years 21.4 23.6 20.9 34.3 26.1 175 35-44 years 22.4 28.9 28.2 28.6 39.3 40.9 45-54 years 29.3 28.2 30.5 18.5 22.4 41.4 55-64 years 28.5 24.9 28.6 20.1 25.6 26.0 65-74 years 24.8 23.1 25.8 n7A 21.6 26.4 Female Age adjusted, 25-74 years 27.3 27.4 27.5 47.3 47.8 49.5 25-34 years 13.9 15.9 17.9 29.6 31.5 33.5 35-44 years 21.2 24.5 24.8 46.1 49.9 40.8 45-54 years 28.5 29.9 29.9 47.8 53.5 61.2 55-64 years 40.5 36.6 34.8 71.4 58.7 59.4 65-174 years 43.2 37.0 36.5 47.8% 49.2 60.8 1 Overweight is defined for men as body mass index greater than or equal to 27.8 kilograms/meter® , and for women as body mass index greater than or equal to 27.3 kilograms/meter?. These cutoff points were used because they represent the sex-specific 85th percentiles for persons aged 20-29 years in the second National Health and Nutrition Examination Survey. Excludes pregnant women. 2 Based on fewer than 45 persons. 109 activities than former smokers or nonsmokers, and that activity level was lowest in those who smoked the most cigarettes. Smokers were also more likely to be heavier drinkers, but less likely to be obese, than former smokers or nonsmokers. Data from the same survey indicated that 40 percent of all American adults exercised or played sports regularly in 1985. For almost all age groups, exercise was more prev- alent in men than in women and was related to level of education. Although many persons are exercising, knowledge of the level of exercise required for cardiovascular fitness is very limited. Dietary Factors Dietary and nutritional factors related to cardiovascular disease (see background of this chapter) are discussed in this section. Information is included on dietary factors that exert effects on serum lipids (food energy, total fat, fatty acids, cholesterol, dietary fiber) and those with established or suspected effects on blood pressure (sodium, potassium, calcium, magnesium, alcohol). With respect to these food components, the NNMS provides data on the amounts and food sources in the food supply, individual dietary intakes and trends in dietary intake, personal and demographic factors that are associated with diet, and characteristics of population groups at various levels of intake. Tables and graphs of the most current available cross— sectional and trend data for supply and intake of the relevant dietary factors are included in appendix II. Although the inferences that may be drawn about food consumption and nutrient intake from the food supply data are limited, these data provide a sound basis for examining trends. Assessing trends in individual intake is less certain because of the lack of comparability in the techniques used to collect dietary intake data among surveys carried out at different times and by different Agencies. Another current limitation in studying dietary factors related to cardiovascular diseases is that the most recent data that provide good estimates of usual intake (CSFII 1985-86 4-day data) are for women aged 19-50 years and young children, groups not considered to be at high risk of cardiovascular diseases. Data on dietary intake alone cannot identify groups at risk because of the multifactorial etiology of cardiovascular diseases, but NNMS data can be used to characterize groups with varying levels of intake of relevant food components. In the following sections, examples are given of uses of NNMS data to examine dietary factors related to cardiovascular diseases. Studies about the 110 relationships of dietary factors to other risk factors are examined in the section on Associations. Food Energy Obesity is an important risk factor for the develop- ment of cardiovascular disease and is associated with many coronary heart disease risk factors such as hypertension, low levels of HDL-cholesterol, elevated plasma glucose levels, high blood cholesterol levels, and hypertriglyceridemia (DHHS, 1988). Food energy is the food component most logically associated with obesity. Calorie intake and physical activity deter— mine an individual's energy balance and, together with genetic factors, ultimately determine whether or not he or she will be obese or overweight. Total calorie consumption has been associated with cor— onary heart disease prevalence in international com— parisons (DHHS, 1988); most studies within popula- tions, however, have shown that high calorie intakes are associated with decreased coronary heart disease risk, but increased body weight is associated with increased risk. This finding suggests that high calorie intakes are related to increased energy expenditure which, in turn, is related to decreased coronary heart disease risk. The paradox of the inverse relationship between calorie intake and body weight noted in chapter 3 argues for the need to develop and include measures of physical activity in future NNMS surveys. The currently available NNMS data cannot be used to estimate an individual's level of physical activity. In view of the high prevalence of overweight noted in chapters 3 and 4 and the increasing evidence that overweight is an independent risk factor for the development of coronary heart disease and hyper- tension, more emphasis should be given in NNMS activities to collecting information that will broaden our understanding of this major public health problem. Another concern, also noted in chapter 3, is the possibility of underreporting of food and alcohol intake in surveys. Fat, Fatty Acids, and Cholesterol There is a strong, continuous, and graded risk between increasing levels of serum cholesterol and risk of coronary heart disease (DHHS, 1988). There is also a large body of evidence showing a relationship of diet to elevated serum cholesterol levels. Intake of dietary fats shows the strongest relationship in this regard. Increasing total fat, especially saturated fat, intake tends to raise serum cholesterol levels, while higher levels of polyunsaturated fat intake tend to decrease serum cholesterol levels. Formulas that have been used to predict the change in serum cholesterol levels resulting from given changes in dietary lipids are presented below. Keys, Anderson, and Grande (1965) formula: ASC = 1.26@2AS - AP) + 1.5A71000 C/E Hegsted et al. (1965) formula: ASC =2.16AS -1.65AP + 0.0677AC - 0.53 Where A = Change in component SC = Serum cholesterol in milligrams/100 ml C = Cholesterol intake in milligrams E = Energy intake in kilocalories S = Percent of kilocalories from saturated fatty acids P = Percent of kilocalories from polyunsaturated fatty acids From these equations, it can be seen that the cholesterol-lowering effect of a decrease in the intake of a given amount of saturated fat is greater than the increase in the intake of the same amount of polyun- saturated fat; dietary cholesterol intake also has a small but definite effect on serum cholesterol levels. Recent studies have suggested that monounsaturated fat may also have an independent effect on lowering cholesterol levels (National Research Council, 1989). Other studies have shown that different saturated fatty acids may have different effects on serum cholesterol levels; not all are hypercholesterolemic (DHHS, 1988). Epidemiological studies have shown that higher intakes of fish, a source of omega-3 fatty acids, are associated with a lower incidence of coronary heart disease. In clinical studies, omega-3 fatty acids lower triglyceride levels, a possible risk factor for coronary heart disease (DHHS, 1988). As previously stated, evidence that triglycerides are an independent coronary heart disease risk factor is inconclusive. Finally, trans fatty acids, isomers of naturally occurring cis unsaturated fatty acids, have little, if any, hypercholesterolemic effect (DHHS, 1988; National Research Council, 1989). Reliable estimates of the total fat content of foods have been available in the NNMS for some time. The same is not true of values for the individual fatty acids or fatty acid groups; analytical values for these components were not available for many foods in ear- lier surveys (Dresser, 1983). Further improvement in 111 the nutrient database, that is, obtaining values for individual fatty acids by the most current gas-liquid chromatographic techniques, would be desirable. Data for the omega—-3 fatty acid content of the food supply have become available only recently (Raper and Exler, 1988). Data on trans fatty acid content in a wide variety of foods are not available in the NNMS. In view of the conclusions that current levels of intake (in conjunction with current levels of intake of linoleic acid) have no significant deleterious effects (LSRO, 1985; British Nutrition Foundation Task Force, 1987; Zevenbergen, 1986), there is little urgency in modifying the USDA nutrient databases in regard to trans fatty acid composition of foods or in monitoring the intake of trans fatty acids in national surveys. Validating and maintaining a nutrient database for fatty acids is very difficult. Composition values represent foods as used throughout the country on a year-round basis for a particular time (Dresser, 1983). Differences may be expected when an analysis of a single food is compared with average values for a group of similar foods reported under a generic name in a database. Another difficulty with respect to maintaining a nutrient database for fatty acids is that manufacturers of processed foods may change the types of fats and oils used in response to changes in price and other demand and supply factors; thus, the fatty acid content of such foods may change frequently and unpredictably. Despite the caveats given above, and as noted below, the increased availability of data from the NNMS on intake of different fats and fatty acids is important to understanding changes in coronary heart disease prevalence and incidence over time. Examples of useful data from the NNMS on dietary fat, fatty acids, and cholesterol and their food sources are highlighted in the discussions below. Trends in the amounts and sources of fats in the food supply Estimates of nutrients in the food supply exclude nutrients from the inedible parts of foods (such as bones, rinds, and seeds) but include nutrients from parts of foods that are edible but not always eaten (such as the separable fat on meat). Estimates do not account for losses after food is measured, such as during processing, marketing, or cooking. Rizek et al. (1983) have calculated that approximately one-fourth of the fat in the 1980 food supply was wasted. Despite the limitations of these data, the trend information they provide is useful. Since the late 1940s, although there has been an absolute increase in the availability of total fat, from 137 to 169 grams per capita per day, the change has been primarily due to increases in monounsaturated fat (24 percent) and polyunsaturated fat (94 percent) (see figure 5-20 and II-5). This change in total fat is attributable to an increase in fat from vegetable sources; the changing levels of monounsaturated and polyunsaturated fats reflect this shift to vegetable sources. Per capita amounts of saturated fat have changed little over this time period. The amount of cholesterol has also decreased from 590 to 500 milligrams per capita per day. During this period of time the percent of calories available from total fat increased from 38 to 43 percent. Within the different fatty acid groups, the percent of calories from saturated fat decreased only slightly from 17 to 16 percent of calories, while the percent of calories from monounsaturated and polyunsaturated fat increased. These shifts among the fatty acid groups, which presumably bear some relation to consumption of these fats, are consistent with current dietary guidelines to avoid consumption of too much saturated fat (USDA/DHHS, 1985). Fats in the U.S. Food Supply Saturated fat Monounsaturated fat Polyunsaturated fat Grams 70 60 50} 40} 30} Ed 20} ro —_—— 10} 0 1 L 1 i 1 A A J. 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure 5-20. Per capita amounts of saturated, monounsaturated, and polyunsaturated fats in the U.S. food supply: U.S. Food Supply Series, 1909-85 The only information available from the NNMS regarding omega-3 fatty acids in the U.S. diet is from the U.S. Food Supply Series (Raper and Exler, 1988). These data indicate that in the period 1947-49 through 1985 the amount of the two major omega-3 fatty acids found in fish increased by 52 percent and the amount of linolenic acid (from plant sources) increased by 65 percent. These changes reflect the increased consumption of fish and soybean oil that occurred over this period of time. Although from the late 1940s to the present there was little overall change in the percentage of total fat that was contributed by the meat, poultry, and fish food group (31.3 to 31.4 percent), the contribution of meat decreased by 3 percentage points while that of poultry and fish increased by 3.1 percentage points. The con— tribution of dairy products to total fat decreased by 6.6 percentage points while within the fats and oils food group, the collective availability of fat from butter, margarine, shortening, lard, and beef tallow decreased by 3 percentage points and that from salad, cooking, and other edible oils increased 11.9 percent- age points. See figures 5-21 and 5-22 for a compar- ison of food supply sources of fat in 1947-49 and 1985. In regard to saturated fat intake, since the late 1940s, the contribution of the meat, poultry, and fish food group increased by 4.7 percentage points, while that of dairy products and butter, margarine, short— ening, lard, and beef tallow decreased by 8.5 percent— age points. The percentage increase of monounsat— urated and polyunsaturated fats contributed by salad, cooking, and other edible oils were 11 and 20 percentage points, respectively. The contribution of eggs and butter, margarine, shortening, lard, and beef tallow to the cholesterol content of the food supply decreased by more than 12 percentage points. Contributions to Fat in Food Supply 1947-49 1985 CJ ANN Percent 50 45+ 40} 35} 30} 25} 20+ 15} 10} I Meat, poultry, Dairy fish products — IN Fruits and ~~ Other foods vegetables Figure 5-21. Food sources of fat in the U.S. food supply: U.S. Food Supply Series, 1947-49 and 1985 Contributions to Fat in Food Supply 1947-49 C1 1985 zZ 7 Percent 30 28 26 24 a Lard and beef tallow 07 Butter and Shortening margarine Salad, cooking, and other edible oils Figure 5-22. Contributions of meat and fats and oils to per capita fat in the U.S. food supply: U.S. Food Supply Series, 1947-49 and 1985 Trends in individual consumption of foods contributing to fat intake Recent trends in individual food consumption con- tributing to fat intake can be examined in women using data from the CSFII 1985 and the NFCS 1977- 78. Such analyses have been performed (Popkin, Guilkey, and Haines, 1989, in press; Popkin, Haines, and Reidy, 1989) examining refined food groupings based on the fat and fiber contents of foods (per 100 grams of food). Overall, the quantity consumed within most food groups decreased, and the diversity of women's diets and the number of lower—fat foods used increased between 1977 and 1985. In terms of average consumption, increases in the amounts of lower—fat milk and lower—fat beef and pork products were associated with decreases in the amounts of the higher—fat products in the same category; however, there was also an increase in the average consump-— tion of higher—fat grain—-based mixed dishes and a small decrease in the average consumption of higher- fiber vegetables. In examining the proportion of women that consumed selected food groups, a slightly different pattern emerges: there was a striking increase in the percentage of the population that con- sumed lower—fat milk, lower—fat high-fiber bread, and lower— or medium-fat poultry, beef, and pork, as well as higher—fat items such as cheeses, mixed grain dishes, and salty snacks. Decreases in the proportion of the population consuming higher-fat milk and higher-fat beef and pork were also observed. Despite a general downward trend in the quantities of foods 113 consumed, there was an increase in the quantity of higher—fat mixed grain dishes consumed and a slight increase in the quantity of higher—fat desserts. Addi- tional analyses, in which sample weights were not considered, suggested that these changes in consump- tion were affected by behavioral changes as well as by changes in the socioeconomic and demographic characteristics of the population over time. Data from NHANES I and NHANES II also indicate a trend toward a decrease in the consumption of high- fat foods (Sempos et al., 1987). These findings are discussed in detail later. Contribution of foods to individual intake of fat Information from the CSFII 1985 on the (4-day) dietary intake of women can be used to study the contributions of different foods and food groups to nutrient intake. One of the problems in reporting consumption by food groups has been classification of mixed dishes, such as casseroles and sandwiches. The contributions of various food groups to fat, saturated fat, and cholesterol intakes were determined in two ways by Krebs-Smith, Cronin, and Haytowitz (1988): first, by classifying food mixtures as single items and assigning them to groups according to their main in- gredient and second, by separating nearly all mix- tures into their component ingredients and assigning the ingredients to their appropriate groups. These analyses suggest that a more complete picture of the contribution of foods to fat intake is obtained when the ingredients of food mixtures are considered sep— arately rather than categorizing each mixture according to its main ingredient. This situation is exemplified by the data presented in table 5-3; when mixtures are separated a major decrease in the con- tribution to fat intake occurs for the grain group (in which fats and oils are major components of many grain—based dishes) together with a corresponding large increase in the contribution of the fats and oils food group. Meat, poultry, and fish dishes con- tributed the largest percentage of fat, but as separate ingredients, fats and oils contributed the largest percentage of fat. Factors associated with fat intake Estimates of fat intake for women in the CSFII 1985- 86 (see tables II-16 and II-17 in appendix II) suggest that total fat intake varies with race, poverty status, and education. Analyzing the CSFII 1985 data from women, Krebs-Smith (1988) found that the characteristics of living in the Midwest or West, Table 5-3. Percentage contribution of selected food groups to total intake of selected food components for women aged 19-50 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals 1985 (Krebs—-Smith, Cronin, and Haytowitz, 1988) Saturated fatty acids Cholesterol As Mixtures As Mixtures reported separated reported separated Total fat Mixtures Food group reported separated Meat, poultry, fish 31 26 Milk and milk products 15 19 Eggs 4 4 Legumes, nuts, seeds 4 4 Grain products 22 9 Vegetables 8 6 Fats and oils 14 30 Sugars, sweets 1 1 30 27 37 36 25 33 12 15 3 3 31 42 2 2 --1 ro 20 6 16 2 7 4 1 — 10 21 3 4 2 2 -— -— I Dashes denote value is less than one, but not zero. having a child less than one year of age, being white, having more than a high school education, and being a current smoker were all associated with a higher probability of having a fat intake in the upper two quintiles. In the future, such analyses may be helpful in identifying target groups for education. Sodium Information about sodium consumption is difficult to collect accurately from surveys owing, in large part, to the multiplicity of sources and wide variations in food preparation and dietary practices. An accurate assessment of sodium intake must take into account that occurring naturally in foods and water, sodium added to prepared foods as salt, sodium added in other capacities (for example, as sodium nitrate or citrate for preservation), and discretionary sodium added by the consumer. The NNMS provides very limited information about the availability and con- sumption of sodium. Data on sodium were not col- lected in the U.S. Food Supply Series or in the NFCS 1977-78. In the NHANES I, NHANES II, and CSFII 1985-86, participants were questioned about their use of salt, but quantitative estimates were not made for salt used at the table. Different assumptions were made in different surveys when the participants did not specify their salting of foods. Thus, because of the complexities discussed above, the data from these surveys are not considered precise and accurate. 114 The FDA has collected information on sodium label- ing in its FLAPS. These surveys have found that the sodium content of established food products has not decreased in recent years, but that the total sodium in labeled foods is lower because of the introduction of low-salt foods. Calcium, Potassium, Magnesium, Fiber Although it has been suggested that there is a relationship between the consumption of these food components and different cardiovascular diseases, these hypotheses are still being tested. Inasmuch as the levels of these nutrients that may be protective or deleterious are not known, it is not useful to discuss NNMS data regarding supply and intake in any more detail than is presented in chapter 3. Alcohol Data available from the NNMS on the availability and intake of this component can be found in the dis- cussion of alcohol in chapter 3. Although there is some evidence of a relationship between alcohol intake and certain cardiovascular diseases, particu- larly hypertension, the difficulty in assessing the intake of this component in dietary surveys makes the currently available NNMS data inadequate for assessing usual levels of intake. Associations Associations among cardiovascular diseases, cardio— vascular risk factors, and dietary factors that have been assessed using NNMS data are examined in this section. Such associations can be analyzed using a number of epidemiological study designs: ecological, cross-sectional, and prospective. A variety of ecolog- ical and cross-sectional studies investigating rela- tionships between dietary variables and cardiovas— cular diseases have been carried out using NNMS data. To date, the only NNMS survey to use a prospective cohort design has been the NHEFS. The hallmark of the ecological study design is that grouped data are used to describe both the dependent and independent variables. One type of ecological study is that in which cross-sectional group data on both a risk factor and a disease or risk factor outcome are compared among a number of different popula- tion groups. A second type of ecological design uses trend data for both the dependent and independent variables to look for associations between changes over time in one variable as compared with changes over time in the other variable. An example would be the investigation of changes in national fat availabil—- ity (from the U.S. Food Supply Series) and changes in mortality due to cardiovascular disease (from the Vital Statistics Data). Ecological associations detected in populations often overestimate the strength of the relationship among individuals. In a cross-sectional study one may attempt to relate differences in the dependent and independent vari- ables. among individuals who have been examined. Data regarding both variables are available from the same individuals. The most common methods of analysis used in these studies are correlation and regression. In some cases, where there is a high degree of intraindividual variation in one of the vari- ables, a quantile analytic approach may be used. One potential drawback of the cross—sectional design is the inability to determine temporal relationships; chronic disease may develop many years after the consumption or exposure that caused the disease. Moreover, persons who have been diagnosed as having a disease may currently be consuming a diet different from the one they consumed before the diagnosis. Misleading associations may be identified if analyses do not account for possible confounding factors. In order to determine whether or not a factor 115 associated with a disease is or is not causally related to the development of that disease, it is necessary that a clear temporal relationship be shown between the presence of the risk factor in individuals without evi- dence of disease and the future development of disease among these same individuals. In cross- sectional studies, where associations are determined at one point in time, it is not possible to identify or infer causal relationships when significant associa— tions are found between risk factors and disease. The cohort, or prospective study design, is one in which individuals in a population who are free of disease are first classified on the basis of exposure to a risk factor (on a yes or no basis in the case of dichotomous variables or on a continuous basis in the case of many biological variables such as blood pres- sure, nutrient intake, or serum cholesterol). These individuals are then followed over time and are reexamined at periodic intervals to ascertain the presence or absence of disease (in the case of death, the cause of death is determined). The risk of developing disease can be compared between the exposed and unexposed groups. A number of analytic techniques may be used, the most common being relative risks calculation, where the proportion of individuals exposed to the risk factor is compared to the proportion of unexposed individuals developing the disease over the same period of time. Ecological Associations of Changes in Cardio- vascular Mortality, Dietary Factors, and Risk Factors Over Time Trends in mortality and food supply The availability of data collected over extended periods of time on cardiovascular disease, diet, and other risk factors makes possible the examination of ecological associations and temporal relationships among the changes observed (based on group charac- teristics of the U.S. population) among these factors. Slattery and Randall (1988) have performed such an analysis using the Vital Statistics data on coronary heart disease mortality, together with data on food "consumption" implied from disappearance data in the U.S. Food Supply Series. Their analysis shows an increase in coronary heart disease mortality for men aged 45-54 and 55-64 years from 1925 to the early 1950s when the rate began to level off. In the mid- 1960s the mortality rate began to decline and the decline persisted through 1978, the last year included in this study. Pertinent changes in the food supply include the following changes in the availability of selected foods: A decrease in eggs beginning around 1950. A decrease in dairy products after the late 1940s. Fluctuations in fruits and vegetables beginning around 1955. A decrease in whole milk and an increase in low— fat milk after 1950. A change to margarine being substituted for butter occurring in the mid—1950s. An increase in poultry starting in the late 1940s. An increase in cheese. The net changes in the use of food groups and the substitutions within groups suggested a diet higher in saturated fat through the 1950s with a decrease thereafter. These changes preceded the national decline in coronary heart disease mortality, lending some support to the diet—heart hypothesis. Trends in serum cholesterol and intake of dietary fats and cholesterol Sempos et al. (1987) compared age-adjusted aggregate trends in the consumption of foods that contributed fats and cholesterol in the diet; calculated the intakes of total fat, saturated fat, linoleic acid (the major polyunsaturated fat), and cholesterol as percent of kilocalories; and predicted and observed changes in serum cholesterol levels between NHANES I (1971- 74) and NHANES II (1976-80). For all adults, there was a consistent pattern leading to a decrease in the consumption of dairy products, meat, poultry, eggs, and fats, oils, and gravies, although the magnitudes of some changes were very small. Within dairy prod- ucts, there was a shift away from consumption of whole milk to consumption of low-fat milk; there were also decreases in the consumption of butter and increases in the consumption of margarines. Changes in the intake of vegetables were not consistent for all race—-sex groups; food composition data for vegetables assumed added saturated fat (butter) in 1971-74 and added unsaturated fat (margarine) in 1976-80, if not specified by respondents. This change in coding, plus the introduction of fatty acid composition data for many new foods in the NHANES II nutrient compo- sition database, suggests that caution should be applied in the interpretation of trends between the two surveys. The results for nutrient intakes (table 5-4) indicated an increase in intake of linoleic acid Table 5-4. Trends in age-adjusted’ mean intakes of energy, fats, and cholesterol for persons aged 20-74 years, by sex and age: first National Health and Nutrition Examination Survey, 1971-74, and second National Health and Nutrition Examination Survey, 1976-80 (Sempos et al., 1987) Dietary component Sex Race 1971-74 1976-80 Change in mean Energy (kilocalories) Male White 2,483 2,481 -2 Female White 1,658 1,541 -17 Male Black 2,233 2,227 -6 Female Black 1,410 1,439 +29 Total fat (percent) Male White 36.9 36.8 -0.1 Female White 36.1 36.0 -0.1 Male Black 36.2 36.2 0 Female Black 35.5 35.8 +0.3 Saturated Male White 13.6 13.3 -0.3 fat (percent) Female White 13.0 12.5 -0.5 Male Black 12.7 12.8 -0.1 Female Black 12.4 12.3 -0.1 Linoleic Male White 3.9 5.1 +1.2 acid (percent) Female White 4.0 5.4 +1.4 Male Black 3.7 4.9 +1.2 Female Black 4.0 5.3 +1.3 Cholesterol Male White 483 436 -47 (milligrams) Female White 310 272 -38 Male Black 535 476 -59 Female Black 313 295 -18 ? Age-adjusted by the direct method to the U.S. population at the midpoint of the NHANES II. 116 (polyunsaturated fat) and a decrease in intake of saturated fat and cholesterol. These changes in nutrient intake were used in the equations of Keys, Anderson, and Grande (1965) and Hegsted et al. (1965) (see page 111) to predict the change that might be expected in mean serum cholesterol. The predicted and observed changes in serum cholesterol levels between the two surveys are shown in table 5-5. The predicted changes in the mean serum cho- lesterol levels, based on changes in the mean con- sumption of fats and cholesterol, are consistent in magnitude and direction with those observed. This finding improves confidence in the trends in dietary intake, despite the caveats noted above. Table 5-5. Change in age-adjusted’ mean serum cholesterol levels (milligrams/dl) for persons aged 20- 74 years, by race and sex: first National Health and Nutrition Examination Survey, 1971-74, and second National Health and Nutrition Examination Survey, 1976-80 (Sempos et al., 1987) Race and sex Formula® Predicted Observed White men Keys et al. -3 -3 Hegsted et al. -6 White women Keys et al. -2 -1 Hegsted et al. -6 Black men Keys et al. -4 -4 Hegsted et al. -6 Black women Keys et al. -3 -4 Hegsted et al. -4 I" Age-adjusted by the direct method to the U.S. population at the midpoint of the NHANES II. 2 See page 111 for formulas. Cross-sectional Associations of Dietary Intake and Other Risk Factors Dietary intake and serum cholesterol level Published studies of the cross—sectional association of dietary variables and serum cholesterol using the NHANES I data (NCHS, 1983a) did not show any 117 consistent associations when mean serum cholesterol was assessed for groups with different levels of dietary intake. Some of the dietary variables exam— ined in this study were frequency of consumption of high-fat foods, total dietary calories, total fat, percent of calories from fat, and the ratio of dietary linoleic acid to saturated fatty acids. The lack of association is not surprising because only a single day of dietary data was available, a situation in which potential associations are attenuated by the high level of intra- individual variation present in these data (see chapter 2). Other factors such as confounding variables and lag time for development of hypercholesterolemia contribute to these findings. Data analyses prepared for the EPONM using NHANES II data examined many of the same associ- ations, but classified individuals into groups based on quartile of serum cholesterol. These analyses also revealed no consistent associations between dietary variables and serum cholesterol levels. These find- ings suggest the relative weakness of this type of cross-sectional design to assess such dietary associa- tions with disease or other risk factors (Jacobs, Anderson, and Blackburn, 1979). Dietary intake and blood pressure Studies similar to those carried out with serum cholesterol as the dependent variable (in the preceding section) have also been carried out with blood pressure (NCHS, 1983b). The results have been similarly nonrevealing, showing positive associations of blood pressure only with alcohol and with dietary sodium to potassium ratio in blacks. Since 1982 a number of papers have been published suggesting an inverse association between calcium intake and blood pressure. Although some of these studies have utilized the NHANES I data set (Harlan et al., 1984; McCarron et al., 1984), other studies using the same data set have shown inconclusive results (Feinleib, Lenfant, and Miller, 1984; Gruchow, Sobocinski, and Barboriak et al., 1985). A recent study (Sempos et al., 1986) utilizing a quantile approach to both the NHANES I and II data did not find any significant association between calcium intake and either blood pressure or the prevalence of hypertension. Reported dietary intake was estimated in both NHANES I and II from a 24-hour recall, a technique that does not take into account the large within—person variation that exists in dietary intake; the use of such data in regression analyses often will greatly underestimate the magnitude of any association that might exist. Because of the potential for misclassification due to the large amount of within—person variation in this data set, "quantile" methods of analysis that decrease the probabilities of such misclassification should be used. It is evident from these and other attempts (preceding section) to utilize the NHANES dietary intake data to search for associations between nutrient intake and chronic disease risk factors that such analyses must be performed and interpreted carefully. Dietary patterns of women smokers and nonsmokers The food consumption patterns and dietary intakes of smokers and nonsmokers differ. Data from CSFII 1985 indicate that women smokers consumed less of fruits and vegetables and more of eggs, sugars, coffee, and alcoholic beverages than nonsmokers (Larkin et al., 1989). Their intakes of carbohydrate, fiber, vitamin C, and thiamin per 1,000 kilocalories were also lower than the intakes among nonsmokers. The total energy and fat intakes of smokers and nonsmokers did not differ significantly. There do not appear to be large differences in the food components associated with cardiovascular disease (such as fat) between smokers and nonsmokers. Smoking is a risk factor independent of dietary patterns. Cross-sectional Associations Among Nondietary Risk Factors Relationship of serum cholesterol to body mass and skinfold thickness Analysis of data from NHANES I (NCHS, 1983a) showed that successively greater guintiles of body mass index, weight (kg)/height (m)“, were associated with higher serum cholesterol levels in adults. The association was found to be independent of age, sex, and race. A similar relationship (in magnitude and direction) was found between serum cholesterol level and the sum of subscapular and triceps skinfold thicknesses. These findings are consistent with other data showing an association between obesity and hypercholesterolemia. Relationship of serum cholesterol to selected bio- chemistries Analyses of data from NHANES I (NCHS, 1983a) showed several unsuspected associations between serum biochemical measurements and serum choles— terol. Serum calcium and magnesium levels and serum glutamic acid transaminase (SGOT) were directly and independently related to serum cholesterol levels. No metabolic activities of calcium or magnesium suggest an explanation for their 118 association with circulating cholesterol levels, and no diseases marked by excesses or deficiencies of these minerals are accompanied by striking changes in serum cholesterol levels. SGOT is frequently used as a marker for liver impairment. Alcohol is the most common liver toxin; thus, the relationship between SGOT and serum cholesterol may be mediated by alcohol intake. The association between serum cholesterol, SGOT, and alcohol intake has not been adequately explored with NNMS data. Relationship of blood pressure to body mass and skinfold thickness An analysis of data from NHANES I (NCHS, 1983b) indicated that body mass index, weight (kg)/height (m)? and the sum of subscapular and triceps skinfolds were significantly and positively related to systolic and diastolic blood pressure in all race, sex, and age groups among adults. Relationship of blood pressure to selected bio— chemistries Analyses of data from NHANES I (NCHS, 1983b) in- dicated that hemoglobin levels were directly related to systolic blood pressure in women and to diastolic blood pressure in men and women in most age groups, even after controlling for body mass differ- ences. Serum cholesterol levels were also found to be directly related to blood pressure. Serum inorganic phosphate levels were inversely related to systolic and diastolic blood pressure, while serum calcium levels were directly related to blood pressure in women but not in men. The ratio of serum calcium to phosphate was strongly and directly related to blood pressure independent of age, sex, race, and body mass index. Multiple regression analysis showed age to be the best predictor of blood pressure, followed by body mass index. Associations of selected measures of body size and body fat distribution with cardiovascular disease risk factors Several studies have reported an increased ratio of waist to hip girth to be associated with increased occurrence of cardiovascular disease and diabetes independent of overall body weight. Gillum (1987b,c) used data from NHES (1960-62 and 1963-65) to analyze associations of various body measures with cardiovascular disease risk factors by race and sex in a nationally representative sample. In adults, the ratio of waist girth (measured directly) to hip girth (estimated from seat breadth and thigh clearance) was found to increase steadily with age; be higher in men than in women; be higher in blacks than in whites; and have higher values associated with higher blood pressure, higher post-load glucose levels, and greater prevalences of hypertension and hypertensive heart disease, independent of multiple confounders. In children (aged 6-11 years) and youths (aged 12-17 years), waist to hip girth declined with age; was higher in boys than in girls; and was significantly associated with systolic blood pressure in youths and with diastolic blood pressure in children, independent of confounders. These studies confirmed and extended findings from smaller studies in a nationally representative sample and indicate the opportunity for further research on the determinants and associations of body fat distribution. Multiple cardiovascular disease risk factors in the same individuals Rowland and Fulwood (1984) studied changes in the prevalence of the three major risk factors for cardiovascular disease (hypertension, cigarette smoking, and elevated serum cholesterol levels) in blacks and whites between NHANES I and NHANES II. The criteria for elevated blood pressure were sys— tolic pressure at least 160 mm mercury, and/or dia- stolic pressure at least 95 mm mercury; elevated serum cholesterol levels were defined as 6.70 mmol/L (260 mg/dl) or more; and a smoker was defined as a person who had smoked at least 100 cigarettes in his or her lifetime and was a current smoker. The prev- alence of elevated blood pressure and of cigarette smoking was found to have declined dramatically in blacks between the two surveys. The proportion of adults with zero, one, and two or more of the specified risk factors was also examined; a striking decrease in the proportion of black men and women with two or more risk factors for cardiovascular disease was observed between the two surveys, while the decline for white women was very modest (see table 5-6). In all sex—race groups the percentage of individuals with none of the three risk factors increased between 1971-75 and 1976-80. Cardiovascular disease risk factors (cigarette smoking, serum cholesterol, blood pressure) and oral contraceptive use Analyses of data from NHANES II (Russell-Briefel et al., 1986) showed that oral contraceptive users had Table 5-6. Age-adjusted’ percent distribution among groups with zero, one, and two or more risk factors for cardiovascular disease for persons aged 25-74 years, by race and sex: first National Health and Nutrition Examination Survey, 1971-74, and second National Health and Nutrition Examination Survey, 1976-80 (Rowland and Fulwood, 1984) Risk factor groups” None One Two or more 1971-1975 1976-1980 1971-1975 1976-1980 1971-1975 1976-1980 Race—sex group Percent SE’ Percent SE Percent SE Percent SE Percent SE Percent SE White males 390 13 425 1.1 46.2 1.3 45 11 148 0.38 130 0.6 White females 454 1.1 470 09 419 1.0 429 08 12.7 0.7 10.1 0.5 Black males 234 32 293 20 443 39 509 23 323 41 19.8 2.1 Black females 28.0 24 414 23 492 25 43.7 2.0 228 1.8 149 1.2 z Age-adjusted by direct method to the total U.S. population estimated at the midpoint of the 1976-80 NHANES II. 2 Risk factors included the following: systolic blood pressure of at least 160 mm mercury and/or diastolic blood 4 Pressure of at least 95 mm mercury, serum cholesterol level of at least 6.70 mmol/L (260 mg/dl), or smoker. SE = Standard error. 119 higher prevalences of elevated serum cholesterol levels and smoking, and a slightly higher age- adjusted prevalence of high blood pressure, than did nonusers. The prevalence of two or more of these risk factors was also found to be higher in users than in nonusers. These data confirm observations on the effects of oral contraceptive agents from clinical studies in a nationally representative population. Risk Factors and Subsequent Morbidity and/or Mortality Experience (Cohort Studies) Data from the NHEFS (Leaverton et al., 1987) were used to evaluate the generalizability of the Framingham risk model for the prediction of death from coronary heart disease. The model, which employed systolic blood pressure, total serum cholesterol level, and cigarette smoking, predicted remarkably well for the nationally representative sample studied in NHANES I and permit the conclusion that the major risk factors for coronary heart disease described in Framingham analyses are applicable to the U.S. white population. Rowland and Fulwood (1984) also employed a model derived from the Framingham study to estimate the expected change in coronary heart disease mortality between NHANES I and NHANES II based on changes in risk factors (hypertension, elevated serum cholesterol level, and cigarette smoking) between the surveys. Comparisons of the expected and observed mortality rates are shown in table 5-7. These results suggested that the changes in risk factors accounted for a significant proportion of the decrease in coronary heart disease mortality. Knowledge, Attitudes, and Behaviors Related to Cardiovascular Disease Prevention Various surveys of the NNMS assess knowledge, attitudes, and behaviors related to the occurrence and prevention of cardiovascular disease. For example, the 1985 NHIS (NCHS, 1986¢c, 1988a) included ques— tions to assess knowledge of factors related to high blood pressure, heart disease, and stroke. The survey results indicated that more than 90 percent of adults knew that having high blood pressure, smoking cig- arettes, or being very overweight increased a person's chances of getting heart disease. Fewer (61 percent) knew that diabetes was also a risk factor. Seventy- seven percent recognized that a person's chance of having a stroke was increased by high blood pressure. Knowledge was greater when family income and level 120 Table 5-7. Percentage decrease in age-adjusted’ rates for observed and expected coronary heart disease mortality (ICD-8, 410-413) between 1974 and 1978 for adults aged 35-74 years, by race and sex: first National Health and Nutrition Examination Survey, 1971-74, and second National Health and Nutrition Examination Survey, 1976-80 (Rowland and Fulwood, 1984) Percent decrease Observed Expected Race and sex mortality mortality White males 14 7 White females 15 8 Black males 13 13 Black females 20 16 I Age adjusted by direct method to the total U.S. population as estimated at the midpoint of the 1976-80 NHANES II. 2 From the Division of Vital Statistics. 3 Predicted from NHANES I and NHANES II survey data based on Framingham risk model. of education were higher. The percentage of elderly subjects (65 years and over) who knew of these associations was smaller than that of younger subjects (but educational levels were lower in the elderly subjects). The survey did not attempt to relate knowledge of the risks associated with high blood pressure to risk—avoidance behavior; however, nearly 85 percent of the population had had their blood pressure checked in the past year. Elderly subjects were more likely than younger subjects to have had their blood pressure checked. The 1985 NHIS (NCHS, 1986¢c) also assessed know- ledge of the associations of dietary and nutritional factors with cardiovascular diseases. In the survey population, 80 percent stated that eating a diet high in animal fat either definitely or probably increases a person's chances of getting heart disease, 58 percent indicated that sodium or salt is the food substance most often associated with high blood pressure, and 86 percent stated that "high cholesterol” increases a person's chances of getting heart disease. The Health and Diet Surveys (Heimbach, 1986, 1987), conducted in 1982, 1984, and 1986, have confirmed the increasing public awareness of the association of diet and disease, including the relationships of fat intake to heart disease and of sodium intake to high blood pressure (tables 5-8 and 5-9). Data from these surveys also suggest that increasing numbers of persons engage in risk—avoidance behaviors such as using ingredient list information to avoid or limit the consumption of salt or sodium and fats or cholesterol (table 5-10). In Health and Diet Surveys of the general public and practicing physicians jointly sponsored by FDA and NHLBI in 1983 and 1986 (Schucker et al., 1987a,b), results indicated increases in the percentages of persons in both groups who believed that reducing high blood cholesterol would have a large effect on heart disease (from 64 to 72 percent in the general population, and from 39 to 64 percent among physicians). Among adults in the general public, 46 percent reported they had had their serum cholesterol level checked in 1986 compared to 35 percent in 1983. In both years, more than 60 percent identified dietary changes (eating less cholesterol and fat) as a way to reduce serum cholesterol. In 1986, physicians reported initiating dietary intervention at lower levels of blood cholesterol in their patients than they had reported in 1983. Together, these surveys show gains in public awareness and action and in physician beliefs and therapeutic interventions in relation to high blood cholesterol risk. Table 5-9. Perceived major dietary factors related to high blood pressure: Health and Diet Surveys, 1982- 86 (Heimbach, 1986) ___Percent of respondents Factor 1982 1984 1986 Salt/sodium 54 49 49 Alcohol 26 31 31 Fats/saturated fats 20 16 22 Cholesterol 17 13 18 Caffeine 9 8 10 Major Limits to Interpretation of Data and Gaps in the Database eo The most recent dietary survey, the Continuing Survey of Food Intakes by Individuals 1985-86, provides estimates of dietary intake of several food components associated with cardiovascular diseases——food energy, fat, saturated fat, mono- unsaturated fat, polyunsaturated fat, cholesterol, dietary fiber, calcium, and potassium. Estimates of Table 5-8. Perceived relationships between diet and disease: Health and Diet Surveys, 1982-86 (Heimbach, 1987) Percent of respondents’ Issue 1982 1984 1986 Heard of health problems linked to sodium 70 83 High blood pressure 51 65 Heard of health problems linked to fat 78 84 Heart/atherosclerotic disease 64 75 Cancer 4 7 Sodium named as cause of high blood pressure 34 Fat named as cause of heart disease 29 43 Cholesterol named as cause of heart disease 26 40 Fat named as cause of cancer 12 19 Food additives named as cause of cancer 25 21 I The denominator for all estimates is all respondents. 121 Table 5-10. Reported food control behavior: Health and Diet Surveys, 1978-86 (Heimbach, 1987) Percent of respondents’ Issue 1978° 1982 1984 1986 Pay attention to ingredient list 77 75 78 79 Use information to avoid/limit 54 57 62 68 Salt/sodium 14 40 38 44 Sugar 27 30 30 32 Additives 18 17 20 26 Fats/cholesterol 7 8 9 15 On sodium-controlled diet 38 39 43 On blood cholesterol-lowering diet 14 24 On weight-loss diet 16 19 , The denominator for all estimates is all respondents. 2 FDA Consumer Food Labeling Survey (personal interviews with 1,374 consumers selected through a national area—based probability sample). Multiple responses were permitted; specific responses may sum to more than the percent of respondents who used any information. sodium intake from food are also available, but these exclude sodium from salt added at the table; thus, total sodium intake is underestimated. Estimates of alcohol intake from self reports are also less certain, because of methodological diffi culties such as underreporting. Because the prop— erties of individual fatty acids differ, estimates of their intake would also be desirable; the nutrient composition databases available during the Panel's review do not contain composition information with respect to individual fatty acids. With respect to implications for cardiovascular disease, the most recent data on the distribution of usual dietary intakes are limited because they are for women and children, groups not thought to be at high risk of cardiovascular disease. The earlier Nationwide Food Consumption Survey 1977-78 obtained dietary intake data for multiple days for both sexes and all ages, but the nutrient composi- tion database was more limited with respect to information on fatty acids and cholesterol at the time. The ability to assess the distribution of usual dietary intakes from the data obtained in the Health and Nutrition Examination Surveys is lim- ited because only 1 day of dietary data is collected. Trends in dietary intake of individuals with respect to fat and cholesterol, as assessed by the 122 dietary and health surveys of the NNMS, must be interpreted with caution because of differences in survey methodology and improvements in the nutrient composition databases over time. The ability to interpret changes in intake over time could be improved by the conduct of methodo- logical studies designed to assess the consequences of changes in survey procedures on the estimates generated. Cross-sectional associations of dietary influences and risk factors for cardiovascular diseases may be examined using data from the Health and Nutri- tion Examination Surveys, but the power of such analyses is severely restricted, because of the large day-to-day differences in the food and nutrient intake of any individual. The results of analyses of 1-day food intakes of individuals do not represent the average usual intake of any individual over a longer period of time. In study- ing diet and disease relationships, it is generally recognized that an estimate of average, or "usual," nutrient intakes is needed. With respect to studying relationships between diet and cardio- vascular disease risk factors in the NNMS, mea- surements of "usual" dietary intake are not obtained for the same individuals in whom mea- surements of risk factors are performed. Other limitations in interpretation of such associations exist because of the nature of cross-sectional data: an "association" between a postulated risk factor and a disease may be identified, not because it is causally related to the disease, but because it is related to another factor which is really one of the causes of the disease. eo Because a single 24-hour recall was used to obtain dietary intake information, there are also limi- tations in the interpretation of relationships between dietary intake and subsequent morbidity and mortality in the longitudinal NHANES I Epidemiologic Followup Study. e Surveys have not assessed directly the impact of knowledge and attitudes about cardiovascular disease risk factors and diet on patterns of food consumption or nutrient intake. Conclusions Based on the analyses discussed in this chapter, the major contributions of the NNMS to the under- standing of dietary and nutritional factors as they relate to cardiovascular diseases are the following: Populations at Risk eo Measurements of body weight, blood pressure, serum lipids, and glucose tolerance, and question— naires in the Health and Nutrition Examination Surveys permit the assessment of the prevalence of several major diet— and nutrition-related risk factors for cardiovascular diseases—-obesity (over— weight), hypertension, elevated serum cholesterol level, and diabetes——in nationally representative samples. By comparing the prevalence estimates of different population groups, some characteristics of groups most affected by each risk factor can be identified. Characteristics of the groups at risk differ depending on the risk factor considered. For example, blacks are at greater risk of hypertension than whites; women of low socio— economic status are at greater risk of obesity than women of high socioeconomic status, and persons above poverty are at greater risk of hypercholes— terolemia than persons below poverty. The char- acteristics of individuals with multiple risk factors can also be examined: black males have a higher prevalence of multiple risk factors than white males and black or white females. However, a model that quantitates the relative contribution of all risk factors, including genetic predisposition, has not been developed for application with the NNMS data to assess overall risk of cardiovascular diseases. 123 eo High intakes of fat, especially saturated fat, and cholesterol constitute risk factors for hypercholes— terolemia, and characteristics of populations with high intakes can be assessed in the NNMS. For example, in women, high fat intake is associated with being white, having more than a high school education, and smoking. Trends eo Data from the U.S. Food Supply Series provide information on trends in the foods and amounts of food components in the food supply over time. Although the inferences about food consumption and food component intake that may be drawn from these data are limited, they are nonetheless useful to demonstrate shifts over time within food supply sources of various food components related to cardiovascular diseases (notably, total fat and fatty acid groups) in the national diet. These data indicate recent shifts from animal sources of fat to vegetable sources of fat, consistent with dietary guidance to avoid too much saturated fat. Data from the surveys that collect information on food consumption indicate a decline in the con- sumption of some high-fat foods and foods con- taining saturated fat. Some changes observed in the past 20 years include a shift from whole milk to low-fat milks, an increased consumption of leaner types and cuts of meat, and an increase in the use of margarine with a concomitant decrease in the use of butter. However, as noted earlier, interpretation of trends in nutrient intake is problematic because of changes in survey methods over time. Biochemical and clinical measurements that per- mit assessment of the prevalence of overweight, hypertension, and elevated serum cholesterol lev- els have been made repeatedly over time (with limited changes in methodology). Thus, trends in the prevalence of these risk factors can be assessed. Data from the NNMS indicate a recent decline in the prevalence of hypertension and ele- vated serum cholesterol levels, but no decline in the prevalence of overweight. Associations be- tween risk factors (for example, body weight and serum cholesterol level) and the occurrence of multiple risk factors in population groups can also be examined. The NNMS trend data are useful for examining concurrent changes in group intakes or status over time. For example, changes in food availability in the food supply contributing to a decrease in the content of saturated fat have been observed to precede the decline in coronary heart disease mortality. Changes in mean serum cholesterol levels consistent with changes in mean dietary intake of fats and cholesterol can also be detected between the first and second National Health and Nutrition Examination Surveys. Determining Factors eo Sex and age are important determining factors for the risk of coronary heart disease and cerebro- vascular disease. Men are at higher risk than women for coronary heart disease and hyper- tension. Although serum cholesterol levels do not vary dramatically with sex, elevated levels con- stitute a greater risk for men than for premeno- pausal women. The dietary intakes of fat, satu- rated fat, and cholesterol are higher in males than in females. eo Race also has an important impact on relative risk of cardiovascular disease. NNMS and related data indicate that blacks are at greater risk than whites for coronary heart disease, cerebrovascular disease, hypertension, and hypercholesterolemia. More black women are significantly overweight than white women or men of either race. The effects of socioeconomic factors such as poverty status and education do not seem consistent for all risk factors related to cardiovascular diseases. Indicators of high socioeconomic status tend to be associated with a higher level of hypercholesterolemia and higher intakes of fat, but with lower prevalences of hypertension and, for women, overweight. Several surveys of the NNMS permit the assessment of knowledge and attitudes about diet and nutrition-related risk factors for cardiovascu- lar diseases. Such surveys have been repeated over time and show a trend for increasing knowledge and changing diet-related practices. Future Concerns and Recommendations Examples highlighted by the EPONM include the following: eo The rapid changes that can be expected to occur in the food supply (increases in the availability and 124 consumption of low-fat and low-salt foods) should be monitored. Nutrient composition databases should reflect changes in food composition and introduction of new food products, as appropriate, as they occur. There is a particular need for information on the specific fatty acid composition of food items. To the extent possible, greater formulation and/or product specificity in the collection of food consumption data and the compilation of food composition data should be sought. Methods for assessing usual dietary intake should be employed in the Health and Nutrition Exam- ination Surveys to enhance the usefulness of longitudinal followup. References Cited British Nutrition Foundation Task Force. 1987. Trans Fatty Acids. London: British Nutrition Foundation Publishers. Dannenberg, A. L., T. Drizd, M. J. Horan, et al. 1987. Progress in the Battle Against Hypertension: Changes in Blood Pressure Levels in the United States from 1960 to 1980. Hypertension. 10:226-233. Dresser, C. M. 1983. From Nutrient Data to a Data Base for a Health and Nutrition Examination Survey: Organization, Coding, and Values——Real or Imputed. Proceedings of the Eighth National Nutrient Data Base Conference. Minneapolis, Minn. July 25. Feinleib, M., R. J. Havlik, R. F. Gillum, et al. 1988. Coronary Heart Disease and Related Procedures: National Hospital Discharge Survey Data. Preliminary Report. Feinleib, M., C. Lenfant, and S. A. Miller. 1984. Hypertension and Calcium. Science. 226:384-386. Gillum, R. F. 1987a. Acute Myocardial Infarction in the United States, 1970-1983. Am. Heart J. 113:804- 811. Gillum, R. F. 1987b. The Association of the Ratio of Waist to Hip Girth with Blood Pressure, Serum Cholesterol and Serum Uric Acid in Children and Youths Aged 6-17 Years. J. Chronic Dis. 40:413-420. Gillum, R. F. 1987c. The Association of Body Fat Distribution with Hypertension, Hypertensive Heart Disease, Coronary Heart Disease, Diabetes and Cardiovascular Risk Factors in Men and Women Aged 18-79 Years. J. Chronic Dis. 40:421-428. Gruchow, H. W,, K. A. Sobocinski, and J. J. Barboriak. 1985. Alcohol, Nutrient Intake, and Hypertension in U.S. Adults. JAMA. 253:1567-1570. Harlan, W. R,, A. L. Hull, R. L. Schmouder, et al. 1984. Blood Pressure and Nutrition in Adults. The National Health and Nutrition Examination Survey. Am. J. Epidemiol. 120:17-28. Hegsted, D. M., R. B. McGandy, M. L. Myers, and F. J. Stare. 1965. Quantitative Effects of Dietary Fat on Serum Cholesterol in Man. Am. J. Clin. Nutr. 17:281-295. Heimbach, J. T. 1986. The Growing Impact of Sodium Labeling of Foods. Food Technol. 40:102-104, 107. Heimbach, J. T. 1987. Risk Avoidance in Consumer Approaches to Diet and Health. Clin. Nutr. 6:159-162. Jacobs, D. R., Jr., J. T. Anderson, and H. Blackburn. 1979. Diet and Serum Cholesterol: Do Zero Correla- tions Negate the Relationship? Am. J. Epidemiol. 110:77-817. Keys, A., J. T. Anderson, and F. Grande. 1965. Serum Cholesterol Response to Changes in the Diet. I. Iodine Value of Dietary Fat Versus 25-P. II. The Effect of Cholesterol in the Diet. III. Differences Among Individuals. IV. Particular Saturated Fatty Acids in the Diet. Metabolism. 14:747-7817. Kovar, M. G., M. I. Harris, and W. C. Hadden. 1987. The Scope of Diabetes in the United States Population. Am. J. Public Health. 77:1549-1550. Krebs—Smith, S. M. 1988. Total Fat in the Diets of American Women. Paper presented at the Conference for Advanced Practice and Research, American Dietetic Association. Chicago, Ill. March 12. Krebs-Smith, S. M, F. J. Cronin, and D. B. Haytowitz. 1988. Food Sources of Energy-Yielding Nutrients, Cholesterol, and Fiber in Diets of Women. Prepared for the Expert Panel on Nutrition Monitoring. Larkin, F. A, P. P. Basiotis, H. A. Riddick, et al. In press. Dietary Patterns of Women Smokers and Nonsmokers. J. Am. Diet. Assoc. Leaverton, P. E.,, P. D. Sorlie, J. C. Kleinman, et al. 1987. Representativeness of the Framingham Risk Model for Coronary Heart Disease Mortality: A 125 Comparison With a National Cohort Study. J. Chronic Dis. 40:775-784. Life Sciences Research Office, F. R. Senti, ed. 1985. Health Aspects of Trans Fatty Acids. Prepared for the Food and Drug Administration under Contract No. 223-83-2020. Bethesda, Md.: Federation of American Societies for Experimental Biology. Linn, S., R. Fulwood, B. Rifkind, et al. 1989. High Density Lipoprotein Cholesterol Levels Among U.S. Adults by Selected Demographic and Socioeconomic Variables: The Second National Health and Nutrition Examination Survey 1976-80. Am. J. Epidemiol. 129:281-94. McCarron, D. A,, C. D. Morris, H. J. Henry, and J. L. Stanton. 1984. Blood Pressure and Nutrient Intake in the United States. Science. 224:1392-1398. National Center for Health Statistics. 1988a. Health, United States, 1987. DHHS Pub. No. (PHS) 88-1232. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, T. Drizd, A. L. Dannenberg, and A. Engel. 1986b. Blood Pressure Levels in Persons 18-74 Years of Age in 1976-80, and Trends in Blood Pressure from 1960 to 1980 in the United States. Vital and Health Statistics. Series 11, No. 234. DHHS Pub. No. (PHS) 86-1684. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, R. Fulwood, W. Kalsbeek, B. Rifkind, et al. 1986a. Total Serum Cholesterol Levels of Adults 20-74 Years of Age, United States, 1976-80. Vital and Health Statistics. Series 11, No. 236. DHHS Pub. No. (PHS) 86-1686. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, W. C. Hadden and M. I. Harris. 1987. Prevalence of Diagnosed Diabetes, Undiagnosed Diabetes, and Impaired Glucose Tolerance in Adults 20-74 Years of Age, United States, 1976-80. Vital and Health Statistics. Series 11, No. 237. DHHS Pub. No. (PHS) 87-1687. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, W. R. Harlan, A. L. Hull, R. P. Schmouder, et al. 1983a. Dietary Intake and Cardiovascular Risk Factors, Part II. Serum Urate, Serum Cholesterol, and Correlates. Vital and Health Statistics. Series 11, No. 227. DHHS Pub. No. (PHS) 83-1677. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, W. R. Harlan, A. L. Hull, R. P. Schmouder, et al. 1983b. Dietary Intake and Cardiovascular Risk Factors, Part I. Blood Pressure Correlates: United States, 1971-75. Vital and Health Statistics. Series 11, No. 226. DHHS Pub. No. (PHS) 83-1676. Public Health Service. Washing- ton: U.S. Government Printing Office. National Center for Health Statistics, C. A. Schoenborn. 1988b. Health Promotion and Disease Prevention: United States, 1985. Vital and Health Statistics. Series 10, No. 163. DHHS Pub. No. (PHS) 88-1591. Public Health Service. Washington: U.S. Government Printing Office. National Center for Health Statistics, C. A. Schoenborn and V. Benson. 1988c. Relationships Between Smoking and Other Unhealthy Habits: United States, 1985. Advance Data from Vital and Health Statistics. No. 154. DHHS Pub. No. (PHS) 88-1250. Hyattsville, Md.: Public Health Service. National Center for Health Statistics, O. T. Thornberry, R. W. Wilson, and P.M. Golden. 1986c. Health Promotion Data for the 1990 Objectives: Esti— mates from the National Health Interview Survey of Health Promotion and Disease Prevention, United States, 1985. Advance Data from Vital and Health Statistics. No. 126. DHHS Pub. No. (PHS) 86-1250. Hyattsville, Md.: Public Health Service. National Center for Health Statistics and National Heart, Lung, and Blood Institute. 1987. Trends in Serum Cholesterol Levels Among US Adults Aged 20 to 74 Years. JAMA. 257:937-942. National Institutes of Health, Expert Panel on Detec- tion, Evaluation, and Treatment of High Blood Cho- lesterol in Adults. 1987. National Cholesterol Educa- tion Program. Bethesda, Md.: National Heart, Lung, and Blood Institute. National Institutes of Health, Consensus Develop- ment Panel. 1985. Lowering Blood Cholesterol to Prevent Heart Disease. JAMA. 253:2080-2086. National Institutes of Health, National Diabetes Data Group. 1985. Diabetes in America: Diabetes Data Compiled 1984. NIH Pub. No. 85-1468. Bethesda, Md.: National Institutes of Health. National Research Council, Committee on Diet and Health. 1989. Diet and Health: Implications for Reducing Chronic Disease Risk. Washington: National Academy Press. Persky, V., W. H. Pan, J. Stamler, et al. 1986. Time Trends in the US Racial Difference in Hypertension. Am. J. Epidemiol. 124:724-7317. Popkin, B. M,, D. K. Guilkey, and P. S. Haines. In press. Food Consumption Changes of Adult Women Between 1977 and 1985. Am. J. Agric. Econ. Popkin, B. M,, P. S. Haines, and K. C. Reidy. 1989. Food Consumption Trends of US Women: Patterns and Determinants Between 1977 and 1985. Am. oJ. Clin. Nutr. 49:1307-1319. Raper, N., and J. Exler. 1988. Levels and Sources of Omega-3 Fatty Acids in the U.S. Food Supply, 1935- 1985. Paper presented at the Annual Meeting of the American Dietetic Association, San Francisco, Calif. October 5. Rizek, R. L., S. O. Welsh, R. M. Marston, and E. M. Jackson. 1983. Levels and Sources of Fat in the U.S. Food Supply and in Diets of Individuals, in E. G. Per- kins and W. J. Visek, eds., Dietary Fats and Health. Champaign, Il.: American Oil Chemists' Society. Rowland, M. L.,, and R. Fulwood. 1984. Coronary Heart Disease Risk Factor Trends in Blacks Between the First and Second National Health and Nutrition Examination Surveys, United States, 1971-1980. Am. Heart J. 108:771-779. Russell-Briefel, R., T. M. Ezzati, R. Fulwood, et al. 1986. Cardiovascular Risk Status and Oral Contra- ceptive Use: United States, 1976-80. Prevent. Med. 15:352-362. Schucker, B., K. Bailey, J. T. Heimbach, et al. 1987a. Change in Public Perspective on Cholesterol and Heart Disease: Results from Two National Surveys. JAMA. 258:3527-3531. Schucker, B., J. T. Wittes, J. A. Cutler, et al. 1987b. Change in Physician Perspective on Cholesterol and Heart Disease: Results From Two National Surveys. JAMA. 258:3521-3526. Sempos, C., R. Cooper, M. G. Kovar, and M. McMillen. 1988. Divergence of the Recent Trends in Coronary Mortality for the Four Major Race-Sex Groups in the United States. Am. J. Public Health. 78:1422-1427. Sempos, C., R. Cooper, M. G. Kovar, et al. 1986. Dietary Calcium and Blood Pressure in National Health and Nutrition Examination Surveys I and II. Hypertension. 8:1067-1074. 126 Sempos, C., C. Dresser, M. Carroll, et al. 1987. Recent Trends in Serum Cholesterol and in Consumption of Dietary Fat and Cholesterol: A Comparison of NHANES I and II. Paper presented at the Annual Meeting of the Federation of American Societies for Experimental Biology. Washington, D.C. April 1. Slattery, M. L., and D. E. Randall. 1988. Trends in Coronary Heart Disease Mortality and Food Consumption in the United States between 1909 and 1980. Am. J. Clin. Nutr. 47:1060-1067. Subcommittee on Definition and Prevalence of the 1984 Joint National Committee. 1985. Hypertension Prevalence and the Status of Awareness, Treatment, and Control in the United States. Hypertension. T: 457-468. U.S. Department of Agriculture, N. Raper and R. Marston. 1988. Nutrient Content of the U.S. Food Supply: Tables of Nutrients and Foods Provided by the U.S. Food Supply. HNIS Administrative Report No. 299-21. Hyattsville, Md.:: Human Nutrition Information Service, USDA. 127 U.S. Department of Agriculture and U.S. Department of Health and Human Services. 1985. Nutrition and Your Health: Dietary Guidelines for Americans. Home and Garden Bulletin No. 232. 2nd edition. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services. 1988. The Surgeon General's Report on Nutrition and Health. DHHS Pub. No. (PHS) 88-50210. Public Health Service. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services, Health Care Financing Administration. 1980. ICD.9.CM. International Classification of Diseases. Clinical Modification. 2nd edition. 9th revision. Vol. 1, Diseases: Tabular List. DHHS Pub. No. (PHS) 80- 1260. Public Health Service. Washington: U.S. Government Printing Office. Zevenbergen, J. L. 1986. Biological Effects of Trans Fatty Acids, in C. Galli and E. Fedeli, eds. Fat Production and Consumption: Technologies and Nutritional Implications. New York: Plenum Press. Chapter 6 Assessment of Iron Nutriture Purpose Iron deficiency is the most common single nutrient deficiency in the United States and is also the most common cause of anemia (a reduction below normal in erythrocytes, hemoglobin, or hematocrit). The JNMEC (DHHS/USDA, 1986) considered iron to be a food component warranting public health monitoring priority. The U.S. Dietary Guidelines (USDA/DHHS, 1985) recommend that Americans eat a variety of foods to provide enough essential nutrients. The Surgeon General's Report on Nutrition and Health (DHHS, 1988) notes the importance of children, ado- lescents, and women of childbearing age consuming foods that are good sources of iron. The conse- quences of iron deficiency include impaired work performance, body temperature regulation, behavior, and intellectual performance; decreased resistance to infections; and increased susceptibility to lead poi- soning (DHHS, 1988). The purpose of this chapter is to demonstrate how NNMS data can contribute to the understanding of iron nutriture as well as identify the strengths and weaknesses of data and information available pri- marily from components of the NNMS in addressing this topic. In this regard, the major strength of the NNMS is the information it contains on iron intake coupled with abundant data on iron status, which provides a basis for reevaluating the reasonableness of current estimates of iron requirements and dietary allowances. The discussions in this chapter focus especially on the NNMS capabilities for identifying 1) populations at risk, 2) trends, and 3) determining factors. Limits to interpretation of data and gaps in the database are also addressed. To accomplish the objectives described above, data from the NNMS and related sources are used to ex- amine the prevalence of iron deficiency and anemia, the distribution of dietary iron intakes, factors affecting dietary iron intake and assimilation, non- dietary risk factors for impaired iron status, and relationships among dietary and other risk factors and iron status. 129 Background and Definitions The discussion below is based mainly on the reviews contained in The Surgeon General's Report on Nutrition and Health (DHHS, 1988) and in the eval- uation of iron data from NHANES II by the Expert Scientific Working Group (LSRO, 1984). Functions of Iron in the Body The iron in the body that functions to transport oxy- gen and utilize it in the production of cellular energy may be categorized as essential iron. Essential iron comprises that in hemoglobin (which contains most of the body's iron), in myoglobin, and in iron—-dependent enzymes. The other category of iron is storage iron. Ferritin and hemosiderin comprise the storage forms of iron that serve as a reserve for the production of hemoglobin and other essential iron compounds. Iron Balance Iron balance is controlled chiefly by absorption, but the control of excretion may also serve as a mechanism for "fine-tuning" balance in normal individuals. Iron absorption is ordinarily fairly low, 5-15 percent of ingested iron, and is affected by a variety of factors, including the combination of foods in each meal and factors associated with intestinal regulation. Heme iron, found principally in animal products, is absorbed to a far greater extent than nonheme iron which is found primarily in plant products. Heme iron accounts for a relatively small proportion of iron in the diet, and its absorption is not greatly influ- enced by other components of the diet. In contrast, absorption of nonheme iron, which accounts for most of the iron in food, is very much influenced by the composition of each meal. The absorption of non- heme iron is greatly enhanced when meat, fish, poultry, or foods containing vitamin C are consumed in the same meal. Conversely, the absorption of non— heme iron is decreased by the concurrent consump- tion of certain vegetables such as spinach and egg- plant, calcium-rich foods, and tea. The iron in breast milk is relatively well absorbed in comparison with that in cow milk. Various forms of iron used for food fortification differ in bioavailability. Important interactions with other minerals also occur. Supplements of calcium and magnesium de-— crease iron absorption. High doses of iron impair the absorption of zinc and copper. In iron deficiency, the absorption of the toxic minerals lead and cadmium is increased. Intestinal regulatory factors also have an effect on absorption. When iron stores are low, intestinal mucosa readily takes up iron, and the normally low proportion of dietary iron absorbed can be increased greatly. Failure to reduce iron absorption when stores are adequate occurs in a genetic disorder called hemochromatosis and leads to a toxic accumulation of iron in the body. On the other hand, malabsorption may occur in the presence of gastrointestinal dis— orders such as celiac disease, atrophic gastritis, and chronic inflammatory disease of the gastrointestinal tract. Iron Deficiency: Assessment Stages of Development and Iron deficiency is likely to result when ingestion or assimilation of iron from the diet is insufficient to match iron losses from the body or the additional requirements imposed by growth or pregnancy. The risk of iron deficiency increases during periods of rapid growth, notably in infancy (especially in pre— mature infants), adolescence, and pregnancy. In addition, iron deficiency commonly may result from excessive blood loss that cannot be compensated by dietary intake arising from heavy menstrual losses in women of childbearing years, frequent blood dona- tions, early feeding of cow milk to infants, frequent aspirin use, or disorders characterized by gastroin-— testinal bleeding. The consequences of iron defi- ciency include reduced work performance, impaired body temperature regulation, impairments in behav— ior and intellectual performance, increased suscepti— bility to lead poisoning, and decreased resistance to infections. The risk of iron deficiency is increased when there is a depletion of iron stores, assessed as a decrease in serum ferritin levels, without other laboratory abnormalities. This first stage of iron depletion is not 130 associated with adverse physiological effects, but it does represent a state of vulnerability. Low stores occur in healthy individuals and appear, in fact, to represent the usual physiological condition for grow- ing children and menstruating women. The risk of developing deficiency is minimized by the adaptive increase in iron absorption that occurs with reduced iron stores. Serum ferritin values fall to low levels during this stage, reflecting directly the decline in iron stores. Identification of this stage can be obscured by infection, which causes a rise in serum ferritin. The second stage of iron depletion can be considered to represent early or mild iron deficiency because, at this stage, there is the first possibility of adverse physiological consequences. This stage is character- ized by changes that indicate a lack of sufficient iron for the normal production of hemoglobin and other essential iron compounds, assessed by a decrease in transferrin saturation and an increase in erythrocyte protoporphyrin levels or by a significant rise in hemoglobin concentration in response to iron admin- istration. At this stage, hemoglobin levels are not greatly affected. The level of erythrocyte protopor- phyrin, a precursor of hemoglobin, increases when too little iron is available for synthesis of hemoglobin at an optimal rate; it also increases in lead poisoning, infection, and chronic disease. Transferrin saturation declines during this stage, but its measurement is subject to diurnal variation and large analytical variation; it also declines with infection and chronic disease. In the third stage of iron depletion, frank iron defi- ciency anemia occurs, assessed by low hemoglobin concentration (below the normal reference range for individuals of the same sex and age) and typically accompanied by reduced mean corpuscular volume (MCV), as well as decreased transferrin saturation, decreased serum ferritin level, and/or increased erythrocyte protoporphyrin level. (Anemia may also result from causes other than iron deficiency includ- ing infection, chronic disease, and deficiencies of other nutrients such as folacin and vitamin B12.) Changes in the body iron stores and the laboratory indicators of iron status that occur during the various stages of iron depletion are illustrated in figure 6-1. Laboratory measurements appropriate to the assess— ment of each stage are noted in table 6-1. No single biochemical indicator has proven to be diagnostic of iron deficiency. The use of several indi— cators in concert provides a much better measure of iron status. The NNMS data permit this type of assessment because of the number of different mea- sures of iron status included in the two most recent HANES: hemoglobin, hematocrit, mean corpuscular INCREASING Depletion of Body Iron ¥ «— Depletion of stores : Lowest level in : <— normal sujbects Iron deficiency Iron stores Serum ferritin Transferrin saturation Erythrocyte protoporphyrin Hemoglobin concentration TT Normal : : Iron deficiency anemia First Third Stages of iron depletion Second Figure 6-1. Changes in body iron compartments and laboratory assessments of iron status during the stages of iron depletion (adapted from International Nutritional Anemia Consultative Group [1977]) Table 6-1. Stages of iron depletion and appropriate laboratory assessments Laboratory Stage Descriptive term assessments First Depleted iron stores ~~ Serum ferritin level Second Iron deficiency Transferrin (without anemia) saturation Erythrocyte protoporphyrin Third Iron deficiency Hemoglobin anemia Mean corpuscular volume (MCV) 131 hemoglobin, MCV, serum iron; and serum total iron— binding capacity (which are used to calculate trans- ferrin saturation), free erythrocyte protoporphyrin, and serum ferritin. For analyses of NHANES II data, the Expert Scientific Working Group (LSRO, 1984) developed two different models for evaluating iron status in which two or more abnormal values for indicators of iron status were considered indicative of iron deficiency or "impaired iron status." The first, termed the ferritin model, employed serum ferritin, transferrin saturation, and erythrocyte protopor- phyrin as indicators; the second, termed the MCV model, employed MCV, transferrin saturation, and erythrocyte protoporphyrin as indicators. The deci- sion to devise two models was guided partially by practical considerations (serum ferritin was not mea- sured in the entire population) and partially by theo— retical considerations. In theory, the ferritin model would be expected to yield higher prevalences indica- tive of an earlier stage of iron deficiency, because it includes ferritin which reflects only depletion of iron stores. The MCV model, on the other hand, includes three laboratory tests all of which reflect altered erythropoiesis. Both models identify individuals in the second and third stages of iron depletion. The cutoff values, developed for use with the NHANES II data, which are indicative of the presence of iron deficiency, are given below: Serum Transferrin Erythrocyte Age ferritin saturation protoporphyrin MCV (years) (pg/L) (percent) (pmol/L RBC) (fl) 1-2 - <12 >1.42 <73 3-4 <10 <14 >1.33 <75 5-10 <10 <15 >1.24 <7 11-14 <10 <16 >1.24 <78 15-74 <12 <16 >1.24 <80 Some of the indicators used (transferrin saturation and erythrocyte protoporphyrin) are also affected by inflammatory disease. Thus, the two models may overestimate the contribution of iron deficiency. However, analyses of NHANES II data showing the increase in the prevalence of low hemoglobin values with increasing numbers of abnormal values for indi- cators of iron status support the usefulness of the models for identifying populations with iron defi- ciency (see table 6-2). Similar analyses also show that the models developed with the NHANES II data are applicable to the three ethnic groups represented in HHANES. Table 6-2. Percent’ of persons with low hemoglobin concentrations” in iron status groups as defined by the MCV model’: second National Health and Nutrition Examination Survey, 1976-80, and Hispanic Health and Nutrition Examination Survey, 1982-84 MCV model 3 normal 1 abnormal 2o0r3 Sex (years) Population values value abnormal values Males and females 5-10 Non-Hispanic white PA 1.5 5.9 62) 519) 4 Mexican American 04 Co 220) 35) Puerto Rican 14 * (217) © (10) Cuban 0 * * (68) (13) 3) Females 20-44% Non-Hispanic white 2.4 6.2 31.3 (1,558) (351) (103) Mexican American 2.0 9.5 37.3 (680) (222) (103) Puerto Rican 2.9 11.0 39.8 (275) (76) (25) Cuban 0 5.8 (138) (33) (15) Males 45-74 Non-Hispanic white 22 8.2 29.4 (2,154) (313) (63) Mexican American 0.6 5.8 * (435) (64) (5) Puerto Rican 3.4 * % (170) (22) (1) Cuban 1.5 * * (210) (16) 1) Females 45-74 Non-Hispanic white 2.0 6.1 24.8 (2,196) (520) (104) Mexican American 12.2 6.4 39.8 #59 (113) (49) Puerto Rican 3.1 alo) (58) (24) Cuban 2.3 * 07) (48) (8) I" All statistics are weighted. 2 See page 134 for criteria for low hemoglobin. 3 The MCV model employs mean corpuscular volume, transferrin saturation, and erythrocyte protoporphyrin. See page 131 for age-specific cutoff values used in this analysis. 4 Value in parentheses is number of examined persons. 5 Pregnant women are excluded. * Indicates a statistic not reported because of small sample size. 132 When data for all four measures (serum ferritin, erythrocyte protoporphyrin, transferrin saturation, and MCV) are available, there may be an advantage in constructing a single, four—variable model in which either the value for serum ferritin or for MCV plus one additional value must be abnormal. This model not only simplifies presentation of information about iron status, but also helps to eliminate the contribu- tion of inflammatory conditions to the prevalence estimates. This model was tested using the data for non-Hispanic persons in NHANES II; comparisons of estimates of iron deficiency obtained with the MCV, ferritin, and four-variable models are shown in figures 6-2, 6-3, and 6-4, for children aged 4-11 years, males aged 12-74 years, and females aged 12-74 years, respectively. These comparisons indicate show that, as expected, prevalence estimates with the four-variable model were similar to those given by the ferritin model in groups with a relatively high or moderate prevalence of iron deficiency. For those groups suspected of having a high prevalence of inflammatory disease (the elderly), the estimates of the prevalence of iron deficiency obtained with the four—variable model were lower than those obtained with the MCV model. The EPONM concluded that the four-—variable model represents a promising approach for the assessment of iron nutriture in the future. Further development and refinement of the Iron Deficiency by Three Models: Children MCV model Ferritin model Four—variable model Percent CC] ANN ay 14 12} 10 8 6 |e 4} i 4-5 6-11 Age in years Figure 6-2. Prevalence of iron deficiency assessed by three models in non-Hispanic children aged 4-11 years: second National Health and Nutrition Exam- ination Survey, 1976-80 (See text for explanation of models.) 133 Iron Deficiency by Three Models: Males Ferritin model Four—variable model ry EE MCV model C1] Percent 14 lem 0° [Na [Sm IN \ \ 16-19 20-29 30-39 40-49 50-59 60-69 70-74 Age in years 12-15 Figure 6-3. Prevalence of iron deficiency assessed by three models in non-Hispanic males aged 12-74 years: second National Health and Nutrition Exami- nation Survey, 1976-80 (A zero in the figure indicates a prevalence estimate of 0.0 percent. See text for explanation of models.) Iron Deficiency by Three Models: Females Four—variable model MCV model Ferritin model C1] MAAN Percent 14 -15 16-19 20-29 30-39 40-49 50-59 60-69 70-74 Age in years Figure 6-4. Prevalence of iron deficiency assessed by three models in non-Hispanic females aged 12-74 years: second National Health and Nutrition Exami- nation Survey, 1976-80 (See text for explanation of models.) model and the cutoffs for indicators are recom- mended. The variables used in each of the three models are summarized in table 6-3. An advantage of all three models in estimating the prevalence of iron deficiency in blacks is that they do not include the hemoglobin concentrations. Hemoglobin concentrations are lower in blacks than in other racial groups, independent of iron status, for reasons that are thought to have a genetic basis. In the remainder of this chapter, estimates of preva— lence of iron deficiency are provided using the MCV model only for the following reasons: e Use of a single model simplified the presentation of data, but nonetheless permitted useful illustrations of the factors that influence iron nutriture. e Serum ferritin was measured only in a subsample of subjects in NHANES II. e Although serum ferritin was measured in all sub- jects in HHANES, the data were not available for review by the EPONM. Data on hemoglobin concentrations from the HANES are also useful for estimating the prevalence of anemia and iron deficiency anemia. The criterion used was the 2.5th percentile of hemoglobin concen— tration in reference populations formed by elim- inating persons with abnormal values for other iron status indicators. The age-specific values derived from this criterion and used to evaluate NHANES II and HHANES data appear in the next column: Hemoglobin cu ue Age Males and (years) females Males Females 1-2 109° = - 3-4 109 ~ - 5-10 112 - - 11-14 - 120 118 15-19 - 131 117 20-44 - 134 119 45-64 = 132 118 65-74 - 136 119 I The cutoff values were estimated using the method of Dallman et al. (1984), for whites only: second National Health and Nutrition Examination Sur- vey, 1976-80 (LSRO, 1984). Because of the small sample size, the value obtained for subjects aged 3-4 years was also used for subjects aged 1-2 years. This value was also in accord with data from other sources. Hemoglobin level or hematocrit is measured in the women and children included in the PNSS and PedNSS. For these surveillance activities, the criterion for low hemoglobin and hematocrit was developed recently by CDC, based on 5th percentile values for a reference population from NHANES II. Table 6-3. Laboratory measurements used in three different models for assessing iron deficiency MCV model’ Ferritin model’ Four-variable model? MCV Serum Transferrin saturation Erythrocyte protoporphyrin Transferrin saturation Erythrocyte protoporphyrin ferritin MCV or serum ferritin Transferrin saturation Erythrocyte protoporphyrin I Two of three values must be abnormal. 2 Values for either MCV or serum ferritin (or both) must be abnormal, in addition to at least one other abnormal value. 134 Hemoglobin and hematocrit values derived from this criterion and used with most of the CDC surveillance data in this report are as follows: Hemoglobin Hematocrit Age (years) (g/L) (percent) 1- 1.9 110 33 2- 4.9 112 34 5- 7.9 114 35 8-11.9 116 36 12-14.9 (females) 118 36 12-14.9 (males) 123 37 15-17.9 (females) 120 36 15-17.9 (males) 126 38 18+ (females) 120 36 18+ (males) 135 40 Pregnancy 1st trimester 110 33 2nd trimester 105 32 3rd trimester 110 33 Iron Overload and Hemochromatosis Iron overload results in the toxic accumulation of storage iron in tissues. The most common cause is the hereditary condition hemochromatosis in which a genetic defect in iron absorption is considered the antecedent of abnormal iron accumulation over years with eventual damage to the liver, heart, pancreas, and adrenal glands. Arthritis is also a frequent man- ifestation of hemochromatosis. Individuals who are homozygous frequently develop symptoms by middle age; heterozygous carriers of the gene may be at increased risk for iron overload. The expression of symptoms may be influenced by age, sex, alcohol intake, and other factors. Iron overload may also result from frequent transfusions or, under extreme circumstances, from excessive oral intake. The ability to assess iron overload using the bio- chemical indicators of iron status available in the NNMS is limited. For analyses from NHANES II conducted by LSRO (1984) and discussed in this report, the presence of transferrin saturation greater than 70 percent plus an elevated serum ferritin value was regarded as indicative of possible iron overload. Age- and sex-specific criteria were used for the evaluation of elevated serum ferritin values. Difficulties arise in assessing both the prevalence and the health consequences of hemochromatosis. 135 Edwards et al. (1980) found that in 35 homozygotes identified through pedigree studies, thirteen were asymptomatic. Symptoms were rare in all women and in men less than 30 years; men usually had no symp- toms until 50 years of age or older. Lindmark and Eriksson (1985) screened 941 men aged 55 years or older and reviewed 8,834 autopsies on males in Sweden (where the level of iron fortification of cereal products is two to three times greater than that in the United States) during a nine-year period. They esti mated that clinically important idiopathic hemo- chromatosis was present in fewer than 1:1,000 males, but concluded that the higher prevalence found in certain parts of northern Sweden, France, and Aus- tralia probably represented real regional differences. (It should be noted that the HANES and other NNMS surveys are not designed to estimate reliably the prevalence of conditions that occur so infrequently in the population.) Edwards et al. (1988) screened 11,065 blood donors in presumed good health. Their calculations indicated that the frequency of fasting serum transferrin values greater than or equal to 62 percent was 0.008 in men and 0.003 in women. Based on HLA typing, the frequency of the homozygous condition was estimated at 0.0045. Conceptual Model Based on the review above, figure 6-5 presents a con- ceptual model which illustrates the relationships among dietary and other factors and iron status, as well as relevant sources of data from the NNMS. Components of the model that are most relevant to iron nutriture are highlighted by the shaded boxes; individual topics that appear in boldface type are those for which some data are available. Potential NNMS and related data sources are represented by the numbers that appear above or below the boxes; numbers that appear in boldface type represent those surveys or studies from which data relevant to assess— ment of iron nutriture were obtained. Dietary iron intakes and factors affecting iron intake were measured in the NFCS 1977-78, CSFII 1985-86, and the HANES. In addition, the data from the NFCS 1977-78 have been used to assess the contri- bution of foods consumed together to the "absorbable iron" (Raper et al., 1984). Supplement intake was as- sessed in a variety of surveys, but quantitative esti- mates of intake are currently available only from the FDA Survey of Vitamin/Mineral Supplement Intake. The iron status indicators used in the various models for assessing iron deficiency were measured in NHANES II and HHANES; various factors that influ- ence iron status have also been assessed. Data on hemoglobin or hematocrit status of low-income 9€T NATIONAL FOOD SUPPLY —> FOOD DISTRIBUTION —— CONSUMPTION —————> NUTRIENT UTILIZATION ———— > HEALTH OUTCOME 125 6+«7+8 9 13 14x 15+ 16 19 Sanitation Housing Occupation Other factors 12 , | Environmental factors | Away-from-home Away—-from—home Away-from-home EE 3 fo yf ! Agricultural factors food available {food acquired {ood consumed + Economic factors , . Policy considerations : 1»2 567839 —_— TP Loh 5 Food preferences, y J 1 cognitions, attitudes : 142044 5004 SRA V7 9 11s (3%) 3 eth SE a ! Storage | 125 64748 9 = 5 \_ Household food Household food Household food 14% 15+ 16 19 B 5742 14r 15+ ——————— available — acquired —) consumed 12 125 6+7 9 12« Primary representative action or consequence ' i Influencing or mitigating factor National Nutrition Monitoring System and other data sources: 1 = CSFIl 1985-86, 2 = NFCS 1977-78, 3 = U.S. Food Supply Series, 4 = National Nutrient Data Bank, 5 = NHANES, 6 = NHANES II, 7 = HHANES, 8 = NHEFS, 9 = NHIS, 10 = FLAPS, 11 = Total Diet Study, 12 = Vit/Min, 13 = Health and Diet Study, 14 = PedNSS, 15 = PNSS, 16 = BRFSS, 17 = U.S. Vital Statistics, 18 = AEDS, 19 = NHES. See appendix lll for definitions of acronyms. Shaded boxes highlight portions of the model discussed; an asterisk (+) indicates data and data sources considered in this chapter. Figure 6-5. Conceptual model for the assessment of iron nutriture (see text for explanation) children and pregnant women are collected in the CDC surveillance activities and permit some limited evaluation of trends in iron nutritional status. The NHEFS offers the possibility to examine the influence of iron nutritional status on subsequent health. Estimates of Prevalence Iron Deficiency The prevalence of iron deficiency determined by the MCV model is presented for non-Hispanic persons aged 4-74 years, by sex, age, and race, from NHANES II in tables 11-98 and 11-99, and for the three ethnic groups from HHANES in tables II-94 and II-95 in appendix II. The results are summarized in figures 6-6, 6-7, and 6-8 for children aged 4-19 years, males aged 20-74 years, and females aged 20-74 years, respectively. Analyses by poverty status for non- Hispanic persons in NHANES II (tables II-100 and 11-101) and Mexican Americans in HHANES (tables 11-96 and 11-97) are also presented (see discussion of poverty below in this chapter). Because these analy- ses do not include data for children under 4 years, results for younger children (1-4 years), based on Iron Deficiency by MCV Model: Children and Adolescents Non-Hispanic Non-Hispanic Puerto Rican white black Bg [1 a Cuban NANNY Mexican American Percent 14 % p ] © J p J bo J © J po J 5 J b J b » J % o M12-15 F 12-15 Sex and age in years N M&F 4-5 M&F 6-11 M 16-19 Figure 6-6. Prevalence of iron deficiency assessed by the MCV model in children and adolescents aged 4-19 years, by ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 (A zero indicates a prevalence estimate of 0.0 percent; an asterisk indicates an unstable statistic or a statistic not reported because of small sample size.) 137 Iron Deficiency by MCV Model: Males Non-Hispanic Non—Hispanic Puerto Rican white black By [C1 am Cuban NNNNN Mexican American Percent 14 20-29 30-39 40-49 70-74 50-59 Age in years 60-69 Figure 6-7. Prevalence of iron deficiency assessed by the MCV model in males aged 20-74 years, by ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 (A zero indicates a prevalence estimate of 0.0 percent; an asterisk indicates an unstable statistic or a statistic not reported because of small sample size.) Iron Deficiency by MCV Model: Females Non-Hispanic Non-Hispanic Puerto Rican white black BRR [C1 a Mexican American Cuban XX] Percent 14 12 10 70-74 60-69 Age in years Figure 6-8. Prevalence of iron deficiency assessed by the MCV model in females aged 20-74 years, by ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 (A zero indicates a prevalence estimate of 0.0 percent; an asterisk indicates an unstable statistic or a statistic not reported because of small sample size.) analyses reported by the Expert Scientific Working Group (LSRO, 1984), are provided separately in iron table 1 following the references in this chapter. Based on these data analyses, the major populations currently at risk are clearly delineated: infants, ado— lescents, and women during the childbearing years. Men, elderly persons of both sexes, and children between infancy and adolescence are much less likely to be iron deficient. The risk in HHANES Mexican Americans was similar to that in non-Hispanic whites in NHANES II, except that the prevalence in women of childbearing years was higher; this difference may be explained by increased parity among Mexican- American women. Among the age and sex groups at risk, there is as much as a threefold difference in the prevalence of iron deficiency according to poverty status. The overall prevalence of iron deficiency in the groups at greatest risk was lower than antici- pated from earlier estimates based on different cri- teria. Nevertheless, iron deficiency remains the most common known nutritional deficiency in the United States. Anemia The prevalence of iron deficiency anemia in females, defined by low hemoglobin values plus evidence of impaired iron status based on the MCV model, is shown in table 6-4 (Looker et al, 1989); the prevalence of iron deficiency anemia in males was very low. Women in the age group 20-44 years were the only ones with relatively high prevalence estimates. The prevalence of anemia from any cause (low hemoglobin values) for white persons in NHANES II and HHANES is reported in iron table 2 following the references at the end of this chapter (Looker, 1988). These analyses indicate that the prevalence of anemia was low in most groups. Hemoglobin criteria are given for whites only because of uncertainty about appropriate values for blacks. The mean hemo- globin concentration in blacks is lower than in whites, irrespective of iron status, probably because of a genetically determined difference in globin synthesis. Table 6-4. Percent of females with iron deficiency anemia’, by age and race or ethnic group: second National Health and Nutrition Examination Survey, 1976-80, and Hispanic Health and Nutrition Examination Survey, 1982-84 (Looker et al., 1989) Age 2 Standard error (years) Population n Percent of the percent 11-14 Non-Hispanic white 110 0.2 0.24 Mexican American 310 0.6 0.42 Cuban 40 o* o* Puerto Rican 124 2.7 1.41 15-19 Non-Hispanic white 197 0.8 0.50 Mexican American 302 1.7 0.71 Cuban 40 o* 0* Puerto Rican 152 2.9 1.31 20-44 Non-Hispanic white 618 1.5 0.25 Mexican American 889 4.1 0.76 Cuban 172 4.8 1.93 Puerto Rican 363 2.8 1.08 45-64 Non-Hispanic white 460 1.0 0.31 Mexican American 478 3.0 0.74 Cuban 193 0 0 Puerto Rican 242 1.8 0.76 65-74 Non-Hispanic white 339 0.7 0.27 Mexican American 97 1.0 1.0 Cuban 53 3.9 2.59 Puerto Rican 48 0 0 I Iron deficiency anemia is defined as having two or three abnormal values for the indicators in the MCV 2 model and a low hemoglobin value (see page 131 for criteria for cutoff values). Number of examined persons. * Indicates statistic that may be unreliable due to sample size. Data on the prevalence of anemia (low hemoglobin or low hematocrit) in low-income children from the PedNSS, 1973-88, are shown in figure 6-9. Interpretation of trends over time is complicated because the number of States participating has increased over time; thus, the base population has changed. To overcome this difficulty, Yip etal. (1987a) have analyzed data from 1975 to 1985 from six States in the PedNSS to assess trends in the prevalence of anemia (low hematocrit or hemoglobin values) in low-income children over time. The criteria for defining anemia in this study were lower than the usual CDC cutoffs for surveillance in order to exclude any doubtful cases of anemia. These data are shown in figure 6-10 and suggest a generalized improvement in child iron status as well as a positive impact of public health programs. The former suggestion is supported by the observation of a declining prevalence of anemia in a group of middle class children in Minneapolis (Yip et al, 1987b). Racial differences in the prevalence of anemia were also evident, with blacks showing the highest preva- lence, Native Americans the lowest, and whites an intermediate level (the same criteria were used for all racial groups, probably resulting in overestimation of the prevalence of anemia in blacks). oo Nee: [nema in Lowfr income Phildremy er: folic: New AL: Low hematocrit ——m— Low hemoglobin —— Percent 40 35} 30+ 25} 20} Oo. ea 1 1 0 ) . \ \ L L 1972 1974 1976 1978 1980 1982 1984 1986 1988 Year Figure 6-9. Prevalence of anemia (hemoglobin or hematocrit <56th percentile) in children 0-59 months: Pediatric Nutrition Surveillance System, 1973-88 Iron Overload Only limited assessment of iron overload is possible with NNMS data. Among the 3,540 adults in NHANES II with serum ferritin values, 22 had 139 Anemia in Low—Income Children: Six Selected States First visit Followup visits ——— —— — Percent 10 9} 8 7 6 5 4} 3 2 1 0 + L L \ \ L 1972 1974 1976 1978 1980 1982 1984 Birth Year Figure 6-10. Prevalence of anemia in children 6-60 months for each birth—year cohort, six selected States: Pediatric Nutrition Surveillance System, 1973-84 (Yip et al, 1987a) (Top line represents the prevalence for children seen at initial visit to qualify for WIC program only; bottom line, prevalence for children covered by the (Pee seen at elevated transferrin saturation and only 9 had both elevated transferrin saturation and serum ferritin values. Of these nine, only five had other biochemi- cal measurements suggesting uncomplicated idio- pathic hemochromatosis. Although the NHANES II was not designed to yield statistically reliable esti- mates of conditions which are expressed at such low frequency in the population, these values were not out of line with estimates of prevalence in the medical literature. Dietary Factors and Supplement Intake Iron Available in the Food Supply The changes in the per capita iron content of the food supply over time are shown in figure II-37 in appen- dix II. Iron content was lower prior to the 1940s when enriched white flour was introduced and rose approximately 24 percent from 1935-39 through the early 1980s, with a further increase since 1982. The iron level achieved a peak in 1985 at 17 milligrams per capita per day. The recent upward trend in iron is the result of increased iron content for enriched flour concomitant with an increase in per capita use of flour and cereal products, especially fortified cereal products. Food Sources of Iron Available in the Food Supply The food sources of iron available in the food supply have also changed over time (see figure 6-11 for comparisons of 1909-13, 1970, and 1985), with a slight decrease in the percentage of iron contributed by meats since 1970 and an increase in the percent- age contributed by grains. The primary reason for these recent changes was an increase in the standard used for enrichment of white flour in 1983, coupled with an increase in the use of flour and cereal. Sources of Iron in Food Supply 1909-13 1970 C1 Percent 45 40} 3st so} Meat, poultry, and fish Grain products Vegetables Legumes, nuts, and soy Figure 6-11. Major food sources of iron in the food supply: U.S. Food Supply Series, 1909-13, 1970, and 1985 Iron Content of Typical U.S. Diets Data from the FDA Total Diet Study (Pennington et al, 1984) indicate that the analyzed mean content of iron in typical U.S. diets changed little in the period 1974-82 for adult male, toddler, and infant diets (table 6-5). The measured iron contents of the adult male and toddler diets agreed well with the cal- culated dietary iron intakes from the NFCS 1977-78 and NHANES II (this finding is reassuring because the foods collected for the Total Diet Study were 140 based on the foods most commonly reported in the NFCS 1977-78). However, the measured iron content of the infant diet (based on 1965 food consumption data) was lower than dietary intake estimates from the two surveys and the level over time does not reflect changes known to have occurred in infant diets. A possible contribution to this discrepancy was that the infant formula collected for the infant total Table 6-5. Iron levels of adult male, infant, and toddler diets: Food and Drug Administration Total Diet Study, 1974-81/82 (Pennington et al., 1984) Diet and Number of Mean! Standard year collections (mg/day) deviation Adult diets® 1974 30 20.0? 3.0 1975 20 18.0° 3.1 1976 20 18.220 5.9 1977 25 18.3% 5.8 1978 20 17.9 41 1979 20 18.320 4.2 1980 20 21.0% 5.1 1981/82 27 18.420 3.2 Infant diets’ 1975 10 742 5.7 1976 10 7.3% 4.0 1977 12 6.8% 3.6 1978 10 4.7% 2.1 1979 10 9.6% 10.3 1980 10 10.2 48 1981/82 13 9.5% 6.3 Toddler diets? 1975 10 11.12 3.8 1976 10 10.12 3.2 1977 12 8.5% 2.6 1978 10 1.72 2.3 1979 10 11.9% 11.5 1980 10 9.1% 2.0 1981/82 13 9.0 1.7 Mean values within each diet category with the same superscript are not significantly different. 2 Adult diets are based on 2,850 kilocalories per day. 3 Infant (6 months) diets are based on 880 kilo— calories per day. 4 Toddler (2 years) diets are based on 1,300 kilo- calories per day. diet was not specified to be iron fortified. The use of iron—fortified formula has increased in the last 20 years (Martinez and Ryan, 1985) The major source of iron in all diets was grain and cereal products (con— tributing 52, 54, and 48 percent of the iron in the adult male, infant, and toddler diets, respectively). Mean Dietary Iron Intakes Among Age and Sex Groups Mean dietary iron intakes by age and sex, based on 1-day data from NHANES I, NFCS 1977-78, NHANES II, and CSFII 1985-86 are shown in table 11-93 in appendix II. These values are remarkably similar over time, with the exception of a slight increase in intake in the youngest children. This observation is not unexpected based on market changes in infant formula (infants are now fed more iron—fortified formulas than previously). In fact, this change would be expected to have a greater impact in younger infants (not included in the table), especially those less than 9 months old. The trends in individual intakes do not seem to reflect the recent increase in the iron content of the food supply. Methodological differences in the food consumption surveys may contribute to this finding. Perloff (1988) has examined the impact of differences in composition databases for iron in the NFCS 1977- 78 and the CSFII 1985-86 on the iron intake of women. The analyses indicated an increase due to product changes (increased fortification) and a decrease due to data changes (revised values for some meats), with little net change. Estimates of the mean dietary iron intakes of women and children, based on 4-day CSFII data are shown in tables II-89 through II-92. For women aged 20- 49 years, mean dietary iron intake was 10.1 milli- grams per day. Mean intake did not differ greatly with age, but did vary by race, poverty status, and education level. Black women had intakes 11 percent lower than white women and 15 percent lower than women of other races, women below poverty had intakes 8 percent lower than women above poverty, and women with less than a high school education had intakes 18 percent lower than those with more than a high school education. The intake of iron per 1,000 kilocalories did not differ by socioeconomic status. Using a logistic regression analysis with these data, Moshfegh (1988) found that the following factors were significantly associated with high iron intake (intake in the highest tertile): e Being a race other than white or black. e Living alone. e Living with a male. e Being a former smoker. e Having no days of less than usual intake. e Being pregnant. The mean dietary iron intake of children aged 1-5 years was 9.8 milligrams per day; older children (3-5 years) had higher intakes than younger ones (1-2 years). Poverty status and education of the mother did not have notable effects on the children's intakes. Relationship of Iron Deficiency and Iron Intake More than 95 percent of the women in the most recent dietary survey (CSFII 1985-86) had iron intakes below the RDA and about half of women in childbearing years had intakes below estimated mean requirements’. This finding would suggest that a very large number of women might be iron deficient, yet only about 5 percent had biochemical evidence of deficient iron status in the NHANES II. Even allowing for substantial underreporting of intakes, this observation suggests that in women with decreased iron stores, homeostatic mechanisms that promote the increased absorption of dietary iron are very effective in preventing iron deficiency. These findings also suggest that the RDA may be overly generous. Mean intakes of iron by young children with and without iron deficiency determined by the MCV model in NHANES II are shown in table 6-6. Intakes 4 Mean iron requirements for women during the childbearing years can be estimated as follows: (1) Basal iron losses average 0.9 milligram per day corrected for smaller body surface area in women, and based on the value of 1.0 milligram per day reported from experiments in men (Green et al, 1968). (2) Menstrual iron losses average 0.5 milligram per day (Hallberg et al, 1966). (3) Total iron losses average 1.4 milligrams per day (0.9 plus 0.5). (4) Absorbed iron has to equal 1.4 milligrams per day to balance total iron losses. (5) An assumed iron absorption of 12.5 percent is proposed in the draft of the 10th edition of the Recommended Dietary Allowances. The figure of 12.5 percent is midway between the values of 15 and 10 percent which were suggested by Monsen et al. (1978) for diets characterized by good and moderately good bioavailability of iron, respectively. The value seems appropriate because intakes of protein and ascorbic acid, which enhance iron absorption, are relatively high in U.S. diets. (6) Dietary iron has to average 11 milligrams per day in order to yield 1.4 milligrams iron absorbed, assuming 12.5 percent iron absorption. 141 Table 6-6. Dietary iron intake (milligrams per day) for non-Hispanic children aged 1-5 years, by iron status determined by the MCV model: second National Health and Nutrition Examination Survey, 1976-80 Iron status Deficient Not deficient Standard Standard Age error error (years) n! Mean of mean n! Mean of mean 1-2 46 175 0.49 434 9.1 0.24 3-5 70 9.5 0.36 1,313 10.2 0.10 7 Number of examined persons. Children with blood collected by finger stick were excluded. Data were combined for all race and sex groups. were slightly higher in the children aged 1-2 years without iron deficiency than in those with poorer status. An earlier study of the association of dietary iron and iron deficiency with NHANES I data (NCHS, 1982) had found no association between dietary intake and status except in the youngest children, probably because of the large contribution of intra- individual variation to the dietary intake data, which were based on a single day's recall. It may be hypo- thesized that the dietary intakes of infants and young children vary less from day to day than those of older individuals, so that the single 24-hour recall used in the HANES may provide a relatively good estimate of usual intake for the youngest children. Food Combinations Using the model developed by Monsen et al. (1978), Raper et al. (1984) calculated bioavailable iron intake in the NFCS 1977-78. For each eating occasion (foods consumed within 60 minutes), total iron, heme iron, nonheme iron, and the enhancing factors, as- corbic acid and meat, poultry, or fish were calculated. Then, for each eating occasion, available heme and nonheme iron were calculated using a factor of 23 percent for heme iron and factors of 3-8 percent (based on logarithmic function with enhancing 142 factors) for nonheme iron. Finally, the available heme iron and nonheme iron were summed to yield an estimate of total available iron. Results of these calculations are shown in tables 6-7 and 6-8; avail- able iron intakes also appeared low in relation to requirements. This approach has promise in provid- ing information about dietary iron adequacy, but needs to be tested to determine whether available iron intake correlates better with iron nutritional status than does total iron intake. Table 6-7. Total, nonheme, and heme iron intakes of persons, by sex and age, 1 day: Nationwide Food Consumption Survey 1977-78 (Raper et al., 1984) Dietary iron intake Total Nonheme Heme Sex and iron iron iron age (years) n (mg) (percent)? (percent)! Males and females 1-2 268 7.7 94 6 3-5 437 9.2 94 6 6-8 469 10.7 92 8 Males 9-11 216 12.7 91 8 12-14 313 15.0 92 8 15-18 400 16.7 90 10 19-22 287 15.1 89 12 23-34 770 14.8 90 11 35-50 784 14.5 90 11 51-64 634 14.0 89 11 65-74 295 13.9 91 0 75 and over 128 12.2 91 8 Females 9-11 241 11.6 92 8 12-14 309 11.0 91 9 15-18 402 10.7 92 9 19-22 337 10.7 90 10 23-34 949 9.9 90 10 35-50 942 9.9 90 10 51-64 792 10.4 90 10 65-74 377 9.9 92 9 75 and over 197 9.9 93 7 1 Components may not add to 100 percent due to rounding. Table 6-8. Available iron intakes for persons, by sex and age, 1 day: Nationwide Food Consumption Survey 1977-78 (Raper et al., 1984) Sex and Available iron age (years) (mg) (percent of total) Males and females 1-2 0.50 6.5 3-5 0.64 7.0 6-8 0.80 7.5 Males 9-11 0.96 7.6 12-14 1.17 7.8 15-18 1.39 8.3 19-22 1.32 8.7 23-34 1.28 8.6 35-50 1.25 8.6 51-64 1.19 8.5 65-74 1.13 8.1 75 and over 0.95 7.8 Females 9-11 0.86 7.4 12-14 0.86 7.8 15-18 0.86 8.0 19-22 0.83 8.2 23-34 0.80 8.1 35-50 0.81 8.2 51-64 0.85 8.2 65-74 0.79 8.0 75 and over 0.73 7.4 Heme and nonheme iron intakes for the women and children in CSFII 1985-86 have also been deter- mined. Heme iron, which is best absorbed, comprises only 8 and 5 percent of the total iron intake of women and children, respectively. Use of Supplements: Iron and Ascorbate Information on iron and ascorbate use by age and sex from the FDA Vitamin/Mineral Supplement Intake Survey is provided in appendix II, tables II-132 and 11-133. Twenty-two percent of the adult population surveyed consumed iron supplements and more than half of all supplement users consumed iron; use was highest in women aged 16-64 years. The majority of supplement users (approximately 90 per- cent) also consumed ascorbic acid. Dietary iron intake and iron status of vitamin/mineral supplement users in NHANES II have been examined by Looker et al. (1987, 1988). The type and amount of the supplement(s) used were not specified. Sup- plement users aged 1-19 years consumed more ascorbic acid and more fruits and vegetables than nonusers, but values for iron status measures were not consistently associated with supplement use. The same was true for persons aged 20-74 years, with the exception that several values for iron status indicators were higher for users than for nonusers in the age group 65-74 years. Prevalences of "impaired iron status" or iron deficiency determined by the MCV model did not differ by supplement use. Possibly, these findings are explained by the fact that supple- ment use is highest among groups with higher socio- economic status whose risk for iron deficiency is low. Iron Fortification of Foods Changes in the type of iron compounds used for forti- fication, as assessed by National Academy of Sciences' surveys of industry use (poundage) of Generally Rec- ognized as Safe (GRAS) substances, suggest increased use of forms that are more bioavailable (Committee on GRAS List Survey [Phase III], 1979; Subcommittee on Review of GRAS List [Phase II], 1972). The contribution of enriched and fortified grain products to iron intake of individuals was assessed in the NFCS 1977-78 (Cook and Welsh, 1986) and was found to be substantial. These sources of added iron provided 20 percent of the total dietary iron. There has also been an increase in the proportion of infant formula that is iron fortified, and the feeding of cow milk to infants has decreased (Martinez and Ryan, 1985). These factors seem to have contributed to improved iron nutritional status in infants, in addition to the increased rates of breastfeeding. Because of the limited current information available for young infants, these recent trends are not readily apparent in the NNMS data. Nondietary Factors Poverty Differences in the prevalence of iron deficiency in women of childbearing age by poverty status are 143 presented in figure 6-12 (see also tables 11-96, 11-97, 11-100, and II-101 in appendix II for other groups). Results in figure 6-12 indicate a consistent trend for a higher prevalence of iron deficiency in women below poverty than in those above poverty, even though all differences between the groups were not great. The effects of poverty on iron intake are shown in tables II-89 through II-92 for women and children in the CSFII 1985-86. For women of childbearing years above and below poverty, mean intakes of iron differ by about 0.8 milligram per day but the iron density of the diet is the same. Women aged 19-50 years in the low-income CSFII 1985 had a mean iron intake (based on 4-day data) of 9.6 milligrams per day, which was slightly lower than the mean for women in the all-income portion of the survey (10.1 milligrams per day) (USDA, 1987, 1988). Older women (35-50 years) in the low-income survey had lower iron intakes than younger women (8.6 and 10.1 milligrams per day, respectively) (USDA, 1988). Intake was slightly higher (9.7 milligrams per day) for women in households participating in the Food Stamp Program than for those in nonparticipating households (9.4 milligrams per day). These findings support concerns expressed in The Surgeon General's Report on Nutrition and Health about iron intake in low-income groups. Iron Deficiency in Females by Poverty Status Below poverty Above poverty C1 NAN 4 ’ 7 Percent 20 18} Mexican American 16 F 14} 12 10+ Non-Hispanic ol N 6} q N | N INN \ dN NLS 16-19 20-29 30-39 40-49 16-19 20-29 30-39 40-49 Age in years Figure 6-12. Percent of Mexican-American and non-Hispanic women aged 16-49 years with iron deficiency assessed by the MCV model, by poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 Education The prevalence of iron deficiency in adults by level of education, based on data from NHANES II, is shown in figure 6-13. The intake of iron by adults by level of education, based on CSFII 1985-86 data, is shown in tables II-89 and II-90 in appendix II. Status and intake improve with increased education. The differ- ence in dietary intake by education level disappeared in multivariable analysis when other factors were controlled for (Moshfegh, 1988). If race and income had not been controlled for, education might have been found to have a significant effect (Moshfegh, 1988). Parity The prevalence of impaired iron status by level of parity in women of childbearing years is presented in figure 6-14, based on data from NHANES II and HHANES. These results generally indicate that greater parity is associated with poorer iron status. However, differences in parity do not fully explain differences in iron status among the different ethnic groups of women during childbearing years. Iron Deficiency by Education High school CJ NNN 14 12} Women Men and women 10+ sl 6 4} 2} 0 20-44 years 45-74 years Figure 6-13. Percent of women aged 20-44 years and men and women aged 45-74 years with iron defi- ciency assessed by the MCV model, by level of education: second National Health and Nutrition Examination Survey, 1976-80 Iron Deficiency in Women by Parity Puerto Non-Hispanic Non-Hispanic Cuban Rican white black YY B28 [1 a Mexican American Percent 20 18 16 14 12 10 8 =o R [> i b> PS > pd B PS > £5 ke PS o, OS %%%%%%%". %' RIS [XX ON » O 5 or more Number of children Figure 6-14. Percent of women aged 20-44 years with iron deficiency assessed by the MCV model, by parity and ethnic group or race: Hispanic Health and Nutrition Examination Survey, 1982-84, and second National Health and Nutrition Examination Survey, 1976-80 (An asterisk indicates an unstable statistic or a statistic not reported because of small sample size.) Growth in Children Rapid growth in individuals would be anticipated to increase iron requirements and, consequently, the risk of developing iron deficiency. Data from the NNMS are not available to assess the effect of growth in children on iron status; longitudinal studies would be required. However, the relatively high prevalence of iron deficiency seen in the NHANES II data for young children and young adolescents (LSRO, 1984) is suggestive of the increased demands on iron stores during these periods of rapid growth. Pregnancy Data on iron status in various stages of pregnancy are not available in the NNMS for a nationally represen- tative sample because of the small number of preg- nant women sampled in such surveys. For women in all stages of pregnancy, the prevalence of iron defi- ciency according to the MCV model was 10.7 percent in NHANES II (LSRO, 1984). Data on the prevalence of low hemoglobin and hematocrit in low-income pregnant women, by trimester, from the PNSS are shown in table 6-9. The data (see also chapter 4) indicated that the prevalence of low hematocrit in these low-income women was highest in the third trimester and was higher in blacks than in any other race or ethnic group. The same hemoglobin criteria were used for blacks and whites; thus, this finding should be interpreted with caution. Studies from the medical literature show that the mean hemoglobin concentration in unsupplemented women drops below that of iron-supplemented women during the second and third trimesters. In an analysis of CSFII 1985-86 data, pregnancy was found to be associated with higher iron intakes (Moshfegh, 1988). Mean dietary intakes were 2 milligrams per day higher in pregnant women than in nonpregnant women (Krebs—Smith, 1988). Blood Loss Studies of blood donors suggest that women who do- nate more than three times per year and men who donate more than five times per year are at high risk of developing iron deficiency. Data from the NNMS are not available regarding blood donation. High menstrual blood losses are responsible for much of the iron deficiency in young women; collection of information on menstrual blood loss by self report is not reliable; other measures are not feasible in field surveys. Table 6-9. The prevalence of low hemoglobin and/or low hematocrit in low-income pregnant women at initial visits: Pregnancy Nutrition Surveillance System, 1987 Low hemoglobin Low hemoglobin Low hematocrit or low hematocrit Trimester n Percent n Percent n Percent First 5,344 6.2 9,917 5.3 15,261 5.6 Second 14,055 9.2 22,372 10.2 36,367 9.9 Third 10,414 26.7 15,729 26.8 26,143 26.8 Use of Medications e No information on blood donation has been col- lected in the NNMS. (Information on the number of donations in the last 12 months and the time of most recent donation is being collected in NHANES III.) Gable et al. (1987) found no relation of iron status to use of aspirin in NHANES II, but did find a positive association with use of oral contraceptives. Major Limits to Interpretation of Data and Gaps in the Database eo Information about selected groups at risk of iron deficiency, pregnant women and infants under age 1 year, is inadequate. In pregnant women, anemia is associated with increased neonatal mortality and a higher prevalence of low birth weight. Iron deficiency during the brief period of infancy is believed to lead to long-term harmful conse-— quences in regard to subsequent development. In both groups, dietary practices differ from other age and sex groups. In both of these groups, dietary intake and use of supplements over a period of 6 to 9 months determine the risk of iron deficiency. More detailed information is needed on the type and amount of supplement intake. Total iron intake could not be determined because quan- titative estimates of iron intake from supplements were not available from any of the surveys in which estimates of intake from food were made. Conclusions Based on the analyses in this chapter, the EPONM drew the following conclusions about assessment of iron nutriture in the United States using NNMS data: These groups require longitudinal assessment over =~ Populations at Risk at least 6 months for an adequate assessment of their nutritional status because iron status changes =e The variety of biochemical and hematological rapidly over a period of a few months. The combination of foods eaten at each meal is the most important determinant of iron absorption. Such information is even more important than the amount of iron consumed and has only been analyzed on a very limited scale. Improvements in the ability to provide such analyses should be incorporated into the NNMS. Distinguishing iron deficiency from mild inflam- matory conditions is difficult because laboratory measurements in mild inflammatory conditions or following infections may mimic iron deficiency, thereby suggesting a higher than actual preva- lence. This problem, which is greatest among the elderly, may be alleviated by using the four- variable model (see table 6-3) and laboratory tests that reflect the presence of inflammation. Criteria for anemia in blacks are uncertain. Blacks have lower hemoglobin concentrations than whites irrespective of iron status. These lower concen- trations lead to misleadingly higher prevalences of anemia among blacks if uniform criteria for low hemoglobin values are used for all races. This problem can be circumvented by using the three- or four—variable models for iron deficiency (that do not include hemoglobin as a variable) and by using laboratory tests that reflect the presence of inflammatory disease (C-reactive protein). 146 measures of iron nutritional status collected in the NNMS permits estimation of the prevalence of iron deficiency and iron deficiency anemia in the U.S. population and some characterization of population groups at risk of iron deficiency. The prevalence of iron deficiency anemia (based on findings of low hemoglobin levels plus evidence of iron deficiency) in NHANES II and HHANES is low (less than 5 percent); however, the prevalence of iron deficiency without anemia is still appreciable (up to 14 percent) in several groups. Groups at greatest risk of iron deficiency, as indicated by the data from the NNMS, are young children, adolescents, and women of childbearing age. Pregnant women and infants under 1 year of age are risk groups not well covered in current nation- ally representative surveys. CDC surveillance data indicate that low-income pregnant women are at high risk of anemia. Dietary iron intake, assessed in the CSFII 1985-86, is very low in women of childbearing years relative to recommended levels. Available iron intake, determined using data from the NFCS 1977-78, is also low for women relative to apparent require— ments. The intake estimates do not include the contribution from iron supplements. In contrast to iron deficiency, iron overload cannot be assessed adequately with current NNMS data to identify groups at risk. The prevalence of hemo- chromatosis is too low to be estimated reliably by available surveys. Trends The best trend data available are those on the nu- trient content of the U.S. food supply, which indi- cate an increase in the level of iron in recent years. Assessing trends in individual intake of iron by various population groups is more difficult because of methodological differences in the surveys over time, including revised nutrient composition data. The available NNMS data indicate little change during the last 20 years. Assessing trends in iron nutritional status is also difficult, because the measures used to estimate the prevalence of iron deficiency have not been used in many surveys. Limited data from the PedNSS suggest recent improvements iron status among low-income children aged 1-5 years. Determining Factors e Sex and age are powerful determining factors rela- tive to iron nutritional status. Evidence of iron deficiency is rare in males, in the elderly of both sexes, and in school children. Univariate analyses of NNMS data indicate that the prevalence of iron deficiency is influenced by race and socioeconomic factors such as poverty status and education. Iron intake also differs with these variables, but not as consistently as iron status, suggesting differences in bioavailable iron. Parity is also observed in NNMS data to be an influence on iron status in women during child- bearing years; women who have given birth to many children are at greater risk of deficiency. Other determining factors, such as iron supple- ment use, blood donation, use of medications, and growth, could not be assessed with current data from the NNMS. Future Concerns and Recommendations Based on their review, the EPONM identified the following needs: 147 e Dietary intake data for multiple days for the same persons for whom biochemical data are obtained to better assess iron adequacy and the appropri- ateness of estimated requirements. Such data will facilitate estimates of the range of usual intakes in subgroups of the population. eo Adjustment of iron intake data for differences in iron absorption from different types of meals. The usefulness of such calculations in relation to iron status, as indicated by the biochemical data, should be tested. Special surveys to monitor longitudinally two high-risk groups that are presently not adequately covered by the NNMS: infants and pregnant women. Even if larger numbers were included in existing cross-sectional surveys, assessments would still be limited because of the rapid changes occurring in these two groups over a short time. References Cited Committee on GRAS List Survey (Phase III). 1979. The 1977 Survey of Industry on the Use of Food Additives. Prepared under FDA Contract No. 223- 77-2025 by the Food and Nutrition Board, National Research Council, National Academy of Sciences, Washington, DC. Cook, D. A,, and S. O. Welsh. 1986. The Effect of Enriched and Fortified Grain Products on Nutrient Intake. Cereal Foods World 32(February):191-196. Dallman, P. R., R. Yip, and C. Johnson. 1984. Prevalence and Causes of Anemia in the United States: 1976 to 1980. Am. J. Clin. Nutr. 39:437-445. Edwards, C. Q., G. E. Cartwright, M. H. Skolnick, and D. B. Amos. 1980. Homozygosity for Hemochroma- tosis: Clinical Manifestations. Ann. Intern. Med. 93:519-525. Edwards, C. Q., L. M. Griffen, D. Goldgar, et al. 1988. Prevalence of Hemochromatosis Among 11,065 Presumably Healthy Blood Donors. N. Engl. J. Med. 318:1355-1362. Gable, C. B,, E. A. Yetley, C. Johnson, and M. McDowell. 1984. Relationships Among Indices of Iron Status, Demographic Factors, Health-Related Behaviors, and Health Problems. Paper presented at the Annual Meeting of the Federation of American Societies for Experimental Biology, St. Louis, Mo. April 1-6. Green, R., R. W. Charlton, H. Seftel, et al. 1968. Body Iron Excretion in Man: A Collaborative Study. Am. J. Med. 45:336-353. Hallberg, L., A.-M. Hogdahl, L. Nilsson, and G. Rybo. 1966. Menstrual Blood Loss——A Population Study. Variation at Different Ages and Attempts to Define Normality. Acta Obstet. Gynecol. Scand. 45:320-351. International Nutritional Anemia Consultative Group (INACGQ). 1977. Guidelines for the Eradication of Iron Deficiency Anemia. Washington: The Nutrition Foundation. Krebs—Smith, S. M. 1988. Letter and attachments to C. Suitor, National Research Council, National Academy of Sciences, dated October 18. Lindmark, B., and S. Eriksson. 1985. Regional Differences in the Idiopathic Hemochromatosis Gene Frequency in Sweden. Acta Med. Scand. 218:299-304. Life Sciences Research Office, S. M. Pilch and F. R. Senti, eds. 1984. Assessment of the Iron Nutritional Status of the U.S. Population Based on Data Collected in the Second National Health and Nutrition Examination Survey, 1976-1980. Bethesda, Md.: Federation of American Societies for Experimental Biology. Looker, A. C. 1988. Letter and attachments to A. I. Mendeloff, Editor-in-Chief, American Journal Clin- ical Nutrition, dated April 14. Looker, A. C., C. L. Johnson, M. A. McDowell, and E.A. Yetley. 1989. Iron Status: Prevalence of Impairment in Three Hispanic Groups in the United States. Am. J. Clin. Nutr. 49:553-558. Looker, A. C., C. T. Sempos, C. L. Johnson, and E. A. Yetley. 1987. Comparison of Dietary Intakes and Iron Status of Vitamin-Mineral Supplement Users and Nonusers, Aged 1-19 Years. Am. J. Clin. Nutr. 46:665-672. Looker, A. C., C. T. Sempos, C. Johnson, and E. A. Yetley. 1988. Vitamin—-Mineral Supplement Use: Association with Dietary Intake and Iron Status of Adults. J. Am. Diet. Assoc. 88:808-814. Martinez, G. A, and A. S. Ryan. 1985. Nutrient Intake in the United States During the First 12 Months of Life. J. Am. Diet Assoc. 85:826-830. Monsen, E. R., L. Hallberg, M. Layrisse, et al. 1978. Estimation of Available Dietary Iron. Am. J. Clin. Nutr. 31:134-141. 148 Moshfegh, A. J. 1988. Iron: Levels of Intake and Associated Factors in the Diets of Women. Paper presented at the 71st Annual Meeting of the American Dietetic Association, San Francisco, Calif. October 4. National Center for Health Statistics, J. D. Singer, P. Granahan, N. N. Goodrich, et al. 1982. Diet and Iron Status, A Study of Relationships: United States, 1971-74. Vital and Health Statistics. Series 11, No. 229. DHHS Pub. No. (PHS) 83-1679. Public Health Service. Washington: U.S. Government Printing Office. Pennington, J. A. T., D. B. Wilson, R. F. Newell, et al. 1984. Selected Minerals in Foods Surveys, 1974 to 1981/82. J. Am. Diet. Assoc. 84:771-782. Perloff, B. 1988. Assessment of Change in Iron Data 1977-1985. Prepared for Expert Panel on Nutrition Monitoring. Raper, N. R., J. C. Rosenthal, and C. E. Woteki. 1984. Estimates of Available Iron in Diets of Individuals 1 Year Old and Older in the Nationwide Food Con- sumption Survey. J. Am. Diet. Assoc. 84:783-7817. Subcommittee on Review of GRAS List (Phase II). 1972. A Comprehensive Survey of Industry on the Use of Food Chemicals Generally Recognized as Safe (GRAS). Prepared under DHEW Contract No. FDA 70-72 by the Committee on Food Protection, Division of Biology and Agriculture, National Research Coun- cil, National Academy of Sciences, Washington, DC. U.S. Department of Agriculture. 1987. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Women 19-50 Years and Their Children 1-5 Years, 4 Days, 1985. NFCS, CSFII Report No. 85-4. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture. 1988. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 4 Days, 1985. NFCS, CSFII Report No. 85-5. Hyattsville, Md.: U.S. Department of Agriculture. U.S. Department of Agriculture and U.S. Department of Health and Human Services. 1985. Nutrition and Your Health: Dietary Guidelines for Americans. Home and Garden Bulletin No. 232. 2nd edition. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services. 1988. The Surgeon General's Report on Nutrition and Health. DHHS Pub. No. (PHS) 88-50210. Public Health Service. Washington: U.S. Government Printing Office. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1986. Nutrition Monitoring in the United States—--A Progress Report from the Joint Nutrition Monitoring Evaluation Committee. DHHS Pub. No. (PHS) 86-1255. Public Health Service. Washington: U.S. Government Printing Office. 149 Yip, R, N. J. Binkin, L. Fleshood, and F. L. Trowbridge. 1987a. Declining Prevalence of Anemia Among Low-Income Children in the United States. JAMA. 258:1619-1623. Yip, R., K. M. Walsh, M. G. Goldfarb, and N. J. Binkin. 1987b. Declining Prevalence of Anemia in Childhood in a Middle-Class Setting: A Pediatric Success Story? Pediatrics. 80:330-334. Iron Table 1. Prevalence of iron deficiency determined by the MCV model in children aged 1-4 years, by age, race, and poverty status: second National Health and Nutrition Examination Survey, 1976-80 (LSRO, 1984) Race/ Age Standard error poverty status (years) n Percent of the percent All races 1-2 542 9.4 14 3-4 989 3.9 0.7 White 1-2 434 84 1.5 3-4 803 3.2 0.6 Black 1-2 89 10.9 3.0 3-4 157 8.5 1.9 Above poverty 1-2 409 6.7 1.2 3-4 720 2.5 0.6 Below poverty 1-2 121 20.6 4.1 3-4 241 9.7 21 Iron Table 2. Percent of persons with anemia (hemoglobin below age-specific cutoffs’ ), by age, sex, and race or ethnic group: second National Health and Nutrition Examination Survey, 1976-80, and Hispanic Health and Nutrition Examination Survey, 1982-84 (Looker et al., 1989) Sex and Standard error age (years) Ethnic group n Percent of the percent Males and females 5-10 Non-Hispanic white 1,299 24 0.51 Mexican American 862 0.7 0.25 Cuban 79 0 0 Puerto Rican 271 1.5 0.70 Males 11-14 Non-Hispanic white 443 1.9 0.49 Mexican American 335 0.5 0.37 Cuban 47 0 0 Puerto Rican 121 2.1 1.25 15-19 Non-Hispanic white 626 2.1 0.57 Mexican American 264 1.4 0.77 Cuban 54 0 0 Puerto Rican 161 3.7 1.46 20-44 Non-Hispanic white 1,934 2.2 0.46 Mexican American 747 1.1 0.50 Cuban 130 0.6 0.87 Puerto Rican 205 2.2 1.33 45-64 Non-Hispanic white 1,561 3.4 0.56 Mexican American 367 2.2 0.75 Cuban 174 1.8 0.98 Puerto Rican 163 7.3 1.89 65-74 Non-Hispanic white 969 3.8 0.71 Mexican American 66 1.6 1.73 Cuban 38 0+ 0* Puerto Rican 21 we "x See footnotes at end of table. 150 Iron Table 2. Percent of persons with anemia (hemoglobin below age-specific cutoffs’), by age, sex; and race or ethnic group: second National Health and Nutrition Examination Survey, 1976-80, and Hispanic Health and Nutrition Examination Survey, 1982-84--continued (Looker et al., 1989) Sex and 2 3 Standard error age (years) Ethnic group n Percent of the percent Females 11-14 Non-Hispanic white 415 3.4 1.02 Mexican American 310 1.9 0.72 Cuban 40 0 0 Puerto Rican 124 54 1.95 15-19 Non-Hispanic white 552 3.1 0.98 Mexican American 302 4.0 1.09 Cuban 40 0+ 0 Puerto Rican 152 5.3 1.756 20-44 Non-Hispanic white 2,012 4.4 0.58 Mexican American 889 7.9 1.03 Cuban 172 59 2.14 Puerto Rican 363 7.6 1.74 45-64 Non-Hispanic white 1,668 3.7 0.56 Mexican American 478 4.8 0.92 Cuban 193 0 0 Puerto Rican 242 49 1.23 65-74 Non-Hispanic white 1,152 3.3 0.76 Mexican American 97 3.7 1.95 Cuban 53 5.9 3.17 Puerto Rican 48 59 3.07 T See page 134 for criteria for low hemoglobin. 2 Number of examined persons. 3 Pregnant women were excluded. * Indicates statistic that may be unreliable due to sample size. ** Statistic not reported due to inadequate sample size. 151 Chapter 7 Recommendations The EPONM was charged to recommend sways strengthen the NNMS based on experiences involved in reviewing data analyses for this report. (In deciding how to frame their recommendations, the EPONM reviewed the recommendations for imp collection, dissemination, and usefulness o data offered by the JNMEC (DHHS/USDA, 1986) and the Coordinating Committee on Evaluation of Food Consumption Surveys (National Research Council, 1984). The EPONM finds that the conclusions of these groups, which dealt mainly with the HANES and NFCS, remain timely and of continued import- ance. In recently completed surveys (NFCS 1987-88) and in surveys currently in the field (NHANES III, CSFII 1989), progress has been made in implement- ing some of these recommendations, but most of the suggested changes were not reflected in the data available to the EPONM. In addition, both the USDA and DHHS have developed plans for survey activities, to a large extent, through 1995; thus, recommenda- tions offered now may not be implemented for some time. As a result of this Panel's deliberations, we wish to emphasize the following general areas: com- parability and compatibility among components of the NNMS; needs for data collection, analysis, and dis— semination; and future reports on the NNMS. (Specific recommendations related to dietary and nutritional status in cardiovascular disease and to the assessment of iron nutriture are included in chapters 5 and 6, respectively.) patibility he NNMS Comparability and Com; Among Components of t! The EPONM's objective of integrating data from the NNMS survey and surveillance activities was con- strained because of differences among the various data-collecting activities. The different program- matic obligations and logistical requirements of the Agencies make it impractical to suggest that data collection methodologies should be identical in all survey and surveillance activities. They do not, 153 however, preclude more serious efforts to improve the comparability of data. Because the EPONM was not informed about the details of content and [Method} ology in the current and planned surveys, it is difficult to be specific in recommending many changes to improve comparability, but some particular concerns and suggestions arising from experiences in preparing this report are discussed below. In the past, USDA and NCHS developed the nutrient composition databases for the NFCS and HANES independently. Because of this and other meth- odological differences, it has been impossible to determine the reason(s) for differences in nutrient intake between surveys conducted at the same time. The situation changed with the CSFII 1985-86 and HHANES; in fact, the most recent NFCS 1987-88 and the current CSFII 1989 and NHANES III all use the same food codes, descriptors, and nutrient composition data. However, there are still differences as to whether various food mixtures (such as casseroles) are coded as standard recipes or as separate ingredients reported by the respondents. Such differences in coding may introduce discrepancies in apparent nutrient intakes (especially for type of fat and amount of sodium) that may or may not be related to real differences in nutrient intakes. = The Panel favors introducing greater similarity in data collection methods. As long as methods differ, questions will continue to arise about the comparability of data. In such cases, studies evaluating the effects of methodological differences on the data gathered should be conducted jointly by USDA and NCHS. If the goal of greater integration of the components of the NNMS is to be achieved, resources must be allocated to the conduct of such studies and the results should be made readily available to the community of data users. The same situation is true for changes in method- ology within surveys over time. Information is needed on the impact of any such changes, which are instituted to improve the accuracy or efficiency of data collection, on the resultant data. For example, comparison studies are usually conducted when new analytical methods are introduced for blood chemistry measurements in the HANES. The analyses per— formed by Perloff (1988a,b) on the impact of changes in the nutrient database between the NFCS 1977-78 and the CSFII 1985-86 and the ongoing USDA 1988 Bridging Survey evaluating methodology changes between the NFCS 1977-78 and the NFCS 1987-88 also represent the types of studies needed if the NNMS is to achieve one of its important objectives of assessing changes in diet and nutritional status over time. Results of these studies should be made more readily available to data users. The Panel recommends introduction of a common core of sociodemographic descriptors in all NNMS surveys to enhance the capacity to establish linkages among surveys (USDA and NCHS currently use many common descriptors). For example, if groups could be characterized by age, race, income, and education on the basis of the same descriptors, knowledge and atti— tudes of specified groups assessed in the BRFSS or the Health and Diet Study could be related ecologi- cally to dietary intakes of the groups in the NFCS characterized in the same fashion. With improved capability for linking survey results, the need to add additional measures to individual surveys could be reduced. The Panel also noted that the capability of more thorough and complete comparisons from various components of the NNMS would be enhanced if there were greater similarities in data reporting from various Agencies, for example, using the same age groups. The Panel recommends that the Agencies coordinate data reporting to the extent possible. Needs for Data Collection, Analysis, and Dissemination Coverage of Groups Currently Excluded The sampling plans of many NNMS surveys neces— sarily exclude some groups of the population. Nationally representative samples of the civilian non— institutionalized population exclude military person- nel, persons living in institutions such as prisons and long-term care facilities, Native Americans living on reservations, and persons without fixed addresses (migrant workers and the homeless). Conclusions of the EPONM about the rarity of nutritional defi- ciencies in the United States were made cautiously because of the exclusion of some groups suspected to be at higher risk of being malnourished (notably the homeless and the institutionalized elderly). 154 Assessments of at least some of these excluded groups are necessary if statements about the nutritional sta- tus of the entire U.S. population are desired. Special- purpose surveys may be a more efficient mechanism for obtaining information about these groups than their inclusion in one of the existing surveys. Special methodologies will need to be developed for sampling (in the case of the homeless) and for collecting dietary intake and medical history information (from the mentally impaired elderly) in order to provide assessments for some of these groups. Improved Coverage of Groups Currently Included The EPONM noted limitations in the information available for several groups of particular interest sur— veyed in the NNMS. These included young infants, children, pregnant women, lactating women, and the elderly. ® Relatively small numbers of infants, pregnant women (including teenagers), and lactating women are included in the nationally representative samples of the entire population. Surveillance activities of the CDC that focus on women and children do not provide representative samples because they generally select only low-income sub- jects. Even if larger numbers of infants, pregnant women, and lactating women were included in existing cross-sectional surveys, assessments would still be limited because the rapid changes under- gone by these groups require analyses by narrow age groups or short time periods (such as trimes- ters) for appropriate analysis of status. Infants double their body weights in the first 3-4 months of life and experience many changes in the types of foods consumed over a short period of time: breast milk or formula in the first 3 months, some solid foods introduced during months 3-6, and table food in months 6-12. They represent a group at high risk because of the potential that long-term adverse developmental consequences may be the result of undernutrition early in life. Similarly, pregnant women undergo many physiological changes in a short period of time and under- nutrition during pregnancy can have a profound influence on the development of the fetus and the health of the pregnant woman, while nutritional requirements and status differ greatly by trimester. High nutritional requirements are also imposed by lactation. These considerations convinced some of the EPONM that longitudinal studies of nutritional status in representative samples of these groups would be useful and advisable. Other members of the EPONM believed that such studies were more appropriate in a clinical setting than in the surveys of the NNMS. The elderly are also a group that may be at risk of malnutrition because of physiological changes, physical or mental impairments, or social factors. This group has not been adequately represented in all NNMS surveys, little is known about how their nutritional requirements may differ from younger adults, and their numbers are increasing rapidly. There has been no upper age limit in most of the dietary surveys conducted by USDA. The CSFII 1985-86 included no adults older than 50 years, but the current and future CSFII will include persons of all ages. In the past, HANES excluded persons older than 74 years, but NHANES III has no upper age limit. Elderly persons should be sampled in sufficient numbers to permit assessment of sub- groups, for example, "elderly" (65-74 years), "aged" (75-84 years), and "very old" (85 years and over). Children aged 1-5 years also represent a group vulnerable to malnutrition and disorders related to diet and nutrition, such as dental caries and lead poisoning. One particular concern of the EPONM with respect to this group was the validity of proxy reports on dietary intake, especially in the case of parents reporting the intakes of children in day care or in school. Efforts should be made to test the validity of such reports or to find alternatives for obtaining intake information. Some members of the EPONM concluded that adolescents also represent a group at nutritional risk. Some of the factors related to concern about this group are the different levels of maturation, the nutritional requirements of the adolescent growth spurt, the frequency of eating away from home, the prevalence of dieting and eating dis— orders, and the nutritional demands of teenage pregnancy. Dietary Data in the HANES One-day dietary data, such as those collected in the HANES, do not provide information on the distribu- tion of usual intakes; thus, a meaningful assessment of the cross-sectional association of dietary intake and health status cannot be made. Increasing the number of days of data collection would permit some assessment of intraindividual variation in intake and improve estimates of usual intake. Limitations would still exist in examining cross—sectional associations of diet and health status (see chapter 5); however, 155 improved dietary intake data would be very helpful for planned followup studies of HANES populations. Knowledge and Attitudes Some members of the Panel recommended that ques— tions on dietary and nutrition knowledge and atti- tudes be included in surveys that estimate usual dietary intake or dietary pattern. In the data avail- able for this report, the EPONM found little informa- tion relating knowledge and attitudes to practices. Vitamin and Mineral Supplements The Panel could not evaluate total nutrient intake using the available data — none of the available sur- veys that assess nutrient intake from food include quantitative estimates of nutrient intake from sup- plements. The ability to examine excessive intakes, and possibly to assess consequences of nutrient tox-— icity, would be enhanced if such measures were included to provide more accurate assessments of total levels of nutrients consumed. Such information would also shed light on whether the dietary intakes of some nutrients are truly marginal. Alcohol Consumption Alcohol intake can influence dietary intake, nutri- tional status, and health status. The EPONM noted limitations in survey data on individual alcohol intake currently available. n The HHANES included an extensive questionnaire on alcohol use, but data were not available to the Panel. The EPONM recommends that efforts be continued to improve the quantitative assessment of alcohol intake in NNMS surveys. Nonresponse Analyses In its analyses of the available data, the EPONM was concerned about the possibility of bias in national estimates because of nonresponse in the surveys. The Panel was reluctant to use some of the data in which response was less than 50 percent. Therefore, the Panel recommends that every effort be made to improve current response rates. Methods for improv- ing response rates, such as monetary and other incentives, should be tested, especially in the USDA surveys in which such techniques have not been attempted. Current efforts to collect as much infor- mation as possible on nonrespondents by increasing the information collected in screening questionnaires, performing followup studies of nonrespondents, and conducting proxy interviews to obtain information on nonrespondents should be extended. The USDA and NCHS have conducted detailed analyses of non- response (discussed in the current report) in their recent surveys; such analyses should be made avail- able to data users. Education of Data Users The EPONM noted that some of the published reports based on NNMS data, primarily those produced by investigators using public release data tapes or working under Agency contracts, failed to use appropriate procedures to account for sample weights and design effects inherent in these complex surveys. Thus, the EPONM strongly recommends that the Agencies continue and increase efforts to educate users on appropriate use of survey data. These efforts may take such forms as publications on statistical issues, workshops, and/or greater docu- mentation for data tapes made available to investi— gators, and may require additional Agency staff with statistical expertise. In those situations where analy- ses are proposed as part of a contractual activity, it is incumbent on the funding Agency to ensure that the review of such proposals include an evaluation of the understanding and ability of the investigators to analyze data from a complex survey. Responsiveness to Needs of State and Local Data Users The Panel was aware of a desire by public health personnel in many States and localities to use NNMS data. The CDC surveillance activities are State— based, but the nationally representative surveys do not generate State— or local-level data. Nonetheless, policy makers in the States and localities need to know how to use national data, the implications of national data in terms of the State or locality, and how to "dovetail" State or local surveys more effi- ciently with national surveys. State and other officials who are responsible for implementation of the National Nutrition Objectives need ways to moni- tor status and progress in achieving these objectives. The Panel encourages more interaction of Federal and State data collection activities and research on the value and validity of synthetic estimates for States and other localities. 156 Research Needs The EPONM noted many research needs in the course of reviewing the analyses included in this report, but wishes to highlight two issues: e Development of methods for the assessment of dietary adequacy (and nutrient excess) to reduce reliance on the RDA as a standard for nutrient intake. Development of measures of status for food com-— ponents identified as current public health con- cerns, such as calcium. Improved and/or validated measures of obesity, energy intake, and physical activity are also needed. Future Reports on the NNMS The integration of the currently available dietary and nutrition-related health status data from the two major surveys of the NNMS (the NFCS and the HANES) to assess the dietary and nutritional status of the U.S. population is a major contribution of reports such as this one and that of the JNMEC. Data from these surveys and other NNMS sources can also be used to provide a detailed analysis and sum- mary of specific issues related to diet, nutrition, and health. The Panel's experiences, as amplified below, indicate that presentations of both updated infor- mation on status and detailed analyses are appro- priate, valuable, and feasible objectives for such reports. Content of Reports The EPONM recommends that reports presenting updated information on dietary and nutritional status and reports presenting detailed analyses on special topics be prepared separately. In the Panel's experi- ence, trying to accomplish both in the same report was overly ambitious. eo Update reports may take a variety of forms. One option is a relatively comprehensive discussion and analysis of available dietary and nutritional status data containing conclusions regarding public health significance and monitoring priority, such as the JNMEC report and chapters 3 and 4 in the current report. Another possibility is the tabulation of the most recent data, with limited interpretation, in a format similar to that used for Health: United States. These two alternatives are especially appropriate in view of the increasing trend for con- tinuous data collection in the NNMS. Considera- tion might also be given to studying the develop- ment of a set of "leading indicators" (similar to leading economic indicators) that potentially could rapidly monitor changes in food consumption and nutritional status of the population. Such indi- cators need not be direct measures of food con- sumption or biochemical measures related to nutri- tional status, but might consist of data already col- lected for other purposes, such as food expenditures or participation in food and nutrition programs, that may reflect dietary and nutritional status. Reports of detailed analyses on special topics could alternate with update reports or could be prepared concurrently (but separately, as the need for such reports is identified and the data become available). For such reports, the assistance of consultants with a wide range of expertise within the specified sub- ject area would be most helpful. The types and depth of expertise needed would differ for different topics. Some topics for consideration are listed below (not all of these topics could be undertaken with existing data). - The impact of supplement use on nutrient intake, nutritional status, and health status. — The nutritional and dietary status of the elderly. - The impact of consumption of food away from home on dietary status. — The impact of social changes such as single- parent and two-income households on dietary practices and nutritional status. - The impact of "dieting" behavior on dietary patterns and nutritional status. Frequency of Reports Difficulties in interpretation arise if update reports are prepared too frequently: little detectable change in dietary or nutritional status of the population would be expected in short time periods and appro- priate data for desired analyses may not be available at short intervals. The latter was true of the EPONM's review; the major sources of data available for the update of dietary and nutritional status were limited in coverage of age and sex groups (CSFII 1985-86) and ethnic groups (HHANES). In addition, analyses of dietary intake data from HHANES were not completed for the EPONM review. These factors limited the ability of the Panel to meet the charge to 157 update information on the nutritional status of the U.S. population and raised concerns about the timeli- ness of data release from some surveys. Planned schedules for release of data from current surveys will resolve concerns about the timeliness of data avail- able, provided that the Agencies are allocated ade- quate staff and resources to meet the planned sched- ules. Reports intended to update information on the nutritional status of the U.S. population should be timed according to the availability of data from the two major components of the NNMS. Thus, the next major update report should be planned to incorporate data from the first half of NHANES III (1988-91), the NFCS 1987-88, and 1989-91 cycles of the CSFII (all of these data should be available in 1993). In addi- tion, the most recent food supply data and data from other NNMS activities should be included. In inter— vening years in which reports on the NNMS are mandated, more limited update reports or reports on special topics should be prepared. The NNMS of the Future The recommendations above are predicated on the Panel's experiences in analyzing the NNMS data in this report, the recognition that most suggested changes cannot be made in ongoing surveys for sev- eral years, and the probability that the basic structure of the NNMS will remain the same. The EPONM believes that it is appropriate now to begin efforts to determine the most useful form of the NNMS in the future. The main considerations that should drive the introduction of changes to make the separate components function more effectively as a system are the needs of data users, especially policy makers. One obvious need is for data that permit the assessment of progress on implementing the National Nutrition Objectives for the Year 2000 at the midcourse review (1995) and at the end of the decade. Planning for the future of the NNMS should include a poll of data users to determine unmet needs and should proceed with cooperation between the Agencies from the highest to the lowest level. With such direction and cooperation, the best features of the existing NNMS can be retained and an even better system can be constructed. References Cited National Research Council, Coordinating Committee on Evaluation of Food Consumption Surveys. 1984. National Survey Data on Food Consumption: Uses and Recommendations. Washington: National Academy Press. Perloff, B. 1988a. Assessment of Change in Total Fat Data 1977-1985. Prepared for Expert Panel on Nutrition Monitoring. Perloff, B. 1988b. Assessment of Change in Iron Data 1977-1985. Prepared for Expert Panel on Nutrition Monitoring. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1986. Nutrition Monitoring in the United States——A Progress Report from the Joint Nutrition Monitoring Evaluation Committee. DHHS Pub. No. (PHS) 86-1255. Public Health Service. ~~ Washington: U.S. Government Printing Office. 158 Appendix I Description of NNMS Surveys Descriptions are included for the following: 8 National Nutrlemt Data Bank ...oreeomoonoonsnnen ors oon mns sn aes ome a0 60m 008900 0.05 5009 50 5 408 555 5145 ® U.S. Food Supply Series .........c.ouuiiniiiiiiiit iti ie etait aaa eo Nationwide Food Consumption Survey 1977-78 ..........cciiuiiiiiniiiiiiiiiieineannenn.. e Continuing Survey of Food Intakes by Individuals 1985-86 ...................ccciiiiirriiinnnn.. eo National Health and Nutrition Examination Survey ...................cooiiiiiiiiiiiiiiiiin. eo Hispanic Health and Nutrition Examination Survey ................. cc iii. eo NHANES I Epidemiologic Followup Study ............coiiuiiiiiiiiiiiiiiiii iii, eo National Health IMervVIeW SUIVEY cuvisssnsuvinonsrmadne sme snes ienss 2o0 5s sem asss ines venss wuss 0 Food Label and Package SUrVEY .vivsssicainsinitisrssissntsbsinssnsisinsinsbssisiotnsssininnmmes o Total Diet Study .......oiinninii iii te eo Vitamin/Mineral Supplement Intake Survey .....ccsvviissinsssiivinsssvinsinsssssssansnvis ssn eo Healthand Diet SUrvey ............uiininiiiiiiiii iii itt iii eae aaeeanneanns e Pediatric Nutrition Surveillance System ............ cc. iii a e Pregnancy Nutrition Surveillance System ............. cc. iii e Behavioral Risk Factors Surveillance System ............... i. 1-1 National Nutrient Data Bank Sponsoring Agency: Human Nutrition Information Service, USDA Conducted: Continuously. Objective: To compile and summarize reliable data on the nutrient composition of foods through development and maintenance of a nutrient data bank. These data are made available in published tables of food composi- tion and on public use data tapes, which include the nutrient data bases used for the NFCS 1977-78, CSFII 1985-86, and HHANES. Data sources: Nutrient composition data are obtained from scientific publications, university and government laboratories, food processors and trade groups, and through Human Nutrition Information Service (HNIS)- funded contracts for purposes of generating needed food composition data. Most values released are supported “by laboratory analyses. Nutrient values not available from laboratory analyses are imputed from data for other forms of the food or from data for similar foods. Measures: Nutrient data bases for use with survey results are of two types: nutrient content of the edible parts of a pound of foods in forms as they enter the kitchen (household food use) and the nutrient content of 100 grams of food as ingested (individual intake). Currently, values are derived for food energy and 28 nutrients and other food components. Included are protein, total fat, saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, cholesterol, total carbohydrate, dietary fiber, vitamin A as International Units, vitamin A as retinol equivalents, carotenes, thiamin, riboflavin, preformed niacin, vitamin B6, vitamin B12, folacin, vitamin C, vitamin E, calcium, phosphorus, magnesium, iron, zinc, copper, sodium, potassium, alcohol, and moisture (water). Activity description: The HNIS maintains the National Nutrient Data Bank (NNDB). The Agency expands NNDB coverage of nutrients and foods as required to provide needed information on the nutrient composition of foods for use by Federal, State, and local food program administrators, researchers, health professionals, nutrition educators, the food industry, and consumers. The products of this NNDB activity are reference values for over 60 food components in thousands of foods Americans consume, presented in published and machine- readable forms. Results from analyses conducted by industry, government, universities, and others are compiled, evaluated, and entered into the NNDB. Research to fill knowledge gaps is planned and sponsored. These studies fall into three general categories: (1) analyses of particular nutrients in foods known to be important sources; (2) analyses of particular foods for which data on many nutrients are lacking; and (3) analyses of new foods or new forms of foods of increasing popularity. Examples of the latter are tropical fruits and fish raised by aquaculture. Much of the research is planned in consultation with the Agricultural Research Service's (ARS) Nutrient Composition Laboratory at Beltsville, Maryland, and, as appropriate, with the National Institutes of Health, the Food Safety and Inspection Service, the Food and Drug Administration, the food industry, and others. Studies utilize national sampling plans appropriate to the national distribution of food types and employ validated analytical methods applied with proper laboratory quality control procedures. Food components believed to be important to health promotion and disease prevention receive emphasis in studies of the composition of foods. Current examples are selenium, lipids, carotenoids, tocopherols, and dietary fiber components. With understanding of the effects of handling and processing on the nutrient content of foods, nutritive values of some foods can be estimated, thereby avoiding more extensive analyses. Examples of such studies are the distribution of nutrients in solid and liquid portions of canned foods, the effects of trimming of fat from meat and poultry prior to cooking on nutrient content of the cooked food, the absorption of cooking fat by fried foods, and the retention of nutrients after cooking and other preparation of mixed dishes. Agricultural Handbook No. 8: The technical reference tables on the nutrient composition of foods, "Composi- tion of Foods ... Raw, Processed, Prepared," Agricultural Handbook No. 8 (AH-8), are being revised. Of the 22 sections, each showing values for a group of foods, five sections are yet to be published. They are lamb, veal, and game; baked products; snacks and sweets; cereal grains and pasta; and mixed dishes. Nutrient values in the 22 sections will be updated as necessary for either individual items or for complete food groups. The computer ized USDA Nutrient Data Base for Standard Reference corresponds to AH-8 and is revised as the handbook is updated. I-2 Special-use tables and data bases: Food composition tables and data bases for special uses are developed and kept up-to-date. For example, special tables are required by HNIS for estimating the nutrient content of U.S. food supplies each year. Also, consumers and many professionals require references that are smaller and less technical than AH-8. They may show the nutritive value of only foods most commonly eaten by Americans or for only selected food components such as calories, lipids, sodium, or dietary fiber. Special data bases will allow (1) school food service managers to assess the nutritional quality of school meals, and (2) Extension agents, other educators, and consumers to assess diets using interactive computer programs. Data bases for assessing the nutrient content of diets: Large nutrient data bases for assessing diets reported in national food consumption surveys conducted by USDA, DHHS, and others are developed and documented in ways that facilitate improved comparability of assessments across surveys and with past surveys. Data bases involved are Nutrient Data Base for Individual Food Intake Surveys and Nutrient Data Base for Household Use Surveys. These data bases must contain values for all nutrients assessed in the survey in all foods reported. Until analytical data are available, best estimates are made by HNIS staff. Computerized files are developed and maintained (1) for linking the Nutrient Data Base for Standard Reference and the two survey data bases to facilitate updating through automatic calculation of values for specified recipes; (2) for converting quantities of foods as reported to weight in grams or pounds; (3) for linking food codes (and associated nutrient data) used in CSFII 1985 and NFCS 1987 with those used in NFCS 1977-78; and (4) for converting home-prepared and commercially prepared food mixtures reported in surveys to their basic ingredients. Data bases for food industry use: The USDA cooperates with the food industry, the Food and Drug Administration, and other USDA Agencies in developing special data bases of the generic nutrient composition of fresh and processed foods for use in point-of-purchase labeling. International data bases: The Department continues to cooperate internationally in the development and exchange of food composition data and technologies. Support of INFOODS involved a cooperative study of statistical methods for handling data and other assistance in development of standard procedures for handling food composition data. The Department also assisted the Food and Agriculture Organization (FAO) with the development and dissemination of data for foods eaten in developing countries. Data user conferences: Annual Nutrient Data Bank Conferences provide USDA with an effective means of communicating with data users. The USDA helps develop uniform and concise guidelines to help nutrient data users to obtain accurate and consistent estimates of the nutritive value of foods and diets. I-3 U.S. Food Supply Series Sponsoring Agency: Economic Research Service, USDA Conducted: Each year since 1909. Target population: U.S. civilian population. Design: The Economic Research Service of USDA provides annual estimates on amounts of about 350 foods that disappear into civilian food consumption at or before the retail level of distribution. This information is derived from data on production, imports and exports, military use, and beginning and year-end inventories. Measures: Food that "disappears" into and is available for civilian consumption on a per capita basis. The nutrient content of these foods is estimated. Such food and nutrient supply estimates, available each year since 1909, are the only source of information on food and nutrient trends since the beginning of the century. Background: The Nutrient Content of the U.S. Food Supply is a historical series providing data on amounts of nutrients per capita per day in food available for consumption each year beginning with 1909. Levels of nutri- ents per capita per day are rapidly and inexpensively derived indicators of diet quality. They are used to assess the potential of the U.S. food supply to satisfy the nutritional needs of the population. These data also have other uses such as in epidemiological studies on the relationship between diet and the prevalence of disease and in studies of the effects of technological, economic, and social changes on the U.S. diet and future food production. Data on the nutrient content of the U.S. food supply are published annually in Agricultural Statistics (USDA, 1988), Statistical Bulletins on food consumption, prices and expenditures (USDA, 1989), and handbooks of agricultural charts in the Agriculture Handbook series (USDA, 1986), and Statistical Abstract of the United States (U.S. Bureau of the Census, 1987). Interpretive analyses of trends in nutrient levels in the food supply are reported annually in the National Food Review (USDA, 1987) and frequently in other publications (Welsh and Marston, 1982; Raper and Marston, 1986). Design: Two sources of information within USDA are used to calculate the nutrient content of the U.S. food supply. The Economic Research Service provides estimates of quantities of food available for consumption per capita per year, and the Human Nutrition Information Service (HNIS) provides data on the nutrient content of food. The nutrient content of the U.S. food supply is calculated by multiplying the pounds of each food con- sumed per capita per year by the nutritive value of the edible portion per pound, totaling the results for all foods, and then converting the total to a per-day basis. Food consumption estimates: The Economic Research Service estimates the quantities of approximately 300- 400 foods that "disappear" into the U.S. food distribution system. The methods used have been described in detail previously (USDA, 1965). In brief, disappearance data are estimated by deducting data on exports, military use, year-end inventories, and nonfood use from data on production, imports, and beginning-of-the- year inventories. The methodology avoids double counting of any food. Data on per capita consumption of food are derived by dividing the weight of food available for use during the year by the population of the 50 States and the District of Columbia, as estimated by the U.S. Bureau of the Census. Disappearance of all foods is not measured at the same point in the distribution system. Some foods are in a raw or primary form and others are retail products when their disappearance is measured. For example, the disappearance of meat, poultry, fish, flour, eggs, sugar, and fat is measured when they are in a primary state, that is, before they are processed into finished products such as bread, bakery products, soft drinks, and frozen casseroles. On the other hand, quantities of fruit, fruit juices, vegetables, and potatoes are measured in several forms——fresh, canned, frozen, or dehydrated. However, these products too may undergo further processing, for example, into pies and jellies. Losses that occur after food is initially measured, such as in further processing, marketing, or home use, are not considered in these estimates. Food disappearance estimates exclude some sources of nutrients: alcoholic beverages and the sugars and grains used in their manufacture; baking powder, baking soda, yeast, and certain vitamins and minerals added to foods for their functional or flavoring properties; and vitamin and/or mineral supplements in tablet, capsule, and liquid form. I-4 Nutrient content of food: The HNIS data on the nutrient content of foods have been sufficient to derive estimates for the U.S. food supply series for the following nutrients and food components: energy; protein; total fat; saturated, monounsaturated, and polyunsaturated fatty acids; cholesterol; carbohydrate; calcium; phosphorus; magnesium; iron; zinc; copper; potassium; vitamin A (international units and retinol equivalents); carotenes; vitamin E; vitamin C; thiamin; riboflavin; niacin; vitamin B6; folacin; and vitamin B12. Because the food supply series requires food composition data for a relatively small number of foods and for the primary (uncombined) state of the foods, food composition data have been sufficient to make rough estimates for simple and complex carbohydrates. Nutrients added to foods commercially through fortification and enrichment are included in the food supply estimates on the basis of periodic surveys of industry conducted for USDA by the U.S. Bureau of the Census. Therefore, data on the nutrient content of the U.S. food supply include quantities of iron, thiamin, riboflavin, and niacin added to flour and cereal products; vitamin A value added to margarine, milk, and milk extenders; vitamin B12 added to cereal; and ascorbic acid added to fruit juices and drinks, flavored beverages, dessert powders, milk extenders, and cereal. Estimates of the nutrient content of the food supply exclude nutrients from the inedible parts of food, such as bones, rinds, and seeds, but include nutrients from portions of food that are edible but not always eaten and include all the separable fat that is left on retail cuts. For example, the nutrient values used for meat are for composite retail cuts. Nutrient estimates also include food and nutrients that may be lost after food disappear- ance is measured, as in processing, marketing, or cooking. Insofar as possible, nutrient estimates reflect chang- es in the composition of individual foods since 1909. For example, the ascorbic acid values applied to fresh potatoes consumed in recent years are higher than the values applied to potatoes consumed at the beginning of the century because of better storage conditions and use of different cultivars. The most recent composition data are used if earlier data are unavailable or considered unreliable. References: Raper, N. R., and R. M. Marston. 1986. Levels and Sources of Fat in the U.S. Food Supply, in C. Ip, D. Birt, A. Rogers, and C. Mettlin, eds., Dietary Fat and Cancer. New York: Alan R. Liss, Inc. U.S. Bureau of the Census. 1987. Statistical Abstract of the United States, 1988. 108th edition. Washington: U.S. Government Printing Office. U.S. Department of Agriculture. 1986. Agricultural Chartbook. Agricultural Handbook No. 663. Washington: U.S. Government Printing Office. U.S Department of Agriculture. 1988. Agricultural Statistics, 1987. Washington: U.S. Government Printing Office. U.S. Department of Agriculture, Economic Research Service. 1965. U.S. Food Consumption. Stat. Bull. No. 364. Washington: U.S. Government Printing Office. U.S Department of Agriculture, Economic Research Service. 1989. Food Consumption, Prices, and Expenditures, 1966-87. Statistical Bulletin No. 773. Washington: U.S. Government Printing Office. U.S. Department of Agriculture, Economic Research Service, R. Marston and N. Raper. 1987. Nutrient Content of the U.S. Food Supply. National Food Review, NFR-36. Washington: U.S. Government Printing Office. Welsh, S. O., and R. M. Marston. 1982. Review of Trends in Food Use in the United States, 1909 to 1980. J. Am. Diet. Assoc. 81:120-125. Nationwide Food Consumption Survey 1977-78 Sponsoring Agency: Human Nutrition Information Service, USDA Conducted: April 1, 1977 to March 30, 1978. Target populations: Private households in the 48 conterminous States and the individuals residing in those households. Special surveys targeted populations in Alaska, Hawaii, and Puerto Rico; low-income households; and households with elderly person(s) in the 48 States. Design: Multistage stratified area probability samples of the defined population. Sample size: 14,930 households; 30,770 individuals. Measures: Food used from home supplies during one week by entire household and food ingested by individual household members at home and away from home for three consecutive days. Nutrients available from food used by the households and nutrients ingested by household members are derived using appropriate food composition data files developed from HNIS's nutrient data bank. See JNMEC report for additional details on the survey. Nonresponse analyses: For the Nationwide Food Consumption Survey (NFCS) 1977-78, household question- naires were completed for about 72 percent of the households contacted. In the participating households, 94 percent of the eligible individuals completed the first-day dietary report, and 85 percent completed all three days of the dietary reports (DHHS/USDA, 1986). Hence the response rate for the group with three days of data is 61.1 percent (72 percent x 85 percent). Comparisons between the NFCS 1977-78 and the March 1977 Current Population Survey of the U.S. Census were performed by Tuszynski and Roidt (1988). Variables examined included tenancy (owned or rented), race (white or nonwhite), urbanization (metropolitan or nonmetropolitan), ethnic origin (Hispanic or non-Hispanic), household size, region (Northwest, Midwest, South, or West), and household income (less than $10,000, $10,000- $19,000, or $20,000 or more). For these variables, most of the differences between the two surveys were small. The largest difference involved race, with the Current Population Survey estimating that 11.9 percent of the U.S. households were nonwhite compared to the estimate of 14.8 percent from the NFCS 1977-178, a difference of approximately 3 percentage points between the races. An investigation of the nonrespondents was not conducted in 1977-78, and it is not reasonable to attempt such an investigation 10 years after the survey. Therefore, these were the only comparisons available on which to base a decision about nonresponse bias in the data. These comparisons were based on the one-day data; it would be useful to have a similar comparison for the three-day data. However, because there was not a large drop in the numbers from the one-to the three-day data, this comparison may not be necessary. The difference in the two racial estimates should not affect the results much unless there is a substantial difference in food or nutrient intake. References Tuszynski, C., and L. Roidt. 1988. Nationwide Food Consumption Survey (NFCS)--Comparisons with Census. Tables prepared for the Expert Panel on Nutrition Monitoring. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1986. Nutrition Monitoring in the United States—-A Progress Report from the Joint Nutrition Monitoring Evaluation Committee. DHHS Pub. No. (PHS) 86-1255. Public Health Service. Washington: U.S. Goverment Printing Office. 1-6 Continuing Survey of Food Intakes by Individuals, 1985-86 Sponsoring Agency: Human Nutrition Information Service, USDA Conducted: 1985 (April 1, 1985 to March 30, 1986) and 1986 (April 1, 1986 to March 30, 1987). Target populations: Persons of selected sex and age residing in the 48 conterminous States in private house- holds with income at any level (basic survey) and with income at or below 130 percent of the poverty guidelines (low-income survey). In 1985: women 19-50 years, their children 1-5 years, and men 19-50 years. In 1986: women 19-50 years and their children 1-5 years. Design: Multistage stratified area probability samples drawn using a sampling frame organized using estimates of the U.S. population in 1985. Panel design: women and children were interviewed 6 times (waves) spread over the year. __ ~~ Responserate(wavel) Eligible Respondent! households households 1985: Basic 1,893 1,341 (71%) Low income 2,176 1,916 (88%) 1986: Basic 1,722 1,351 (79%) Low income 1,386 1,223 (88%) 1 In some households 2 or more women provided data. For example, in basic survey, Spring, 1985, 1,459 women in 1,341 households provided data. Measures: Food intakes from six 24-hour recalls collected by interview at about 2-month intervals during the year. Nutrient intakes derived using food intakes and special food composition data files developed from HNIS's nutrient data bank. The food components assessed included food energy, protein, total fat, saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, cholesterol, total carbohydrate, dietary fiber, vitamin A (IU and RE), carotenes, thiamin, riboflavin, preformed niacin, vitamin B6, vitamin B12, folacin, vitamin C, calcium, phosphorus, magnesium, iron, zinc, vitamin E, copper, sodium, alcohol, and water. Description: The Continuing Surveys of Food Intakes by Individuals (CSFII) are part of USDA's system of Nationwide Food Consumption Surveys (NFCS). They are conducted between the larger decennial NFCS. The primary purpose of the CFSII 1985-86 was to provide timely information on U.S. diets and the diets of population groups of concern, and to indicate differences in information provided by previous surveys. Another purpose was to provide the basis for assessing "usual" diets as measured by several days' data spread over the year and for studying how diets vary over time for individuals and groups of individuals. The CSFII 1985-86 focused on two groups considered at possible nutritional risk because earlier surveys had found them to have intakes of some nutrients well below recommended levels. Groups are women 19-50 years and children 1-5 years. (In Summer 1985, men 19-50 years were also surveyed.) Women and children con- tacted as part of the CSFII 1985-86 were asked to provide 6 days (waves) of dietary data over a 1-year period. Each wave of data was collected at about 2-month intervals using a 1-day dietary recall. National Analysts (a division of Booz, Allen, and Hamilton, Inc.) a private firm in Philadelphia, conducted the CSFII 1985-86 under contract with the HNIS, USDA. National Analysts designed the sample; collected the information; edited, coded, and keyed the data; and prepared the data tape. The HNIS defined the target populations and the information to be collected; provided technical information such as food codes, gram weights of household measures, and the nutrient composition of foods; and monitored all aspects of the survey operation. Sample design: The CSFII 1985-86 samples were drawn from all private households in the United States. The survey was designed to provide a multistage stratified area probability sample representative of the 48 conterminous States. (Alaska and Hawaii were not included because their diets and factors affecting them differ significantly from those in the mainland States and, therefore, require separate surveys.) The sampling frame was organized using estimates of the U.S. population in 1985. The stratification plan took into account geographic location, degree of urbanization, and socioeconomic considerations. The 48 States were grouped into the 9 census geographic divisions; then all land areas within the divisions were divided into 3 urbanization classifications: central city, suburban, and nonmetropolitan. The stratification process resulted in a total of 60 strata-—17 were central city, 28 suburban, and 15 nonmetropolitan--which correspond to the geographic distribution, urbanization, and density of the population within the 48 States as defined by the Bureau of the Census. Within each stratum two relatively homogeneous units—-primary sam- pling units (PSU)--were randomly selected for a total of 120 PSU. For the basic sample, each selected PSU was then divided geographically along census boundaries into smaller clusters, known as area segments, containing a minimum of 100 housing units. A total of 206 area segments was drawn into the sample. The number of area segments selected from each PSU varied, depending on the size of the PSU. For the basic sample, within the 206 area segments, existing housing units were prelisted. The number of prelisted housing units in the area and census information were used to determine the number of housing units to be selected from that area. A sampling rate was derived from the expected occupancy rate, the expected eligibility rate (women 19-50 years living there), and the expected completion rate and applied to the target of 1,200 completed interviews by women 19-50 years. Data collection: To contact individuals in housing units selected as part of the sample, trained interviewers made a minimum of three personal visits plus up to eight telephone calls to each household having a telephone. To contact households without telephones, interviewers made a minimum of six personal visits (five in rural areas). At each household, the interviewer conducted a screening interview to determine if the household was eligible to participate in the survey. Eligible households contained at least one woman 19-50 years of age at the time of initial contact. In eligible households, all women within this age range and their children 1-5 years of age, if any, were invited to be interviewed and to participate in a year-long survey panel. A letter of introduction was provided, and respon- dents were informed that the full survey involved the collection of 6 individual days of intake data. In each wave, the interviewing process included two major steps: (1) collection of information about the household and (2) collection of information on food intake. Separate intake records were used for each woman and for each child. Interviewers were instructed to complete all interviews in a single household during the same visit or call, to complete the household schedule first and then the required intake records, and to obtain intake data about a woman and her children for the same 24-hour period. Interviewers were provided with instructions on what to do if deviation from this pattern was necessary. Multiple contacts were made when needed to complete interviews in eligible households. Interviewing of a household was not considered complete until the household schedule and intake records for all eligible individ- uals who agreed to participate were obtained. The first wave of data (wave 1) was collected by personal interview from 1,341 households for the basic sample; subsequent waves of data were collected by telephone, if possible. The proportion of households interviewed by telephone in waves 2 and 3 was 91 percent; in waves 4-6, it was 90 percent. In households without telephones or where the respondent requested to be interviewed in person, the information for waves 2-6 was collected in a personal interview. Only households that contained a member who was interviewed in wave 1 were recontacted in subsequent waves. Within these households, only women and children who completed interviews in wave 1 were eligible for reinterview in waves 2-6. Respondents were retained in the survey even if they missed one or more waves. Respondents who moved out of their area during the survey were not followed. Individuals who became mem- bers of participating households after wave 1 were not eligible for participation regardless of their age. In wave 1, information on the characteristics of the household was collected from the primary age-eligible woman in the household (the household informant). The female head of the household was always the house- hold informant if she was age—eligible. In households where the female head did not participate in the survey, interviewers collected data on household characteristics from the participant who was the main meal plan- ner/preparer or from the participant who could best answer questions about the household. Household charac- teristics included the previous month's household income by source; the previous year's household income before taxes; participation in food programs; age, education, occupation, and employment status of the male head of the household; household size; tenancy; usual amount of money spent on food; and each household member's sex, age, and relationship to the female head of the household. In waves 2-6, interviewers were instructed to attempt a reinterview within 10 days of two month's time from the date of the last interview. At each interview in waves 2-6, the interviewer first obtained limited informa- tion on the household from the woman who provided the household information in wave 1. This included information on changes in household membership since the last interview, usual amount of money spent on food, program participation, and changes in monthly household income. In all waves, each woman interviewed provided information on her own food intake as well as that of her elig- ible children. Information was collected on all food eaten either at home or away, the time of day food was eaten, what the eating occasion was called, and the use of salt at the table. The main meal planner/preparer was asked about the use of fat (including type) and salt in food preparation and about the form in which the food was brought into the home (commercially frozen, canned, or bottled, or in another form). Foods were designated as coming from the home food supply or as obtained and eaten away from home. A Food Instruction Booklet, developed by National Analysts based on information provided by HNIS, was used by the interviewers to help respondents adequately describe foods and amounts eaten. The interviewers used standard household measuring cups and spoons and a ruler during the interview to help respondents estimate quantities of foods and beverages consumed. Respondents kept the cups, spoons, and ruler for use during subsequent interviews. Each woman interviewed also provided information on her age, race, physiological status (pregnancy and lacta— tion); employment, occupation, and education. Children were assigned the race of their mother/caretaker. In wave 1, eligible households were scheduled for interviews in a manner designed to provide representative— ness of intake data by day of the week over all households. In subsequent waves, interviewers were instructed to collect data for a household on different days of the week. For example, if the data for a household were collected for a Tuesday in wave 1 and for a Friday in wave 2, the household was asked to provide data for one of the other 5 days in wave 3, if possible. For the basic sample, the largest proportion of dietary intakes was collected for Tuesday (18 percent) and the smallest proportion for Saturday (9 percent). Many participants were reluctant to be interviewed on a Sunday. Seventy-two percent of all respondents provided at least 1 day of intake data for Tuesday; 34 percent provided at least 1 day of intake for Saturday. In wave 1 of CSFII 1985, a total of 1,459 women 19-50 years in 1,341 households completed the food intake interview. Of these women, 1,032 completed at least 3 of the 5 additional waves. Food and nutrient intakes by women reporting at least 4 or more days are presented in some tables. If a woman reported 4 days' data, all were used; if a women reported more than 4 days, the first day's data and three of the remaining days selected randomly were used. For more information about response over the six days, see CSFII Report No. 85-4 (USDA, 1987a). Sample weights: The sample was designed to be self-weighting. That is, the proportion of eligible persons in the sample with a particular characteristic was designed to represent the same proportion of eligible persons in the population. However, adjustments to the sample were required because not all eligible households I-9 participated, not all eligible women and children in eligible households participated, not all interviews yielded complete information, and not all participants in wave 1 completed each subsequent wave. Weighting factors were applied to data from completed intake records to adjust for these sources of nonresponse. See CSFII reports for more information on weights and their derivation (USDA, 1985, 1986a,b, 1987a,b,c, 1988). Data processing: Each food and beverage reported as ingested during the 24-hour survey period was assigned a code number, and amounts of foods were converted to their weight in grams. The amount of each of 30 nutri- ents and other food components was calculated using the gram weight of the food and the nutritional value of that food per 100 grams from a nutrient data base developed by HNIS for use with these survey data and with survey data from the Hispanic HANES. Amounts of each nutrient in all foods reported by an individual were summed to obtain the nutrient intake for the day. Data were subjected to computer-assisted screening and checking by the contractor. Dietary intake records that were known to be incomplete were eliminated. The gram weight of each individual's total intake of food and intakes of food energy, protein, fat, carbohydrate, calcium, iron, and vitamin C were compared with the 2nd and 98th percentiles for individuals of the same age and sex in the NFCS 1977-78 as a check for reasonable— ness. Also, the gram weight of food reported was checked against reasonable maximums established by HNIS on a food group basis. Data that fell outside the limits set as reasonable were checked against the original questionnaire and were corrected if in error. Data presentation: Data tapes provided by the contractor were further processed by HNIS to generate tables for the CSFII reports. Data from the basic surveys of CSFII 1985-86 were combined for this report (see appendix II). Income levels: Tables present results by income. Households are classified by their income for the previous calendar year expressed as a percentage of the appropriate Federal Poverty Income guideline. Food intakes: Data on food intakes in the tables are weighted means (averages) for the group of individuals identified in the stub. If no food from a group or subgroup was reported on the survey day(s), that individual's total was zero; the zero was included in the calculation of the group mean. The mean intakes presented in the tables, therefore, include intake values for both users and nonusers. Nutrient intakes: The nutrient intakes in the table are weighted means for the group of individuals identified in the stub. Nutrient intakes do not include intakes from vitamin and mineral supplements for which informa- tion on only the frequency and type used were collected. Sodium intake does not include the amount of sodium from salt added at the table, for which quantity information was not collected. Nutrient intakes at selected percentiles: (Presented only for averages of 4 days’ intakes.) Intakes for 4 days by each individual were totaled and divided by four to obtain a mean intake per day for the individual. Values for individuals were arranged from lowest to highest and intakes were identified at specified weighted percentiles (10th, 25th, 50th, 75th, 90th). Food energy from protein, total fat, fatty acids, and carbohydrate: For each individual, intakes of protein (in grams) were multiplied by 4, fat and fatty acids by 9, and carbohydrate by 4 to estimate the calorie contribution of each energy-providing nutrient. Values were divided by the individual's total food energy intake, then multiplied by 100 to obtain the percentage of an individual's total food energy provided by each nutrient. Weighted means were then determined for different groups of individuals. The general factors 4, 9, and 4 give estimates for a typical mixed diet. Alcohol is also an energy source and was included in determining total energy, but the percentage of food energy contributed by alcohol was not calculated. Nonresponse analyses: For the CSFII 1985-86, at least four separate samples can be identified: the basic and low-income samples for the years 1985 and 1986. Overall CSFII response rates include the screening rate, rate of participation among eligible households, and rate of interviews among eligible women in participating households. The response rates for these samples ranged from a low of 57.5 percent (83.75 percent x 70.88 per- cent x 96.9 percent) for the basic sample in 1985 to a high of 77.3 percent (89.3 percent x 88.2 percent x 98.2 percent) for the low-income sample in 1986. These percentages are based on a one-day response; if the four-day data are considered, the response rate for the basic sample in 1985 is reduced from 57.5 percent to 39.2 percent (Basiotis and Pao, 1987). Individuals who moved from the area in which they were originally contacted were not followed up; this accounted for one-fifth of the dropouts between the first and sixth day of data collection. I-10 Tuszynski and Roidt (1988) provided tables comparing the distribution of selected variables from the March 1986 Current Population Survey with distributions of those variables in the CSFII 1986 basic sample. Variables examined were urbanization, region, division (9 divisions ranging from New England to the Pacific), race, ethnic origin, proportion working, household size, tenure, household income, money spent on food, food stamp participation, and age. There was reasonable agreement between the Current Population Survey and the CSFII 1986 for most variables. The largest differences occurred with tenure, household income, and money spent on food. The CSFII 1986 had approximately 63.3 percent of households owning their residences whereas the Census had 56.7 percent (a difference of 7 percentage points), a mean income approximately $5,000 less ($28,179 versus $33,243), and approximately 28 percent more dollars spent on food at home per capita ($17.6 versus $13.8). Perhaps some of these differences can be accounted for by the lack of exact comparability between the target populations used for the comparison of the two surveys. For example, the CSFII 1986 sample was drawn only from households containing a woman 19-50 years of age and separate U.S. level census data for these households and for women aged 19-50 are not available. It is difficult to determine how much of the difference can be attributed to the lack of comparability and how much may be due to nonresponse. Two studies examined the 1- and 4-day data from the basic group for the CSFII 1985. Basiotis and Pao (1987) did not consider the sample design and the sample weights in their analysis. This unweighted analysis could not assess whether reweighting for nonresponse adequately accounted for the observed differences between respondents and nonrespondents, but the results of Basiotis and Pao (1987) are nevertheless interesting. The 4-day group contains 995 respondents of the 1,459 1-day respondents. The likelihood that a respondent would complete 4 or more days of the CSFII 1985 was estimated to increase with the following characteristics: nonmetropolitan household. participation in Food Stamp Program. participation in WIC. household food supply reported not sufficient. older. Hispanic ethnic origin. a short initial interview. The likelihood of a respondent completing 4 or more days decreased with the following characteristics: did not report an income for last year. one or more children 1-5 years of age. no male head present. physical activity at work reported as light. race other than white or black. smokes cigarettes. contacted in person any time after the first interview. These results were based on a multivariate logistic regression analysis and, hence, could differ from results found when one explanatory variable at a time is examined in relation to the membership in the 1- or 4-day group. This second type of study was conducted by Kott (1988) who performed a weighted analysis and took the sample design into account in the analysis. Fewer variables were considered in this study and means or proportions were compared for each variable for the 1- and 4-day groups. When a weighted analysis was done and design effects were considered, the only significant difference was for the age variable, but this variable does not appear to be of practical importance. The mean intakes for 60 published food groups and 28 nutrients within six age groups were also examined, and very few differences were found between the 1- and 4-day groups in their first day intakes. The six statistically significant, but probably not practically significant, differences were: One-day Four-day Nutrient or sample sample food group ams ams Fat 68.9 70.9 Carbohydrates 191.1 195.9 Polyunsaturated fats 13.8 14.2 I-11 Nonalcoholic beverages 838.4 876.3 Carbonated beverages 286.7 311.5 Diet carbonated beverages 107.6 124.4 The results from the studies of Basiotis and Pao (1987) and Kott (1988) provide different perspectives on the issue of nonresponse bias. They both focused on the comparison of 1- and 4-day data. Basiotis and Pao (1987) suggested that the two groups differed on a number of variables whereas Kott (1988) found that these differ— ences did not translate into many large differences in food or nutrient intake data. These results, comple- mented by the tables from Tuszynski and Roidt (1988), provide a more complete look at the representativeness of the CSFII 1985-86 for some of the four groups from the basic and low-income surveys of 1985 and 1986. The final examination of the nonresponse issue by the USDA was a followup survey of the nonrespondents to the CFSII 1986, wave 1 (first day of data collection). In the followup survey described by Tuszynski and Roidt (1988), the USDA attempted to obtain information on the following: the number of persons in the household. the number of children 1-5 years of age. the number of women 19-50 years of age. race. origin. food stamp participation. WIC participation. money spent on food. a household assessment of food sufficiency. household income. This survey was conducted during November and December 1986, approximately six months after wave 1 of the CFSII 1986 interviewing was done. Three different approaches were used to gather information from the nonrespondents. The overall response rate from the people who were nonrespondents in the CFSII 1986, wave 1 was 37 percent. This value of 37 percent includes some persons for whom other people, for example, a neighbor, provided information. This low response rate raises the question of whether the respondents to this followup survey are representative of the total group of nonrespondents. Based on the limited information available for comparison, the respondents to the followup survey did not differ substantially from the entire group of nonrespondents to the original survey. There were some differences for household composition variables, but these may be somewhat inflated owing to problems with recall. Comparison of the nonrespondent followup survey participants with the participants in the CFSII 1986 showed little or no differences for most variables examined when the sampling weights were not used. However, there were differences for the proportion of metropolitan residents (greater for the followup group), for the propor- tion participating in food stamp programs (lower for the followup group), and for the proportion reporting having adequate quantities of the kinds of foods they like as determined from the food sufficiency question (greater for the followup group). There was also a suggestion of differences for the proportion of Hispanics (greater in the followup group) and for the proportion of WIC users (lower in the followup group). Lastly, a weighted analysis was conducted and comparisons were made to the results one would obtain from incorporating the information from the followup survey. These results suggest that adjusting the sampling weights for nonresponse may have dealt with the disparities mentioned above except for food stamp usage. These investigations performed by the USDA have suggested the following: e Some characteristics were related to whether 4-day data were available or only 1-day data were available. e Even though there may be differences in the composition of the groups with 4-day data or only 1-day data, they did not translate into major differences in dietary intake. e There were some differences between the nonrespondents and the participants in the CFSII 1986, but these differences, except for food stamp use, disappeared when the sampling weights were adjusted to account for the nonresponse. I-12 These investigations have not uncovered major problems with the data, except possibly for the use of food stamps. What is the effect, if any, of too high an estimate of the use of food stamps on dietary intake? Although these studies have not found any major problem, nonresponse bias remains a potential problem. The followup survey of nonrespondents only had a response rate of 37 percent. Consequently, followup participants may not completely represent all nonrespondents, but that response rate is probably about what would be expected from a survey of nonrespondents. With the NFCS 1977-78 and the CSFII 1985-86 both having at least 40 percent nonresponse, there is a strong possibility of a bias in the data. The investigations performed have not discov- ered large differences between the respondents and nonrespondents except for the use of food stamps. Howev- er, even though no other differences were found, drawing conclusions from these data must be done with caution because of the large nonresponse. Although the examinations performed by the USDA do not guaran- tee a lack of bias, they do provide a basis for the use of the data from these surveys which provides some reassurance that nonresponse bias is not likely to have generated strongly misleading results. References: Basiotis, P. P., and E. M. Pao. 1987. Survey Sample Integrity: Experience with the Panel Approach. Research on Survey Methodology. Proceedings of a Symposium Held at the 71st Annual Meeting of the Federation of Ameri- can Societies for Experimental Biology. Administrative Report No. 382. Kott, P. S. 1988. Comparing Estimates Based on the One and Four Day 1985 CSFII Samples. Tables prepared for the Expert Panel on Nutrition Monitoring. Tuszynski, C., and L. Roidt. 1988. Characteristics of the CSFII '86 Core Wave 1 Participants and the U.S. Population. Tables prepared for the Expert Panel on Nutrition Monitoring. U.S. Department of Agriculture. 1985. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII Report 85-1. Hyattsville, Md: U.S. Department of Agriculture. U.S. Department of Agriculture. 1986a. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1985. NFCS, CSFII Report 85-2. Hyattsville, Md: U.S. Department of Agriculture. U.S. Department of Agriculture. 1986a. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Men 19-50 Years, 1985. NFCS, CSFII Report 85-3. Hyattsville, Md: U.S. Department of Agriculture. U.S. Department of Agriculture. 1987a. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Women 19-50 Years and Their Children 1-5 Years, 4 Days, 1985. NFCS, CSFII Report 85-4. Hyattsville, Md: U.S. Department of Agriculture. U.S. Department of Agriculture. 1987b. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Women 19-50 Years and Their Children 1-56 Years, 1 Day, 1986. NFCS, CSFII Report 86-1. Hyattsville, Md: U.S. Department of Agriculture. U.S. Department of Agriculture. 1987c. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 1 Day, 1986. NFCS, CSFII Report 86-2. Hyattsville, Md: U.S. Department of Agriculture. U.S. Department of Agriculture. 1988. Nationwide Food Consumption Survey, Continuing Survey of Food Intakes by Individuals, Low-Income Women 19-50 Years and Their Children 1-5 Years, 4 Days, 1985. NFCS, CSFII Report 85-5. Hyattsville, Md: U.S. Department of Agriculture. I-13 National Health and Nutrition Examination Survey Sponsoring Agency: National Center for Health Statistics Conducted: 1976-80. Target population: The civilian noninstitutionalized population of the United States, 6 months through 74 years of age. Design: Stratified, multistage, probability cluster of households throughout the United States. Sample size/response rate (see also tables I-1 through 1-4): Sample size Interviewed Examined 27,801 25,286 (91%) 20,322 (73%) Measures: Dietary interviews, body measurements, hematological tests, biochemical analyses of whole blood and serum, oral glucose tolerance tests, blood pressures, electrocardiograms; urine tests; X-rays of cervical and lumbar spine and chest (see information in tables I-5 and 1-6). See JNMEC report for additional details on the survey. Nonresponse analyses: See discussion in section on HHANES. I-14 1-1 Table I-1. Sample size and response rates for persons by race, age, and sex: second National Health and Nutrition Examination Survey, 1976-80 T T T T T | | | | Total Number | Percent Number ; Percent Race, age, and sex | sample size interviewed interviewed examined | examined 1 1 1 1 1 ALL RACES Age Total............... 27,801 25,286 90.95 20,322 73.10 4-74 years.......... 24,669 22,271 90.28 17,765 72.01 20-74 years. ......:x 17,390 15,364 88.35 11,864 68.22 6 months-3 years.... 3.132 3.015 96.26 2.557 81.64 4-5 years... ..c.vvvix 1,937 1,861 96.08 1,561 80.59 6-11 years.......... 2,085 1,963 94.15 1,725 82.73 12-15 years... xn 1,616 1,521 94.12 1,320 81.68 16-19 years......... , 1,641 1,562 95.19 1,295 78.92 20-29 years......... 3,522 3,262 92.62 2,627 74.59 80-39 years...:.vvss 2,541 2,300 90.52 1,857 73.08 40-49 years. ........ 2,148 1,882 87.62 1,461 68.02 50-59 years. ........ 2,230 1,913 85.78 1,484 66.55 60-69 years. ........ 5,080 4,380 86.22 3,308 65.12 70-74 years. ........ 1,869 1,627 87.05 1,127 60.30 Sex Male ............... 13,406 12,164 90.74 9,983 74.47 Female. ...cccovevnns 14,395 13,122 91.16 10,339 71.82 Race WOTLE. « cv vv tie iv vivin on wis 23,537 21,350 90.71 17,105 72.67 BlacK. cc. .unsinussnms 3,653 3,389 92.77 2,763 75.64 other. .:nsvipusimns 611 547 89.53 454 74.30 Continued on page I-16 91-1 Table I-1. Sample size and response rates for persons by race, age, and sex: second National Health and Nutrition Examination Survey, 1976-80--continued T T T T T | | | I | Total Number Percent Number Percent Race, age, and sex | sample size interviewed | interviewed examined | examined 1 1 1 1 1 Non-Hispanic white TOtAY.. vvinwusmsvins 22,050 18,912 - 90.30 15,921 72.20 4-74 years.......... 19,802 17,754 89.66 14,109 71.258 20-74 years... ..:+» 14,434 12,678 87.83 9,794 67.85 6 months-3 years.... 2,248 2,158 96.00 1,812 80.60 4-5 years......-:.55 1,348 1,289 95.62 1,070 79.38 6-11 yORaPS..ccsnssns 1,547 1,451 93.79 1,271 82.16 12-15 years......... 1,203 1,125 93.52 976 81.13 16-19 years......... 1,270 1,211 95.35 998 78.58 20-28 years.....: x= 2,803 2,587 92.29 2,077 74.10 30-39 years. ........ 2,061 1,855 90.00 1,505 73.02 40-49 years. ........ 1,734 1,520 87.66 1,187 68.45 50-59 years. ........ 1,844 1,872 85.25 1,224 66.38 60-69 years. ........ 4,373 3,746 85.66 2,841 64.97 70-74 years. ........ 1,619 1,398 86.35 960 59.30 Sex Male ...::eoi:massmums 10,626 9,562 89.99 7,797 73.38 Female... ........... 11,424 10,350 90.60 8,124 71.11 Non-Hispanic black Tota). ciunesnwssnnns 3,610 3,347 92.71 2,725 75.48 4-74 years. ......... 3,108 2,862 92.08 2,290 73.68 20-74 years. ........ 1,935 1,744 90.13 1.318 68.11 6 months-3 years.... 502 485 96.61 435 86.65 4-5 years. .......... 332 323 97.29 277 83.43 8-11 years. . .....s- 342 325 95.03 285 83.33 12-15 years. ........ 256 241 94.14 214 83.59 16-19 years......... 243 229 94.24 196 80.66 20-29 years. ....:.:: 437 412 94.28 339 77.57 30-39 years. ........ 290 263 90.69 203 70.00 40-49 years. ........ 249 211 84.74 157 63.05 50-59 years......... 267 236 88.39 178 66.67 60-69 years. ........ 505 450 89. 11 321 63.56 70-74 years. ........ 187 172 91.98 120 64.17 Sex Male ..::::cci:nvssn 1,695 1,576 92.98 1,324 78.11 Female... ........... 1,915 1,771 92.48 1,401 73.16 LT-1 Table I-2. Number examined and estimated population for non-Hispanic persons 6 months-74 years of age, by sex, age, race, and poverty status: second National Health and Nutrition Examination Survey, 1976-40 oh Me Non-Hispanic white Non-Hispanic black T T | | | | | T | T | Below poverty | Above poverty | Below poverty ; Above poverty | T | T | T | T Estimated | Estimated | Estimated | Estimated Number of population; Number of | population) Number of population, Number of | population examined in examined in examined in examined | in Sex and age persons | thousands persons | thousands persons thousands | persons thousands 1 1 1 1 1 1 1 1 Both sexes 6 months-3 years............. 264 968 1,510 6,645 234 737 174 760 48 YOAIrS.: inn: misnusnmas Ea 154 537 897 3,826 134 408 128 505 6-11 years. .................. 158 1.725 1,076 13,573 136 1,208 139 1,782 Male 12-15 years... ............... 53 560 442 5,284 44 437 54 642 16-19 Years... uv nss musa 71 885 417 5,247 39 462 50 538 20-29 YEAS. «vcvmws vwmswnmsmin 102 1,256 882 13,050 43 487 107 1,291 30-39 years. .............. 45 598 645 10,056 17 * 71 1.022 GO=49 YEAS. ou: inns sans sm min 19 * 531 8,253 i5 * 40 706 BO~59 YOAIrS..c: xv: inunsinwson 33 433 514 7,892 17 * 53 720 60-69 years. ................. 108 501 1,202 6,716 44 173 96 441 70-74 y8AIrsS... . err ass ca3s ss 61 290 350 1,889 24 88 31 148 Female 12-15 yoarS., «ov vii vms smws 2a 49 604 394 5,118 61 585 45 460 16-19 years... ............... 30 1,064 378 5,141 58 618 33 380 20-29 YOAIrS. cvs smmurmnstmuis 2 138 1,784 899 12,690 69 776 108 1,661 30-39 years. .........c.... 80 1,061 701 10, 245 38 428 68 999 40-49 years. ............. 44 583 554 8,616 33 413 59 778 50~59 Years. oi nnvimuns comes 49 632 578 8,870 31 357 57 725 60-69 years. ................. 185 901 1.225 7,135 64 240 94 472 70-74 y8aArsS.. cc: :cvi:snnscnms sa 97 486 411 2,468 25 102 32 151 81-1 Table I-3. Percent of examined persons with missing blood assessments for non-Hispanic persons by age: second National Health and Nutrition Examination Survey, 1976-80 I T I T I T T T | Number | : | | Mean of I | | Red blood jwiite Dood) Mean | Mean corpuscular jexamingd | | cell cell jSorpuscular hemoglobin hemoglobin Race and age jpersons {Hemoglobin Hematecr iv) count count jhemogiobin concentration concentrat fon 1 1 1 1 1 1 1 1 Non-Hispanic white Total. .............. 15,921 6.1 6.1 6.6 6.5 6.8 9.3 89.0 4-74 years. ......... 14,109 2.8 2.8 3.4 3.2 3.5 6.0 5.7 20-74 years. ........ 9,794 0.9 0.9 1.4 1.3 1.7 4.1 3.8 6 months-3 years.... 1,812 31.2 31.2 31.7 31.7 31.9 34.4 34.2 4-85 years. ....«.xxxs 1,070 15.14 15.1 15.6 15.4 15.6 17.9 17.6 6-11 years. ......... 1,271 7.7 7:7 8.3 8.0 8.3 11.3 10.9 12-15 years. .....=s- 976 3.1 3.1 3.7 3.6 3.8 6.6 6.1 16-19 years. ........ 998 1.8 1.8 2.5 2.4 2.5 5.4 4.8 20-29 years......... 2,077 1:2 1.2 1.8 1.7 2.0 4.7 4.3 20-39 years... :«xs« 1,505 1.1 1.1 1.3 1.3 1.8 3.9 3.9 40-49 years. ........ 1,187 0.6 0.6 1.3 1.3 1.3 3.8 3.3 50-59 years. ........ 1,224 0.6 0.6 1.3 1.2 1.5 4.1 3.6 60-69 years. ........ 2,841 0.9 0.9 1.3 1.2 1.6 3.9 3.9 70-74 years. ........ 960 1.0 1.0 1.7 1.5 2.0 8.7 3.4 Non-Hispanic black Total............... 2,725 8.2 8.2 9.9 10.2 10. 12.0 10.8 4-74 years. .us.nnsns 2,290 5.3 5.3 7.0 7.3 7. 9.0 7.8 20-74 years......... 1,318 2.1 2.1 3.7 3.9 3. 5.2 4.1 6 months-3 years.... 435 23.7 23.7 25.5 25.7 25.5 28.0 26.4 4-5 years... cv. nn+ 277 13.4 13.4 15.2 15.5 15.2 17.7 16.2 6-11 years. ......... 285 10.9 10.9 13.7 14.7 13.7 15.8 13.7 12-18 years. ..v. nus 214 7.5 7:5 8.9 9.3 9.3 12.1 11.7 16-19 years......... 196 4.6 4.6 5.6 5.6 5.6 8.2 7.7 20-29 years......... 339 1.8 1.8 3.5 3.8 3.5 5.9 4.4 30-39 years. ........ 203 2.0 2.0 3.0 3.0 3.4 4.4 4.4 40-49 years. ......«. 157 1.3 1.3 3.2 3.2 3.2 4.5 3.2 50-59 years......... 178 2.2 2.2 3.9 3.4 3.9 5.6 3.9 60-69 years. ........ 321 2.8 2.8 4.7 5.0 5.0 5.3 4.0 70-74 years......... 120 2.5 2.5 3.3 4.2 3.3 5.0 4.2 61-1 Table I-3. Percent of examined persons with missing blood assessments for non-Hispanic persons by age: second National Health and Nutrition Examination Survey, 1976-80--continued 1 T 1 1 T 1 | | | | een Lorem | | | | Serum | iron-binding Transfers iy Erythrocyte | Serum Race and age I iron I capacity {Eaturation Vitamin A [protoporphyrin cnolesters) 1 1 1 1 1 1 Non-Hispanic white Total. .............. 7.5 16.1 16.2 36.2 8.2 2.0 4-74 years.......... 6.1 14.6 14.8 32.5 5.3 2.0 20-74 years. ........ 3.0 10.8 10.9 $ 3.3 2.0 6 months-3 years.... 40.9 81.7 51.7 51.5 28.6 $ 4-5 years. .......... 27.5 35.2 35.2 38.9 20.1 $ 6-11 years. .... scx» 14.2 23.7 23.8 27.1 10.6 $ 12-15 years... x: 5.5 16.4 16.5 $ 4.5 $ 16-19 years. ........ 4.3 17.2 17.6 $ 3.4 $ 20-29 years. .. «x: «x 3.0 10.0 10.2 $ 3.5 2.0 30-39 years. .......« 3.5 10.6 10.8 $ 3:5 2.7 40-49 years. ........ 2.4 10.4 10.5 $ 2.9 1.7 50-59 years. ........ 2.6 10.5 10.6 $ 2.7 1.5 60-69 years......... 2.9 11.6 11.8 $ 3.2 2.0 70-74 years. ........ 3.9 11.1 11.2 $ 3.9 2.3 Non-Hispanic black Tota: :iviinmennmen 12.2 24.7 24.9 45.9 8.7 3.9 4-74 YOAPrS.. cs onvwen 10.2 23.0 23.2 42.9 7.2 3.9 20-74 years. ........ 3.9 17.1 17.83 $ 3.3 3.9 6 months-3 years.... 46.9 55.5 55.5 59.4 34.4 $ 4-5 years. .......... 33.2 45.5 45.8 48.7 19.1 $ 6-11 years. ..cocxws + 20.0 31.6 31.6 37.2 15.4 $ 12-15 years. ........ 9.8 23.8 24.3 $ 7:9 $ 16-19 years......... 6.6 17.9 17.9 $ 4.6 $ 20-29 years......... 4.4 15.0 15.0 $ 3.5 3.5 30-39 years. ... .... 2.8 12.3 12.3 $ 3.0 3.0 40-49 years. ........ 2.5 17.8 18.5 $ 1.9 1.9 50-59 years... cc. 3.4 21.9 21.9 $ 2.8 4.5 60-69 years......... 5.3 18.1 18.7 $ 4.0 5.0 70-74 years......... 3.3 20.0 20.0 $ 3.3 5.8 NOTE: $ Assessment not performed on this age group. 03-1 Table I-4. Number examined and estimated population for non-Hispanic persons 6 months—74 years of age, by sex, age, and race: second National Health and Nutrition Examination Survey, 1976-80 Non-Hispanic white Non-Hispanic black Estimated Estimated aie i rr ee as i os T T | | Number of population; Number of population examined in examined in Sex and age persons thousands persons thousands 1 1 Both sexes 6 months-3 years. ............ 1,812 7,770 435 1,586 B=D YOArS. = crwmss wwe ewwswmws 1,070 4,429 277 949 6-11 years. .................. 1,271 15,825 285 3,069 Male 12-15 years... 517 6,116 101 1,112 16-19 Years: . i: ss susr ern emms 511 6,423 96 1,078 20-29 YRAPS.: sums inwrsmne cwmn 1,011 14,727 158 1,881 30-39 years. ................. 707 10, 901 93 1,258 40-48 Years. .: isi usa inte smms 572 8,926 62 980 50-89 years. ci us wus ravewns 575 8,798 77 970 60-69 years. ................. 1,354 7,419 151 6855 70-74 YeOrS. ivi cnn: sxMs sums nma 427 2,263 56 242 Female 12-15 years. ...... cou eoonn. 459 5,931 113 1,106 16-19 YRArS. ..: sushi ions ems sms 487 6,479 100 1,129 2029 years. viunssnwnsmmemn 1,066 14,914 181 2,486 30-39 years. ............. 798 11,548 110 1,474 80-49 YRArS....:vvsnwrsinmnsin 615 9,495 95 1,215 BOBS YOArS. vvicwmnsivmmsnwosmm 649 9,863 101 1,245 60-69 years. ............. 1,487 8,471 170 761 70-74 YyOBIrS. «con vssnmusmmasins 533 3,098 64 286 12-1 Table I-5. Examination components in the second National Health and Nutrition Examination Survey, 1976-80 6 months-2 years 3-11 years 12-19 years 20-74 years 20-74 years (bile acids test group) (glucose tolerance test group) Urine: 6-11 yr only Urine Urine Urine Body measurements Body measurements Body measurements Body measurements Body measurements Physician exam Physician exam Physician exam Physician exam Physician exam Venipuncture Venipuncture Venipuncture Venipuncture Venipuncture Dietary interview Dietary interview Dietary interview Dietary interview Dietary interview Audiometry: 4-11 yr only Audiometry Speech test: 4-6 yr only Allergy test: 6-11 yr only Allergy test Allergy test Allergy test Spirometry: 6-11 yr only Spirometry Spirometry: 20-24 yr only Spirometry: 20-24 yr only Electrocardiogram: 25-74 yr only Chest and neck X-rays: 25-74 yr only Back X-ray: 25-74 yr for men; 50-74 yr for women Bile acids test: 35-74 yr only Electrocardiogram: 25-74 yr only Chest and neck X-ray: 25-74 yr only Back X-ray: 25-74 yr for men, 50-74 yr for women Glucose tolerance test ¢c-1 Table I-6. Blood and urine assessments in the second National Health and Nutrition Examination Survey, 1976-80 6 months-2 years 3-11 years 12-19 years 20-74 years (bile acids group)! 20-74 years (glucose tolerance test group) Lead: all examinees Lead: all examinees 3-6 yr; odd- WHOLE BLOOD Lead: odd-numbered Lead: odd-numbered Lead: odd-numbered numbered examinees 7-11 yr examinees examinees examinees Carboxyhemoglobin: even— Carboxyhemoglobin: even- Carboxyhemoglobin: even— Carboxyhemoglobin: even— numbered examinees numbered examinees numbered examinees numbered examinees Protoporphyrin 2 Protoporphyrin Protoporphyrin Protoporphyrin Protoporphyrin 2 Red blood cell folate Red blood cell folate? Red blood cell folate? Red blood cell folate? Red blood cell folate SERUM Ferritin? Ferritin? Ferritin? Ferritin? Bile acids: 35-74 yr only 3 Cholesterol Cholesterol Triglyceride oe. es High density lipoprotein Pesticides: even-numbered Pesticides: all examinees — examinees Creatinine Creatinine Creatinine > wn ee. Syphilis Syphilis Syphilis won Iron mon Iron Iron otal igon binding capacity Total iron binding capacity otal igon binding capacity Total iron binding capacity Total iron binding capacity Foe Folge? vy Pole? Pole? Bl Bl Bl B1 Bl Vitamin A . ee. op Copper Copper Copper Copper Zinc Zinc Zinc Zinc Glucose tolerance: 75 gram load at 0—, 1-, and 2-hr intervals Vitamin C Vitamin C Vitamin C Vitamin C URINE N-Multistix: 6-11 yr only N-Multistix N-Multistix N-Multistix ee. Gonorrhea Gonorrhea: 20-40 yr only Gonorrhea: 20-40 yr for men; 20-24 yr for women Microscopy cee #6 3 Specific gravity Pesticides Pesticides ce. 1 Bilirubin, SGOT, and alkaline phosphatase performed only on those samples with elevated bile acids. Performed only on those samples with abnormal complete blood count, hemoglobin, hematocrit, or mean corpuscular volume. Hispanic Health and Nutrition Examination Survey Sponsoring Agency: National Center for Health Statistics, DHHS Conducted: 1982-84. Target population: Civilian noninstitutionalized "eligible" Hispanics; aged 6 months through 74 years; that is, Mexican Americans in five Southwestern States, Cubans in Dade County, Florida, and Puerto Ricans in New York, New Jersey, and Connecticut. Design: Complex, multistage, stratified, clustered samples of the defined populations. Sample size/response rate (see also tables I-7 through I-10): Sample size Interviewed Examined Mexican—Americans 9,894 8,554 (87%) 7,462 (75%) Cubans 2,244 1,766 (79%) 1,357 (61%) Puerto Ricans 3,786 3,369 (89%) 2,834 (75%) Measures: Dietary interviews, body measurements, hematological tests, biochemical analyses of whole blood and serum, oral glucose tolerance tests, blood pressures, electrocardiograms (see tables I-11 and I-12). Survey description: The National Center for Health Statistics (NCHS) collects, analyzes, and disseminates data on the health status of Americans. The results of surveys, analyses, and studies are made known primarily through publications and the release of computer data tapes. From 1960 through 1980 NCHS conducted five population-based, national health examination surveys. Each survey involved collecting data by direct physical examination, the taking of a medical history, and laboratory and clinical tests and measurements. Questionnaires and examination components have been designed to obtain and support analyses of data on certain targeted conditions such as diabetes, hypertension, and anemia. Beginning with the first National Health and Nutrition Examination Survey (NHANES I) a nutrition compo- nent was added to obtain information on nutritional status and dietary practices. The numbers of Hispanics in these samples were, however, insufficient to enable adequate estimation of their health conditions. From 1982 through 1984 a Hispanic Health and Nutrition Examination Survey (HHANES) was conducted to obtain data on the health and nutritional status of three Hispanic groups: Mexican Americans from Texas, Colorado, New Mexico, Arizona, and California; Cubans from Dade County, Florida; and Puerto Ricans from the New York City area, including parts of New Jersey and Connecticut. The general structure of the HHANES sample design was similar to that of the previous National Health and Nutrition Examination Surveys. All of these studies have used complex, multistage, stratified, clustered samples of defined populations. The major difference between HHANES and the previous surveys is that HHANES was a survey of three special subgroups of the population in selected areas of the United States rather than a national probability sample. A detailed presentation of the design specifications is found in chapter 5 of "Plan and Operation of the Hispanic Health and Nutrition Examination Survey, 1982-84" (NCHS, 1985). Table I-7. Sample size and response rates for persons of specified Hispanic origin, by age and sex: Hispanic Health and Nutrition Examination Survey, 1982-84 Hispanic origin, Total Number Percent Number Percent age, and sex sample size interviewed interviewed examined examined Mexican American Total. . conve runs tans 9,455 8,222 86.96 7,197 76.12 4-74 years.......... 8,482 7,327 86.38 6,386 75.29 20-74 years. ........ 4,735 3,935 83.10 3,328 70.24 6 months-3 years. ... 973 895 91.98 811 83.35 4-5 years. .......... 519 482 92.87 439 84.59 6-11 years.......... 1,508 1,384 91.78 1,287 85.34 12-15 years. ........ 896 807 80.07 732 81.70 16-19 years. ........ 824 719 87.26 602 73.06 20-29 years. ........ 1,407 1,207 85.79 1,003 71.29 30-39 years. ........ 1,093 957 87.56 837 76.58 40-49 years. ........ 822 661 80.41 558 67.88 50-59 years. ........ 847 653 77.10 561 66.23 60-69 years. ........ 420 338 80.48 266 63.33 70-74 years. ........ 146 119 81.51 101 69.18 Mate wus iwasemme vm 4,589 3,926 85.55 3,385 73.76 Female. ............. 4,866 4,296 88.29 3.812 78.34 Cuban Yota). ons: inv: sms run 2,125 1,677 78.92 1,291 60.75 4-74 years.......... 2,012 1,582 78.63 15,225 60.88 20-74 years. ........ 1,481 1,134 76.57 865 58.41 6 months-3 years. ... 113 95 84.07 66 58.41 =D YBAPS, suwwii vw ii 52 44 84.62 29 55.77 6-11 years.......... 178 182 85.39 126 70.79 12-15 years. ........ 145 123 84.83 104 71.72 16-19 years. ........ 156 129 82.69 101 64.74 20-29 years. ........ 244 173 70.90 127 52.05 30-39 years. ........ 247 196 79.35 152 61.54 40-49 years. ........ 311 237 76.21 186 59.81 50-59 years. ........ 356 286 80.34 223 62.64 60-69 years. ........ 226 167 73.89 117 51.77 70-74 years. ........ 97 75 77.32 60 61.86 MBI@ is viws vm swma 999 786 78.68 608 60.86 Female. ............. 1,126 891 79.13 683 60.66 Puerto Rican TOA. uu suns runs ews + 3,525 3,137 88.99 2,645 75.04 4-74 years.......... 3,195 2,835 88.73 2,387 74.71 20-74 y8ars., ..... «x» 1,764 1,519 86.11 1,220 69.16 6 months-3 years. ... 330 302 91.52 258 78.18 4=8 YRBAIrS., «+ iv vs vous 166 149 89.76 131 78.92 B~11 YRArS. , «uve cn 501 463 92.42 420 83.83 12-18 y8ars. ....: +» 382 358 93.72 316 82.72 16-19 years. ........ 382 346 90.58 300 78.53 20-29 years. ........ 454 394 86.78 317 69.82 30-39 years......... 370 317 85.68 266 71.89 40-49 years. ........ 365 310 84.93 259 70.96 50-59 years. ........ 337 295 87.54 234 69.44 60-69 years. ........ 186 157 84.41 119 63.98 70-74 years. ........ 52 46 88.46 25 48.08 MRIS vis rmms mms «nw 1,575 1,385 87.94 1,188 73.33 Female. ............. 1,950 1,782 89.85 1,490 76.41 I-24 gc-1 Table I-8. Number examined and estimated population for Mexican-American persons 6 months-74 years of age, by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Below poverty Above poverty Estimated Estimated T T | | Number of : population; Number of | population examined in examined in Sex and age persons thousands persons thousands 1 1 Both sexes 6 months-3 years............. 287 268 475 481 4-5 YEArS. «i tiie 147 134 261 252 611 YeArS. uss inmnip ny mmesy 467 414 720 686 Male 12-15 years. ................. 121 128 216 246 16-19 years... ............... 85 110 156 204 20-29 YeArS. uu: sums ans sma 101 225 306 682 BO-39 YEAPS. .uvs rms iwimmy mrmmn om 70 131 284 577 40-49 years... ............... 49 64 173 262 BO=89 YRArS. ..o: cai sani sms 53 55 161 192 60-69 YyeAIrS. ...c.cnvrsnnnirmron 41 54 71 89 70-74 years. ................. 14 * 21 * Female 12-15 years. spurs in moe mms 140 145 183 202 16-19 yoarsS....ccivvesmmnemmas 127 141 175 194 20-29 years. ......... oo... 160 256 358 586 30-39 years. x sivsrsnavecers 134 188 296 445 40-49 years... ........i.. 93 106 189 233 50-59 years... nisms isnint amis 98 101 190 201 B0-6Y years. .:urns :susinmuswess 46 56 77 92 70-74 years. ...........uuun.. 37 45 18 * 92-1 Table I-9. Percent of examined persons with missing blood assessments, by specified Hispanic origin and age: Hispanic Health and Nutrition Examination Survey, 1982-84 T T T T T T T T | | | | | | | | | umber | | | | ] Mean yred blood White blood, Mean Mean corpuscular Hispanic origin [examined cell cell j Sorpuscular hemoglobin hemoglobin and age joer sons jHemoglobin Hematocrit, count count j Femogliobin jeencentration) coneantration | | | | | | | | 1 1 1 1 1 1 1 1 Mexican American Tota)... ..sccams inns 7,197 8 7.5 9.1 8.6 9.1 9.2 8.5 4-T4 yearS. « « ««cuwwrn 6,386 6.8 5.9 7.4 7.0 7.4 7.5 6.8 20-74 years. ........ 3,326 2.8 2.0 3.4 2.9 3.4 3.5 2.8 6 months-3 years.... 811 21.9 20.2 22.4 21.6 22.7 23.2 22.1 4-5 years. .......... 439 831.0 29.8 31.2 31.2 31.2 31.2 31.0 6-11 years... x-xxs 1,287 11.0 9.6 11.7 11.2 11.7 11.7 11.0 12-15 years... .v.» 732 5.3 4.4 5.9 5.6 5.9 5.9 5.3 16-19 years. ........ 602 3.8 3.7 5.1 4.0 5.1 5.1 3.8 20-29 years... : ua 1,003 2.8 2.3 3.2 3.0 83.2 3.2 2.8 30-39 years. ......«« 837 3.0 2.2 3.7 3.2 8.7 3.7 3.0 40-49 years. ........ 558 3.8 2.2 3.8 3.6 3.8 3.9 3.8 50-59 years. ........ 561 2.5 2.0 3.6 2.7 3.6 3.7 2.5 60-69 years......... 266 1.1 0.8 2.6 1.8 2.6 2.8 1.1 70-74 years. ........ 101 2.0 2.0 2.0 2.0 2.0 2.0 2.0 Cuban Total. ..svonnssnnsnn 1,291 7.4 7.4 7.7 7.7 7.7 7.7 7.4 4-74 years. ......... 1,225 6.3 6.2 6.5 6.5 6.5 6.5 6.3 20-74 years. ........ 865 2.3 2.2 2.5 2.5 2.5 2:5 2.3 6 months-3 years.... 66 28.8 28.8 28.8 28.8 28.8 28.8 28.8 4-85 years. ......ex:» 29 34.5 34.5 34.5 34.5 34.5 34.5 34.5 6-11 years. ......... 126 22.2 22.2 22.2 22.2 22.2 22.2 22.2 12-15 years......... 104 141.8 11.5 12.5 12.5 12.5 12.5 11.5 16-19 years. ........ 101 6.9 6.9 6.9 6.9 6.9 6.9 6.9 20-29 years... ...«.» 127 5.5 5.5 5.5 5.5 5.5 5.5 5.5 30-39 years. ........ 152 4.6 3.9 4.6 4.6 4.6 4.6 4.6 40-49 years......... 186 1.1 1.1 1.1 1.1 1.1 1.1 1.1 50-59 years... ...«.. 223 .0 0.0 0.4 0.4 0.4 0.4 0.0 60-69 years. ........ 117 2.6 2.6 2.6 2.6 2.6 2.6 2.6 70-74 years. . «va.» 60 1.7 1.7 3.3 3.3 3.3 3.3 1.7 L3-1 Table I-9. Percent of examined persons with missing blood assessments, by specified Hispanic origin and age: Hispanic Health and Nutrition Examination Survey, 1982-84 — continued T T T T T T T T | | | | | | | | | Number | | | ; Mean | of | jRed Bloc wii te Boos) Mean Mean corpuscular Hispanic origin jexamined, | cell | cell jeorpuseular hemoglobin | hemoglobin and age jPersons {Hemoglobin Hematocrit) count count |hemogliobin |coneentrat ion) concentration | | | | | | | | 1 1 1 1 1 1 1 1 Puerto Rican Total ws snus same cwmn » 2,645 12.6 12.4 14.6 14.6 14.7 14.7 12.7 4-74 years. ......... 2,387 11.4 11.2 13.5 13.5 13.5 13.5 11.4 20-74 years. ........ 1,220 4.8 4.8 7.4 7.4 7.4 7.8 4.9 6 months-3 years.... 258 24.0 23.6 25.2 24.4 26.0 25.2 24.8 4-5 years. .......... 131 43.5 43.5 44.3 44.3 44.3 44.3 43.5 6-11 years....«. iww. 420 21.2 21.0 23.3 23.3 23.3 23.3 21.2 12-15 years......... 316 11.7 11.1 13.6 13.6 13.6 13.6 11.7 16-19 years. ........ 300 10.0 9.7 11.0 11.0 11.0 11.0 10.0 20-29 years... ..c..««. 317 5.0 4.7 7.9 7.9 7.9 7.9 5.0 30-39 years......... 266 5.6 5.6 7.9 1+.9 7.9 7.9 5.6 40-49 years. ........ 259 5.0 5.0 7.7 7.7 7.7 7-7 5.0 50-59 years........» 234 2.6 2.6 5.86 5.6 5.6 6.0 3.0 60-69 years......... 119 6.7 6.7 7.6 7.6 7.6 7.6 6.7 70-74 years. ........ 25 4.0 4.0 8.0 8.0 8.0 8.0 4.0 8¢-1 Table I-9. Percent of examined persons with missing blood assessments, by specified Hispanic origin and age: Hispanic Health and Nutrition Examination Survey, 1982-84 — continued T T T | | lL totar | | | 1 1 1 cn is: en es: as: ed Hispanic origin Serum, iron-binding, Transferrin Erythrocyte Serum and age iron capacity saturation Vitamin A Vitamin E protoporphyrin cholesterol Mexican American Tota). cin wusvwinmmen 12.6 9.3 12.6 11.0 11.0 6.8 3.7 4-74 years. ......... 12.6 9.3 12.86 11.0 11.0 6.8 3.7 20-74 years......... 6.4 3.4 6.4 4.5 4.6 3.2 3.7 6 months-3 years.... $ $ $ $ $ $ $ 4-5 years... vx vx 47.2 42.6 47.4 46.7 46.9 30.5 $ 6-11 years. ......... 19.7 17.3 19.7 20.1 20.0 9.8 $ 12-15 years. ........ 10.2 5.5 10.2 6.7 6.7 4.8 $ 16-19 years... ee: mms 9.3 5.5 9.3 6.3 6.3 5.0 $ 20-29 years. .....««- 6.3 3.7 6.3 5.1 5.1 3.6 4.0 30-39 years......... 6.7 3.2 6.7 4.4 4.4 3.1 3.8 40-49 years. ........ 5.9 3.4 5.9 4.7 4.7 3.0 3.8 B0-859 years... :wws 6.2 3.2 6.2 4.1 4.1 3.2 8.7 60-69 years......... 6.8 2.6 6.8 3.0 3.4 2.3 2.3 70-74 years. ........ 8.9 5.0 8.9 5.0 5.9 5.0 4.0 Cuban Tota. .o: swe smn s smu 6.2 6.2 7.6 6.2 6.2 7.8 4.5 4-74 years. ......... 6.2 6.2 7.6 6.2 6.2 7.8 4.5 20-74 years. ........ 2.2 2.2 3.1 2.2 2.2 4.4 4.5 6 months-3 years. ... $ $ $ $ $ $ $ 4-5 YyeArS. cc xiuvssne 34.5 34.5 51.7 34.5 34.5 34.5 $ 6-11 years. ......... 22.2 22.2 23.0 22.2 22.2 22.2 $ 12-15 years. ..q:»: 15x 11.5 11.5 14.4 11.5 11.5 11.5 $ 16-19 years... ... «x 6.9 6.9 6.9 6.9 6.9 6.9 $ 20-29 years......... 5.5 5.5 7.9 5.5 5.5 7.1 10.2 B0O-39 years... «x:ws 3.9 3:8 4.6 3.9 3.9 5.9 5.3 40-49 years. ........ 1.1 1.1 1.6 1.1 1; 3 2.2 2.2 50-59 years. ........ 0.0 0.0 0.9 0.0 0.0 3.1 2.7 60-69 years......... 2.6 2.6 3.4 2.6 2.6 4.3 5.1 70-74 years... ««:««« ¥7 1.7 1.7 1.7 1.7 6.7 3.8 NOTE: $ Assessment not performed on this age group. 62-1 Table I-9. Percent of examined persons with missing blood assessments, by specified Hispanic origin and age: Hispanic Health and Nutrition Examination Survey, 1982-84 - continued ——————— T T | | | | Total | | | | | | | | 1 1 T T | | | | | | | | | | | | | | Hispanic origin Serum; iron-binding, Transferrin Erythrocyte Serum and age iron capacity saturation Vitamin A Vitamin E protoporphyrin, cholesterol | do Puerto Rican TofA). ns cnucimurnms 11:1 11.1 13.4 11.1 11.1 11.1 6.7 4-74 years. ......... 14.4 11.1 13.4 11.1 13.1 1.9 6.7 20-74 years. ........ 4.8 4.8 6.4 4 4.8 4.8 6.7 6 months-3 years.... $ $ $ $ $ $ $ 4-5 years. .......... 42.7 42.7 53.4 42.7 42.7 42.7 $ 6-11 years... .xv: sms 21.0 24.0 23.1 21.0 21.0 21.0 $ 12-15 years.: .««: sx 11.1 11.1 12.7 14:14 41.1 11.1 $ 16-19 years......... 9.7 9.7 12.0 9.7 9.7 9.7 $ 20-29 years. ........ 4.7 4.7 7.9 4.7 4.7 4.7 8.2 80-39 years... vs sex 5.6 5.6 6.8 5.8 5.6 5.6 6.4 40-49 years. ........ 5.0 5.0 5.4 5.0 5.0 5.0 6.6 50-59 years. ........ 2.6 2.6 4.7 2.6 2.6 2.6 4.3 60-69 years. ........ 6.7 6.7 7.6 6.7 6.7 6.7 9.2 70-74 years. .: .x«:«v 4.0 4.0 4.0 4.0 4.0 4.0 4.0 NOTE : $ Assessment not performed on this age group. 08-1 Table I-10. Number examined and estimated population for Hispanic persons 6 months-74 years of age, by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 i sec oir ai opm io] Mexican American Cuban Puerto Rican Estimated Number of population ens ed J i —— Number of ET __ Estimated population Number of T Estimated population examined in examined in examined in Sex and age persons thousands persons thousands persons thousands Both sexes 6 months-3 years. ............ 811 795 66 18 258 96 A-B YOArS.. :covsmnnsm irene 439 412 29 8 131 49 6-11 years. .................. 1,287 1,187 126 38 420 157 Male 12-15 YOAIrS iar int ism mmniton 379 418 58 16 154 60 16-19 years. .... uur snmsnmass 275 356 56 16 147 57 20-29 years... ..... i... 444 991 57 30 114 76 30-39 years. vi: rns anne amn ig 376 750 56 28 90 61 40-49 YOAPrS. .. sows nmin 243 358 82 32 88 46 50-59 years... ............... 235 270 109 34 101 35 60-69 YyOars. in. .ussnsninns+® 124 159 45 14 41 15 70-74 years. . ..: xs:sumsmes in 39 56 28 9 11 * Female 12-18 years. .....cvsnnvivnviw 353 375 46 14 162 60 16-19 years... ............... 327 360 45 13 153 58 20-29 years. ...: «us: wimms sn 559 903 70 32 203 130 30-39 years. ................. 461 680 96 44 176 111 40-49 years... .... a... 315 378 104 39 171 81 B0-59 years... .. vixvssnniesn 326 344 114 36 133 44 60-69 years. ................. 142 169 72 22 78 24 70-74 YOArS....s:iintiimeasmbsini 62 76 32 10 14 * 16-1 Table I-11. Examination components in the Hispanic Health and Nutrition Examination Survey, 1982-84 6 months-5 years 6-11 years 12-19 years 20-74 years (nonfasting) 20-174 years (fasting) Physician exam Dental exam Dietary interview Body measurements TB skin test! Tympanic impedance Venipuncture Physician exam Dental exam Dietary interview Body measurements TB skin test! Tympanic impedance Audiometry Vision test Venipuncture Urine Physician exam Dental exam Dietary interview Body measurements TB skin test! Tympanic impedance Audiometry Vision test Venipuncture Urine Hair collection Physician exam Dental exam Dietary interview Body measurements TB skin test! Tympanic impedance Audiometry Vision test Venipuncture Urine Posterior-anterior chest X-ray Lateral chest X-ray® Electrocardiogram Physician exam Dental exam Dietary interview Body measurements TB skin test! Tympanic impedance Venipuncture Urine Hair collection Posterior-anterior chest X-ray Lateral chest X-ray® Electrocardiogram Oral glucose tolerance test Ultrasound examination 1 Conducted under special arrangements in the California and Dade County, Florida sites. Fingerstick only for 6 months-3 years of age. For persons 45-74 years of age. NOTE: ... = Category not applicable. Table I-12. Blood and urine assessments in the Hispanic Health and Nutrition Examination Survey, 1982-84 Age Test 6 months-5 years 6-11 years 12-19 years 20-74 years (nonfasting) 20-74 years (fasting) Lead Protoporphyrin Red blood cell folate? Complete blood counts® Iron Total iron binding capacity Ferritin Folate? Differential count? Vitamins A and E Glucose Cholesterol Triglycerides High density lipoprotein Pesticides (organic) Syphilis serology Albumin Total protein Alkaline phosphatase LDH SGOT Phosphorus Uric acid Total bilirubin Calcium Urea Nitrogen Creatinine Total CO? Chloride Sodium Potassium SGPT Tetanus Selected trace metals’ Oral glucose tolerance test N-Multistix Pesticides PHM PALL RR P44 pd Da KX Whole blood TEETE | LR Re Peo Serum and hair samples X Plasma Pad MM PAPE PEE DED DA DA DE Dd DA DE DE Dd DA DE DDE DE Dd Dd Dd PE HM PAPE DADE PA PE DE KK sR a Ra Ral R Rl 1 Fingerstick only for persons 6 months-3 years of age. Special hematological subsample only. Includes hematocrit, hemoglobin, red and white cell counts, mean corpuscular volume, mean corpuscular hemoglobin, and mean corpuscular hemoglobin concentration. 4 4-5 years of age. 5 Performed on a subset of the samples collected as a pilot test only. NOTE: X = category applicable; ... = category not applicable. Data collection began with a household interview. Several questionnaires were administered: eo A Household Screener Questionnaire (HSQ), administered at each selected address, for determining household eligibility and for selecting sample persons. e A Family Questionnaire (FQ), administered once for each family containing sample persons, which included sections on family relationships, basic demographic information for sample persons and head of family, Medicare and health insurance coverage, participation in income assistance programs, and housing charac- teristics. e An Adult Sample Person Questionnaire (ASPQ), for persons 12 through 74 years which, depending on age, included sections on health status measures, health services utilization, smoking (20 through 74 years), meal program participation, and acculturation. Information on the use of medicines and vitamins in the past two weeks was also obtained. e A Child Sample Person Questionnaire (CSPQ), for sample persons 6 months through 11 years which included sections on a number of health status issues, health care utilization, infant feeding practices, participation in meal programs, school attendance, and language use. Information on the use of medicines and vitamins in the past two weeks was also obtained. At the Mobile Examination Center two questionnaires were administered and an examination performed: eo An Adult Sample Person Supplement (ASPS), for sample persons 12 through 74 years, which included sections on alcohol consumption, drug abuse, depression, smoking (12 through 19 years), pesticide exposure, and reproductive history. e A Dietary Questionnaire (DQ), for persons 6 months through 74 years, by which trained dietary interviewers collected information about "usual" consumption habits and dietary practices, and recorded foods consumed 24-hours prior to midnight of the interview. e An examination which included a variety of tests and procedures was performed. Age at interview and other factors determined which procedures were administered to which examinees. A dentist performed a dental examination and a vision test. Technicians took blood and urine specimens and administered a glucose tolerance test, X-rays, electrocardiograms, and ultrasonographs of the gallbladder. Technicians also per— formed hearing tests and took a variety of body measurements. A physician performed a medical examina- tion focusing especially on the cardiovascular, gastrointestinal, neurological, and musculoskeletal systems. The physician's impression of overall health, nutritional and weight status, and health care needs were also recorded. Some blood and urine specimen analyses were performed by technicians in the examination center; others were conducted under contract at various laboratories. & Descriptions of the special clinical procedures and tests follow: Ultrasound examination of the gallbladder: For the purpose of estimating the prevalence of gallstones, an ultrasound examination was conducted on a specified subsample of examinees 20 through 74 years of age. Dental examination: All examined persons received an examination that included the following measures of dental health: (1) a decayed, missing, filled (DMF) surface index, (2) a dental restorative treatment needs index, (3) a simplified oral hygiene index, (4) a periodontal index, (5) an assessment of need for and quality of full dentures, and (6) an assessment of malocclusion. Vision screening: Examined persons 6 through 74 years of age were tested for visual acuity. The near vision and distance vision tests involved reading test cards with Sloan letters or Landolt rings set at standard distances from the eyes. Binocularity of vision was tested by using the Random Dot E (RDE) test. Tympanic impedance: For the purpose of assessing levels of effusive and noneffusive middle ear disease, impedance tympanometry was performed on all examined persons. In this procedure, the mobility of the tympanic membrane is induced and recorded electronically under varied air pressures in the ear canal. Puretone audiometry: This test, conducted on examined persons between the ages of 6 and 74 years, permitted determination of threshold levels of hearing for frequencies of 500, 1000, 2000, and 4000 hertz for each ear. 1-33 Electrocardiograms: Electrocardiographic signals, for examined persons 20 through 74 years of age, were digitized and recorded on magnetic tape. This provided normative data on amplitude, duration, interval and axis measurements, and permitted interpretations of heart disease according to the Minnesota classification code. Body measurements: Measurements were made on all examinees and included standing height and/or recum- bent length, depending on age; body weight; triceps and subscapular skinfolds; and various other measure— ments. Hair collection: A small sample of hair was collected from each examined person 12 through 74 years of age and analyzed for selected trace elements. These data, which were collected for the Centers for Disease Control for methodological purposes, can be related to body burdens or stores of the elements and to overall nutritional and health status. Tuberculin skin test: In the California and Dade County, Florida PSUs, examinees were injected with 5 tuberculin units of purified protein derivative (PPD) to test for exposure to tuberculosis. Examinees were examined at the examination center or at home 2 to 3 days later by a trained nurse who read and recorded the test results. X-rays: Two chest X-rays were made, as follows: e Posterior—anterior (PA) - This X-ray of persons 20 through 74 years of age was used for the determination of heart size and diagnosis of cardiovascular conditions, lung and chest conditions, and structural deformities. e Lateral — Taken of persons 45 through 74 years of age, this X-ray provided an additional parameter for the determination of heart size. No X-rays were taken of pregnant women. Urine tests: The following tests were performed on casual samples of urine: e N-Multistix tests - These urinary dipstick tests for qualitative protein, glucose, ketones, bilirubin, blood, urobilinogen, pH, and bacteriuria (nitrite test) were done for examined persons 6 through 74 years of age. e Urinary sediments - Sediments including red cells, white cells, and casts were measured for persons 6 through 74 years of age. e Analysis for pesticide levels — Urine samples from a half sample of examined persons 12 through 74 years of age were tested for the presence of alkyl phosphate residues and metabolites, carbamate residues, phenolic compound residues, and malathion metabolites. Tests on blood samples: Tests on blood samples provide a broad range of information related to health and nutrition. The particular tests performed varied with the specific target condition and age group. ® Oral glucose tolerance test (OGTT) - This test involved the collection of blood specimens from examined persons while they were in a fasting state as well as at 1 and 2 hours after the glucose challenge. The test was performed on a specified half sample of examined adults 20 through 74 years of age to provide estimates of the prevalence of diabetes and impaired glucose tolerance. e Liver function tests — Biochemical liver tests, performed on examined persons 20 through 74 years of age, included bilirubin, SGOT, SGPT, and alkaline phosphatase. e Anemia-related laboratory tests — For the diagnosis of anemia, tests on blood samples included protoporphyrin, iron, total iron-binding capacity, red cell folates, serum folates, serum ferritin, and abnormal hematological indices. o Other biochemical nutritional tests — These tests included alpha-tocopherol and retinol. e Serum lipids — Because of their relevance to cardiovascular disease, determinations were made of serum cholesterol, triglycerides, and high density lipoprotein (HDL). I-34 e Biochemical tests for body burden from environmental exposures — Levels of lead (all persons) and organo- chlorine pesticide residues and metabolites (half sample of persons 12 through 74 years of age) were determined. Tests for carboxyhemoglobin and thiocyanate were performed on a half sample of persons 3 through 74 years of age for the first 12 examination sites only. e Hematology — The hematological determinations included hemoglobin, hematocrit, red blood cell count, white blood cell count and differential analysis, and red blood cell morphology. eo Kidney function —'The serum creatinine test for kidney function was performed on blood samples. eo Syphilis serolggy — The serology determinations for syphilis for examined persons 12 through 74 years of age included qualitative and quantitative ART, a FTA-ABS, and MHA-TP. Because the HHANES sample is not a simple random one, it is necessary to incorporate sample weights for proper analysis of the data. These sample weights are a composite of individual selection probabilities, adjust— ments for noncoverage and nonresponse, and poststratification adjustments. Because of the complex sample design and the ratio adjustments used to produce the sample weights, commonly used methods of point and variance estimation and hypothesis testing which assume simple random sampling may give misleading results. Nonresponse analyses: The response-nonresponse situation for NHANES and HHANES was somewhat differ— ent from the surveys conducted for the USDA. The NHANES and HHANES consisted of a screening compon- ent, an interview component, and an examination component. In NHANES II and HHANES, the screening rates ranged from 99.5 to 99.9 percent. Unlike CSFII 1985-86, proxy responses were accepted for the screening questionnaire which determines if an eligible respondent resides in the household. The extremely small non- response to the screeners is taken into account in the post-stratification adjustments in which ratio adjustments were made within each age—sex—race/ethnicity cell to independent estimates provided by the U.S. Bureau of the Census for the population as of the midpoint of the survey. Because screening is completed in virtually the entire sample and the extremely small nonresponse to the screener is built into the post-stratification adjustment calculations, NCHS has not found it necessary to show screening rates in the response rate calculations. The interview response rate was calculated by dividing the number of sample persons completing the interview by the total sample size. Similarly, the examination response rate was calculated by dividing the number of sample persons completing the examination by the total sample size. The nonresponse bias analyses considered in this report covered both NHANES II and HHANES. In both surveys, the first stage of nonresponse (refusal of the screener) constituted less than 0.5 percent of the target sample, and therefore is not an important problem. For the NHANES II analysis, the second stage of non- response (nonresponse to the interview) was addressed through comparison of identical questions in the National Health Interview Survey which at that time had a response rate in the high 90s. For HHANES, information from the screeners was used to assess bias in the interviewed sample. After demonstrating that there did not appear to be significant bias in the interviewed sample, the third stage of nonresponse (nonresponse to the examination) was assessed in both surveys by comparing the characteristics of the examined and interviewed samples. For each of the specific analyses presented in this report, an additional nonresponse bias analysis was performed to determine if the analytical subsamples differed from the examined sample. The NHANES II had an overall interview rate of 91 percent and an examination rate of 73 percent. The examination rate is of most interest, and it ranged from a low of 62 percent for persons aged 65-74 years to a high of 82.7 percent for children 6-11 years old (DHHS/USDA, 1986). The existence of these components made it possible to use information from those interviewed but not examined to address the nonresponse bias issue. Additionally, the National Health Interview Survey that was conducted at approximately the same time as NHANES II had a response rate of 97 percent and included many of the same questions as NHANES II, thereby facilitating the investigation of nonresponse bias in NHANES II. Results of this investigation were I-35 published by Forthofer (1983). There was little indication of bias due to nonresponse for the variables examined. The HHANES 1982-84 consisted of three separate surveys. For the Mexican Americans, the response rate for those examined was 75 percent compared to rates of 61 percent for the Cubans and 75 percent for the Puerto Ricans. Clearly, there was substantial nonresponse for these households surveys, consequently raising the possibility of nonresponse bias. Therefore, it was necessary to compare the samples to the target populations based on what- ever information was available. Even if differences between the samples and the populations were absent, non— response bias could be present in the data. If differences were found, it may be possible to adjust the data for the differences that were identified. The NCHS conducted an examination of the potential for nonresponse bias in HHANES. The investigation consisted of several components: e Comparison of those who provided only limited demographic data (age, sex, household size, and language of the screener) with those who provided a medical history. e Comparison of the total selected sample with only those who provided the medical history. e Comparison of those who provided only limited demographic data with those who were examined. e Comparison of the weighted interviewed and examined samples: = for those 6 months-74 years of age based on the eight variables: birthplace, marital status of the head of household, education of the head of household, health insurance coverage, perceived health status, report of never having had a routine physical, report of ever having vision problems, and report of ever having hearing problems; — for those 12-74 years of age based on five variables: previous diagnosis of high blood pressure or hypertension, report of previous physician diagnosis of anemia, report of previous physician diagnosis of diabetes, report of previous physician diagnosis of bronchitis, and self-perception of overweight; — for those 20-74 years of age based on the variables of report of previous physician diagnosis of gallstones and current smoking status; and ~ for those 6 months-11 years of age based on three variables: previous physician diagnosis of anemia, perception of overweight, and previous physician diagnosis of asthma. Interviewed and examined persons tended to be younger than those not interviewed or examined. For the Mexican Americans and Puerto Ricans, females were more likely than males to be interviewed and examined. Household size was positively associated with participation, and if the language of the interview was Spanish, the household was more likely to participate. In the comparison of the interviewed and examined, Cubans who were examined were more likely to have a family income below $10,000 than those who were interviewed. Cuban females who were examined were more likely to have no health insurance, not to have had a routine exam, and to have hearing problems. The Cubans examined were more likely to state that they had fair or poor health status than the interviewed persons. Cuban males aged 45-74 years who were examined were more likely to report a previous diagnosis of hyperten- sion and examined Cuban females were more likely to be current smokers than those in the interviewed group. For children aged 6 months-11 years, Mexican-American girls in the examined sample were more likely to be called overweight than those in the interviewed sample, and Puerto Ricans in the examined sample were more likely to have a previous physician diagnosis of anemia than those in the interviewed group. The adjustment of the sample weights for nonresponse and the poststratification done for the Mexican- American sample corrected for differences in the marginal distributions of age, sex, income, and household size. There is no guarantee that because the marginal distributions have been made to agree, bias no longer exists; however, it is a reasonable approach. These adjustments have not dealt with the differences that were found in the Cuban sample between the examined and interviewed groups. These differences are of concern, suggesting the data from the Cuban sample should be interpreted with caution. 1-36 References: Forthofer, R. N. 1983. Investigation on Nonresponse Bias in NHANES II. Am. J. Epidemiol. 117:507-15. National Center for Health Statistics. 1985. Plan and Operation of the Hispanic Health and Nutrition Examination Survey 1982-84. Programs and Collection Procedures. Series 1, No. 191. Public Health Service. Hyattsville, Md. U.S. Department of Health and Human Services and U.S. Department of Agriculture. 1986. Nutrition Monitoring in the United States——A Progress Report from the Joint Nutrition Monitoring Evaluation Committee. DHHS Pub. No. (PHS) 86-1255. Public Health Service. Washington: U.S. Government Printing Office. I-37 NHANES I Epidemiologic Followup Study Sponsoring Agency: National Center for Health Statistics, DHHS Conducted: 1982-84, 1986. Target population: 14,407 persons examined in NHANES I who were 25 through 74 years old at baseline. Sample size/response rate: Total subjects in cohort Traced (1982-84) Interviewed 1982-84) 14,407 13,380 (92.9%) 12,220 (84.8%) Measures: Personal interview for survivors and proxy interviews for decedents and incapacited persons includ- ing medical history, history of hospitalization, functional status, medications usage, smoking history, alcohol history, psychological status, food frequency, and physical activity; physical measurements of pulse, blood pressure, and weight; death certificates; hospital and nursing home records for overnight stays. Background: NHANES I was carried out in 1971-75 and was designed to collect extensive health-related information on a probability sample of the U.S. civilian noninstitutionalized population. To increase sample size in selected population subgroups, there was oversampling of the elderly, of women of childbearing age, and of persons living in poverty areas. The NHANES I Epidemiologic Followup Study (NHEFS) builds on the baseline data collected in NHANES I. The objectives of the followup are to study (a) mortality, morbidity, and institutionalization associated with suspected risk factors; (b) changes in the participants’ characteristics between NHANES I and the followup survey; and (c) the progression of chronic disease and functional impairments. The study population comprised the 14,407 examinees in NHANES I who were 25-74 years old at the time of that survey. An attempt was made to trace all these examinees to their current addresses. Personal interviews, which included weight, blood pressure, and pulse measurements, were conducted with survivors. Personal interviews were also conducted with suitable proxies for deceased or incapacitated participants. In order to be accepted as a proxy respondent, the individual had to answer correctly the verification questions which were used to establish the identity of the deceased NHANES I participant. Persons who had lived with the subject were the preferred proxies. Thirty-seven percent of the proxy respondents were spouses of the subject, 39 percent were children, 10 percent were siblings, and the remaining 14 percent had various connec- tions with the subject such as friend, neighbor, or other relative. Death certificates for deceased persons were also obtained. Records of hospitalizations and nursing home stays were collected for both surviving and deceased participants. The results of data collection activities in the NHEFS are summarized in the diagram (figure 1-1). Ninety- three percent of the cohort was successfully traced—-11,358 were alive at followup, and 2,022 were deceased. Interviews were successfully completed for 93 percent of the traced examinees who were alive at followup. Interviews with proxy respondents were completed for 84 percent of those deceased. Death certificates were obtained for 96 percent of decedents. Both a death certificate and a proxy interview were available for 1,610 participants (80 percent of all decedents). A proxy interview only was available for 87 decedents (4 percent), and a death certificate only was available for the remaining 325 decedents (16 percent). Interviewed 10,523 (92.6 percent) Alive: 11,358 Sow] Not interviewed: 835 (7.4 percent) Traced: 13,380 Total Subjects (92.9 percent) In cohort: 14,407 — Not traced: 1,027 (7.1 percent) eee] Measured Pulse 10,144 (96.4 percent) Blood Pressure 9,955 (94.6 percent) Weight 10,005 (95.1 percent) Proxy interview Deceased: 2.022 Proxy interview only: 87 (4.3 percent) Death certificate only: 325 (16.1 percent) and death certificates: 1,610 (79.6 percent) - Figure I-1. Summary of data collection procedures, NHANES I Epidemiologic Followup Study. References Cornoni-Huntley Epidemiologic Followup Survey. Publ. Health Rep. 98:245-251. J., H. E. Barbano, J. A. Brody, et al. 1983. National Health and Nutrition Examination I-- National Center for Health Statisics, B. B. Cohen, H. E. Barbano, C. S. Cox, et al. 1987. Plan and Operation of the NHANES I Epidemiologic Followup Study, 1982-84. Vital and Health Statistics. Series 1, No. 22. DHHS Pub. No. (PHS) 87-1324. Public Health Service. Washington: U.S. Government Printing Office. 1-39 National Health Interview Survey Sponsoring Agency: National Center for Health Statistics, DHHS Specific survey name: National Health Interview Survey on Health Promotion and Disease Prevention Conducted: January-December 1985. Target population: Civilian noninstitutionalized population aged 18 years and over in the United States. Design: Complex, multistage, stratified, clustered sample. Sample size/response rate: Sample size Interviewed Response rate 38,000 34,000 90% Survey description: This survey was designed to measure progress toward the 1990 Health Objectives for the Nation and was collaboratively designed, sponsored, and analyzed by the Agencies in the Public Health Service which have responsibility for monitoring progress toward the Objectives. It is planned to repeat this survey in 1990, and to include new questions as baseline measures for the Year 2000 Health Objectives for the Nation. Detailed list of variables: Topical areas Pregnancy and smoking Occupational safety and health Injury control Child safety and health High blood pressure Stress Exercise Smoking Alcohol use Dental care General health habits Nutrition-related items Breakfast regularity Snacking Doctor's advice on diet Height and weight Weight loss knowledge Weight loss practice Perceived relative weight Breastfeeding Diet to reduce hypertension Fluoride use Specific survey name: National Health Interview Survey on Vitamin and Mineral Supplements Conducted: January-June 1986. Target population: Civilian noninstitutionalized children aged 2-6 years and adults aged 18 years and over in the United States. Design: Complex, multistage, stratified, clustered sample. Sample size/response rate (provisional estimates): Sample size Interviewed Examined Children, 2-6 years 4,000 3,600 90% Adults, 18 years 12,500 11,250 90% and over I-40 Measures: Self-reports (for adults) and proxy-reports (for children) of vitamin or mineral supplements to the diet used during the two weeks before interview, including product name and manufacturer, nutrient informa- tion from product label, and dosage. Survey description: Questions were designed to determine the prevalence and quantitative level of vitamin and mineral supplement intake among adults and young children in the United States. The project was initiated by the Food and Drug Administration's Division of Consumer Studies, Center for Food Safety and Applied Nutrition. The data obtained will provide baseline data aimed at meeting several priority objectives of the 1990 Health Objectives for the Nation. Detailed list of variables: For each sample person it was determined how many vitamin or mineral supplement products were taken in the 2 weeks before interview. For each product, the interviewer obtained the product name and manufacturer, the number of days during the 2 weeks on which the product was taken, the form in which it was taken, the number of doses taken, the duration for which the product had been taken, and whether or not it was prescribed by a doctor. If the product container was available at the time of interview, the inter- viewer also recorded from the label each nutrient and its dosage; however, FDA later matched product informa- tion to its data files on product nutrient content, and those nutrient data were written onto the data tape. The pregnancy status and lactation status of sample women 18-44 years of age were also determined. General description Background: The National Health Interview Survey (NHIS) is the principal source of information on the health of the civilian noninstitutionalized population of the United States. The NHIS is one of the major data collec- tion programs of the National Center for Health Statistics (NCHS). The National Health Survey Act of 1956 provided for a continuing survey and special studies to secure accurate and current statistical information on the amount, distribution, and effects of illness and disability in the United States and the services rendered for or because of such conditions. The survey referred to in the Act, now called the National Health Interview Survey, was initiated in July 1957. Since 1960, the survey has been conducted by NCHS, which was formed when the National Health Survey and the National Vital Statistics Division were combined. Purpose and scope: The objective of the survey is to address major current health issues through the collection and analysis of data on the civilian noninstitutionalized population of the United States. National data on the incidence of acute illness and injuries, the prevalence of chronic conditions and impairments, the extent of disability, the utilization of health care services, and other health-related topics are provided by the survey. A major strength of this survey lies in the ability to display these health characteristics by many demographic and socioeconomic characteristics. The NHIS data are obtained through personal interviews with household members. Interviews are conducted each week throughout the year in a probability sample of households. The interviewing is performed by a per- manent staff of interviewers employed by the U.S. Bureau of the Census. Data collected over the period of a year form the basis for the development of annual estimates of the health characteristics of the population and for the analysis of trends in those characteristics. The survey covers the civilian noninstitutionalized population of the United States living at the time of the interview. Because of technical and logistical problems, several segments of the population are not included in the sample or in the estimates from the survey. Persons excluded are: patients in long-term care facilities, persons on active duty with the Armed Forces (though their dependents are included), and U.S. nationals living in foreign countries. Sample design: The NHIS is a cross-sectional household interview survey. Sampling and interviewing are continuous through each year. The sampling plan follows a multistage area probability design that permits the representative sampling of households. The first stage consists of a sample of about 200 primary sampling units (PSU) drawn from approximately 1,900 geographically defined PSU that cover the 50 States and the District of Columbia. A PSU consists of a county, a small group of contiguous counties, or a Metropolitan Statistical Area. Within PSU, intermediate stage units called segments are defined in such a manner that each segment contains approximately 40 households. Within these segments, a systematic sample of eight households is selected for the NHIS sample. I-41 The NHIS sample implemented with the 1985 data collection year was a complete redesign from previous years. A feature that was added for the 1985 design is the formation of panels of PSU. The total NHIS sample of PSU is subdivided into four separate panels such that each panel is a representative sample of the U.S. population. This design feature has a number of advantages, including flexibility for the total sample size. The 1985 NHIS sample included three of the four panels and the 1986 NHIS sample included two panels. Another design feature implemented in 1985 was the oversampling of black persons to improve the precision of estimates for that population. This resulted in an increase in the number of black persons in the NHIS sample by approximately 75 percent and an increase in the precision of most related statistics by more than 20 percent. The new sample design also facilitates followup studies of respondents and linkage with other national health— related data sets such as the National Death Index. The households selected for interview each week are a probability sample representative of the target popula- tion. With four sample panels, data are collected from approximately 50,000 households including about 135,000 persons in a calendar year. Participation is voluntary and confidentiality of responses is guaranteed. The annual response rate of NHIS is over 95 percent of the eligible households in the sample. The nonresponse is divided equally between refusals and households where no eligible respondent could be found at home after repeated calls. Data collection procedures: Data are collected through a personal household interview conducted by interview- ers employed and trained by the U.S. Bureau of the Census according to procedures specified by the National Center for Health Statistics. All adult members of the household 17 years of age and over who are at home at the time of the interview are invited to participate and to respond for themselves. For children and for adults not at home during the interview, information is provided by a responsible adult family member (19 years of age and over) residing in the household. Between 65 and 70 percent of the adults 17 years of age and over are self-respondents. Gener- ally, a random subsample of adult household members is selected to respond for themselves to questions on current health topics that are added each year. Nationally, there are approximately 150 interviewers, trained and directed by health survey supervisors in each of the 12 Census Bureau Regional Offices. The supervisors are career Civil Service employees whose primary responsibility is the NHIS. The interviewers are part-time employees, selected through an examination and testing process. Interviewers receive thorough training in basic interviewing procedures and in the concepts and procedures unique to the NHIS. Depending on the family size and the nature and extent of health conditions of family members, the length of interview ranges between 20 and 90 minutes. On the average, the interviews require about 50 minutes in the household. Content of the questionnaire: The questionnaire consists of two basic parts: (1) a set of basic health and demographic items, and (2) one or more sets of questions on current health topics. The basic items constitute approximately 50 percent of the questionnaire and are repeated each year. These items provide continuous information on basic health variables. Questions on current health topics facilitate a response to changing needs for data and coverage of a wide variety of issues. This combination yields a unique national health data base. The questionnaire includes the following types of basic health and demographic questions: e Demographic characteristics of household members, including age, sex, race, education, and family income. e Disability days, including restricted-activity and bed-disability days, and work- and school-loss days occurring during the 2-week period prior to the week of interview, as well as 12-month bed days. e Physician visits occurring during the same 2-week period, interval since the last physician visit, and the number of visits in the last 12 months. ® Acute and chronic conditions responsible for these days and visits. I-42 e Long-term limitation of activity resulting from chronic disease or impairment and the chronic conditions associated with the disability. eo Short-stay hospitalization data, including the number of hospital episodes during the past year and the number of days for each stay. In addition, each of six representative subsamples is asked to respond to questions about one of six lists of selected chronic conditions. Questions on special health topics change in response to current interest and need for data. The 1983 ques- tionnaire contained questions on alcohol, dental care, physician services, and health insurance. The 1984 current health topic questionnaire was devoted entirely to issues of aging, 1985 covered health promotion and disease prevention, and 1986 included questions on health insurance, dental health, vitamin and mineral intake, longest job worked, and functional limitations. The 1987 NHIS included an extensive questionnaire on cancer risk factors and questions on child adoption. Suggestions and requests for special health topics are solicited and received from many sources. These include university-based researchers, administrators of national organizations and programs in the private and public health sectors, and other parts of the U.S. Department of Health and Human Services (for example, the National Institutes of Health and the Centers for Disease Control). Topics are selected after consultation with Agencies within the Public Health Service and after an assessment of priority health issues and the related need for population-based data. A lead time of at least 18 months is required to develop and pretest questions for new topics. Reference: National Center for Health Statistics, C. A. Schoenborn. 1988. Health Promotion and Disease Prevention, United States, 1985. Vital and Health Statistics. Series 10, No. 163. DHHS Pub. No. (PHS) 88-1591. Public Health Service. Washington: U.S. Government Printing Office. I-43 Food Label and Package Survey Sponsoring Agency: Food and Drug Administration, DHHS Purpose of survey: The survey is conducted to monitor labeling practices of U.S. food manufacturers. Excluded from the survey are fresh fruit and vegetables and USDA regulated meat products. The survey also includes a surveillance program to identify levels of accuracy of selected nutrient declarations compared with values obtained from nutrient analysis of products. Last survey conducted: 1986. (Previous surveys conducted in 1978, 1980, 1982, 1983, and 1984.) Design: eo Biennial probability survey of retail packaged foods using commercial market research data bases (A. C. Nielsen Co.). The survey involves about 1,500 individual food brands representing about 44 percent of the packaged food supply in retail dollar terms. Label observations are interpreted on a share-of-the-market sales basis. e Biennial analysis of a representative sample of the 55 percent of packaged foods which bear nutrition labels. Approximately 300 foods are analyzed for an average of eight nutrients. Measures: Prevalence of nutrition labeling in general as well as declaration of selected nutrients/ingredients (for example, cholesterol and sodium content, fats and oils, food additives). Sampling methodology: Represented in the 1986 survey were 143 packaged food classes chosen to account for a minimum of 75 percent of the annual retail sales of each of the 51 major supermarket food groups regulated by FDA. In addition, another 40 product classes were selected at random without regard to sales importance, bringing the total number of product classes included in the survey to 183. Excluded from the sample were fresh fruit and vegetables and fresh meat. For each product class selected for the survey, the three leading brands (on the basis of share of market) were predesignated for inclusion in the product sample, using sales information contained in the A. C. Nielsen Com- pany syndicated data base NEIS. An additional three brands (depending on total number of brands available) were sampled at random from each product class. A third set of three brands per category was selected at random and designated as substitutable replacements for selected products not available after reasonable search at retail. In total, 948 preselected brands were included in the sample. In addition to prior specified brands, convenience samples of off-premises—baked bread, milk and ice cream were picked up in three major markets, one in the East, one in the West, and one in the Midwest. Field agents visited one chain and two independent stores in each market and purchased all brands available in each store. A total of 541 unduplicated products were obtained in this manner. Field agents of the A. C. Nielsen Company purchased sample products at retail between February and July 1986. I-44 Total Diet Study Sponsoring Agency: Food and Drug Administration, DHHS Purpose: The Total Diet Study (TDS) is conducted yearly to assess the levels of various nutritional elements and organic and elemental contaminants in the U.S. food supply and in representative diets of specific age-sex groups, and to monitor trends in the levels and consumption of these substances over time. The Selected Minerals in Foods Survey is the component of the TDS that estimates levels of 11 essential minerals in repre- sentative diets of specific age—sex groups. Last survey conducted: 1987. (The TDS has been conducted annually since 1961. Collection of nutritional elements began in 1974.) Target population: Eight age—sex groups: infants, young children, male and female teenagers, male and female adults, and male and female older persons. (Between 1975 and 1982, diets were collected for 3 groups: 6-month-old infants, 2-year-old toddlers, and teenaged males. Prior to 1975, only the teenaged male diet was used.) Design: Two hundred and thirty-four foods are collected from retail markets in urban areas, prepared for consumption, and analyzed for nutritional elements and contaminants 4 times each year. The representative diets of specific age—sex groups which are used to estimate intake levels of nutrients and contaminants are based on the food consumption patterns indicated by NFCS 1977-78 and NHANES II. (Prior to 1982, food lists and diets were based on the 1965 and 1955 USDA Household Food Consumption Surveys. Also prior to 1982, foods were prepared for consumption and composited into 11 or 12 food groups before analysis for nutrients and contaminants.) Sample size/response rate: Not applicable. Measures: Assessment of food composition and dietary intake. List of variables: Variables are listed only for the Selected Minerals in Foods Survey component of the TDS. Calcium Phosphorus Magnesium Sodium Potassium Iron Zinc Copper Manganese Selenium Iodine References: Food and Drug Administration, J. A. T. Pennington and E. L. Gunderson. 1986. History of the Food and Drug Administration's Total Diet Study--1961 to 1987. J. Assoc. Off. Anal. Chem. 70:772-782. Pennington, J. A. T. 1983. Revision of the Total Diet Study Food List and Diets. J. Am. Diet Assoc. 82:166- 173. I-45 Vitamin/Mineral Supplement Intake Survey Sponsoring Agency: Food and Drug Administration, DHHS Purpose of survey: The survey was conducted to quantitatively assess nutrient intake from vitamin/mineral supplements in the United States and to examine characteristics of supplement users by supplement intake patterns. Last survey conducted: 1980. Target population: Civilian noninstitutionalized adults, aged 16 years and over. Design: Telephone interviews with a national probability age-stratified sample selected by a random digit dialing method. One person 16 years or older from each household contacted was randomly selected to partici- pate in the survey. Design limitation: Sample excludes households without telephones. Sample size/response rate: Residential telephone Number screened for sample size vitamin/mineral supplement use 7,986 6,409 (80%) Number vitamin/mineral supplement users Interview interviewed completion rate 2,991 (47%) (95%) Measures: Assessment of supplement intake, attitudes, and behaviors among supplement users. Sample selection and subjects: Between August and December 1980, a random digit dialing method was used for telephone interviews of a national probability sample of 2,991 individuals. Up to four attempts were made for each telephone number at different times of the day and evening. On contacting a residence, the inter- viewer explained the purpose of the survey, stated that it was being done under the auspices of the FDA, and asked the age and the sex of all members of the household 16 years old and older. One person 16 years old or older was then randomly selected to participate in the survey. Respondents were selected until 1,000 indi- viduals in each age group (16-24 years, 25-64 years, and 65 years and older) had been interviewed. No control over gender of respondent was attempted. A total of 12,577 telephone numbers was dialed; 7,986 resulted in contacts with residences. Of the contacted individuals, 6,409 (80.3 percent) agreed to participate in the screening questionnaire. Among the 2,991 individuals who met the sampling requirements, the interview completion rate was 95 percent. Estimates of general supplement usage are based on 2,866 (96 percent) of the 2,991 respondents; 77 otherwise acceptable respondents were excluded because of special supplementation needs owing to pregnancy/lactation, and 48 were excluded because the interview ended before the questions on vitamin/mineral supplement use were asked. Incomplete interviews that did determine general supplement use were included. The more specific analyses (for example, type of supplement and amount of nutrient consumed) are based on 2,751 respondents (92 percent); the 115 additional respondents were excluded because of later interview termination or lack of quantitative information regarding supplement content. The further respondent elimination was necessary because the more definitive tabulations required specific information about the supplement type and its complete nutrient profile. The 2,751 respondents who provided full information included 1,012 men (16-24 years = 433, 25-64 years = 315, and 65 years and older = 264) and 1,739 women (16-24 years = 502, 25-64 years = 604, and 65 years and older = 633). Data collection: The screening and survey questionnaires were pretested to ensure respondent comprehension. The survey questionnaire determined whether the selected respondent was currently taking a vitamin, mineral, I-46 or other type of nutritional supplement (excluding fortified, ready-to-eat cereals). Respondents answering "yes" were asked a series of questions to establish the brand name, manufacturer, and retail source of each supplement; the form of the product; frequency of use per week; quantity per day; and other information related to the source of influence in the purchase of the product. Each respondent was asked to bring all supplement bottles to the telephone and to read the potency of individual nutrients from the labels. Respon- dents unable or unwilling to do that were requested to complete and return a mail questionnaire for each supplement. In addition, respondents were asked for demographic information (for example, income, education, race). State of residence was determined from area code. Interviewers were dietitians trained in telephone interview techniques. The average interview length was approximately 19 minutes for vitamin/mineral supplement users and 11 minutes for all others. Data analyses: "Supplement" denotes the product purchased by the respondent (for example, multiple vitamins); "nutrient" refers to each of the constituents of a product (for example, vitamin A, thiamin). Various analyses were conducted: 1. Examination of bias owing to noncooperation (terminated interviews or incomplete quantitative information for nutrients) using the Mann-Whitney U Test for each of the six groups formed by the factorial combina- tion of two levels of sex (male, female) and three age levels (16-24 years, 25-64 years, and 65 years and over). 2. Evaluation of sex—age group differences in prevalence of use for each of nine mutually exclusive supplement types with the chi-square test used for the 6—cell contingency table derived from the factorial combination of sex and age levels: single vitamins/miscellaneous dietary components, single minerals, vitamin combina- tions, mineral combinations, vitamin/mineral combinations, multivitamins, multiminerals, multivita- min/multimineral combinations, and multivitamins plus iron. 3. Evaluation of differences in prevalence of use by census region, education, income, and race, with the chi- square test used as in No. 2. 4. Evaluation of sex—age group differences in prevalence of use for individual nutrients, with the chi-square test used as in No. 2. 5. Calculation of statistics on central tendency and variability of quantitative intake for selected nutrients. To make possible reporting of combined sex—age group estimates in Nos. 2 through 5, the data were weighted according to the 1979 population census (U.S. Bureau of the Census, 1980). Noncooperative bias: Of those respondents identified as vitamin/mineral supplement users, 10.6 percent were classified as noncooperators (that is, the interview was prematurely terminated or the respondent did not sufficiently identify supplements for quantitative analysis). Only 2 percent of nonusers were noncooperators. The number of supplements consumed was significantly higher (p<.05) for cooperating than for noncooperating supplement users for two of the six sex—age groups (males 25-64 years and females 65 years and older). Mean daily intake among cooperating supplement users was 2.3 and 2.1 supplements for males 25-64 years and females 65 years and older, respectively; mean daily intake among noncooperating supplement users was 1.5 and 1.4 supplements, respectively. Therefore, estimates of number of supplements taken may slightly over- estimate true values for those two sex—age groups. References: Levy, A. S,, and R. E. Schucker. 1987. Patterns of Nutrient Intake Among Dietary Supplement Users: Attitudinal and Behavioral Correlates. J. Am. Diet. Assoc. 87:754—-760. Stewart, M. L., J. T. McDonald, R. E. Schucker, and D. P. Henderson. 1985. Vitamin/Mineral Supplement Use. A Telephone Survey of Adults in the United States. J. Am. Diet. Assoc. 85:1585-1590. U.S. Bureau of the Census. 1980. Statistical Abstract of the United States, 1980. Washington: U.S. Department of Commerce. 1-47 Health and Diet Survey Sponsoring Agency: Food and Drug Administration (Cosponsored by National Institutes of Health/National Heart, Lung and Blood Institute), DHHS Purpose of survey: The survey is conducted to assess public knowledge, attitudes, and practices about food and nutrition, particularly as they relate to such health problems as hypertension, hypercholesterolemia, coronary heart disease, and cancer. The survey also assesses the public's use of information on food labels including the use of ingredient lists to avoid or limit food substances. Last survey conducted: 1986. (Previous surveys conducted in 1982 and 1984.) Target population: Civilian noninstitutionalized adults aged 18 years and over. Design: Telephone interviews with a national probability Waksberg sample selected by a random-digit dialing method. One adult from each household contacted was randomly selected to participate in the survey. Design limitation: Sample excludes households without telephones. Sample size/response rate: 4,000 (comprising 4 replicate samples of 1,000 each); response rates are 70 to 75 percent. Measures: Awareness (perceptions), attitudes (concerns), knowledge, and behaviors regarding food and nutri- tion; height, weight, and household health status and history (as reported by household members). eo List of Variables: e Awareness (perceptions) causes of coronary heart disease causes of cancer causes of breast cancer causes of colorectal cancer causes of lung cancer diet links with coronary heart disease diet links with hypertension diet links with cancer (beneficial and detrimental) diseases related to sodium diseases related to fat diseases related to fiber diseases related to calcium e Concerns (attitudes) sodium intake blood cholesterol levels e Knowledge dietary control of coronary heart disease cause(s) of hypercholesterolemia control of hypercholesterolemia outcome(s) of hypercholesterolemia early detection of breast, colorectal, and cervical cancer fatty acids and cholesterol fiber I-48 eo Behavior (for some variables, information is collected on households, as well as individuals overall dietary change to reduce cancer or coronary heart disease barriers to dietary change weight loss diets sodium-modified diets blood cholgsterol reducing diets tobaccS Hag 5 ENG exercise food label use purchase of sodium-modified products eo Health status (as rted by household member. Information is collected on households as well as individuals) blood cholesterol checks/level hypercholesterolemia hypertension heart disease stroke cancer I-49 Pediatric Nutrition Surveillance System Sponsoring Agency: Division of Nutrition, Centers for Disease Control, DHHS Conducted: Continuous data collection. Target population: Low-income, high-risk children 0-17 years, especially those 0-5 years. Design: Simple, key indicators of nutritional status are continuously monitored using readily available clinic data from participating States. The data are collected on a convenience population of low-income children who participate in publicly funded health, nutrition, and food assistance programs. These data are rapidly analyzed, with reports available to States monthly, quarterly, and annually. Sample size/response rate: Sample size and response rates do not apply in the Pediatric Nutrition Surveillance System (PedNSS). The coverage of PedNSS reflects the count of clinic visits in participating programs and can be separated by initial and followup visits. Over 2.4 million records from 36 States including the District of Columbia and Puerto Rico were submitted for analysis during FY 1987. Measures: Anthropometry (height, weight), birthweight (below 2,500 grams), hematology (hemoglobin, hemat- ocrit, erythrocyte protoporphyrin) are measured. Description: The PedNSS is designed to monitor continuously the status of major nutrition problems among high-risk infants and children 0-17 years of age. The system is based on information collected routinely in health, nutrition, and food assistance programs such as the Special Supplemental Food Program for Women, Infants, and Children (WIC), Early and Periodic Screening, Diagnosis and Treatment (EPSDT) and Maternal and Child Health (MCH). The data consist prima- rily of measurements such as height, weight, birthweight, and hemoglobin and/or hematocrit. This information is rapidly analyzed and comparisons are made with reference population data. In this way, the nutritional status of the surveillance population can be characterized by the prevalence of short stature (low height for age), underweight (low weight for height), overweight (high weight for height), anemia (low hematocrit, low hemoglobin), and low birthweight. Trends in the prevalence of these nutrition status indicators are also monitored. The linkage of Pregnancy Nutrition Surveillance System (PNSS) data to birth certificates has been explored in one State and was found to be a potentially useful method for assessing program coverage and targeting and the evaluation of program impact. The PedNSS reports are returned to States monthly, quarterly, and annually for use in patient followup, program planning, management, and the evaluation of health resource allocation at the State and local level. Prevalent nutrition-related problems are identified by age and ethnic groups for the targeting of high-risk groups for intervention. The data are available by individual clinic, county, region, or the State as a whole, thus, various State needs are met for geographic—specific nutrition surveillance data. States may also incorpo- rate the CDC PedNSS software into their own computer system to independently produce these reports. Tech- nical assistance, consultation, and training are available for States from CDC for the collection, processing, analysis, interpretation, and application of PedNSS data. Demographic: Nutrition risk indicators: e Basic identification eo Height e Date of birth eo Weight e Ethnic group e Hemoglobin eo Sex eo Hematocrit eo Protoporphyrin eo Birthweight I-50 Pregnancy Nutrition Surveillance System Sponsoring Agency: Division of Nutrition, Centers for Disease Control, DHHS Conducted: Continuous data collection. Target population: Low-income, high-risk pregnant women. Design: Simple, key indicators of pregnancy nutritional status, behavioral risk factors, and birth outcome are monitored using readily available clinic data. The data are collected on a convenience population of low income women who participate in publicly funded health, nutrition, and food assistance programs. These data are analyzed and made available to participating States quarterly and annually. The Pregnancy Nutrition Surveillance System will be enhanced to collect additional behavioral risk factor data which relate to low birthweight during FY 1988. Cooperative agreement funding was awarded to nine States during September 1987 to implement this enhanced PNSS. Sample size/response rate: Sample size and response rates do not apply in PNSS. The coverage of PNSS reflects the number of pregnant women who participate in the programs contributing to the surveillance system. During FY 1986, approximately 70,000 records were received from 12 States including the District of Columbia. Including the nine cooperative agreement States, 18 or more States will be participating in FY 1988. Measures: Pregravid weight status, anemia (hemoglobin, hematocrit), pregnancy behavioral risk factors (smoking), low birthweight (<2,500 grams) and other indicators are currently monitored. With enhancement of the PNSS, additional behavioral risk factor data (quantity and frequency of smoking, drinking alcohol, other) will be collected. Description: The PNSS is designed to monitor the prevalence of nutrition-related problems and behavioral risk factors among high-risk prenatal populations. The PNSS is based on data collected from health, nutrition, and food assistance programs for pregnant women such as the Special Supplemental Food Program For Women, Infants, and Children (WIC) and prenatal clinics, including such programs as the Maternal and Infant Care Project (MIC) and Improved Pregnancy Outcome Project (IPO). Nutrition-related problems currently monitored include pregravid underweight or overweight and anemia (low hemoglobin, low hematocrit). The primary behavioral risk factor currently monitored is the rate of smoking among pregnant women and its association to low birthweight. Low birthweight is also exam- ined relative to pregravid weight and smoking status as well as to weight gain during pregnancy. With the enhancement to the PNSS, additional behavioral risk factor data will be collected such as the quantity and frequency of smoking and drinking alcohol. The emphasis is to quantify prevalent preventable nutri- tion-related problems and behavioral risk factors among low-income pregnant women for targeting low birth- weight intervention efforts. Trends in the prevalence of these nutrition status indicators and behavioral risk factors can also be monitored. The future linkage of PNSS data to birth certificates may also allow for the assessment of program coverage and targeting and the evaluation of program impact. The PNSS reports are returned to States quarterly and annually for use in program planning, management, and the evaluation of health resource allocation at the State and local level. Prevalent nutrition-related and behavioral risk factor problems are identified by age and ethnic groups for targeting high-risk women for intervention. The PNSS data can be available by geographic sub-areas (for example, region, county); however, routine reports summarize the data Statewide. In the future, CDC will develop exportable software so that States will be able to produce their PNSS reports independently. Technical assistance, consultation, and training are available to States for the collection, processing, analysis, interpretation, and application of PNSS data. I-51 Behavioral Risk Factors Surveillance System Sponsoring Agency: Division of Nutrition, Centers for Disease Control, DHHS Conducted: Continuous data collection. Target population: Adults (=18 years) residing in participating States (35 in 1987) in households with tele- phone. Design: The Behavioral Risk Factors Surveillance System (BRFSS) uses a multistage cluster telephone survey design based on the Waksberg method. Respondents are selected randomly from adult civilian residents with telephones. After a household is contacted, an adult aged 18 years or older is randomly selected from among the adults residing in the household and interviewed. If the adult selected is not available, the interview is done during a followup telephone call. To improve efficiency in contacting eligible respondents, the interviews are conducted primarily weekday evenings, but also during the day and on weekends. Beginning in 1985, most States began using computer assisted telephone interviewing (CATI) to facilitate the interview, data coding and entry, and quality control procedures. Sample size/response rate: Average State Number of Total Average Year(s) sample size States sample size response rate 1981-83 797 29 23,113 86% 1984 675 17 11,480 83% 1985 1,174 22 25,830 83% 1986 1,182 26 30,730 86% Measures: Height, weight, dieting practices, table salt use, cigarette smoking, smokeless tobacco use, alcohol use, and cholesterol screening practices, awareness, and treatment. Description: The State-based behavioral risk factor surveillance system uses standard telephone survey methods and questionnaires to assess the prevalence of personal health practices. These behaviors are related to the leading causes of death and include cigarette smoking, alcohol use, drinking and driving and seatbelt use, physical activity, weight control, high blood pressure, and preventive health practices, such as cholesterol and breast cancer screening. Between 1981 and 1983, the Centers for Disease Control (CDC) collaborated with 29 State health departments (including the District of Columbia) to conduct one-time random-digit dialed tele- phone surveys of adults. Beginning in 1984, most States began collecting data continuously throughout the year, completing approximately 100 interviews per month (range 50-250), with an average of 1,200 completed interviews per year (range 600-3,000). In 1987, 35 States were collecting data on an ongoing basis——bringing to 46 the number of States which have participated in behavioral risk factor surveillance since 1981. This system has proven to be flexible--by providing data on emerging public health problems such as smokeless tobacco use and AIDS, timely--by providing data within months of the completion of data collection, and affordable--by operating at a fraction of the cost of comparable Statewide in-person surveys. Behavioral risk factor surveil- lance has been used by State health departments to plan, initiate, and guide Statewide health promotion and disease prevention programs, and to monitor their progress over time. Background: During the 1960s and 1970s, the role of personal behaviors——such as smoking, alcohol abuse, obesity, and physical inactivity—-as risk factors for disease became recognized. Accordingly, many State health departments launched health education and risk reduction programs directed towards reducing the prevalence of these "behavioral risk factors" in the population. However, data on which to plan or guide these efforts were either unavailable, or obtained by conducting household surveys or by using synthetic estimates based on the prevalence of behaviors at the national level. Because of the expense of conducting household surveys and the uncertainty of applying national prevalence estimates at the State level, an alternate method was sought. I-52 By 1980, telephone surveys had emerged as both a reliable and affordable alternative method for determining the prevalence of behavioral risk factors in the population. Accordingly, the CDC began working with State health departments to develop a system for the surveillance of behavioral risk factors in the population using random-digit dialed telephone techniques. The goal of the system was to collect, analyze, and interpret behav- ioral risk factor data, in order to plan, implement, and monitor public health programs. Between April 1981 and October 1983, random-digit dialed telephone surveys were conducted in 29 States (including the District of Columbia) over a 1-6 week period by State health department personnel. These sur— veys were supported, in part, through funds provided in the Health Education and Risk Reduction Grants and through training, coordination, and standard methods provided by the CDC. Beginning in 1984, the surveys were conducted during a 1-week period every month continuously throughout the year. The CDC continued to provide training, coordination, and standard methods but, in addition, provided funds directly to the participating State health departments through cooperative agreements. Several State health departments have conducted the surveys using the standard telephone survey methods and questionnaires, but without CDC funds. Since 1981, 43 States have participated with the CDC in the BRFSS. In the BRFSS, respondents are selected randomly from adult civilian residents with telephones. The telephone number is selected using a multistage cluster design procedure based on the Waksberg method. After a house- hold is contacted, an adult aged 18 years or older is randomly selected from among the adults residing in the household and interviewed. If the adult selected is not available, the interview is done during a followup telephone call. To improve efficiency in contacting eligible respondents, the interviews are conducted primarily weekday evenings, but also during the day and on weekends. Beginning in 1985, most States began using com- puter assisted telephone interviewing (CATI) to facilitate the interview, data coding and entry, and quality control procedures. The questionnaire used in the BRFSS has two components. One is "core" of questions developed jointly by the CDC and the participating States and asked by all States. For comparability, the questions in the core are selected from national surveys, such as the National Health and Nutrition Examination Surveys and the National Health Interview Surveys. States which have interest in subject areas not covered by the core questionnaire add specific questions. These "modules" are added at the end of the questionnaire in order to maintain the comparability of the core between States and over time. Upon completing the interviewing cycle each month, the data are keyed and sent to CDC for editing. After editing, the data are weighted to provide representative population-based estimates of risk factor prevalence (accounting for telephone noncoverage, nonresponse, refusals, and the cluster survey design). The weighted and unweighted tabulations are provided to the States within two months of completion of the last December interview. In addition, the CDC publishes the annual summary and selected risk—factor specific reports in the Morbidity and Mortality Weekly Report throughout the year. The information gathered under the BRFSS is expected to be of use to State health departments to support risk reduction and disease prevention activities. Because comparable methods are used from State to State and from year to year, States can compare risk factor prevalence with other States and monitor the effects of interven- tions over time. Also, the use of consistent methods in a large group of States permits the assessment of geographic patterns of risk factor prevalence. These telephone survey techniques can also be applied at the community level to guide local efforts in reducing risk factor prevalence. Taken together, the behavioral risk factor survey and surveillance data provide a new resource to guide State and local disease prevention efforts. 1-53 Appendix II Summary Data on Food Components Explanatory Notes Table Notes for Data from the Continuing Survey of Food Intakes by Individuals (CSFII) 1985-86, 4 Nonconsecutive Days N represents number of individuals who provided at least 4 days of dietary data. The number of children excludes 2 breastfed children. The mean nutrient intake estimates for women were adjusted to reflect a specified proportion of individuals in the ten—year age groups 20-29, 30-39, and 40-49. The adjustment was done to the means for all 20- to 49-year-old women and the means for each category of race, poverty status, education, region, and urban- ization. The standard errors were computed for these adjusted means. The specified proportions were derived from the 1980 Census counts for all races, both sexes in the United States population. Thus, both males and females were adjusted to the same standard proportions. These counts are: 1980 Census Percent of Age count 20- to 49-year—olds 20-29 40,839,623 42.93 30-39 31,526,222 33.14 40-49 22,759,163 23.93 20-49 95,125,008 100.00 The sample was designed to be self-weighting. That is, the proportion of eligible households in the population with a particular characteristic was designed to represent the same proportion of eligible households in the population. However, adjustments to the sample were required because not all eligible households agreed to participate, not all eligible women and children in eligible households agreed to participate, not all interviews yielded complete information, and not all participants in wave I completed each subsequent wave. Weighting factors were applied to data from completed intake records to adjust for these sources of nonresponse. For more information on how the weighting factors were derived, see NFCS, CSFII Report No. 85-4. Estimated coefficients of variation of the mean were less than 20 percent for all cells for the women and most cells for the children. Exceptions are shown with an asterisk and should be interpreted with caution. The following minimum cell sizes were used in presentation of various percentile values: 5th & 95th percentiles: N = 140 10th & 90th percentiles: N = 80 25th & 75th percentiles: N = 40 Where cell sizes are below this minimum, the fields are left blank. Individuals are classified into age and poverty status based on the responses provided on the first day of the survey. 11-1 e Education status and race of a child are that of the mother/caretaker (that is, the mother or guara.an of a child respondent or the person most responsible for that child). For race, poverty status, and education, the number of persons does not add to the number of all children because some individuals had missing values. eo For race, poverty status, and education, the number of women does not add to the number of all women because some individuals had missing values. e The nutrient intakes represent the nutrient content of all foods and beverages (except water) reported to be ingested by the respondent and do not include intakes from vitamin and mineral supplements, for which information on only the frequency and type used was collected. Sodium intake does not include sodium from salt added at the table, for which information was not collected. e To estimate the mean nutrient intakes for each dietary component, the intakes calculated for each individual over the 4 days of observation were totaled and divided by 4 to obtain a mean intake per day for the individual. Individual mean intakes were then totaled and divided by the number of individuals in the group to obtain the mean intake per individual per day for that group. e To estimate nutrient intakes at selected percentiles for each dietary component, the intakes calculated for each individual over the 4 days of observation were totaled and divided by 4 to obtain a mean intake per day for the individual. Values for each individual were then arrayed from lowest to highest, and intakes were identified at specified population (weighted) percentiles (5th, 10th, 25th, 50th, 75th, 90th, 95th). eo To estimate the percent of calories from various components, each individual's intakes of protein, total fat, saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, and carbohydrate were summed over 4 days. Intakes of protein were multiplied by 4 kilocalories per gram; fat, by 9 kilocalories per gram; and carbohydrate, by 4 kilocalories per gram. Those values were divided by the sum of the individual's energy from reported food intakes over 4 days then multiplied by 100 to obtain the percentage of an individual's total food energy intake provided by each nutrient. Individual percentages were totaled and divided by the number of individuals in the group to obtain the mean percentage per individual for that group. Alcohol is also an energy source and was included in determining total energy, but the percentage of food energy contributed by alcohol was not calculated. e To estimate the nutrient densities at selected percentiles for each dietary component, nutrient density values for each individual were arrayed from lowest to highest, and intakes were identified at specified population (weighted) percentiles (5th, 10th, 25th, 50th, 75th, 90th, 95th). Definitions: Age--Calculated from date of birth as reported by the household informant. Carotenes——Beta—carotene and other provitamin A carotenoids (see Vitamin A). Dietary fiber—-Total dietary fiber including both the insoluble fraction (neutral detergent fiber) and the soluble fraction (for example, gums in cereal grains and pectin in fruits and vegetables). Educational level--Adult respondents were categorized according to the highest grade of formal schooling they completed. High school refers to 4 years or high school equivalency. Formal schooling does not include trade or vocational schooling or company training unless credit is given which would be accepted at a regular school or college. Folacin—--Total folate activity. Niacin--Nicotinic acid and nicotinamide present in foods. Does not include niacin converted from dietary tryptophan, a niacin precursor. Poverty status--Tables presenting results by income level use household income for the previous calendar year expressed as a percentage of the Federal Poverty Income guidelines. Each household's income before taxes was expressed as a proportion of the poverty guideline for households of the appropriate size. Individuals were then grouped according to their household income as a percentage of the poverty 11-2 guidelines. The poverty guidelines, provided by the U.S. Department of Health and Human Services, are adapted from the poverty thresholds published by the Bureau of the Census. They are used by many Federal Agencies to determine whether a person or family is financially eligible for assistance under a particular Federal program. The guidelines (which are based on the previous calendar year's income) are as follows: Household Poverty guidelines _ size 1985 1986 1 $ 5,250 $ 5,360 2 7,050 7,240 3 8,850 9,120 4 10,650 11,000 5 12,450 12,880 6 14,250 14,760 7 16,050 16,640 8 17,850 18,520 For households with more than eight members, $1,800 was added for each additional member in 1985 and $1,880 was added for each additional member in 1986. Several major Federal food assistance programs use 130 percent of poverty to determine eligibility for benefits. These include the Food Stamp Program, the National School Lunch Program, the School Breakfast Program, the Child Care Food Program, the Special Milk Program, the Temporary Emergency Feeding Program, and the Food Distribution Program to Indian Reservations. Race--Race of women was self-reported as white, black, Asian/Pacific Islander, or Aleut/Eskimo/American Indian. Children were assigned the race of their mother/caretaker (that is, the mother or guardian of a child respondent or the person most responsible for taking care of the child). Region-—An area of the conterminous United States as defined by the U.S. Department of Commerce for the 1980 Census of Population. The four census regions and their States are as follows: Northeast: Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont. Midwest: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin. South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia. West: Arizona, California, Colorado, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming. Urbanization--Based on metropolitan statistical areas (MSA) defined by the U.S. Department of Commerce for the 1980 Census of Population. The degrees of urbanization used in this report are as follows: Central city: A city which has a population of 50,000 or more and is the main city within an MSA. Suburban area: Generally within the boundaries of an MSA but not within the legal limits of the central city; includes some farms and rural areas. Nonmetropolitan area: Any area not within an MSA; includes some towns and urban areas. 11-3 Vitamin A--Vitamin A activity derived from both preformed vitamin A (retinol) and provitamin A carotenoids. Values in tables are expressed as retinol equivalents (RE). One RE equals 1 microgram of retinol, 6 micrograms of beta-carotene, or 12 micrograms of other provitamin A carotenoids. Vitamin E--Vitamin E activity derived from alpha-, beta—, and gamma-tocopherol and alpha-tocotrienol. Values in tables are expressed as alpha-tocopherol equivalents. One alpha-tocopherol equivalent equals 1 milligram of alpha-tocopherol, 2 milligrams of beta-tocopherol, 10 milligrams of gamma-tocopherol, or 3.3 milligrams of alpha-tocotrienol. Table Notes for Data from the Nationwide Food Consumption Survey (NFCS) 1977-78 and the Continuing Survey of Food Intakes by Individuals (CSFII) 1985-86, 1 Day e Number of respondents by sex and age NFCS CSFII Sex and age (years) 1977-78 1985-86 Males and females 1-2 1,113 363 3-5 1,838 633 6-11 4,107 rete Males 12-15 1,613 —-—= 16-19 1,284 ——= 20-29 1,752 222 30-39 1,338 249 40-49 1,082 164 50-59 1,166 me 60-69 941 ——= 70+ 684 a Females 12-15 1,596 == 16-19 1,397 =m 20-29 2,282 1,000 30-39 1,901 1,049 40-49 1,496 735 50-59 1,698 pi 60-69 1,412 —-—= 70+ 1,192 — e Data for children aged 1-2 years exclude 12 breastfed infants in 1977 and 2 breastfed infants in 1985-86. e Data for NFCS 1977-78 were obtained throughout all seasons of the year; data for CSFII 1985-86 were obtained in the spring of each year. e Data for men from CSFII were collected in 1985 only. e Data for males in CSFII 1985 include men from both an all-income sample and a low-income sample. The all-income and the low-income samples were drawn independently. The data from both samples were merged and additional weights were computed for the purpose of treating the two samples as one combined sample. The weights were designed to keep the distribution of low-income households across strata in the combined sample the same as that of low-income households in the all-income sample and to keep the proportion of non-low-income households within strata in the combined sample to be the same as that of non-low-income households in the all-income sample. II-4 eo The NFCS 1977-78 and CSFII 1985-86 samples were designed to be self-weighting. That is, strata, primary sampling units, area segments, and housing units were selected to be representative of the population. Thus, the number of eligible households in each of these divisions or cells in the sample was designed to represent the same proportion as the respective number of households in each cell in the population. However, adjustments to the sample were required because not all eligible households agreed to participate, not all eligible individuals in eligible households agreed to participate, and not all interviews yielded complete information. Weighting factors were applied to data from completed intake records to adjust for these sources of nonresponse. For more information on how the weighting factors were derived, see NFCS, CSFII Report Nos. 85-1 and 85-3 and NFCS 1977-78 Report No. I-2. e Dashes indicate data are not available. In NFCS 1977-78 food composition data were not available for fatty acids, cholesterol, dietary fiber, vitamin A in retinol equivalents, folacin, vitamin E, zinc, sodium, copper, or potassium. In CSFII 1985-86, age groups other than 1- to 5-year-olds and 19- to 50-year-olds were not sampled. eo Estimated coefficients of variation of the mean were less that 20 percent for most cells. Exceptions are shown with an asterisk and should be interpreted with caution. e The nutrient intakes represent the nutrient content of all foods and beverages (except water) reported to be ingested by the respondent and do not include intakes from vitamin and mineral supplements, for which information on only the frequency and type used was collected. Sodium intake does not include sodium from salt added at the table, for which information was not collected. eo To estimate the percent of calories from various components, intakes of protein were multiplied by 4 kilocalories per gram; fat, by 9 kilocalories per gram; and carbohydrate, by 4 kilocalories per gram. Those values were divided by the individual's food energy intake then multiplied by 100 to obtain the percentage of an individual's total food energy intake provided by each nutrient. Individual percentages were totaled and divided by the number of individuals in the group to obtain the mean percentage per individual for that group. Alcohol is also an energy source and was included in determining total energy, but the percentage of food energy contributed by alcohol was not calculated. Definitions: Carotenes——Beta—carotene and other provitamin A carotenoids (see Vitamin A). Dietary fiber--Total dietary fiber including both the insoluble fraction (neutral detergent fiber) and the soluble fraction (for example, gums in cereal grains and pectin in fruits and vegetables). Folacin——Total folate activity. Niacin—-Nicotinic acid and nicotinamide present in foods. Does not include niacin converted from dietary tryptophan, a niacin precursor. Vitamin A--Vitamin A activity derived from both preformed vitamin A (retinol) and provitamin A carotenoids. Values in tables are expressed as international units (IU) and as retinol equivalents (RE). One IU equals 0.3 micrograms of retinol, 0.6 micrograms of beta-carotene, or 1.2 micrograms of other carotenoids having vitamin A activity. One RE equals 1 microgram of retinol, 6 micrograms of beta- carotene, or 12 micrograms of other provitamin A carotenoids. Vitamin E--Vitamin E activity derived from alpha-, beta—, and gamma-tocopherol and alpha-tocotrienol. Values in tables are expressed as alpha-tocopherol equivalents. One alpha-tocopherol equivalent equals 1 milligram of alpha-tocopherol, 2 milligrams of beta-tocopherol, 10 milligrams of gamma-tocopherol, or 3.3 milligrams of alpha-tocotrienol. II-5 Table Notes for Data from the Second National Health and Nutrition Examination Survey (NHANES II) and the Hispanic Health and Nutrition Examination Survey (HHANES) eo The number of examined persons is given in the tables; for the number of persons represented in the population, see tables in appendix I. e Sample weights, which incorporate the selection probabilities, a nonresponse adjustment, and poststratifica— tion, were used to produce the population estimates included in the tables.. e If the sample size is less than 25, the mean or percent (prevalence estimate) is not presented and an asterisk is placed in the cell. If the sample size is 25-44, the mean or percent is presented but with an asterisk. If the sample size is 45 or more, the estimated mean or percent is presented without caveat. e Age-adjustment was calculated by the direct method using the 1980 census population 20-74 years of age for both sexes and all races. e Body mass index (BMI) is the ratio of body weight in kilograms to height in meters squared. eo Non-Hispanic whites were persons whose race was observed and recorded as "white" and whose self- reported family ancestry or national origin was not Chicano, Mexicano, Mexican American, other Spanish, countries of Central or South America, Puerto Rican, or Cuban. e Non-Hispanic blacks were persons whose race was observed and recorded as "black" and whose self-reported family ancestry or national origin was not Chicano, Mexicano, Mexican American, other Spanish, countries of Central or South America, Puerto Rican, or Cuban. e Mexican American includes persons residing in Southwest primary sampling unit areas who specified their national origin or ancestry as Mexican/Mexicano, Mexican American or Chicano/Hispano/Spanish- American/Spanish. e Cuban includes persons residing in Dade County, Florida, primary sampling unit areas who reported their national origin or ancestry as Cuban or Cuban American. e Puerto Rican includes persons residing in New York, New Jersey, and Connecticut primary sampling unit areas who reported their national origin or ancestry as Puerto Rican or Boricuan. e Poverty status is defined by poverty income ratio (PIR), a measure of income relative to a poverty index. A PIR less than 1.0 can be described as "below poverty level", while a ratio equal to or greater than 1.0 describes a status that is "at or above the poverty level." e Poverty Income Ratio formulas vary by survey. For the National Health and Nutrition Examination Survey II, 1976-80, PIR was calculated as follows: PIR = Family income Weighted average poverty threshold where family income equaled the median for the income group for incomes of $7,000 or more or the sum of the component parts of the income questions for incomes less than $7,000, and poverty threshold equaled the value for the year of the interview published by the U.S. Bureau of the Census. For the Hispanic Health and Nutrition Examination Survey, 1982-84, PIR was calculated as follows: PIR = Family income Adjusted poverty threshold where family income equaled the median for the income group, with $50,000 assigned to the highest income category; and poverty threshold was calculated from the full threshold matrix value for the calendar year preceding the interview, updated by the rate of inflation that prevailed between the calendar year and the 12-month income reference period prior to the date of the interview. 1-6 e Overweight is defined as a sex-specific BMI equal to or higher than the BMI of the 85th percentile for men and nonpregnant women 20-29 years of age examined in the NHANES II. Severely overweight is defined as a BMI equal to or greater than the 95th percentile for the same reference population. (For men, a BMI of 217.8 or greater and 31.1 or greater defines overweight and severely overweight, respectively. For women, overweight and severely overweight are defined by BMI values of 27.3 and 32.2, respectively.) e Serum cholesterol status is indicated by both mean serum cholesterol level (in mmol/L) and the percent with "high-risk" serum cholesterol. "High-risk" serum cholesterol is defined as a level of more than 5.69 mmol/L (220 mg/dl) for persons 20-29 years of age, more than 6.21 mmol/L (240 mg/dl) for persons 30-39 years of age, and more than 6.72 mmol/L (260 mg/dl) for persons 40 years of age or older. eo Serum vitamin A status is indicated by both mean serum retinol levels (in pmol/L) and the percent with serum retinol level less than 0.7 pmol/L (20 pg/dl). e Serum vitamin E status is indicated by mean serum alpha-tocopherol levels (in pmol/L). e® Mean corpuscular volume (MCV) equals hematocrit (in percent red blood cells/L x 100 e Iron deficiency is assessed by means of the MCV model. The MCV model uses mean corpuscular volume, transferrin saturation, and erythrocyte protoporphyrin as indicators of iron deficiency, and requires that at least two of three values for these indicators be abnormal. See chapter 6 for the cutoff values used and a detailed description of the model. eo Hypertension is defined as a condition in which a person examined had a systolic blood pressure equal to or higher than 140 mm mercury or a diastolic blood pressure equal to or higher than 90 mm mercury or was taking antihypertensive medication. Notes for Data from the U.S. Food Supply Series eo Nutrient estimates are based on Economic Research Service (ERS) estimates of per capita quantities of food available for consumption (retail weight), on imputed consumption data for foods no longer reported by ERS, and on Human Nutrition Information Service estimates of quantities of produce from home gardens. No deduction is made in food supply estimates for loss of food or nutrients in further processing, in marketing, or in the home. Data include iron, thiamin, riboflavin, niacin, vitamin A, vitamin B6, vitamin B12, and ascorbic acid added by enrichment and fortification. eo Data are given for the following food groups: meat, poultry, and fish; dairy products; eggs; fats and oils; fruits; vegetables; legumes, nuts, and soy; grain products; sugars and sweeteners; and miscellaneous foods. eo Percentages are based on unrounded data. Components may not add to 100 because of rounding. eo Meat: Reported as fresh retail cut equivalent. Includes game, edible offal, and fat cuts of pork. e Poultry: Reported on ready—to—cook basis. Excludes game birds. eo Fish: Reported on edible-weight basis. Includes game fish. eo Eggs: Reported on shell basis. e Dairy products: Includes fluid milk (whole and lowfat), cheese, and other. Reported as calcium equivalent. e Fluid milk: Reported as calcium equivalent. eo Other milk products: Reported as calcium equivalent. Includes cream; canned, evaporated, and dry milk; whey; yogurt; and ice cream and frozen desserts. 11-7 Fats and oils: Includes butter and margarine; shortening; lard and beef tallow; salad, cooking, and other edible oils. Lard and beef tallow: Excludes use in margarine and shortening. Total fruits: Includes citrus and noncitrus fruits. Reported as product weight except for concentrated juices, which are on a single-strength basis. Total vegetables: Includes white potatoes; dark green and deep yellow vegetables; and others. Miscellaneous: Includes coffee, tea, chocolate liquor equivalent of cocoa beans, spices, and fortification not assigned to a specific food group. Coffee, tea, and cocoa: Includes instant and regular coffee reported on roasted basis; tea reported as leaf equivalent; cocoa reported as chocolate liquor equivalent of cocoa beans. n-8 Figure II-1. Figure II-2. Table II-1. Table 11-2. Table II-3. Table 11-4. Table II-5. Table II-6. Table 11-17. Table II-8. Table 11-9. Figure II-3. List of Figures and Tables Food Energy Food energy: per capita amount in the U.S. food supply, 1909-85: U.S. F008 SUPPLY Series .uuunvinsesinmsna sass sumas stones Sanson nsRus seas sams nis 11-24 Food energy: food sources in the U.S. food supply, 1985: U.S. Food Supply Series .........ouiiiiiiniiiniiiiiiiiiiiiiiiiiiiiieteeeeeeeenens 11-24 Food energy: mean intake in kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by INAIVIAUALE, OBB =86 05:2 w5i0 550 55.0 350 006 10.9.3 54 91008 5950.32 15.0: 4.30900 1 90 0 00 AE 400.04 0016 0004 36 04 0 A 1 9 11-25 Food energy: mean intake in kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 . .......ciiiiiiiiiiit tiie ies 11-26 Food energy: mean intake in kilocalories, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nation— wide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ............c.cciriiiiiiiiiiiiiiiiiiireeennnnns I-27 Percent of overweight persons 20-74 years of age, number examined and standard error of the percent by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination SUIVEY, 1082-84 vcr vremrman st st ssn sat 48s hE bok HAE HE 3 Hh Ek Fh Er RE EW Rk Fk 11-28 Percent of overweight Mexican-American persons 20-74 years of age, number examined and standard error of the percent by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 .................. 11-29 Percent of severely overweight persons 20-74 years of age, number examined and standard error of the percent by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 ......... 11-30 Percent of overweight non-Hispanic persons 20-74 years of age, number examined and standard error of the percent by sex, age, and race: Second National Health and Nutrition Examination Survey, 1976-80 ................. 11-31 Percent of overweight non-Hispanic persons 20-74 years of age, number examined and standard error of the percent by sex, age, and poverty status: Second National Health and Nutrition Examination Survey, 1976-80 ........... 11-32 Percent of severely overweight non-Hispanic persons 20-74 years of age, number examined and standard error of the percent by sex, age, and race: Second National Health and Nutrition Examination Survey, 1976-80 ................. 11-33 Protein Protein: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series .........coiiiiiuiiini ities 11-34 11-9 Figure 11-4. Table II-10. Table II-11. Table II-12. Table 11-13. Table 11-14. Table II-15. Figure II-5. Figure II-6. Table II-16. Table II-17. Table II-18. Table II-19. Table 11-20. Table 11-21. Protein: food sources in the U.S. food supply, 1985: U.S. Food SUPPLY Series... oon Protein: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 . ...... cout a Protein: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Sees cece tse ss ses enna Protein: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 “ees se se sss sess ene ne Protein: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Ses eee es cesses eee Protein: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ese ee eee enn Protein: mean percent of kilocalories, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ® 8 8 8 8 se es sess eee se ee esses sees sees esses ees eee eee Total Fat Fat: per capita amount in the U.S. food supply, 1909-85: U.S. Food SUDDIY SOLIS cu vvin 000500 505 005105 55 595 8 58 3154 0 or on rm www wm or 908 540 9 0 0 02 8 Fat: food sources in the U.S. food supply, 1985: U.S. Food Supply Series Fat: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Ses ees ee ttt essere Fat: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 “eee ee ce cee Fat: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Fat: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Fat: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 Fat: mean percent of kilocalories, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-178; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 “es es ee ee tsetse II-10 Figure II-7. Figure 11-8. Table 11-22. Table 11-23. Table 11-24. Table 11-25. Table 11-26. Figure II-9. Figure 11-10. Table II-27. Table 11-28. Table 11-29. Table II-30. Saturated Fat Saturated fat: per capita amount in the U.S. food supply, 1909-85: U.S. F008 SUPPLY SOII08 4. oui visas nnn ssn ames oem en fom eww sb ve @ mwas eames s 8s meen ws Saturated fat: food sources in the U.S. food supply, 1985: U.S. Food SUDDIY SOTIBE + 2 v 50 0750 #90 100 #550 # 130 10 6 128 0 420 10 50 4 Hh 0 Bt 50 HET 0 0 0 0 on 0 0 0 0 Saturated fat: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes hy Individuals, JO85-86 ..cxsinsearmnrsninmtnernrnrsnevosmrmmvsnnsansanes ss ss ren ns Saturated fat: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes hy InAIVidUnlS, 1088-88 .uucnsoninntnsnnsmnsnnsn muss stior names nte nse vin ren mn ens Saturated fat: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes hy INAVIAUALS, 1085-88 ,uu:nomsrnntsmummarssnrasemrmniuns gin vee rensm noes wm wsnss Saturated fat: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............ciiiinniiiiiiiiiii iii ea Saturated fat: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ..............cciiiiiiiiiiiiiiiiiiinn.. Monounsaturated Fat Monounsaturated fat: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series .........coiuiniiiiiiiiiiiiii iii. Monounsaturated fat: food sources in the U.S. food supply, 1985: U.S. Food Supply Series .........ouiuuniiiiiii iii iii iii iii Monounsaturated fat: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes hy Individuals, 1088-86 .....ccouisiiaiiniimimrisssninmess sravinvnsemesssessswnensns Monounsaturated fat: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 .......... ccc Monounsaturated fat: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, YOBB~BB ...xieussiiniassnss semanas sss ins sssrsntssessraona mares Monounsaturated fat: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1085-88 ........ccincrmuinivvisvrnrnsrnvevnimvivasvsnsenven II-11 Page Table II-31. Figure II-11. Figure 11-12. Table 11-32. Table II-33. Table 11-34. Table 11-35. Table 11-36. Figure 11-13. Figure 11-14. Table 11-37. Table II-38. Monounsaturated fat: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 .............c.iintiiiiii i aa Polyunsaturated Fat Polyunsaturated fat: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ............ouiuiiriirineiriiieieiieanennnnn. Polyunsaturated fat: food sources in the U.S. food supply, 1985: T08. Pood SUPPIY SOMIOE 1vuunnssniavmassnannsmsensesinns in sere ssnssenssntensssss Polyunsaturated fat: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 .............c.iiuiinitiitiii aa Polyunsaturated fat: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............iiuiiiiii a Polyunsaturated fat: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...............c.iiitiiitii ta Polyunsaturated fat: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............c.iuiiiiiii iii iiiii reir raraereraeraenas Polyunsaturated fat: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ................. iii Cholesterol Cholesterol: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series .............c.ouiuinirini ieee Cholesterol: food sources in the U.S. food supply, 1985: U.S. Food BUPDIY SOPIE «vc vc04 000 553 604 5 500.505 5005 0:0 5 40 400 0 0m 14100 mw 20m 0 408 8408 830 0 000 9 89 BR Cholesterol: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes bY INAIVIAUALS, 2088-86 crus 0s wie 510.5505 0005.8 0s #008 0 0 wk wo wen mw 29 00 418 270 18 500 0 970 430 5: 181 Cholesterol: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...............iiuiitii a 11-12 Page Table 11-39. Table I1-40. Table 11-41. Table 11-42. Table 11-43. Figure II-15. Figure 11-16. Table 11-44. Table 11-45. Table 11-46. Table 11-47. Table 11-48. Cholesterol: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 .............c.ciiiiiiiiiiiiiiiiiiinennn, Serum cholesterol status of persons 20-74 years of age by sex, specified Hispanic origin and age: Hispanic Health and Nutrition Examination Survey, 1982-84 ..........ciiiiiiiiiiiiiiiiii iia Serum cholesterol status of Mexican-American persons 20-74 years of age by sex, poverty status and age: Hispanic Health and Nutrition Examination Survey, 1982-84 ...........cciiiiiiiiiiiiiiiiiiiiiiiiiie Serum cholesterol status of non-Hispanic persons 20-74 years of age by sex, race and age: Second National Health and Nutrition Examination Survey, 1976-80 ..........c.oiiuiiiiiiiiiiiiiiiaiii iii Serum cholesterol status of non-Hispanic persons 20-74 years of age by sex, poverty status and age: Second National Health and Nutrition Examination Survey, 1982-84 .............ccciiiiiiiiiiiiiiiiiiiinne Carbohydrate Carbohydrate: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ........ccairiiiiiirnesiiivrrsssssersssscscensssacsnnrsnees Carbohydrate: food sources in the U.S. food supply, 1985: US. Food SUPPLY SOTIBE ouvunivmmnssnimusnntnymmonssmrss ons snwes ssid sok ssh nvaseny Carbohydrate: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............c.oiiiiiiiiiiiiiiiiiiiiiiii iii Carbohydrate: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............uiiuiiuintatiiiita iii Carbohydrate: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............cutiniiiit iii Carbohydrate: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............iiuuiiiii iii iis Carbohydrate: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ................cooiiiiiiiiinn., 11-13 Page Table 11-49. Table II-50. Table II-51. Figure I1-17. Figure 11-18. Table 11-52. Table II-53. Table 11-54. Table II-55. Table II-56. Table 11-57. Table II-58. Page Carbohydrate: mean percent of kilocalories, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ...................... 11-81 Dietary Fiber Dietary fiber: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............c.uuitiiiiniii ieee 11-82 Dietary fiber: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............euiuniunin innit 11-83 Vitamin A Vitamin A: per capita amount (in retinol equivalents) in the U.S. food supply, 1909-85: U.S. Food Supply Series ............oueueueuenenennnnnn... 11-84 Vitamin A: food sources in the U.S. food supply, 1985: U.S. Food Supply Series .........c.ouiiiiniit ii 11-84 Vitamin A: mean intake in retinol equivalents, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............cuiuiniiiitee ee 11-85 Vitamin A: mean intake in retinol equivalents, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............ouiiiininiii ieee 11-86 Vitamin A: mean intake in International Units, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ..............ouiiiininiii ieee 11-87 Serum vitamin A status of children 4-19 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 ..................iiiii 11-88 Serum vitamin A status of persons 20-74 years of age by sex, specified Hispanic origin and age: Hispanic Health and Nutrition Examination Survey, 1982-84 ...............c.iuiiiininineannnnnnn., 11-89 Serum vitamin A status of Mexican-American children 4-19 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 ...............couviiiiinenineinannnnnn., 11-91 Serum vitamin A status of Mexican-American persons 20-74 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 ...............cccoiuiiniuniininiianannnn., 11-93 11-14 Figure II-19. Figure 11-20. Table II-59. Table 11-60. Figure 11-21. Figure 11-22. Table II-61. Table 11-62. Table 11-63. Table 11-64. Table 11-65. Table II-66. Figure 11-23. Carotenes Carotenes: per capita amount (in retinol equivalents) in the U.S. food supply, 1909-85: U.S. Food Supply Series ..............cooiiiiiiiiiinnn Carotenes: food sources in the U.S. food supply, 1985: U.S. Food SUPP Series ............oviuiuiniiieirirntnearastrteristetsasasananansnenes Carotenes: mean intake in retinol equivalents, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 .............c..ciiiiiiiiiiiiiiiiiiiinnenn Carotenes: mean intake in retinol equivalents, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...............coiiiiiiiiiiiiiiiiiiieen. Vitamin E Vitamin E: per capita amount in the U.S. food supply, 1909-85: U.S. Food SUPP Series ..........ccvviiriienerisnrnesnsaesassersssasstsnsanennans Vitamin E: food sources in the U.S. food supply, 1985: U.S. Pood Supply Series .........ccierereerersnssrsnrasinsassnrensareartesnrsvsnranras Vitamin E: mean intake in milligrams alpha tocopherol equivalents, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 .............c.coiiiiiiiiiiiiiiiiiiinn. Vitamin E: mean intake in milligrams alpha tocopherol equivalents, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............cooiiiiiiiiiiiiiiiienn Serum vitamin E status of children 4-19 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 ........... iii Serum vitamin E status of persons 20-74 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 ........... iii ie Serum vitamin E status of Mexican-American children 4-19 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 .............cciiiiiiiiiiiiiiiiii iii. Serum vitamin E status of Mexican-American persons 20-74 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 ........ cc... Thiamin Thiamin: per capita amount in the U.S. food supply, 1909-85: U.S. Food SUPPly Series ..........ouiiiiuiiririieiiinenieneentiieanineaiennanann, II-15 Page Figure 11-24. Table II-67. Table II-68. Table II-69. Figure 11-25. Figure 11-26. Table II-70. Table II-71. Table II-72. Figure 11-27. Figure 11-28. Table II-73. Table II-74. Thiamin: food sources in the U.S. food supply, 1985: U.S. Food BUPPIY SOIR i vsinv imu rn0095m 500000000 0b 015855 4 05 000k man 300 200 99°49 450.0 70 0 474 94.13 BA 6 4 0 Thiamin: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1085-88 . ix. ovsvmes ss smusmme corn snes ota ssnwes snes senses Thiamin: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............ciuiniiitiiiite Thiamin: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ........................ Riboflavin Riboflavin: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ..............iuiuinieiiinie eee eee Riboflavin: food sources in the U.S. food supply, 1985: U.S. Food Supply Series ...........ouiiiniiniit iit Riboflavin: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............cuiuiniiiiiii a Riboflavin: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 108888 .....cuuvrsuiaminirsiimmrererensnssressnsnssnsssotesssose Riboflavin: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ........................ Niacin Niacin: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ..............oiuiiiiuiiiiiie eee eee, Niacin: food sources in the U.S. food supply, 1985: U.S. Food Supply Series ...........ouiiininie ii Niacin: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...............iiiniitiiii Niacin: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ................iuiniiiiiiii aa Table II-75. Figure 11-29. Figure II-30. Table 11-76. Table 11-77. Table 11-78. Table II-79. Figure 11-31. Figure 11-32. Table II-80. Table II-81. Figure 11-33. Niacin: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ........................ Vitamin B6 Vitamin B6: per capita amount in the U.S. food supply, 1909-85: U.S. Food SUPPLY Series ........ouiuirinrnununeerataenueueieitarateneneneaeeenns Vitamin B6: food sources in the U.S. food supply, 1985: U.S. Food SUPPLY Series ........ouiuirtnrnununneneenueueueieararaneneeneaeannns Vitamin B6: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...........couiuieiiiiniiiiiiiiii i Vitamin B6: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............cocovieininininennne. Vitamin B6: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...........oouinineriiiiiiiiiiiiii ie Vitamin B6: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ................cooiiiiiieninnnes Vitamin B12 Vitamin B12: per capita amount in the U.S. food supply, 1909-85: U.S. Food SUPPLY Series ........ouiuirinrnrnrnnanetieie i iiitatatatenenenencanns Vitamin B12: food sources in the U.S. food supply, 1985: U.S. Food SUPPLY Series .........couiurirenreneneneeiiiiintineaeeneneeneanaeenn. Vitamin B12: mean intake in micrograms, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............ouiuiuininiiiiiiiiiiiiiii ees Vitamin B12: mean intake in micrograms, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ............oiiiiiiiiiiiiiiiiii ie Vitamin C Vitamin C: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ..........ouiuiiuineireneneiiiieiireieananinneaneneeene. 1-17 Page Figure 11-34. Table II-82. Table II-83. Table 11-84. Figure II-35. Figure 11-36. Table II-85. Table 11-86. Table II-87. Table 11-88. Figure 11-37. Figure 11-38. Table IT-89. Vitamin C: food sources in the U.S. food supply, 1985: U.S. Food Supply Series ...........cuuuirinininiiiiiee eee eens Vitamin C: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...........cc.iiiiiiiiiii iia Vitamin C: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............coiiriiitiiii iia Vitamin C: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ..................cccvvuivueuennnn... Folacin Folacin: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ............uuiuinininiiiii ite Folacin: food sources in the U.S. food supply, 1985: U.S. Food Supply Series ...........c.ouiuuintirinii ieee eee eae, Folacin: mean intake in micrograms, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............oiiiniiitii Folacin: mean intake in micrograms per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 .................oovueeenunnnnnn... Folacin: mean intake in micrograms, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............oiiiiiitiii Folacin: mean intake in micrograms per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...............covureurnnrnnennnn.. Iron Iron: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ..............iuiuininiiiit ieee eee eae Iron: food sources in the U.S. food supply, 1985: U.S. Food Supply Series ............c.iuuiniiiiniii eee eee eae, Iron: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..................iuiniiiitii aa 11-18 Table 11-90. Table 11-91. Table 11-92. Table 11-93. Table 11-94. Table II-95. Table II-96. Table 11-97. Table 11-98. Table II-99. Table II-100. Table II-101. Figure 11-39. Figure 11-40. Iron: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 © ee se se sees ess eee ses es esse eee eee Iron: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ........ c.count iii i ea Iron: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 © 8 es ee ss ses eee ee sees ee ee ees esses Iron: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 Iron deficiency determined by the MCV model of children 4-19 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 Iron deficiency determined by the MCV model of persons 20-74 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 Iron deficiency determined by the MCV model of Mexican-American children 4-19 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 “ees see se ass sass eee Iron deficiency determined by the MCV model of Mexican-American persons 20-74 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Iron deficiency determined by the MCV model of children 4-19 years of age by sex, age, and race: Second National Health and Nutrition Examination Survey, 1976-80 © 8 8 8 8 8 8 8 se ea es es ees ss ee es esses sees ee se sees eee ee eee Iron deficiency determined by the MCV model of persons 20-74 years of age by sex, age, and race: Second National Health and Nutrition Examination Survey, 1976-80 © 6 6 8 8 8 6 6 a a ss ee es ss see ss se se sss es ee ee se ee see eee eee Iron deficiency determined by the MCV model of children 4-19 years of age by sex, age, and poverty status: Second National Health and Nutrition Examination Survey, 1976-80 © 6 4 8 4 80 a a a ass ee ee esse se ee se ss ee ese eee Iron deficiency determined by the MCV model of persons 20-74 years of age by sex, age, and poverty status: Second National Health and Nutrition Examination Survey, 1976-80 © oo sa ss 0 0 0 ee se ee ees ss sees ee ee sees essen Calcium Calcium: per capita amount in the U.S. food supply, 1909-85: US. Pood Supply Seton «overs nnn vunnss ta 60s asst h mb mai ads ow owns sewn sss ae Calcium: food sources in the U.S. food supply, 1985: U.S. Food Supply Series I I I ICICI Table II-102. Table II-103. Table 11-104. Table II-105. Table II-106. Figure 11-41. Figure 11-42. Table 11-107. Table 11-108. Table II-109. Figure 11-43. Figure 11-44. Calcium: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............uiiniitiiii ii a Calcium: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...............c.coviueneenennnnnn.. Calcium: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ..............oiuiiiiiiiiii i a Calcium: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ................c.coouiueenennnnnn.. Calcium: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ........................ Phosphorus Phosphorus: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ..........c.iiuiintiit iii ieee eee eee Phosphorus: food sources in the U.S. food supply, 1985: US. Pood Supply Soros c.cvissisissiosssisirrsrenrmmenonmrnwenssnsssssssss messes Phosphorus: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by INAIVIAUALE, JOBE 8B ..vuuvvsnssnsinssmssmsinssnmnnmennesnsnnssssssssrsasnsss Phosphorus: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-88 ....iuvsusissnninsinsamissssnremsnnrnnsvrmasnssnnsnssssss Phosphorus: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ........................ Magnesium Magnesium: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series ............ouuiuiitiriiti iii ieee Magnesium: food sources in the U.S. food supply, 1985: UB. Food Supply Series «cu uuvuiiniivririnsinrimisrerumvrsvsresensesssnmsnrsnsnns 11-20 Table 11-110. Table II-111. Table 11-112. Table II-113. Table 11-114. Table 11-115. Table II-116. Table 11-117. Table 11-118. Table 11-119. Table 11-120. Table 11-121. Table 11-122. Magnesium: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 .........c.ivuiuiiiiiiiiaiiaiiiti iii Magnesium: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Magnesium: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Magnesium: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ees sss 8 ss ee ee ee ess sees es eee Sodium Sodium: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 sem mE AB sSNA SENNA SSE EVER IE LIAL Rts ssaN sess Sodium: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 CRANE R ESSER ISG e Ears Nese R GIANG seen Sodium: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 ees eee see sess sss sess Sodium/potassium ratios, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 aE EASES EIEN NOSES SAI ENING SSSA, Sodium/potassium ratios, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ams A LENS SEY S AEE r ESE Tsetse sES SAMs ess Hypertension among persons 20-74 years of age by sex, specified Hispanic origin and age: Hispanic Health and Nutrition Examination Survey, 1982-84 es se 6s ss sess sees esses sees esses sees Hypertension among Mexican-American persons 20-74 years of age by sex, poverty status and age: Hispanic Health and Nutrition Examination Survey, 1982-84 ec es sess ss eee esses ss sss esses sess sess Hypertension among persons 20-74 years of age by sex, race and age: Second National Health and Nutrition Examination Survey, 1976-80 CS EAE EERE AEE II EEE IINI IEE teny Hypertension among non-Hispanic persons 20-74 years of age by sex, poverty status and age: Second National Health and Nutrition Examination Survey, 1976-80 ce ees ss ee sees ee ese sees 11-21 Figure 11-45. Figure 11-46. Table II-123. Table 11-124. Table 11-125. Figure 11-47. Figure 11-48. Table II-126. Table 11-127. Figure 11-49. Figure 11-50. Table 11-128. Table 11-129. Potassium Potassium: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series SBS IIE EEE PIII LEI EIEIO Ess t Etter ere Potassium: food sources in the U.S. food supply, 1985: U.S. Food Supply Series TERS BITE 40080R C00 III NPTERN RDO OTS ESWC Ne ee ee Potassium: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ................coiiiiuiinninee ee Potassium: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ...............coiuniieeeine aie Potassium: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey, 1971-74; Nationwide Food Consumption Survey, 1977-78; Second National Health and Nutrition Examination Survey, 1976-80; and Continuing Survey of Food Intakes by Individuals, 1985-86 © 8 0 eee eee tees eee sets Copper Copper: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series CHIBLABIB LEVI HEIN STII NII REDCAR SRT ARNT e Brrr swe inw Copper: food sources in the U.S. food supply, 1985: U.S. Food Supply Series ...........coouueenneeneee eee Copper: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 ................couiueiiniee ie Copper: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 sess s sss LNAI RtI EMILE LCI N LEHI ES Messe Zinc Zinc: per capita amount in the U.S. food supply, 1909-85: U.S. Food Supply Series SHA EA BIB SERIO SANSA R Ar BS eB cms ssn sews EEE NE Zinc: food sources in the U.S. food supply, 1985: U.S. Food Supply Series SS StI EE I lIlEsEPslI IITs Zinc: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 BAWABA ETI OIA CIOL FFIEC WS SN ens ss em view inn Zinc: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 8% ss esse eee tess teeta I-22 Page Table II-130. Zinc: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by InAIVIAUALS, OBE~86 1. 0 00 0005005 0m 0105 500 0 wis 000 810 000 0 000 0 910 0 00 9 #10 wi 1 0 6 0 ow ip im 11-182 Table II-131. Zinc: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 .................ccoiiiiiiiinnnn... 11-183 Vitamin/Mineral Supplement Use Table II-132. Percent of supplement users consuming specific nutrients: Vitamin/Mineral Supplement Intake Survey, 1980 ..............ccoiiiiiiiinnin... 11-184 Table II-133. Median and 95th percentile levels of intake (as percent of RDA) for individual nutrients from supplements: Vitamin/Mineral Supplement Intake Survey, 1980 ...... oii ie 11-185 11-23 ve-11 U.S. Food Supply Food energy Kilocalories 4000 3000 + 2000 + 1000 0 ~~ 3 1 1 1 1 1 x 1 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-1. Food energy: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Food Energy Food energy Grain products 19.9% Fats and oils 20% Meat, poultry, fish 19% Fruits and vegetables 7.9% Sugars and sweeteners 17.8% Dairy products 9.97% Figure II-2. Food energy: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include eggs; legumes, nuts, and soy; and miscellaneous foods) gg-11 Table II-1. Food energy: mean intake in kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 15 90 95 All women 2056 1517 16 783 904 1157 1454 1814 2393 2431 Age 20-29 years 661 1572 25 788 955 1207 1503 1886 2314 2488 30-39 years 812 1503 22 791 891 1138 1445 1810 2177 2337 40-49 years 583 1438 26 713 895 1120 1400 1702 2048 2243 Race 1/ White 1775 1538 16 804 940 1181 1477 1827 2201 2441 Black 167 1376 57 576 724 959 1295 1753 2087 2271 Other 76 1405 53 * * 1141 1413 1745 * * Poverty status 1/ < 100 315 1420 38 670 758 1034 1314 1746 2099 2434 > 100 1575 1542 15 806 939 1191 1488 1834 2182 2392 < 131 414 1431 39 661 758 1059 1350 1771 2087 2377 > 131 1476 1547 15 821 946 1191 1493 1834 2184 2399 Education 1/ < High school 305 1309 33 654 715 968 1241 1591 1943 2111 High school 854 1507 20 788 909 1172 1451 1802 2120 2337 > High school 891 1589 20 838 994 1229 1548 1885 2254 2489 Region Northeast 448 1447 29 824 885 1109 1406 1731 2025 2303 Midwest 564 1577 32 794 926 1205 1528 1924 2284 2488 South 660 1497 25 713 870 A127 1428 1810 2130 2403 West 384 1542 39 824 948 1219 1490 1834 2177 2441 Urbanization Central city 499 1536 35 783 905 1154 1476 1859 2287 2489 Suburban 1039 1515 22 786 900 1158 1462 1809 2168 2396 Nonmetropolitan 518 1488 23 737 905 1148 1432 1796 2084 2301 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 93-11 Table II-2. Food energy: mean intake in kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 30 95 All children 1/ 647 1426 22 920 1005 1178 1374 1620 1909 2064 Age 1-2 years 224 1305 31 867 916 1105 1266 1506 1687 1827 3-5 years 423 1492 26 990 1047 1216 1462 1719 2013 2145 Race 2/ White 559 1440 23 938 1025 1185 1395 1635 1914 2082 Black 53 1397 67 * * 1178 1381 1601 * x Other 26 1222 61 * * x 1216 * * * Poverty status 2/ < 100 140 1396 45 878 960 1156 1352 1601 1829 2038 2 100 471 1437 26 931 1021 1189 1372 1631 1914 2120 < 131 192 1387 40 888 961 1150 1312 1597 1900 2048 > 131 419 1446 27 938 1021 1193 1391 1635 1914 2120 Education 2/ < High school 99 1421 46 * 980 1158 1332 1659 2029 * High school 252 1421 3a 920 9 1166 1364 1634 1914 2052 > High school 295 1431 30 940 1030 1200 1400 lel6 1836 2067 Region Northeast 111 1444 44 x 1023 1200 1433 1683 1914 x Midwest 199 1433 44 963 1033 1218 1369 1622 1983 2073 South 187 1399 43 920 980 1142 1316 1609 1889 2120 West 150 1438 43 934 1021 1209 1409 1635 1836 2022 Urbanization Central city 171 1478 51 931 1036 1234 1435 1708 1960 2145 Suburban 310 1395 28 920 1010 1166 1336 1588 1836 2029 Nonmetropolitan 166 1434 45 920 960 1150 1413 1683 1938 2149 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. Lg-11 Table II-3. Food energy: mean intake in kilocalories, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES 11 \ CSFII 1971-74 1977-78 1976-80 ' | 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM ) Mean SEM Se Both sexes 1-2 1,350 15 1,196 14 1,287 11 1,309 36 3-5 1,676 14 1,454 17 1,569 9 1,527 29 6-11 2,045 22 1,876 16 1,960 22 - - Male 12-15 2,625 51 2,431 32 2,490 55 - - 16-19, 3,010 77 2,629 39 3,048 83 - - 20-29, 2,850 60 2,501 34 2,899 64 2,806 96 30-395 2,668 67 2,382 32 2,554 56 2,484 88 40-49 2,428 56 2,341 37 2542165 2,384 91 50-59 2,157 55 2,240 30 2,203 55 - - 60-69 1,967 23 2,041 33 1,961 19. - - 70+ 1,747 19 1,875 36 1,734 29 - - Female 12-15 1,910 40 1,870 25 1,821 40 - - 16-19 1,735 45 1,721 27 1,687 46 - - 20-29 1,681 18 1,634 18 1,675 30 1,674 31 30-39 1,610 19 1,571 29 1,596 36 1,648 27 40-49 1,552 23 1,562 19 1,531 35 1,541 31 50-59 1,466 33 1,548 20 1,417 36 - - 60-69 1,352 15 1,475 20 1,340 12 - - 70+ 1,239 14 1,386 17 1,270 18 - - 1 CSFII data for 1985 only. Ages 70-74 years only for NHANES I and NHANES II. 8C-1I Table 11-4. Percent of overweight persons 20-74 years of age, number examined, and standard error of the percent by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 T T T Mexican American | Cuban Puerto Rican | T T | T T | I | ; : | Standard | } Standard | ; Standard Number of error of [Number of error of Number of error of lexan ined | the examined the examined | ; the Sex and age jpersons a Percent | percent [persons a Percent percent jpersons vy Percent | percent 1 1 1 1 1 1 1 1 ee i Male 20-74 years. ................. 1,454 29.6 1.4 376 29.4 2.4 443 25.2 2.1 20-74 years, age adjusted ... Ln 30.9 27.6 25.6 20-29 years ................. 441 20.4 2.6 57 23.5 6.7 113 15.6 3.9 30-39 years. ................. 376 34.6 3.1 56 19.9 6.1 30 24.8 5.2 40-49 Y@AIrS.:. . x: snsssmniimes 243 38.2 3.4 82 38.3 5.5 87 35.6 5.2 50-59 years... ............... 234 36.0 3.0 109 33.1 4.1 101 33.1 3.8 60-69 years. ................. 122 34.0 4.4 44 25.9 6.1 41 $23.5 5.7 70-74 years. ......cv: cums smu 38 +26.8 7.9 28 +38.4 B.5 11 + Female 2/ 20-74 years. ................. 1,797 39.1 1.2 484 34.1 2.1% 758 37.3 1.7 20-74 years, age adjusted ... Ln 41.6 LL s 5 3i.6 40.2 20-29 years ................. 514 24 4 3.2 67 16.2 5.0 192 22.5 3.4 30-39 years... ..ovinvvsmus 2 442 39.5 2.6 a5 26.4 5.0 172 36.2 4.1 40-49 years... ............... 314 47.9 2.8 104 41.3 4.8 171 46.4 3.7 50-59 years. ................. 324 53.4 2.6 114 41.5 4.2 132 50.2 3.5 60-69 years. ....cvvsus smuinw 141 59.9 4.1 72 46.8 5.3 77 59.3 4.3 70-74 years... ............... 62 43.0 6.3 32 +40.0 7.9 14 * + 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Excludes pregnant women. 6¢-11 Table II-5. Percent of overweight Mexican-American persons 20-74 years of age, number examined, and standard error of the percent by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Below poverty Above poverty ——————— T T T T } : Standard | Standard Number of | I error of [Number of | error of examined | | the examined | the Sex and age persons Vy Percent | percent jpersons vy Percent percent 1 1 i —— 1 1 1 Male 20-74 years... ............... 324 29.9 3.1 1,013 29.6 1.7 20-74 years, age adjusted ... LL. 31.8 31.4 20-29 years ................. 100 20.2 5.6 304 20.6 3.1 30-39 years. ...::vs i innnmns io 70 38.0 7.4 284 33.3 3.6 40-48 years. ...: un venz snes 2 49 46.2 7.6 173 36.7 4.1 50-59 years... ........... ... 52 35.0 6.3 161 37.0 3.8 60-69 years... ........ ....... 40 +31.8 7.9 70 36.3 5.9 70-74 years. Ph EwiE sm SWE EE 13 + + 21 * * Female 2/ 20-74 years... ...... 549 44.0 2.3 1,084 36.8 1.6 20-74 years, age adjusted ... CL 46 1 40.1 R20=29 YBAPrS uv: ims wins sms ima 152 26.0 4.2 326 23.9 2.7 30-39 years. .......svsvvns sna 125 45.1 4.9 286 37.3 3.2 40-49 years... ................ 92 57.8 5.1 189 45.1 3.6 BO-59 YBArS. . w: cnsimnsimns cms 98 61.9 4.6 188 49.4 3.4 BOB years. ..... uos0ms sams sun 45 59.4 7.5 27 57.4 5.6 70-74 years. ................. 37 *46.1 8.6 18 * * 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Excludes pregnant women. 0e-1I Table II-6. Percent of severely overweight persons 20-74 years of age, number examined, and standard error of the percent by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 T T T | Mexican American Cuban Puerto Rican | T T | T T | 1 ro TTT | Standard ; | | Standard | Standard {Number of error of Nunes of | error of jNusper of error of jeraninsd | | the jFranineg | the jSxaninsd | the Sex and age jparsons i/) Percent | percent {persons V Percent | percent |persons VV) Percent percent 1 1 1 1 $ 1 1 1 —_ Male 20-74 years. css xu sans aoe 1,454 10.3 0.9 376 10.6 1.6 443 7.7 20-74 years, age adjusted ... CL } 10.8 ‘i 10.7 8.0 20-29 Years ..u:isusiewsimws vn 441 7.4 1.5 57 14.9 5.6 113 4.4 3.2 30-39 years. ................. 376 11.9 1.9 56 5.6 3.5 90 7.6 3.2 40-49 years. ........... 243 10.6 1.9 82 9.4 83.3 87 B.2 3.0 BO=-BY yBAIrS: « «m+ swims amas wun vo 234 15.3 2.0 109 12.7 2.9 101 12.2 2.7 60-69 years. ................. 122 12.9 2.8 44 +10.5 4.3 41 +13.8 4.7 70-74 yRAPrS. . .:: iv: ives enn 38 +5.0 8.5 28 +7.4 4.6 11 + + Female 2/ 20-74 y@aArs.. .«. ves sums snns 1,797 15.6 0.8 484 7.7 1.2 758 14.4 1.3 20-74 years, age adjusted ... saw 16.9 6.6 15.7 20-29 yeArS ...c. unas 514 9.4 1.3 67 1.5 1.6 192 8.2 2.2 30-39 years. ........... 442 14.9 1.7 95 7.1 2.9 172 16.6 3.1 80-49 y@Ars. .... xi invinnss 314 20.5 2.0 104 9.3 2.8 171 15.4 2.7 BO~-89 years... crus vas cnmasmins 324 22.7 1.9 114 13.9 3.0 132 17.2 2.6 60-69 years... ............... 141 23.4 3.1 72 4.1 2.1 77 24.1 3.8 70-74 YEAS. ... svn: curs inns 62 17.0 4.3 32 +91 4.6 14 + + 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Excludes pregnant women. 1€-1I Table 11-7. Percent of overweight non-Hispanic persons 20-74 years of age, number examined, and standard error of the percent by sex, age, and race: Second National Health and Nutrition Examination Survey, 1976-80 Non-Hispanic white Non-Hispanic black ——— ———— T I | | | | | | T T T T | | Standard | ; Standard Number of error of [Number of I error of examined | the examined | the Sex and age persons V| Percent percent jpersons a Percent percent 1 1 A. 1 1 — Male 20-74 yeBPS. ui; iuss sus sums sums 4,646 24.4 0.7 597 25.6 2.0 20-74 years, age adjusted. ... : 5 24.2 26.0 2029 Years ...uu:nms cas sans 1,011 15.1 1.4 158 12.4 3.2 30-39 years. ................. 707 24.5 2.1 93 24.3 5.8 40-49 years... ............... 572 31.7 2.6 62 46.0 8.9 BO~59 YRAPrS. .: sus iwvi swus emus 575 28.9 2.5 77 30.5 6.6 60-69 years. ................. 1,354 28.1 1.0 151 29.2 2.7 TO-74 years. .: .vs nes tnps ams + 427 24.7 1.6 56 24.3 4.2 Female 2/ 20-74 years... ............... 5,069 24.8 0.7 711 43.5 2.1 20-74 years, age adjusted.... LL 23.9 711 44.4 20-29 years ................. 1,007 12.2 1.3 173 27.5 4.4 30-39 years. ................. 779 20.6 1.8 108 36.4 6.0 40-49 years. ......vs ions imme 614 27.4 2.4 95 50. 1 6.5 80-59 years... ....:cssamasmiaw 649 32.7 2.4 101 62.6 6.0 60-69 years. ................. 1,487 35.6 1.0 170 61.4 2.8 10-74 YeAIS. ; snus ns svi tims 533 34.9 1.7 64 54.0 4.7 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Excludes pregnant women. ce-11I Table II-8. Percent of overweight non-Hispanic persons 20-74 years of age, number examined, and standard error of the percent by sex, age, and poverty status: Second National Health and Nutrition Examination Survey, 1976-80 Below poverty Above poverty ——————— T | | | | | | | 1 T T T T ; Standard ; ; Standard Number of | | error of [Number of | | error of examined | | the examined | the Sex and age persons ) Percent | percent persons Vy Percent | percent 1 1 1 i J Male 20-74 years... .:. is is8simgi ims 528 19.4 1.7 4,522 24.8 0.7 20-74 years, age adjusted. ... swe 21.5 24.6 20-29 Yy@ArS .....»isn:idis rns 145 9.6 2.8 989 15.3 1.5 80-38 years... .ssssnms sums smn 62 22.2 6.3 716 24.8 2.1 40-48 Y@APS . «ov cv cvmis rmms cn 34 *35.5 10.3 S71 33.6 2.6 50-59 years. ................. 50 24.4 7.0 567 28.7 2:5 BO0-69 years. .: wx» mms rms ams 152 27.4 2.6 1,298 28.2 1.0 70-74 years... ............... 85 17.2 2.9 381 25.2 1.7 Female 2/ 20-74 YOBPrS. ii: nv:ivws imme ans 845 36.9 1.7 4,707 25.3 0.7 20-74 years, age adjusted.... oo 38.5 4,707 24.6 20-29 Years ..:sxrniwss ins 2ea 201 15.5 3.0 948 13.9 1.4 30-39 years. ... cise ns veins 117 40.8 5.4 749 20.2 31.9 40-49 years. ................. 76 53.4 6.9 613 27.5 2.4 BO-59 y@ArS...: xvi ivsrmms smn 80 54.2 6.6 635 33.7 2.4 80-89 YBAIrsS. ..:wvssnims emma sins 249 42.6 2.3 1,319 37.0 1.1 70-74 years. ................. 122 50.3 3.3 443 33.0 1.8 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Excludes pregnant women. €e-11 Table II-9. Percent of severely overweight non-Hispanic persons 20-74 years of age, number examined, and standard error of the percent by sex, age, and race: Second National Health and Nutrition Examination Survey, 1976-80 Non-Hispanic white Non-Hispanic black T | | | T T T T | Standard | | Standard jf aner of | error of [Number of | | error of {2reningd the examined I | the Sex and age persons VV Percent I percent jrersons Vv) Percent | percent Lo 1 1 1 1 le ——— Male 20-74 years... ............... 4,646 7.7 0.4 597 9.9 1.4 20-74 years, age adjusted.... C 7.7 10.0 20-29 years ................. 1,011 4.7 0.8 158 5.7 2.2 30-39 years. ................. 707 8.2 1.4 93 10.0 4.0 40-49 'YBAPS , ... ums th nidng ink 572 9.1 1.6 62 15.8 6.5 80-89 years. ... «ws: -masams smn 575 10.1 1.6 77 12.3 4.7 60-69 years... ............... 1,354 8.1 0.6 151 8.7 1.8 70-74 years. .........-::0::0+ 427 9.5 1.1 56 10.1 3.0 Female 2/ 20-74 YBAPS. .. cui comms mms mma 5,069 9.7 0.5 711 19.2 1.7 20-74 years, age adjusted. ... LL 9.4 A. 19.8 20-29 years ................. 1,007 4.5 0.8 173 9.0 2.8 30-39 years.... u.issiswss ans 779 9.1 1.3 108 23.7 5.3 40-49 years. .......vv:suvsvms 614 11.6 1.7 95 22.9 5.4 50-59 years. ................. 649 12.8 1.7 101 24.5 5.3 60-69 years... ....:.nrxnousnms 1,487 12.8 0.7 170 30.0 2.6 70-74 years. ................. 533 11.0 1.1 64 12.7 3.1 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The Criteria variable is discussed in the table notes. 2/ Excludes pregnant women. ye-11 Protein U.S. Food Supply Protein Protein Grams Meat, poultry, fish 43.5% 120 100 ~~ eee sor Other foods 17% Dairy products 20.6% J Eggs 4.17% 60 + Legumes, nuts, and soy 5.4% 40r Fruits and vegetables 6.4% Grain products 19% 20 + 1 d 1 1 1 1 1 i 1 1 1 0 A i 1 i i 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-3. Protein: per capita amount per day in the U.S. food Figure II-4. Protein: food sources in the U.S. food supply, 1985: supply, 1909-85: U.S. Food Supply Series U.S. Food Supply Series (other foods include fats and oils; sugars and sweeteners; and miscellaneous foods) ge-1I Table II-10. Protein: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 61 7 30 36 47 59 72 87 97 Age 20-29 years 661 62 1.0 30 36 47 59 74 91 104 30-39 years 812 60 «9 30 36 47 59 71 85 96 40-49 years 583 60 1.0 30 36 47 58 71 83 92 Race 1/ White 1775 61 .8 31 36 48 60 73 86 97 Black 167 56 2.1 24 28 41 54 67 84 98 Other 76 61 3.5 * * 44 60 76 * * Poverty status 1/ < 100 315 58 1.5 25 30 42 55 70 83 94 > 100 1575 62 7 31 37 49 60 73 87 97 < 131 414 58 1.4 25 31 43 56 71 82 93 > 131 1476 62 7 32 38 49 60 73 88 98 Education 1/ < High school 305 53 1.5 25 30 37 52 67 79 93 High school 854 60 .8 30 36 46 58 71 86 96 > High school 891 64 9 34 39 50 62 75 90 101 Region Northeast 448 61 1.4 31 37 48 59 73 85 93 Midwest 564 62 1.8 33 38 48 59 73 91 104 South 660 59 1-3 27 33 45 57 11 86 96 West 384 62 1.4 30 36 48 61 73 86 96 Urbanization Central city 499 63 1.5 30 37 46 59 75 92 106 Suburban 1039 60 9 30 36 48 59 72 86 95 Nonmetropolitan 518 59 3.2 30 34 46 59 71 83 95 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. Table II-11. Protein: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 16 .1 1a 12 14 16 19 21 23 Age 20-29 years 661 16 .2 11 12 14 16 18 20 22 30-39 years 812 16 wl 11 12 14 16 19 21 23 40-49 years 583 17 .2 12 12 14 17 19 22 23 Race 1/ White 1735 16 .1 11 12 14 16 18 21 23 Black 167 17 .4 11 12 14 17 19 23 25 Other 76 18 .8 * * 14 17 20 * x Poverty status 1/ < 100 315 17 3 1} 12 14 17 19 22 23 = > 100 1575 le = 11 12 14 16 18 21 23 | 8 131 414 16 2 11 12 14 16 19 21 23 > 131 1476 le i 11 12 14 16 19 21 23 Education 1/ < High school 305 17 -3 11 12 14 17 19 22 23 High school 854 le 2 11 12 14 16 19 21 23 > High school 891 16 wd 11 12 14 16 19 21 23 Region Northeast 448 17 2 12 13 15 17 19 21 23 Midwest 564 le .2 11 12 14 le 18 20 23 South 660 16 .2 11 12 13 le 18 21 23 West 384 16 "2 11 12 14 16 18 21 23 Urbanization Central city 499 17 2 1 12 14 17 19 21 23 Suburban 1039 le .1 11 12 14 16 19 21 23 Nonmetropolitan 518 le .3 11 12 14 16 18 20 22 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. Table II-12. Protein: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food es by Individuals, rere "Ss 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 30 95 All children 1/ 647 54 +9 32 37 43 52 61 73 83 Age 1-2 years 224 50 1.2 31 35 41 49 58 66 70 3-5 years 423 56 1.1 32 39 45 53 64 78 86 Race 2/ White 559 54 .8 32 37 43 52 & 73 ¢ Black 53 56 3.8 * * 44 54 INR di? oO Other 26 51 1.6 * * * 51 (L NOASS Poverty status 2/ < ood 140 55 2.0 32 37 45 53 64 80 86 > 00 471 53 1.0 33 37 43 52 60 72 81 — — & < 131 192 54 1.8 33 37 44 52 61 80 86 EN > 131 419 53 .9 33 37 43 52 61 12 80 Education 2/ < High school 99 56 2.0 * 35 45 53 65 80 AD High school 252 55 1.3 33 36 43 52 64 74 < > High school 295 52 1.1 32 38 43 51 59 70 Region Northeast ill 55 2.5 * 34 41 53 66 79 x Midwest 199 54 1.3 35 39 45 52 61 73 85 South 187 51 1.3 29 34 43 50 57 71 80 West 150 55 2.0 37 40 46 53 63 71 82 Urbanization Central city 171 56 2.2 34 39 46 53 65 78 86 Suburban 310 52 .9 32 36 43 51 59 69 79 Nonmetropolitan 166 56 1.6 32 35 45 54 64 77 83 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 8€-11 Table II-13. Protein: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 15 «1 11 12 13 15 17 18 19 Age 1-2 years 224 15 .2 11 13 13 15 17 19 20 3-5 years 423 15 v2 12 12 13 15 16 18 19 Race 2/ White 559 15 .1 11 12 13 15 16 18 19 Black 53 16 #5 * * 14 16 18 * * Other 26 17 .7 * % % 17 * * * Poverty status 2/ < 100 140 l6 .3 12 13 14 16 17 19 20 > 100 471 15 -2 11 12 13 15 17 18 19 < 131 192 16 .2 12 13 14 16 17 is 20 > 131 419 15 .2 11 12 13 15 le 18 19 Education 2/ < High school 99 le .4 x 13 14 15 18 20 ® High school 252 15 .2 12 i3 14 15 17 19 20 > High school 295 15 «2 11 12 i3 15 16 18 19 Region Northeast 111 15 .4 * 2 13 15 17 19 * Midwest 199 15 «3 12 12 14 15 17 18 20 South 187 15 3 11 12 13 15 16 18 19 West 150 15 “2 12 i3 14 15 17 19 19 Urbanization Central city 171 15 «3 12 13 13 15 16 18 19 Suburban 310 15 .2 11 12 i3 15 17 18 19 Nonmetropolitan 166 16 . 11 12 14 16 17 19 20 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 66-11 Table I-14. Protein: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES I NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 53 0.7 48 0.6 48 0.5 51 1.4 3-5 61 0.6 56 0.8 56 0.4 60 1.5 6-11 76 0.9 72 0.7 71 1.0 - - Male 12-15 97 2.2 94 1.2 92 2.3 - - 16-19, 118 3.5 106 1.6 122 3.9 - - 20-29, 112 2.7 102 1.4 113 2.9 105 4.1 30-39, 107 3.3 96 1.4 99 2.7 96 4.6 40-49 100 2.8 95 1.5 95 2.8 95 3.8 50-59 89 2.4 93 1.2 89 2.5 - - 60-69 80 1.1 84 1.3 79 0.9 - - 70+ 69 0.9 76 1.5 69 1.3 - - Female 12-15 73 1.8 72 1.0 66 1.6 - - 16-19 67 2.2 69 1.2 63 2.1 - - 20-29 67 0.8 67 0.9 64 1.3 65 2.3 30-39 65 0.9 65 0.9 63 1.7 66 1.4 40-49 65 1.2 66 0.9 62 1.8 63 1.2 50-59 63 1.8 66 1.0 56 1.7 - - 60-59 57 0.8 62 0.9 54 0.6 - - 70+ 51 0.7 57 0.8 49 0.9 - - i CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. ov-1I Table II-15. Protein: mean percent of kilocalories, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 16 0.1 16 0.1 15 0.1 16 0.3 3-5 is 0.1 16 0.1 14 0.1 16 0.2 6-11 i5 0.1 16 0.1 15 0.1 - - Male 12-15 15 0.2 16 0.1 15 0.2 - - 16-19, 16 0.3 16 0.1 16 0.3 - - 20-29 16 0.3 16 0.1 i5 0.2 15 0.5 30-39 16 0.4 16 0.2 15 0.3 16 0.4 40-49 16 0.3 17 0.2 16 0.3 17 0.6 50-59 16 0.3 17 0.2 16 0.3 - - 60-69 16 0.1 17 0.2 16 0.1 - - 70+ 16 0.1 17 0.2 16 0.2 - Female 12-15 15 0.2 16 0.1 14 0.3 - - 16-19 15 0.3 16 0.2 15 0.3 - - 20-29 16 0.1 17 0.1 15 0.2 16 0.3 30-39 16 0.2 17 0.1 15 0.3 17 0.2 40-49 17 0.2 17 0.2 16 0.3 17 0.2 50-59 17 0.3 17 0.2 16 0.3 - - 60-69 17 0.2 17 0.2 16 0.1 - - 70+ 16 0.2 17 0.2 15 0.2 - - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. ¥-1I U.S. Food Supply Fat Grams 200 180 + 160 140 + 120+ 100 + 80 + 60 40 + 20+ 1 1 1 1 1 1 1 0 1 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-5. Fat: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Fat Fat Fats and oils 47.2% Eggs 2.3% Other foods 3.7% Legumes, nuts, and soy 3.8% t fi 3 . ‘0 . Meat, poukiy, Ssh 31.42 Dairy products 11.6% Figure II-6. Fat: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include fruits; vegetables; grain products; and miscellaneous foods) ov-11 Table II-16. Fat: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 62 J 27 33 45 59 75 96 107 Age 20-29 years 661 64 1.0 27 34 47 60 79 98 13 30-39 years 812 62 1.0 28 33 44 59 76 96 107 40-49 years 583 60 1.3 27 32 44 58 72 90 105 Race 1/ White 17715 64 .17 29 34 46 60 77 96 108 Black 167 56 2.9 24 26 37 52 70 91 98 Other 76 53 2.4 * * 38 50 62 * * Poverty status 1/ < 100 315 58 1.8 25 28 38 53 72 91 101 > 100 1575 63 7 28 35 47 60 717 96 107 < 131 414 58 1.9 25 29 39 55 73 94 103 > 131 1476 64 7 29 35 47 60 717 96 108 Education 1/ < High school 305 52 1.6 22 27 35 48 64 86 95 High school 854 62 1.0 28 34 46 60 74 93 105 > High school 891 65 -9 29 38 48 62 79 99 11) Region Northeast 448 59 1.4 26 32 43 57 71 89 99 Midwest 564 67 1.4 28 34 48 63 81 102 113 South 660 60 1.2 26 31 42 58 74 93 106 West 384 65 1.3 29 37 47 61 77 97 112 Urbanization Central city 499 63 1.7 25 33 44 58 76 101 115 Suburban 1039 62 .8 28 33 46 60 75 94 107 Nonmetropolitan 518 62 1.5 28 32 44 59 75 90 102 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. ev-11 Table II-17. Fat: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 37 “2 26 29 33 37 41 44 47 Age 20-29 years 661 36 «2 26 29 32 37 41 43 45 30-39 years 812 37 .3 26 28 33 37 41 45 48 40-49 years 583 37 -3 217 29 33 37 41 45 47 Race 1/ White 1775 37 .2 217 29 33 37 41 45 47 Black 167 36 «7 26 29 33 36 40 44 45 Other 76 33 «7 * * 30 34 36 * * Poverty status 1/ < 100 315 36 «5 26 29 33 37 41 44 46 > 100 1575 37 "2 26 29 33 37 41 45 47 < 131 414 36 .4 25 29 32 37 41 44 46 > 131 1476 37 2 21 29 33 37 41 45 47 Education 1/ < High school 305 35 .4 24 27 31 36 40 43 45 High school 854 37 «3 27 29 33 37 41 45 47 > High school 891 37 2 27 29 33 37 41 44 47 Region Northeast 448 36 .4 26 29 33 36 40 44 45 Midwest 564 38 .2 27 30 34 38 42 45 47 South 660 36 3 26 28 32 36 40 44 47 West 384 37 5 27 29 33 37 42 45 48 Urbanization Central city 499 36 .4 25 27 32 36 41 44 47 Suburban 1039 37 “2 27 29 33 37 41 45 47 Nonmetropolitan 518 37 -4 26 29 33 37 11 44 46 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. Table II-18. Fat: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 55 1.2 33 36 43 53 66 77 87 Age 1-2 years 224 51 1.7 29 33 40 48 61 71 76 3-5 years 423 58 1.3 33 38 45 56 68 82 90 Race 2/ White 559 56 1.2 33 35 43 53 67 77 87 Black 53 57 3.4 * 44 56 66 * * Other 26 47 2.4 * * 47 * * * Poverty status 2/ < 100 140 57 1.7 33 38 44 55 66 77 83 > 100 471 55 1.4 32 36 43 52 67 76 89 < 131 192 56 1.8 33 37 44 53 66 78 85 > 131 419 56 1.4 32 36 43 53 68 76 89 Education 2/ < High school 99 58 2.4 * 38 44 54 70 83 * High school 252 56 1.5 33 36 44 55 67 78 90 > High school 295 54 1.6 32 35 42 51 64 74 85 Region Northeast 111 55 2.4 % 33 42 51 66 77 * Midwest 199 56 2.3 32 37 45 54 64 78 87 South 187 55 2.1 32 37 43 52 65 78 89 West 150 56 2.5 3 35 44 55 68 76 82 Urbanization Central city 171 58 2.6 34 39 47 56 70 8 89 Suburban 310 53 1.4 33 35 41 50 63 14 85 Nonmetropolitan 166 58 2.5 32 36 44 57 67 83 95 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. Sv-11 Table II-19. Fat: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 35 3 27 29 31 35 38 41 43 Age 1-2 years 224 35 +5 26 29 31 34 39 42 43 3-5 years 423 35 «4 217 28 31 35 38 41 43 Race 2/ White 559 35 .4 27 28 31 34 38 41 42 Black 53 36 .6 * * 33 35 38 x * Other 26 35 1.3 * * * 34 * x x Poverty status 2/ 100 140 36 .6 29 31 34 36 39 42 44 > 100 471 34 .4 26 28 31 34 38 41 43 < 131 192 36 .6 28 30 33 36 39 42 44 > 131 419 34 .4 26 28 31 34 38 41 43 Education 2/ < High school 99 37 .7 * 31 33 36 40 44 * High school 252 35 «5 28 29 32 35 39 42 43 > High school 295 34 .4 26 28 31 34 37 39 42 Region Northeast aia 34 .8 * 28 31 34 36 40 * Midwest 199 35 +5 28 29 32 35 38 41 43 South 187 35 .6 27 29 32 35 39 41 42 West 150 35 7 28 29 31 35 38 42 43 Urbanization Central city 171 35 .6 28 30 32 35 38 41 42 Suburban 310 34 .4 26 28 31 34 37 41 42 Nonmetropolitan 166 36 .9 27 28 34 36 39 43 44 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 9v-11 Table II-20. Fat: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANESI), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-178; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES I NFCS NHANES II CSFI1 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 56 0.8 50 0.7 51 0.6 51 1.6 3-5 68 0.7 62 1.0 62 0.5 60 1.4 6-11 83 1:1 81 0.8 79 1.1 - - Male 12-15 108 2.6 108 1.7 102 2.8 - - 16-19, 127 4.0 120 2.1 126 4.0 - - 20-29, 119 3.1 115 1.9 118 3.2 115 4.5 30-397 111 3.6 112 1.8 105 3.0 103 5.5 40-49 101 3.1 111 2.0 103 3.5 101 4.6 50-59 91 3.3 107 2.7 91 2.8 - - 60-69 81 1.3 96 2.0 82 1.0 - - 70+ 73 1.0 86 2.2 71 1.6 - - Female 12-15 80 2.1 83 1.5 76 2.1 - - 16-19 72 2.4 77 1.5 69 2.3 - - 20-29 68 0.9 75 1.1 67 1.5 68 1.4 30-39 67 1.0 73 1.0 66 1.9 68 1.4 40-49 65 1.2 74 1.1 64 2.1 65 1.5 50-59 59 1.8 72 2.2 58 2.9 - - 60-69 53 0.8 67 2x2 53 0.7 - 70+ 49 0.8 60 0.9 48 0.9 - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. Ly-11 Table II-21. Fat: mean percent of kilocalories, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 37 0.2 37 0.3 36 0.2 35 0.4 3-5 36 0.2 37 0.3 36 0.1 35 0.4 6-11 37 0.2 38 0.2 36 0.2 - - Male 12-15 37 0.4 39 0.2 37 0.4 - - 16-19 38 0.5 40 0.2 38 0.5 - - 20-29 37 0.5 41 0.3 36 0.4 36 0.7 30-39 37 0.6 42 0.3 37 0.5 36 0.8 40-49 37 0.6 42 0.3 38 0.5 37 0.8 50-59 38 0.6 42 0.3 37 0.5 - - 60-69 37 0.3 42 0.4 38 0.2 - - 70+ 37 0.3 41 0.5 37 0.4 - - Female 12-15 38 0.4 39 0.3 37 0.5 - - 16-19 37 0.5 40 0.3 37 0.5 - - 20-29 37 0.2 40 0.3 36 0.4 36 0.4 30-39 a7 0.2 41 0.3 37 0.5 37 0.3 40-49 37 0.3 42 0.3 38 0.5 37 0.4 50-59 36 0.5 41 0.3 37 0.5 - - 60-69 36 0.3 40 0.3 36 0.2 - - 70+ 36 0.3 38 0.4 35 0.4 - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. 8v-1I U.S. Food Supply Saturated fat Grams 80 70} 60 nT TTT 50 } 40 30 20+ 10+ 1 i 1 1 4 Xk 1 1 i 1 L 1 1 1 0 i i 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-7. Saturated fat: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Saturated Fat Saturated fat Meat, poultry, fish 39.2% Legumes, nuts, and soy 1.8% Eggs 1.9% Fats and oils 34.1% Other foods 3.4% Dairy products 19.6% Figure II-8. Saturated fat: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include fruits; vegetables; grain products; and miscellaneous foods) 6v-11 Table II-22. Saturated fat: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 23 .3 9 11 16 21 28 36 41 Age 20-29 years 661 24 .5 9 12 17 22 30 38 44 30-39 years 812 23 .4 9 11 16 21 28 36 40 40-49 years 583 22 «5 9 11 i5 20 27 33 39 Race 1/ White 1775 24 «3 10 12 16 22 29 37 42 Black 167 19 1.1 8 9 11 19 24 33 40 Other 76 19 9 * * 14 19 22 * * Poverty status 1/ < 100 315 21 17 8 10 13 20 27 33 39 > 100 1575 23 .3 9 12 16 22 29 36 41 < 3131 414 21 «7 7 10 14 20 27 34 39 > 131 1476 23 «3 9 iz 16 22 29 36 41 Education 1/ < High school 305 19 6 7 9 12 17 23 32 36 High school 854 23 .5 9 11 16 21 28 36 40 > High school 891 24 .4 10 12 17 23 30 37 44 Region Northeast 448 22 .6 9 11 15 20 27 33 40 Midwest 564 25 .6 10 12 17 23 31 38 44 South 660 22 .6 9 10 14 20 28 34 40 West 384 24 .5 10 12 17 22 28 37 42 Urbanization Central city 499 23 .6 9 11 15 21 28 39 44 Suburban 1039 23 .4 10 11 16 22 29 36 41 Nonmetropolitan 518 23 .6 10 13 15 21 28 35 38 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 0g-11 Table 11-23. Saturated fat: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Individuals, 1985-86 Continuing Survey of Food Intakes by All women Age 20-29 years 30-39 years 40-49 years Race 1/ White Black Other Poverty status 1/ 100 > 100 < 131 > 131 Education 1/ < High school High school > High school Region Northeast Midwest South West Urbanization Central city Suburban Nonmetropolitan 1/ Some women did not report race, poverty status, or education. do not add to the number of all women. 661 812 583 1775 167 76 315 1575 414 1476 305 854 891 448 564 660 384 499 1039 518 13 13 13 14 13 12 13 13 13 13 i3 13 13 13 14 13 14 — © © @® © © © WYO 10 10 10 10 11 11 11 12 11 10 11 11 11 11 11 11 ai 11 12 22 12 13 13 13 14 12 12 13 13 13 13 13 13 14 13 14 13 14 15 15 15 15 15 i5 15 15 15 15 15 15 15 15 16 15 le 15 18 17 17 17 16 17 17 17 17 16 17 17 17 17 17 18 17 Therefore, the numbers of women in each category 19 18 18 19 17 18 18 18 18 17 13 18 18 19 18 19 T1e-11 Table 11-24. Saturated fat: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 22 +5 12 14 17 21 26 31 35 Age 1-2 years 224 21 +7 10 13 16 20 25 30 32 3-5 years 423 23 +5 13 14 17 22 27 32 36 Race 2/ White 559 22 .5 12 14 17 21 27 31 36 Black 53 21 3.2 x * 17 20 24 * * Other 26 19 .9 * * * 19 * * * Poverty status 2/ < 100 140 22 .6 13 14 18 23 26 30 34 > 100 471 22 .6 12 13 17 21 26 32 36 < a3 192 22 .6 13 14 17 22 26 31 35 > 131 419 22 .6 12 13 17 21 27 31 36 Education 2/ < High school 99 23 .9 * 15 18 22 28 32 * High school 252 22 .6 12 13 16 22 26 31 36 > High school 295 22 .6 12 14 17 21 25 30 34 Region Northeast 111 22 .9 * 14 17 21 27 32 * Midwest 199 23 1.0 13 15 18 22 26 32 38 South 187 22 .9 11 12 17 21 25 31 34 West 150 22 1.13 13 14 17 20 27 30 33 Urbanization Central city 171 23 1.0 14 15 18 22 26 31 34 Suburban 310 21 .6 12 13 16 20 25 30 34 Nonmetropolitan 166 23 1.1 12 14 17 23 28 33 37 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. ¢s-11 Table II-25. Saturated fat: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 14 2 10 11 12 14 l6 17 18 Age 1-2 years 224 14 .2 9 10 12 14 le 18 18 3-5 years 423 14 2 10 1 12 14 15 17 18 Race 2/ White 559 14 v2 10 10 12 14 16 17 18 Black 53 14 «2 * * 13 14 15 * * Other 26 15 .7 * ® * 13 * % * Poverty status 2/ < 100 140 14 .3 11 11 13 14 16 18 18 > 100 471 14 .2 10 10 12 14 16 17 18 < 131 192 14 .3 11 11 13 14 16 18 18 > 131 419 14 .2 9 10 12 13 16 17 18 Education 2/ < High school 99 15 .3 * 12 13 14 16 18 * High school 252 14 -2 10 1 12 14 16 17 18 > High school 295 14 «2 9 10 12 13 15 17 18 Region Northeast 111 14 .4 x 10 12 13 16 17 A Midwest 199 14 «2 10 11 13 14 16 17 18 South 187 14 »3 9 10 12 14 16 17 18 West 150 14 .3 10 10 12 13 15 17 18 Urbanization Central city 171 14 .3 10 11 12 14 15 17 18 Suburban 310 14 .2 9 10 12 13 16 17 18 Nonmetropolitan 166 15 .4 10 11 13 14 16 18 19 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. €a-11 Table 11-26. Saturated fat: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES 11 CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 22 0.3 - - 19 0.2 21 0.7 3-5 26 0.3 - - 23 0.2 24 0.6 6-11 32 0.5 - - 29 0.4 - - Male 12-15 40 1.1 - - 38 1.1 - - 16-19, 46 1.6 - - 48 1.6 - - 20-29 43 1.3 - 43 1.2 42 1.8 30-39 40 1.4 - - 38 1.1 37 2.0 40-49 37 1.3 - - 38 1.3 37 2.2 50-59 33 1.3 - - 33 1.0 - 60-69 30 0.5 - 29 0.4 - 70+ 26 0.4 - 25 0.6 - - Female 12-15 30 0.9 - - 27 0.8 - 16-19 26 1.0 - - 25 0.8 - - 20-29 25 0.4 - - 24 0.5 25 0.6 30-39 24 0.4 - 23 0.7 25 0.5 40-49 23 0.5 - 23 0.7 23 0.6 50-59 22 0.8 - - 20 0.7 - - 60-69 19 0.3 - 18 0.2 - - 70+ 18 0.3 - 16 0.3 - - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. ya-11 U.S. Food Supply Monounsaturated fat Grams 80 40} 20 1 1 i 1 1 1 0 i 1 i 1 i 1 i i 1 i 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-9. Monounsaturated fat: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Monounsaturated Fat Monounsaturated fat Fats and oils 48.4% | Eggs 2.3% Other foods 2.4% Legumes, nuts, and soy 4.4% Meat, poultry, fish 34.5% Daly products 82 Figure II-10. Monounsaturated fat: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include fruits; vegetables; grain products; and miscellaneous foods) Table II-27. Monounsaturated fat: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 15 90 95 All women 2056 23 .3 10 12 16 22 28 35 40 Age 20-29 years 661 23 .4 10 12 17 22 29 37 41 30-39 years 812 23 .4 10 12 16 21 28 35 39 40-49 years 583 22 +5 10 12 l6 21 27 34 38 Race 1/ White 1775 23 2 10 12 17 22 28 36 40 Black 167 21 1.1 8 10 14 19 27 33 38 Other 76 20 1.0 * x 13 19 23 * * Poverty status 1/ < 100 315 22 .8 9 10 14 21 27 35 39 > 100 1575 23 +3 10 12 17 22 28 36 40 < 131 414 22 .8 9 11 15 21 27 35 40 > 131 1476 23 “3 10 13 17 22 28 35 40 Education 1/ < High school 305 20 +7 8 10 13 18 24 33 37 High school 854 23 .4 10 12 17 22 28 35 38 > High school 891 24 3 10 i3 17 23 29 37 41 Region Northeast 448 21 5 10 12 16 20 26 33 37 Midwest 564 24 +5 10 12 18 23 30 37 41 South 660 22 .4 9 12 16 22 27 34 40 West 384 24 .6 10 13 17 22 28 37 41 Urbanization Central city 499 23 .6 9 12 16 21 28 37 41 Suburban 1039 23 .3 10 12 16 22 28 35 39 Nonmetropolitan 518 23 .6 10 12 16 22 28 34 37 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. Table 11-28. Monounsaturated fat: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 13 .1 9 10 12 13 15 17 18 Age 20-29 years 661 13 .1 9 10 12 13 15 16 17 30-39 years 812 13 .1 9 10 12 13 15 17 18 40-49 years 583 14 .1 10 11 12 14 16 17 18 Race 1/ White 1775 14 .1 10 10 12 14 15 17 18 Black 167 14 .4 9 10 12 13 15 17 18 Other 76 i3 .3 x = 10 13 14 * * Poverty status 1/ < 100 315 14 2 9 10 12 14 16 13 18 = > 100 1575 13 .1 10 10 12 13 15 17 18 ; = & < 131 414 14 el 9 10 12 14 16 17 18 > 131 1476 13 .1 10 10 12 13 15 17 18 Education 1/ < High school 305 13 «2 9 10 12 13 15 17 18 High school 854 14 .1 10 1 32 14 15 17 18 > High school 891 13 .1 9 10 12 13 15 17 17 Region Northeast 448 13 2 9 10 12 13 15 16 17 Midwest 564 14 -3 10 11 12 14 15 17 18 South 660 13 .d 9 10 12 13 15 17 18 West 384 14 .2 10 10 12 14 15 17 18 Urbanization Central city 499 i3 .2 9 10 12 13 15 17 x7 Suburban 1039 a3 «1 10 10 12 13 15 17 18 Nonmetropolitan 518 14 .2 9 11 12 14 15 17 18 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. Table 11-29. Monounsaturated fat: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 20 .4 11 13 16 19 24 28 33 Age 1-2 years 224 18 .6 11 12 14 18 22 26 27 3-5 years 423 21 .5 12 14 16 20 25 31 33 Race 2/ White 559 20 .5 12 13 16 19 24 28 33 Black 53 22 1.4 % ® le 22 26 * * Other 26 17 1.0 ® x 18 * * * Poverty status 2/ < 100 140 21 «7 12 13 16 20 25 29 32 > 100 471 20 .5 11 13 15 19 24 28 33 = o < 131 192 21 1 12 13 16 20 24 29 32 = > 131 419 20 «5 11 13 16 19 24 28 33 | Education 2/ < High school 99 22 «9 * 13 16 20 217 31 * High school 252 21 .6 11 13 16 20 25 28 33 > High school 295 20 .6 11 12 15 18 23 27 31 Region Northeast 111 20 1.0 * 12 15 18 24 28 * Midwest 199 20 .9 11 13 16 19 24 29 32 South 187 20 «8 12 13 16 19 24 28 34 West 150 20 .9 11 12 15 20 26 28 32 Urbanization Central city 171 21 1.1 11 13 16 20 26 29 34 Suburban 310 19 .6 12 12 15 18 24 28 31 Nonmetropolitan 166 21 9 11 12 16 21 25 31 34 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 8¢-1I Table II-30. Monounsaturated fat: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 13 -1 9 10 11 13 14 15 16 Age 1-2 years 224 13 .2 9 10 11 13 14 15 16 3-5 years 423 13 .1 10 10 11 13 14 15 16 Race 2/ White 559 13 .2 9 10 11 13 14 15 16 Black 53 14 .3 * 12 14 15 * * Other 26 12 .5 * * * 13 * * * Poverty status 2/ < 100 140 14 .3 1a a1 12 14 15 16 le > 100 471 12 .1 9 10 11 13 14 15 16 < 131 192 13 «3 10 12 12 13 15 16 16 > 131 419 12 2 9 10 11 12 14 15 16 Education 2/ < High school 99 14 «3 * 31 2 14 15 16 ” High school 252 13 2 10 10 11 13 15 15 16 > High school 295 12 “2 9 10 11 12 13 15 15 Region Northeast 111 12 -4 * 10 11 12 14 15 x Midwest 199 13 2 10 10 11 13 14 15 16 South 187 13 .3 9 10 11 13 15 16 16 West 150 13 «3 9 10 12 13 14 15 15 Urbanization Central city 171 13 2 10 11 12 13 14 15 15 Suburban 310 12 se 9 10 11 12 14 15 16 Nonmetropolitan 166 13 .3 9 10 12 13 15 16 16 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 69-11 Table II-31. Monounsaturated fat’: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 20 0.3 - 18 0.2 18 0.7 3-5 25 0.3 - - 22 0.2 22 0.5 6-11 30 0.4 - - 28 0.4 - - Male 12-15 39 1.1 - - 37 1.0 - 16-19, 46 1.5 - - 47 1.4 - - 20-29 44 1.2 - - 44 1.1 44 1.7 30-39, 42 1.4 - - 39 1.2 40 2.5 40-49 39 i.3 - - 39 1.3 38 1.8 50-59 35 1.3 - 34 1.0 - - 60-69 32 0.5 - - 31 0.4 - - 70+ 28 0.4 - 26 0.6 - - Female 12-15 30 0.8 - - 27 0.8 - 16-19 26 1.0 - - 25 0.8 - - 20-29 26 0.4 - - 24 0.5 25 0.5 30-39 26 0.4 - - 24 0.7 25 0.5 40-49 25 0.5 - - 24 0.7 24 0.6 50-59 23 0.7 - - 21 0.7 - - 60-69 21 0.3 - - 20 0.2 70+ 19 0.3 - - 17 0.4 i Oleic acid measured in NHANES I and NHANES II. 2 CSFII data for 1985 only. 3 Ages 70-74 years only for NHANES 1 and NHANES II. 09-11 U.S. Food Supply Polyunsaturated fat Grams 35 30+ 20+ 1 1 1 1 1 1 1 0 1 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-11. Polyunsaturated fat: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Polyunsaturated Fat Polyunsaturated fat Eggs 1.6% Other foods 1.7% Dairy products 2% Grain products 3% Legumes, nuts, and soy 5.9% Meat, poultry, fish 17.6% Figure I-12. Polyunsaturated fat: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include fruits; vegetables; and miscellaneous foods) 19-11 Table 11-32. Polyunsaturated fat: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 12 .1 5 6 8 11 15 19 23 Age 20-29 years 661 12 .3 5 6 8 11 15 19 23 30-39 years 812 12 .2 5 6 8 11 15 19 23 40-49 years 583 12 .3 5 6 8 11 15 19 23 Race 1/ White 1775 12 2 5 6 9 12 16 19 23 Black 167 11 .6 4 5 7 10 13 18 22 Other 76 10 .1 * * 6 9 13 * * Poverty status 1/ < 100 315 11 .3 4 5 7 9 13 17 20 > 100 1575 13 .2 5 6 9 7 16 20 23 < 131 414 11 .4 4 5 7 10 14 17 22 > 131 1476 13 .2 5 6 9 12 16 20 23 Education 1/ < High school 305 10 .4 3 4 6 9 12 17 20 High school 854 12 .2 5 6 9 11 15 19 22 > High school 891 13 .2 6 7 9 12 16 20 24 Region Northeast 448 11 .2 5 6 8 11 14 18 20 Midwest 564 13 .3 5 6 9 12 16 21 24 South 660 12 .3 5 6 8 11 15 18 21 West 384 13 .3 5 6 8 13 16 20 24 Urbanization Central city 499 12 .4 4 6 8 11 15 19 24 Suburban 1039 12 2 5 6 9 12 15 19 23 Nonmetropolitan 518 12 .3 5 6 8 11 15 19 21 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. Table II-33. Polyunsaturated fat: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 7 .1 4 5 6 7 9 10 i1 Age 20-29 years 661 7 al 4 5 6 7 8 10 ii 30-39 years 812 7 ed 4 5 6 7 9 10 12 40-49 years 583 7 «1 4 5 6 7 9 10 11 Race 1/ White 1775 7 +1 4 5 6 7 9 10 11 Black 167 7 +2 4 5 6 7 9 10 11 Other 76 6 .3 * * 5 6 7 * * Poverty status 1/ < 100 315 7 2 4 4 5 6 8 9 11 ne > 100 1575 7 «1 4 5 6 7 9 10 11 ; 2 aN ND < 131 414 7 «2 4 4q 5 7 8 10 11 > 131 1476 7 .1 4 5 6 7 9 10 11 Education 1/ < High school 305 7 v2 4 4 5 6 8 9 11 High school 854 7 .1 4 5 6 7 8 10 11 > High school 891 7 ol 4 5 6 7 9 10 12 Region Northeast 448 7 +d 4 5 6 7 8 10 11 Midwest 564 7 «1 4 5 6 7 9 10 11 South 660 7 " 4 5 6 7 8 10 11 West 384 7 .1 4 5 6 7 9 11 12 Urbanization Central city 499 7 .1 4 5 6 7 8 10 11 Suburban 1039 7 .1 4 5 6 7 9 10 12 Nonmetropolitan 518 7 1 4 5 6 7 8 10 11 | 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. Table 11-34. Polyunsaturated fat: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 9 .2 5 5 7 9 11 14 16 Age 1-2 years 224 8 .4 4 5 6 8 11 12 14 3-5 years 423 10 .2 5 6 7 9 12 15 16 Race 2/ White 559 9 2 5 5 7 9 11 14 16 Black 53 10 «7 * * 7 9 11 * * Other 26 7 .4 * * * 7 * * * Poverty status 2/ < 100 140 9 .4 5 6 7 8 11 13 15 > 100 471 9 Es 4 5 7 9 11 14 le < 131 192 9 .4 4 5 6 8 1 13 15 > 131 419 10 .3 4 5 7 9 12 14 16 Education 2/ < High school 99 9 .4 * 5 7 8 11 14 * High school 252 10 «3 5 7 9 11 15 16 > High school 295 9 .3 4 5 7 9 11 14 15 Region Northeast 11) 9 .5 x 5 6 9 11 13 * Midwest 199 9 .3 4 5 7 9 ia 13 15 South 187 9 .4 5 5 7 8 11 15 16 West 150 10 .4 5 5 7 10 12 15 16 Urbanization Central city 171 10 «5 4 5 7 10 32 16 16 Suburban 310 9 2 5 5 6 9 11 13 14 Nonmetropolitan 166 9 .5 4 5 7 8 11 15 17 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. Table 11-35. Polyunsaturated fat: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 6 .1 4 4 5 6 7 8 9 Age 1-2 years 224 6 1 3 4q 5 6 © 8 9 3-5 years 423 6 «1 4 4 5 6 7 8 8 Race 2/ White 559 6 3 4 4 5 6 7 8 9 Black 53 6 2 * * 5 6 7 * * Other 26 5 72 * * 5 * x * Poverty status 2/ < 100 140 6 .1 4 4 5 6 6 7 8 > 100 471 6 .1 4 4 5 6 7 8 9 < 131 192 6 wl 4 4 5 6 6 7 8 > 131 419 6 .1 4 4 5 6 7 8 9 Education 2/ < High school 99 6 “ld * 4 5 6 6 7 * High school 252 6 .1 4 5 6 7 8 9 > High school 295 6 «1 4 4 5 6 6 7 8 Region Northeast 112 5 el ¥ 4 5 5 6 7 * Midwest 199 6 | 4 4 5 6 6 7 8 South 187 6 wd 4 4 5 6 7 7 8 West 150 6 .1 4 4 5 6 7 8 9 Urbanization Central city 171 6 .1 4 4 5 6 7 8 9 Suburban 310 6 «1 4 4 5 6 6 7 8 Nonmetropolitan 166 6 2 4 4 5 6 7 7 8 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 99-11 Table II-86. Polyunsaturated fat': mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES I NFCs NHANES II ¢ CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 4 0.1 - 6 0.1 8 0.3 3-5 6 0.1 8 0.1 10 0.3 6-11 8 0.2 - 10 0.2 - - Male 12-15 10 0.4 - - 14 0.5 - - 16-19 13 0.7 - - 16 0.6 - - 20-29 13 0.6 - - 16 0.5 21 2.2 30-39, 12 0.7 - - 15 0.5 19 1.0 40-49 10 0.5 - - 14 0.5 18 0.9 50-59 10 0.6 - - 12 0.5 - - 60-69 8 0.2 - - 11 0.2 - - 70+ 7 0.2 - - 10 0.3 - - Female 12-15 8 0.3 - - 31 0.4 - 16-19 8 0.5 - - 10 0.4 - - 20-29 8 0.2 - - 10 0.3 13 0.3 30-39 8 0.2 - - 10 0.4 13 0.4 40-49 7 0.2 - - 10 0.5 13 0.4 50-59 6 0.3 - - 8 0.3 - - 60-49 6 0.1 - - 8 0.1 - - 70+ 5 0.1 - - 7 0.2 - - 1 Linoleic acid measured in NHANES I and NHANES II. 2 CSFII data for 1985 only. 3 Ages 70-74 years only for NHANES I and NHANES II. 99-11 U.S. Food Supply Cholesterol Milligrams 800 700 + 600 500 400 + 300 t+ 200 + 100 + oor 1 1 1 1 AL. 1 1 0 1905 1915 1925 1935 1945 Year 1955 1965 1975 1985 Figure II-13. Cholesterol: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Cholesterol Cholesterol Meat, poultry, fish 42.7% Figure II-14. Cholesterol: food sources in the U.S. food supply, 1985: U.S. Food Supply Series L9-11 Table II-37. Cholesterol: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 2717 4.3 97 122 174 251 354 467 544 Age 20-29 years 661 2717 6.7 90 118 168 245 363 470 550 30-39 years 812 276 6.7 102 126 176 251 346 463 542 40-49 years 583 279 6.5 99 123 178 255 359 462 528 Race 1/ White 1775 271 4.5 95 122 173 247 345 455 542 Black 167 315 18.2 102 122 175 303 431 514 614 Other 76 2717 14.9 * x 175 259 377 * * Poverty status 1/ < 100 315 297 13.4 86 119 172 257 391 518 592 > 100 1575 274 4.6 99 122 176 251 347 455 533 < 131 414 287 10.8 83 111 171 250 364 503 575 > 131 1476 275 4.5 100 125 176 253 349 460 532 Education 1/ < High school 305 262 11.0 81 101 155 231 368 508 582 High school 854 285 7.1 95 122 176 253 362 475 550 > High school 891 275 5.8 107 130 177 253 347 457 532 Region Northeast 448 285 7.2 106 127 171 253 382 471 550 Midwest 564 278 10.0 99 120 174 251 351 453 556 South 660 270 7.4 84 118 173 243 337 465 537 West 384 279 8.2 100 127 185 262 349 468 534 Urbanization Central city 499 305 10.3 100 129 180 2717 385 503 563 Suburban 1039 267 5.7 97 122 171 238 345 454 541 Nonmetropolitan 518 266 7.8 87 112 171 233 337 451 528 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 89-11 Table I-38. Cholesterol: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 15 90 95 All children 1/ 647 228 5.7 87 108 147 201 292 378 445 Age 1-2 years 224 219 8.9 75 100 143 195 276 364 429 3-5 years 423 233 6.4 94 114 148 204 300 382 449 Race 2/ White 559 222 6.2 86 103 145 195 292 365 426 Black 53 260 20.6 * * 179 236 285 * * Other 26 296 36.5 * * * 295 * * * Poverty status 2/ < 100 140 269 5.2 105 124 169 236 330 458 564 > 100 471 220 6.5 94 106 145 194 285 358 419 < 131 192 257 12.2 103 124 168 229 318 415 533 > 131 419 219 6.9 90 103 143 194 285 358 422 Education 2/ < High school 99 273 18.6 * 140 180 243 328 523 * High school 252 234 7.9 95 115 165 211 299 382 426 > High school 295 211 7.6 87 100 137 183 266 350 418 Region Northeast 111 238 15.6 x 94 131 205 313 429 * Midwest 199 215 11.3 89 100 140 199 268 342 380 South 187 225 9.3 98 114 154 194 287 365 416 West 150 238 9.9 94 117 151 206 303 415 477 Urbanization Central city 171 239 12.4 108 124 165 217 292 416 458 Suburban 310 214 8.0 87 99 138 188 264 358 414 Nonmetropolitan 166 249 8.9 94 114 173 228 307 407 445 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 69-11 Table II-39. Cholesterol: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES I NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 284 6.3 - 227 4.2 247 12.5 3-5 290 4.9 - - 245 3.2 261 10.6 6-11 314 6.8 - 278 7.0 - - Male 12-15 379 13.6 - - 359 15.4 - - 16-19 518 23.8 - - 493 21.4 - - 20-29, 513 19.6 - - 453 16.7 466 30.8 30-39 517 22.8 - - 453 18.5 423 24.9 40-49 497 20.0 - - 442 19.5 435 26.1 50-59 470 18.9 - - 437 17.8 - - 60-69 424 8.6 - - 415 7.2 - - 70+ 424 10.7 - - 365 11.1 - - Female 12-15 305 11.3 - - 256 11.2 - 16-19 297 14.2 - - 255 12.4 - - 20-29 304 5.9 - - 270 9.1 302 10.6 30-39 312 6.5 - - 289 11.6 304 8.6 40-49 340 10.0 - - 305 14.7 298 8.8 50-59 317 14.4 - - 263 12.1 - - 60-69 298 6.6 - 262 4.9 - 70+ 256 5.7 - 216 7.3 - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. oL-1I Table II-40. Serum cholesterol status of persons 20-74 years of age by sex, specified Hispanic origin, and age: Hispanic Health and Nutrition Examination Survey, 1982-84 T T TT T T | ; | Percent | | | with Standard | Number of I Mean serum | Standard | high-risk | error Sex, Hispanic origin, | examined | cholesterol error of | serum of the and age persons Vv (mmol/L) 2/ | the mean jsholesterol percent 1 1 1 1 1 MALE Mexican American 20-74 y@ArS. ...:ixsvivvisinn us 1,407 5.26 0.03 16.2 1.1 20-74 years, age adjusted.... L 5.35 15.5 20-29 years... :xw:smmpsmns 429 4.87 0.06 18.0 2:3 30-39 years. ................. 358 5.33 0.06 16.6 2.3 BO0=48 YEAS. : us sms sais swiss 232 5.70 0.07 18.1 2.2 BO-59 years... uous smuusnws 229 5.68 0.06 14.5 2.1 60-69 years. ...... i s40s ns 121 5.58 0.08 12.6 2.8 70-74 y@aArs. . vu: :uwws sums imme» 38 *5.41 0.19 15.8 5.9 Cuban 20-74 years. ................. 366 5.40 0.06 14.6 1.9 20-74 years, age adjusted.... 240 5.28 vi 13.7 20-29 years. ................. 52 4.73 0.16 12.2 5.5 30-39 years. .: ::zi:cnns smn 53 5.12 0.17 12.2 5.3 A0~48 YRBIrS, . cor vamsvwwrnwysn 81 5.70 0.12 14.8 4.1 50-59 years. ................. 107 5.65 0.09 18.4 3.5 60-69 years. ......s:nsassnnen 45 5.55 0.14 8.4 3.9 70-74 years. .....:ccivvuwsinnwwes 28 *5.95 0.16 x24 .1 7.6 Puerto Rican 20-74 y8ArsS. ....ciciuinssmman 422 5.18 0.07 13.3 2.1 20-74 years, age adjusted... .. C. 5.26 .. 13.5 20-29 yBAPrS. cnn invs swap res 110 4.63 0.12 8.5 4.0 30-39 years. ................. 85 5.10 0.16 11.8 5.0 40-49 YEAS. ..::.u: ins: inn: is 83 5.55 0.16 16.3 5.1% BO-59 yeArS....:su5: v5: mwas 97 5.82 0.12 20.6 4.2 60-69 years... ............... 37 *5.73 0.16 *11.3 5.6 10-74 YOArsS. uu: ies suns swns 5 10 * * * * 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ mmol/L = mg/d1+0.02586. 12-1 Table II-40. Serum cholesterol status of persons 20-74 years of age by sex, specified Hispanic origin, and age: Hispanic Health and Nutrition Examination Survey, 1982-84--continued T T T T T Percent I | | I with | Standard | Number of | Mean serum | Standard | high-risk I error Sex, Hispanic origin, | examined | cholesterol | error of | serum | of the and age | persons a (mmol/L) 2/ the mean jcholesterol, percent 1 1 1 1 A FEMALE Mexican American 20~74 years. ci uns inns env smnss 1,797 5.23 0.03 14.6 0.8 20-74 years, age adjusted. ... C. 5.35 . 15.5 20-29 YEAYS.: : us. mms ams sms + 534 4.85 0.05 16.7 1.7 30-39 years. ..........a.. 449 5.02 0.05 10.1 1.4 40-49 years... .......... 305 5.45 0.06 9.1 1.5 BOOBY YBAFS. ox: ewe sums smmanmms » 311 5.81 0.06 15.8 1.7 60-69 years. ................. 139 6.00 0.09 23.1 3.2 70-74 y8aArS. si: wn: sinms sma s wus» 59 8.19 0.14 33.4 5.6 Cuban 20-74 years. ......... cu... 460 5.23 0.05 11.3 1.5 20-74 years, age adjusted... .. 5.16 4 11.2 20-29 years... ......iii.. 62 4.60 0.10 8.7 4.1 BO-39 YOAPS: .: sna wms sms umes 91 4.67 0.09 5.9 2.8 40-49 YRBI'S, «iv uviwums suis vwws 101 5.13 0.07 4.1 2.0 50-59 years. ............ 110 5.98 0.10 20.4 3.6 B0~69 YEAS. vw: swans ipus ams 66 5.94 0.11 18.9 4.4 70-74 years... . c:sviinsxisnws 30 *5.82 0.20 *22.9 7.2 Puerto Rican 20-74 years. ........c.ouuiiun.. 716 5.25 0.06 15.4 1.7 20-74 years, age adjusted. ... 5.39 17.5 20-29 years... ........ in. 181 4.84 0.13 16.2 3.8 30-39 years. ................. 164 4.97 0.11 10.8 3.3 40-49 YBAIS. .. cv swsiuvismra 159 5.53 0.11 13.7 3.2 50-59 years. ................. 127 5.96 0.09 19.2 3.5 60-89 Years. .: vs smsinns imme 71 6.13 0.12 21.6 4.7 FO~78 YEAS. «i. vmtmmuiswme smme 14 * * * * 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ mmol/L = mg/d1+0.02586. Table II-41. Serum cholesterol status of Mexican-American persons 20-74 years of age by sex, poverty status, and age: Hispanic Health and Nutrition Examination Survey, 1982-84 | | | Percent with Standard Number of Mean serum Standard high-risk error examined cholesterol error of serum of the Sex, poverty status, and age persons 1/| (mmol/L) 2/ the mean cholesterol percent MALE Below poverty 20-74 YAS. ci tv was snus pwns 310 5.14 0.06 12.2 2.0 20-74 years, age adjusted... .. 5.27 ‘3 12.1 20-29 YAS vt stv nv ssw sw 98 4.71 0.11 11.2 4.0 30-39 vears. ............n 65 5.28 0.13 18.3 5.2 A0=49 YRArS 1 sc vws s vows spa 46 5.59 0.13 9.9 4.3 S0-59 years. ................. a7 5.59 0.15 11.0 4.0 80-88 VYRArS .u. cons + summa: diss s 40 *5.62 0.13 *12.2 5.0 70-74 years... ........c. oo... 14 * * * * Above poverty 2074 YBAPS. ox sv viii ome hans 981 5.30 0.04 17.5 1.3 20-74 years, age adjusted. ... 5.39 16.9 20-29 YRAPrS. «i suas vines sama 294 4.93 0.07 20.2 2.8 30-39 years. ............... 272 5.36 0.07 17.4 2.7 RO~49 YRArS. «+ sons ssn as swwus 166 5.7% 0.08 14.4 2.7 BOB YBBIS . .. civnv conmns cmmns 158 5.67 0.07 13.7 2.4 80-69 YEArsS. i: visser i sawn 70 5.69 O.11 14.8 3.9 0-74 YOBPS , vu sume s sw cummin 21 i. % x = FEMALE Below poverty 20-74 years. ........ 550 5.18 0.04 12.2 1.4 20-74 years, age adjusted. ... 5.258 “wy 12.6 20-29 years... ... 154 4.72 0.08 10.7 2.7 30-30 years. vu: vn sins rams 131 5.098 0.08 9.1 2.5 40-49 years. ..........i.. 88 5.31 0.11 8.9 2.8 BO=BY Y@ArS. «uv rns samns biases 96 5.59 0.09 13.6 3.0 60-69 years. ........... 46 5.85 0.14 20.1 5.6 70-74 years. ........ uu... 35 *5.27 0.19 «28.6 7.2 Above poverty 20-74 YOBIrS. , c+: suis s cmma so yun 1,078 5.25 0.03 15.7 1.1 20-74 years, age adjusted.... 5.42 Ep 17.95 20-29 339 4.94 0.06 19.5 2:2 30-39 285 4.99 0.06 10.1 1.8 40-49 183 5.54 0.08 9.0 1.9 50-59 180 5.82 0.08 17.8 2.4 60-69 74 6.09 0.14 25.1 4.5 70-74 17 = * = * 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ mmol/L = mg/d1*0.02586. 11-72 Table II-42. Serum cholesterol status of non-Hispanic persons 20-74 years of age by sex, race, and age: Second National Health and Nutrition Examination Survey, 1976-80 | Percent with Standard Number of Mean serum Standard high-risk error examined cholesterol error of serum of the Sex, race, and age persons 1/| (mmol/L) 2/ the mean cholesterol percent MALE Non-Hispanic white 20-78 YRBAPrS wus suv + saws vam 4,646 5.47 0.02 19.1 0.7 20-74 years, age adjusted. . 5.45 ig 19.0 20-20 YRBIS .. «i «uv s vimnn iin 1,011 4.85 0.04 17.14 1.6 30-39 years. ............... 707 5.41 0.06 20.1 2.1 0~40 YRAIrS. .. vcr iervs svn 572 5.78 0.07 18.1 2.3 50-59 years. ............... 57% 5.92 0.07 21.9 2.4 BO~6O YRArS. .. ivr annie na 1,354 5.85 0.03 21.0 0.9 70-748 y@ArS. .: uur: svn vwmn 427 5.58 0.04 13.7 1.4 Non-Hispanic black 20-74 years. ............... 5897 5.37 0.07 20.8 2.0 20-74 years, age adjusted. . 5.39 La 20.5 CL P0~29 YRAPS sus swe i emma tip 158 4.79 0.13 17.1 3.9 BO-39 YBBIrS. .. coun swmuv mn 33 5.36 0.20 20.8 5.8 40-49 years. ............... 62 5.63 0.28 18.7 7.6 BOBO YBBIS vi: wns imursionsy 77 5.96 0.19 30.3 7.0 60-69 years. ............... 151 5.77 0.07 18.4 2:5 TO~78 YRArS..: + cvs: swvs sewn 56 5.48 0.13 14.7 3.7 FEMALE Non-Hispanic white 20-74 YRBIS ss vv: saws sw 5,148 5.59 0.02 22.4 0.7 20-74 years, age adjusted. . 5.54 Tr 231.7 20-29 years... 1,066 4.83 0.04 17.0 1.8 30-39 YOBrS cu: swims » vies i 4 798 5.09 0.0% 14.5 1.7 40-49 YRAIrS. .... viv vreau 615 5.65 0.07 13.6 1.9 50-59 years. ............... 649 6.30 0.07 31.4 2.8 60-89 yBRIrS. .. nsx russ 1,487 6.44 0.03 36.4 1.1 70-74 years. ............... 533 6.42 0.05 37.4 1.8 Non-Hispanic black 20-74 YBAIrS. «viv vv swans in 721 5.48 0.06 19.9 1.8 20-74 years, age adjusted. . 5.82 3% 4 20.2 20-29 years. ............... 181 4.88 0.11 17.4 3.9 BO~39 VRArS., ; va: umn: cmme sin 110 3.11 0.13 14.9% 4.1 AO~40 YOBLrS. vo swuw is vs vin 95 5.61 0.19 17.9 5.3 50-59 years. ............... 101 6.28 0.20 30.0 6.0 BOBO. YOArS nu sums sume sn 170 6.22 0.08 27.2 2.7 70-74 years. ............... 64 6.46 0.14 33.4 4.7 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ mmol/L = mg/d1*0.02586. II-73 Table II-43. Serum cholesterol status of non-Hispanic persons 20-74 years of age by sex, poverty status, and age: Second National Health and Nutrition Examination Survey, 1982-84 Percent with Standard Number of Mean serum Standard high-risk error Sex, poverty status, examined cholesterol error of serum of the and age persons 1/| (mmol/L) 2/ the mean cholesterol percent MALE Below poverty 20-74 years. ............... 541 5.28 0.06 18.0 1.8 20-74 years, age adjusted.. 5.34 . 4 14.7 20-29 years... ...:. svi inns 147 4.85 0.13 16.9 4.3 30-39 years. ............... 67 5.31 0.20 16.0 6.4 40-49 years. ............... 35 *5.34 0.29 *8.8 7.3 BO~89 Y8ArS. vs: vn sums 50 5.96 0.22 15.8 7.1 60-69 years................ 156 5.70 0.07 15.8 2.5 T7O0~74 YRBIrS., ....:. 05s: ins 86 5.34 0.10 9.9 2.7 Above poverty 20-74 years. ............... 4,595 5.48 0.02 19.7 0.7 20-74 years, age adjusted. . 5.47 19.7 ‘ve 20-29 years. ............... 1,008 4.85 0.04 17.5 1.6 30-39 Vears. .... oma sammy 728 5.42 0.06 21.1 2.1% 40-49 years............. .. 580 5.79 0.07 18.6 2.3 BO~B9 y0ArS. ..1 ums suis 576 5.93 0.07 22.7 2.4 60-69 vears................ 1,317 5.86 0.03 21.3 0.9 70-74 YOAPS uw + snub i btmb sb 386 5.63 0.05 14.8 1.5 FEMALE Below poverty 20-74 YBBPS . vv: sums s smmn ¥ : 865 5.45 0.05 20.8 1.7 20-74 years, age adjusted. . 5.47 20.3 20-29 years. ............... 213 4.80 0.09 16.9 3.6 30-38 years. ............... 120 8.4% 0.14 12.2 4.3 40-49 years. ............... 77 5.54 0.22 13.1 5.6 S0-59 years. ........... Co. 81 6.17 0.21 31.1 7.3 80-69 VRBIFrSB. ... i uric 252 6.29 0.07 31.4 2.8 70-74 years. ............... 122 6.28 0.11 34.7 3.8 Above poverty 20-74 yYBArS . .: canis ive in 4,861 5.58 0.02 22.1 0.7 20-74 years, age adjusted. . 5.54 21.8 20-29 1,034 4.84 0.04 17.1 1.6 30-39 779 8.11 0.05 14.5 1:7 40-49 621 5.64 0.07 14.3 1.9 50-59 644 6.30 0.07 31.1 2.5 60-69 1,336 6.42 0.03 35.9 1.1 447 6.46 0.05 37.8 2.0 70-74 1/ Includes persons for whom usable measurements for criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ mmol/L = mg/d1*0.02586. 11-74 qL-11 U.S. Food Supply Carbohydrate Grams 700} soo} s00} 400} CY TN 300} 200+ 100 + 0 1 1 1 1 1 he Xl 1 1 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure 11-15. Carbohydrate: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Carbohydrate Carbohydrate Sugars and sweeteners 39.6% y—Other foods 1.17% Legumes, nuts, and soy 27% Dairy products 5.7% \ Fruits 6.6% Vegetables 9.27% Grain products 35.87% Figure II-16. Carbohydrate: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include meat, poultry, and fish; eggs; fats and oils; and miscellaneous foods) 9L-1I Table II-44. Carbohydrate: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 175 2.2 79 98 127 168 212 262 290 Age 20-29 years 661 184 3.7 86 105 135 179 225 274 296 30-39 years 812 173 3.0 75 94 122 167 214 260 287 40-49 years 583 162 3.4 78 91 123 157 195 238 272 Race 1/ White 1775 177 2.3 84 99 130 169 214 262 292 Black 167 157 7.4 63 72 110 145 197 260 274 Other 76 172 6.5 * * 135 175 209 * * Poverty status 1/ < 100 315 165 4.8 68 86 117 148 207 260 289 > 100 1575 178 2.2 83 100 131 170 215 264 292 < 131 414 le8 4.6 71 87 118 156 210 260 290 > 131 1476 178 2.2 84 100 131 170 215 264 294 Education 1/ < High school 305 154 3.8 68 80 110 144 193 242 264 High school 854 173 2:5 81 101 126 165 209 255 288 > High school 891 184 3.0 85 101 135 178 225 270 296 Region Northeast 448 163 3.7 79 97 120 158 193 236 265 Midwest 564 180 3.7 80 100 132 171 223 265 292 South 660 178 3.2 80 92 128 167 220 27 302 West 384 176 71.3 78 100 132 171 212 259 283 Urbanization Central city 499 176 5.0 75 100 127 168 213 264 292 Suburban 1039 175 3.2 81 98 126 166 212 262 290 Nonmetropolitan 518 175 3.0 79 94 131 169 210 257 284 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. LL-11 Table 11-45. Carbohydrate: mean percent of calories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 46 “3 33 36 41 46 51 56 59 Age 20-29 years 661 47 .4 34 37 42 47 52 56 59 30-39 years 812 46 +3 32 35 41 46 51 56 60 40-49 years 583 45 .4 33 36 40 45 51 55 59 Race 1/ White 1775 46 3 33 36 41 46 51 56 59 Black 167 46 «9 32 34 41 47 51 54 56 Other 76 49 3:2 * * 44 49 56 x * Poverty status 1/ < 100 315 47 .6 34 35 42 47 52 56 61 > 100 1575 46 +3 33 36 41 46 51 56 59 < 131 414 47 “5 34 36 42 47 52 58 61 > 131 1476 46 «3 33 36 41 46 51 56 59 Education 1/ < High school 305 47 +5 34 36 41 47 52 58 61 High school 854 46 .4 33 36 41 46 51 55 58 > High school 891 46 .4 33 36 41 46 51 56 60 Region Northeast 448 45 .4 31 35 40 46 50 54 57 Midwest 564 46 «3 34 36 41 45 50 55 58 South 660 47 5 33 36 42 47 53 58 61 West 384 46 .8 34 35 41 46 50 56 59 Urbanization Central city 499 46 .6 32 35 41 46 51 56 60 Suburban 1039 46 .4 33 36 41 46 51 56 59 Nonmetropolitan 518 47 -4 34 37 42 47 52 56 61 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 8L-1II Table 11-46. Carbohydrate: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 184 3.0 114 124 149 179 214 251 267 Age 1-2 years 224 le7 3.9 108 116 136 165 190 221 254 3-5 years 423 192 3.4 122 131 154 187 226 259 275 Race 2/ White 559 187 3.1 117 127 151 181 218 254 271 Black 53 170 6.7 * * 138 169 196 * * Other 26 153 11.2 * * 154 * * * Poverty status 2/ < 100 140 171 6.9 94 118 131 165 207 233 251 > 100 471 187 3.3 114 127 153 181 218 259 275 < 131 192 171 5.8 96 116 136 167 206 233 253 > 131 419 189 3.5 117 129 154 183 220 259 276 Education 2/ < High school 99 173 6.3 * 114 140 161 208 239 * High school 252 179 4.5 114 122 146 176 207 254 262 > High school 295 190 3.6 117 134 159 184 220 251 271 Region Northeast 111 188 6.0 * 129 153 184 221 254 * Midwest 199 184 5.6 113 126 148 177 214 244 275 South 187 180 6.2 113 118 141 169 213 261 275 West 150 185 5.1 119 128 154 181 213 240 255 Urbanization Central city 171 189 5.9 119 129 155 185 220 261 268 Suburban 310 183 3.9 116 126 151 177 214 252 265 Nonmetropolitan 166 178 6.6 98 114 136 174 209 245 269 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 6L-11 Table 11-47. Carbohydrate: mean percent of calories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 g5 All children 1/ 647 52 .4 41 44 47 52 56 59 61 Age 1-2 years 224 51 .6 41 42 47 52 56 59 61 3-5 years 423 52 .4 42 44 47 52 56 59 61 Race 2/ White 559 52 .4 42 44 48 52 56 60 62 Black 53 49 1.0 * ® 46 49 53 * Other 26 49 1.8 * * 50 * * Poverty status 2/ < 100 140 49 .8 38 42 45 49 53 57 58 > 100 471 52 .4 42 44 48 53 56 60 63 < 131 192 49 .8 39 41 45 49 54 57 59 > 131 419 52 .4 42 45 48 53 56 60 63 Education 2/ < High school 99 49 1.0 * 40 44 49 53 58 * High school 252 51 .6 41 44 47 50 55 58 61 > High school 295 53 .4 43 46 50 54 57 60 62 Region Northeast iia 52 1.1 x 43 48 53 57 60 * Midwest 199 51 .6 42 44 47 52 56 58 59 South 187 51 .8 41 44 47 51 55 60 62 West 150 52 8 42 44 47 52 56 59 61 Urbanization Central city 171 51 17 42 44 47 51 55 57 61 Suburban 310 53 .5 42 44 48 53 57 60 62 Nonmetropolitan 166 50 1.0 40 41 45 49 54 58 61 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 08-11 Table II-48. Carbohydrate: mean intake in grams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES 11 CSF11 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 161 2.1 140 2.3 162 1.6 166 4.9 3-5 209 2.0 172 2.2 200 1.2 192 3.8 6-11 252 3.0 219 2.1 246 3.0 - - Male 12-15 319 6.6 276 4.0 304 7.0 - - 16-19, 341 8.9 282 4.3 340 9.6 - - 20-29 301 7.1 252 3.5 305 6.6 317 12.9 30-39 271 7.7 233 3.4 272 6.5 279 9.3 40-49 245 6.6 225 4.0 245 6.8 263 10.7 50-59 222 5.7 218 3.7 229 7.0 - - 60-69 214 2.7 205 3.6 211 2.2 - - 70+ 199 2.7 199 4.1 198 3.5 - - Female 12-15 228 5.0 213 2.9 223 5:2 - - 16-19 205 5.4 190 3.1 198 5.7 - - 20-29 193 2.2 170 2.0 195 3.9 198 4.2 30-39 178 2.2 160 2.3 177 4.5 188 3.9 40-49 168 2.9 154 2.1 168 4.3 173 4.0 50-59 163 4.1 156 2.3 158 4.3 - - 60-69 156 2.1 158 2.2 158 1.6 - - 70+ 148 1.8 156 2.6 159 2.6 - - 2 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. 18-11 Table II-49. Carbohydrate: mean percent of kilocalories, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES I NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 48 0.3 47 0.3 50 0.2 51 0.5 3-5 50 0.2 48 0.3 51 0.2 51 0.6 6-11 49 0.3 47 0.2 50 0.3 - - Male 12-15 49 0.5 46 0.3 49 0.5 - - 16-19, 46 0.6 43 0.3 45 0.6 - - 20-29, 42 0.6 41 0.3 42 0.4 46 0.9 30-39 41 0.7 40 0.3 43 0.6 46 1.1 40-49 41 0.7 39 0.4 41 0.6 44 1.0 50-59 41 0.6 40 0.3 41 0.6 - - 60-69 44 0.3 41 0.4 43 0.3 - - 70+ 45 0.3 43 0.5 46 0.5 - - Female 12-15 48 0.5 46 0.3 49 0.6 - 16-19 47 0.7 45 0.4 47 0.6 - - 20-29 46 0.3 42 0.3 47 0.5 48 0.5 30-39 44 0.3 41 0.3 45 0.6 46 0.4 40-49 44 0.4 40 0.4 44 0.7 45 0.5 50-59 45 0.7 41 0.3 45 0.6 - - 60-69 47 0.3 44 0.4 47 0.3 - - 70+ 48 0.3 45 0.5 50 0.4 - - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. Table II-50. Dietary fiber: mean intake in grams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 11 «2 4 5 7 10 14 17 20 Age 20-29 years 661 11 «3 4 5 7 10 13 17 19 30-39 years 812 11 .2 4 5 7 10 14 18 21 40-49 years 583 11 .2 4 6 7 10 13 17 18 Race 1/ White 1775 11 .2 4 6 8 11 14 17 20 Black 167 9 -5 4 5 8 10 15 17 Other 76 10 .5 # £ 7 10 13 * * Poverty status 1/ < 100 315 9 «3 3 4 6 8 11 15 17 = > 100 1575 11 «2 5 6 8 11 14 18 20 | 3 < 131 414 9 .3 3 4 6 8 11 15 17 > 131 1476 11 .2 5 6 8 11 14 18 20 Education 1/ < High school 305 8 : 2 3 6 7 11 15 18 High school 854 10 .2 4 5 7 10 13 16 18 > High school 891 12 +2 5 6 9 12 15 18 22 Region Northeast 448 10 .3 4 5 7 9 13 16 19 Midwest 564 1a .4 4 6 8 1 14 17 21 South 660 10 .3 4 5 7 10 13 17 19 West 384 12 .4 5 6 8 11 15 18 20 Urbanization Central city 499 11 .4 4 5 7 10 13 17 22 Suburban 1039 11 “2 4 5 8 10 14 17 19 Nonmetropolitan 518 11 .3 4 5 7 10 13 17 19 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. €8-11 Table II-51. Dietary fiber: mean intake in grams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 15 920 95 All children 1/ 647 10 .2 4 5 7 10 12 15 17 Age 1-2 years 224 9 «3 4 5 7 9 11 13 14 3-5 years 423 10 «3 4 6 7 10 13 15 17 Race 2/ White 559 10 .2 5 6 7 10 12 15 17 Black 53 10 1.1 * * 6 9 1 * * Other 26 7 1 * * 7 * ¥* Poverty status 2/ < 100 140 9 .9 3 4 6 9 1: i5 17 > 100 471 10 .2 5 6 7 10 12 15 17 < 131 192 9 «7 3 4 6 9 11 16 17 > 131 419 10 «2 5 6 8 10 12 15 17 Education 2/ < High school 99 9 .4 * 4 6 9 11 14 * High school 252 10 .5 4 5 7 9 12 15 18 > High school 295 10 2 5 6 8 10 13 15 17 Region Northeast 111 9 .6 * 6 6 8 11 15 * Midwest 199 10 3 4 6 7 9 11 14 17 South 187 9 .4 4 5 7 9 12 14 15 West 150 11 .6 5 6 8 10 a3 16 18 Urbanization Central city 171 11 .6 5 6 8 10 13 17 20 Suburban 310 10 .2 5 6 7 9 12 14 15 Nonmetropolitan 166 9 «5 3 4 7 9 11 14 15 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. U.S. Food Supply Vitamin A RE 2000 1800 + 1600 1400 + 1200 + 1000 800 600 400 + 200+ 0 i 1 i 1 1 1 1 1 1 1 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year 8-11 Figure II-17. Vitamin A: per capita amount (in retinol equivalents) per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Vitamin A Vitamin A Vegetables 35.2% Meat, poultry, fish 26.4% ruits 2.7% KF ~Eqgs 3.4% Other foods 4.7% Dairy products 15.8% Fats and oils 11.8% Figure II-18. Vitamin A: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include legumes, nuts, and soy; grain products; and miscellaneous foods) Table II-52. Vitamin A: mean intake in retinol equivalents, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 832 26 170 248 386 614 1000 1515 1959 Age 20-29 years 661 832 50 171 250 381 593 1000 1546 1892 30-39 years 812 814 32 162 246 392 645 999 1486 1920 40-49 years 583 857 57 182 249 379 611 1005 1512 2124 Race 1/ White 1775 842 26 191 260 408 651 1042 1546 1959 Black 167 818 110 145 171 318 407 627 1014 2162 Other 76 591 64 * * 292 550 785 * * Poverty status 1/ < 100 315 855 98 121 lel 291 448 725 1314 2095 > 100 1575 811 21 198 271 408 643 1013 1477 1921 < 131 414 800 73 121 168 314 484 769 1300 1866 > 131 1476 821 21 207 278 413 647 1015 1512 1958 Education 1/ < High school 305 685 61 104 150 248 397 699 1199 2221 High school 854 1755 45 150 209 369 559 882 1307 1679 > High school 891 950 23 289 340 481 744 1169 1738 2145 Region Northeast 448 853 53 166 248 365 558 942 1552 2145 Midwest 564 892 53 191 245 391 674 1064 1546 2047 South 660 765 50 143 212 354 544 875 1437 2054 West 384 846 49 245 332 479 725 1089 1546 1762 Urbanization Central city 499 867 41 161 248 379 601 966 1524 1879 Suburban 1039 802 31 194 270 414 643 1027 1515 2019 Nonmetropolitan 518 855 73 147 207 327 548 983 1487 2054 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 98-11 Table II-563. Vitamin A: mean intake in retinol equivalents, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 85 All children 1/ 647 816 21 323 400 545 729 965 1297 1628 Age 1-2 years 224 811 31 335 359 545 729 892 1273 1676 3-5 years 423 818 26 323 420 545 719 1003 1301 1587 Race 2/ White 559 816 20 337 421 552 731 970 1285 1562 Black 53 851 97 * * 430 5717 8717 * * Other 26 7178 105 * * % 681 J * * Poverty status 2/ < 100 140 808 70 302 385 479 697 874 1281 1677 > 100 471 816 20 331 397 550 731 985 1299 1587 < a3} 192 819 58 331 368 483 720 877 1390 1677 > 131 419 812 21 323 420 552 730 988 1285 1587 Education 2/ < High school 99 743 51 * 375 476 631 812 1221 > High school 252 821 38 321 358 530 720 944 1393 1671 > High school 295 832 22 350 452 574 747 1010 1292 1551 Region Northeast 111 846 52 * 518 567 772 997 1407 * Midwest 199 859 41 354 460 589 732 970 1380 1667 South 187 765 42 300 337 457 602 944 1345 1686 West 150 814 34 358 439 577 744 940 1264 1498 Urbanization Central city 171 870 49 359 467 604 746 1024 1403 1587 Suburban 310 784 24 321 415 534 712 944 1273 1667 Nonmetropolitan 166 825 47 321 349 493 663 970 1345 1676 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. L8-11 Table II-54. Vitamin A: mean intake in International Units, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFI1 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 3,427 88 3,511 138 3,618 54 4,489 332 3-5 3,753 132 3,958 158 4,008 45 4,411 338 6-11 4,319 156 4,936 144 4,989 130 - - Male 12-15 4,951 266 5,946 21 5,663 341 - - 16-19, 5,272 393 6,101 367 6,295 584 - - 20-29 5,510 401 5,823 324 5,437 220 5.711 669 30-39, 5,092 574 5,799 207 5,917 511 5,933 612 40-49 5,058 387 6,578 417 6,036 460 7.114 716 50-59 5,628 439 6,953 350 6,020 350 - - 60-69 5,684 287 6,889 440 6,163 191 - - 70+ 5,830 294 6,441 319 6.731 376 - - Female 12-15 3,899 296 4,449 189 4,018 210 - - 16-19 3,725 288 4,369 242 3,777 228 - - 20-29 3,891 133 4,462 153 4,207 189 5,241 250 30-39 4,403 190 4,836 217 4,691 364 5,573 384 40-49 4,950 353 5,090 251 5,250 438 5,442 314 50-59 5,337 496 5,990 295 5.331 365 - - 60-69 6,030 483 6,722 327 5,400 139 - - 70+ 4,527 186 5.942 279 5,469 258 - - 1 CSFI1 data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. Table II-55. Serum vitamin A status of children 4-19 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 Percent Mean with Standard Number of vitamin A Standard vitamin A error of Sex, Hispanic origin, examined level error of < 0.7 the and age persons 1/ umol/L 2/ the mean umol/L 3/ percent BOTH SEXES Mexican American 4-5 years. ........ a... 234 1.00 0.02 4.6 1.2 6-11 years. ................. 1,028 1.11 0.01 2.7 0.4 Cuban A~B YOAr'S., .. 1 vus somws sump ns 12 * oe * * 6-11 years.................. 96 1.19 0.02 3.3 1.7 Puerto Rican 4-5 years... ................. 62 1.02 0.03 8.8 3.4 6-11 years. ................. 322 1.14 0.01 1.4 0.6 MALE Mexican American 12-15 years... ............... 351 1.32 0.02 0.0 0.0 18<19 YEArS., «vy sown i saws sams 253 1.53 0.03 0.0 0.0 Cuban 12-15 years. ................ 48 1.32 0.04 0.0 0.0 1819 YBArS.,. o. uv vssmws smn S1 1.58 0.04 0.0 0.0 Puerto Rican 12-15 years. ................ 138 1.41 0.02 0.0 0.0 16-19 years................. 137 1.58 0.03 0.0 0.0 FEMALE 4/ Mexican American 12-18 Y@APr'S, «vs uss c imma cmns 328 1.26 0.02 0.3 0.3 16-19 years. ................ 291 1.37 0.03 0.4 0.3 Cuban 1218 YOAPrS., i suns imma innss 38 *1.29 0.04 *0.0 0.0 16-19 years................. 41 *1.28 0.04 *0.0 0.0 Puerto Rican 1218 y@APrS., . i avs s ums sn ve 140 1.33 0.02 0.0 0.0 16-19 years....... Cay smd 123 1.43 0.04 0.0 0.0 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ umol/L = ug/d1*0.03491. 3/ 0.7 umol/L = 20 ug/dl. 4/ Excludes pregnant women. Table II-56. Serum vitamin A status of persons 20-74 years of age by sex, specified Hispanic origin, and age: Hispanic Health and Nutrition Examination Survey, 1982-84 Percent Mean with Standard Number of vitamin A Standard vitamin A error of Sex, Hispanic origin, examined level error of < 0.7 the and age persons 1/ umol/L 2/ the mean umol/L 3/ percent MALE Mexican American 20748 YRArS. cos vous sang tas 1.392 1.84 0.02 0.3 0.2 20-74 years, age adjusted... .. 1.85 0.5 ‘ 20-20 YRArS.. os uni ana sans 421 1.74 0.04 0.0 0.0 30<38 YRArS. sus: mms sons stim 354 1.89 0.05 0.0 0.0 40-49 years... i 230 1.96 0.05 0.3 0.4 BO~89 Y@ArS.... vos eiintens 229 1.88 0.04 0.0 0.0 60-69 years... ........ i... 122 1.84 0.07 2.6 1.4 70-74 years. ......... in 36 *1.84 0.13 2.4 2.7 Cuban 20-74 years... 372 2.02 0.03 0.0 0.0 20-74 years, age adjusted... .. ' %e 1.86 _- 0.0 20-20 YRAPrS. ... savas canis smd 55 1.82 0.06 0.0 0.0 30-39 yBArsS.,.. ins seme s aww 83 1.99 0.05 0.0 0.0 40-49 years. ..... 82 2.13 0.06 0.0 0.0 BOB YRAPrS. cons sumns tomas twas 109 2.07 0.04 0.0 0.0 BOB YRAPS. . uss svmms sama s sms 45 2.14 0.08 0.0 0.0 T7O=T4 years... ..... «o.oo. 28 *1.90 0.10 *0.0 0,0 Puerto Rican 20°78 YBAPS. ; i vvs sums dt sus 2 £Ws 423 1.86 0.083 0.0 0.0 20-74 years, age adjusted..... Co 1.99 0.0 20-29 YOArsS: . vv: vm wr tammy swe 109 1.76 0.085 0.0 0.0 30-39 years. ......... 86 1.88 0.06 0.0 0.0 40-49 yBArS. . vv swans amas 4 viva 84 1.96 0.07 0.0 0.0 SO~B0 YBArS., . iuvs vnmns amass vs 97 1.89 0.05 0.0 0.0 60-69 years. ............ 37 *1.96 0.06 =0.0 0.0 70~74 YBArS: : vv: swwos imme v wus 10 * * * = 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ umol/L = ug/d1*0.03491. 3/ 0.7 umol/L = 20 ug/dl. 11-89 Table II-56. Serum vitamin A status of persons 20-74 years of age by sex, specified Hispanic origin, and age: Hispanic Health and Nutrition Examination Survey, 1982-84--continued | | | Percent Mean with Stanaard Number of vitamin A Standard vitamin A error of Sex, Hispanic origin, examined level error of < 0.7 the and age persons 1/ umol/L 2/ the mean umol/L 3/ percent FEMALE 4/ Mexican American 20-74 years. .................. 1,784 1.53 0.02 0.6 0.2 20-74 years, age adjusted..... “ss 1.57 Ca 0.5 20-29 years. ............ 531 1.46 0.083 0.9 0.5 30-39 years... . vo ives ann. 446 1.48 0.03 0.2 0.2 40-49 years. .................. 302 1.53 0.03 0.0 0.0 BO~59 years. .......ooce nu... 309 1.62 0.03 0.9 0.5 60-69 years... ................ 136 1.76 0.05 0.7 0.7 70-74 years. .................. 60 1.93 oO. 11 0.0 0.0 Cuban 20-74 years. ................. . 467 1.64 0.02 0.0 0.0 20-74 years, age adjusted... .. LL 1.63 2029 YBAIrS. civ v5 s tmmnr smmne 63 1.52 0.05 0.0 0.0 30-39 years. .................. 92 1.50 0.03 0.0 0.0 40-49 years. .................. 101 1.88 0.03 0.0 0.0 BOY YGAPrS., vs smn + cms 18 bmn vw 112 1.79 0.04 0.0 0.0 60-69 years... ................ 68 1.88 0.05 2.0 0.0 70-74 years. .................. 31 *1.91 0.09 *0.0 0.0 Puerto Rican 20-74 years. .................. 721 1.56 0.02 0.4 0.3 20-74 years, age adjusted... .. LL 1.61 20-29 Y@BIrS, uss imum son ni vas 184 1.49 0.04 0.0 ©.0 30-39 years. .................. 164 1.952 0.04 1.0 1.0 40-49 years. .................. 161 1.56 0.03 0.6 0.6 BO=B9 YBAPrS, , uuu + sms vr chiws 2xms 126 1.714 0.03 0.0 0.0 60-69 years. .................. 72 1.87 0.05 0.0 0.0 70-74 years. .................. 14 * = * * 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ umol/L = ug/d1*0.03491. 3/ 0.7 umol/L = 20 ug/dl. 4/ Excludes pregnant women. 16-11 Table II-57. Serum vitamin A status of Mexican-American children 4-19 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Below poverty T | | | | | Percent | | | Mean | | with | Standard | Number of | vitamin A Standard | vitamin A j Sreon of I examined I level jerror of | < 0.7 I the Sex and age | persons 1/ jumol/L 2/ j the mean | umol/L 3/ Percent 1 3 1 1 1 Both sexes 4=5 y@aArsS,. . :xuiwmesn 87 0.97 0.04 10.1 2.7 6-11 years. .......««. 378 1.10 0.02 2.6 0.7 Male 12-185 YOBrS.: x vs vows « {115 1.32 0.03 0.0 0.0 16-19 years. ........ 80 1.48 0.05 0.0 0.0 Female 4/ 12-15 years. ........ 124 1.25 0.03 0.8 0.7 16-19 years. ....«+» 107 1.35 0.04 0.0 0.0 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ umol/L = ug/d1+0.03491. 3/ 0.7 umol/L = 20 ug/dl. 4/ Excludes pregnant women. Table II-57. Serum vitamin A status of Mexican-American children 4-19 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84--continued Above poverty T | | | | | Percent | | | Mean | with j Sxandary | Number of | vitamin A j Standard | vitamin A jerror of I examined I level (SL ror of I < 0.7 | the Sex and age | persons 1/ [umel/L 2/ j the mean | umol/L 3/ jpercent 1 1 1 te 1 Both sexes 4-5 years. .......:.x 134 1.02 0.02 1.8 1.0 6-11 years... ..... .. 569 1.12 0.01 2.7 0.6 Male 12-15 years. ........ 200 1.33 0.03 0.0 0.0 16-19 years. ........ 142 1.56 0.04 0.0 0.0 Female 4/ a, ' | 12-15 years. ........ 175 1.27 0.03 0.0 0.0 8 16-19 years......... 160 1.40 0.04 0.7 0.6 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ umol/L = ug/dl1+0.03491. 3/ 0.7 umol/L = 20 ug/dl. 4/ Excludes pregnant women. €6-11 Table II-58. Serum vitamin A status of Mexican-American persons 20-74 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Below poverty 1 | | | | Percent | | | Mean | | with j Standard jumper of | vitamin A | Standard I vitamin A jerror of § Sxaninag | level yerror of | < 0.7 | the Sex and age | persons | umol/L 2/ | the mean I umol/L 3/ | percent 1 1 1 1 1 Male 20-74 years... ................ 307 1.75 0.04 0.7 0.5 20-74 years, age adjusted..... Baw 1.76 0.8 20-29 years. ........... nu... 96 1.72 0.07 0.0 0.0 30-39 years... .suninurinni sus 64 1.78 0.09 0.0 0.0 40-49 years... ................ 45 1.92 oO. 11 1.5 1.8 50-59 years................... 48 1.69 0.07 0.0 0.0 BOB yRAPrS. .....nns cnmismmrame 40 *1.67 0.12 *5.0 3.4 70-74 years. .................. 14 * * * * Female 4/ 20-74 years. ......:svs cus ens 520 1.53 0.03 1. 0.5 20-74 years, age adjusted..... ‘mo 1.54 1. 20-29 yoArS.: s:uurin niin niin 141 1.38 0.05 1.6 1.2 30-39 years. .................. 118 1.585 0.06 0.9 0.9 40-49 years... ...... i... 86 1.49 0.06 0.0 0.0 50-89 years. ....visssscvuinnns 94 1.57 0.06 1.2 1:0 60-69 years... ................ 45 1:73 0.10 2.2 2.1 70-74 years. .................. 36 *2 01 0.16 *0.0 0.0 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ umol/L = ug/d1+0.03491. 3/ 0.7 umol/L = 20 ug/dl. 4/ Excludes pregnant women. Table II-58. Serum vitamin A status of Mexican-American persons 20-74 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84--continued Above poverty T | | | ] ; Percent | | | Mean | | with | Standard june of | vitamin A j Standard | vitamin A jerror of j Sram ine | level SC ror of I < 0.7 | the Sex and age |persons 1 umol/L 2/ | the mean | umol/L 3/ percent 1 1 1 1 1 Male 20-74 years. ..........oiian.. 966 1.87 0.03 0.2 0.1 20-74 years, age adjusted... .. “5 1.88 0.4 20-29 years... ....... ll... 287 1.75 0.05 0.0 0.0 B0-39 YBArS. uu: cvs: tos sams ia a 269 1.91 0.05 0.0 0.0 40-49 years... ....... 164 1.97 0.06 0.0 0.0 50-59 years... .... 157 1.94 0.05 0.0 0.0 B0-69 years... ... :::x:: 85: mwa 70 1.98 0.09 1.7 1.5 70-74 years... ......... 19 + * * * Female 4/ 20-78 yBAIS. i: icc ame snns mas 1,029 1.56 0.02 0.2 0.1 20-74 years, age adjusted... .. I. 1.59 0.1 20-29 Years... .:smrianssmmeimas 303 1.54 0.04 0.4 0.4 30-39 years... ......... a... 276 1.47 0.04 0.0 0.0 40-49 years... .......ii.. 182 1.56 0.04 0.0 0.0 50-59 YeArS... ..« ws esmamms emmy 179 1.65 0.04 0.0 0.0 60-69 years. .................. 72 1.78 0.07 0.0 0.0 10-74 years. ......::ccnsmmaians 17 * * * * 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ umol/L = ug/d1+0.03491. 3/ 0.7 umol/L = 20 ug/dl. 4/ Excludes pregnant women. Carotenes Carotenes U.S. Food Supply Carotenes RE 1000 Vegetables 85.8% 800} 600 1 Dairy products 2.3% Fats and oils 2.67% Other foods 2.6% 400} Fruits 6.7% 200} © {505 1915 1925 1938 Te45 1955 1965 1975 1985 Year Figure II-19. Carotenes: per capita amount (in retinol Figure II-20. Carotenes: food sources in the U.S. food supply, equivalents) per day in the U.S. food supply, 1909-85: U.S. Food 1985: U.S. Food Supply Series (other foods include meat, poultry, Supply Series and fish; eggs; legumes, nuts, and soy; grain products; and miscellaneous foods) Table II-59. Carotenes: mean intake in retinol equivalents, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 353 12 44 63 118 227 459 773 1116 Age 20-29 years 661 330 17 44 61 111 208 422 734 1057 30-39 years 812 368 17 40 62 119 230 490 832 1163 40-49 years 583 372 19 53 68 122 256 472 772 1103 Race 1/ White 1775 366 13 48 69 123 239 479 811 1129 Black 167 235 17 35 45 91 171 262 533 668 Other 76 321 51 * * 86 289 514 * * Poverty status 1/ < 100 315 246 17 27 37 73 151 292 570 741 > 100 1575 371 13 53 74 131 251 484 800 1126 < 131 414 258 17 27 40 80 167 333 571 741 > 131 1476 376 13 53 75 132 252 487 843 1149 Education 1/ < High school 305 233 21 17 34 66 137 259 563 783 High school 854 306 13 44 57 103 204 399 674 215 > High school 891 434 16 74 96 164 298 561 979 1267 Region Northeast 448 351 23 51 62 122 214 422 757 1191 Midwest 564 341 27 43 56 108 213 440 747 1126 South 660 310 17 42 63 104 207 411 699 977 West 384 430 22 44 81 159 321 593 938 1331 Urbanization Central city 499 346 24 44 63 113 214 433 741 1129 Suburban 1039 377 17 49 71 136 255 500 803 1152 Nonmetropolitan 518 295 17 33 53 94 182 371 703 944 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. L6-11 Table II-60. Carotenes: mean intake in retinol equivalents, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 252 41 54 84 141 314 589 808 Age 1-2 years 224 230 15 52 59 85 133 289 530 678 3-5 years 423 264 20 36 48 84 150 334 662 851 Race 2/ White 559 263 14 45 55 86 148 337 615 820 Black 53 148 19 * * 65 126 194 * * Other 26 *313 x92 x * x 163 * * * Poverty status 2/ < 100 140 216 21 39 46 74 139 284 547 595 > 100 471 264 16 49 60 87 141 330 662 847 < 131 192 207 17 39 46 79 129 238 530 595 > 131 419 273 18 52 60 89 148 352 6717 855 Education 2/ < High school 99 183 20 * 52 87 126 212 372 * High school 252 242 22 36 46 73 134 314 582 808 > High school 295 281 20 48 62 90 162 358 669 855 Region Northeast 111 246 34 * 36 87 139 298 585 * Midwest 199 200 14 43 53 80 128 251 402 615 South 187 279 30 43 51 73 141 352 751 1002 West 150 276 23 59 67 93 185 383 581 736 Urbanization Central city 171 260 27 43 60 90 174 352 562 836 Suburban 310 260 19 40 55 87 139 314 590 808 Nonmetropolitan 166 220 21 39 46 68 126 290 581 710 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 86-11 Milligrams Vitamin E U.S. Food Supply Vitamin E 18 16 14 12 1 T 1 1 1 a 1 1 1 1 1 0 8 6 4+ 2 0 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-21. Vitamin E: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Vitamin E Fats and oils 66.4% fii | fi 0 i Em . . Fruits 3.4% \ J Grain products 4.1% Other foods 5.4% Meat, poultry, fish 6.1% Vegetables 7.9% Legumes, nuts, and soy 6.7% Figure 1-22. Vitamin E: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include dairy products; eggs; sugars and sweeteners; and miscellaneous foods) 66-11 Table II-61. Vitamin E: mean intake in milligrams alpha tocopherol equivalents, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 7.0 .14 2.5 3.0 4.3 6.1 8.2 11.2 13.8 Age 20-29 years 661 7.0 «25 2.6 2.9 4.2 6.0 8.2 31.7 15.5 30-39 years 812 7.0 .23 2.5 3.0 4.3 6.1 8.4 11.1 13.2 40-49 years 583 6.9 .24 2.4 3.3 4.3 6.1 8.0 11.1 13.8 Race 1/ White 1775 7.2 .17 2.6 3.2 4.4 6.3 8.4 11.4 14.2 Black 167 5.6 .31 1.7 2.3 3.4 4.9 7.1 9.3 11.9 Other 76 5.6 .46 * * 3.0 5.0 7.1 * * Poverty status 1/ < 100 315 5.6 .22 2.1 2.5 3.5 4.8 6.7 9.4 11.6 > 100 1575 7.2 .14 2.7 3.3 4.6 6.3 8.5 11.3 14.0 < 131 414 5.7 .23 2.0 2.5 3.6 4.9 6.9 9.4 11.7 > 131 1476 7.2 .15 2.8 3.3 4.6 6.4 8.5 11.3 14.2 Education 1/ < High school 305 5.3 .36 1.5 2.1 2.9 4.3 6.0 8.3 10.3 High school 854 6.5 .16 2.6 3.1 4.2 5.9 7.8 10.1 12.8 > High school 891 7.9 22 3.1 3.6 5.0 6.9 9.0 12.4 16.1 Region Northeast 448 6.5 .23 2.4 2.9 4.0 5.6 7.6 10.2 13.0 Midwest 564 7.1 .26 2.5 3.1 4.4 6.2 8.4 11.5 13.8 South 660 6.9 «31 2.3 2.9 4.1 5.8 8.0 10.6 14.0 West 384 7.6 .28 2.9 3.4 4.9 6.8 8.7 11.6 14.3 Urbanization Central city 499 7.0 29 2.4 3.1 4.2 5.9 8.2 11.5 14.5 Suburban 1039 7.0 «20 2.6 3.2 4.5 6.3 8.3 11.1 13.6 Nonmetropolitan 518 6.7 .19 2.2 2.8 4.0 5.6 7.7 10.7 13.2 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 00T-1II Table I-62. Vitamin E: mean intake in milligrams alpha tocopherol equivalents, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 5.5 22 2.4 2.9 3.5 4.7 6.2 8.2 10.2 Age 1-2 years 224 5.4 .41 2.3 2.6 3.3 4.3 5.9 8.5 311.7 3-5 years 423 5.6 .22 2.6 3.0 3.8 4.8 6.4 8.0 10.1 Race 2/ White 559 5.3 .18 2.5 2.9 3.5 4.7 6.3 7.9 9.2 Black 53 *6.8 *1.41 * * 3.6 4.4 6.1 * * Other 26 6.0 1.13 * * * 4.9 x * ® Poverty status 2/ < 100 140 6.4 .76 2.4 2.9 3.4 4.7 6.5 9.2 16.8 > 100 471 5.3 .17 2.5 2.9 3.6 4.7 6.2 7.9 8.9 < 131 192 6.1 .60 2.3 2.6 3.3 4.6 6.1 9.2 16.5 > 131 419 5.3 .18 2.5 3.0 3.6 4.7 6.3 7.8 8.7 Education 2/ < High school 99 6.5 1.00 * 2.4 3.4 4.6 6.5 10.6 * High school 252 5.5 «25 2.3 2.9 3.4 4.4 6.2 8.5 14.2 > High school 295 5.2 .24 2.7 2.9 3.6 4.8 6.2 7.6 8.6 Region Northeast 111 5.3 .35 * 2.1 3.3 4.5 5.9 8.1 * Midwest 199 5.9 .65 2.3 2.8 3.5 4.6 6.0 8.9 15.9 South 187 4.7 .16 2.3 2.8 3.3 4.4 5.8 7.1 7.6 West 150 6.2 .41 2:5 3.2 4.2 5.4 7.3 8.6 10.6 Urbanization Central city 171 6.5 «53 2.5 3.2 4.1 5.5 6.9 8.9 16.5 Suburban 310 4.9 .20 2.5 2.7 3.4 4.5 5.8 7.4 8.1 Nonmetropolitan 166 5.6 .49 2.3 2.7 3.4 4.6 6.3 8.9 11.6 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. T0T-1I Table II-63. Serum vitamin E status of children 4-19 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 T T T ) Mexican American : Cuban : Puerto Rican | T T | T T | T I - | Mean ; | Mean Mean jNunber of | vitamin E j Standard y Number of | vitamin E i Standand Nunter of | vitamin E § Standard j Sxaminey | level j error of jSxaminsy | level |error of | examined | level [stor of Sex and age |persons 1/ umol/L 2/ | the mean |persons 1/ | umol/L 2/ | the mean |persons 1/ | umol/L 2/ | the mean 1 1 1 d L 3. 1 id a i ea Both sexes 4-5 years... 0005 233 18 0.4 12 * * 62 15 0.6 6-11 years. ......... 1,029 17 0.2 96 17 0.5 322 15 0.3 Male 12-15 years......... 351 15 0.3 48 15 Q.5 138 5 0.4 16-19 years... ux» 253 16 0.4 51 16 0.7 137 0.9 Female 3/ 12-15 years......... 328 16 0.3 38 “15 0.7 140 14 0.4 16-19 years... .. . «x 291 17 0.4 41 +17 0.7 123 15 0.6 1/ Includes persons for whom usable measurements for 2/ umol/L = mg/d1%23.22. 3/ Excludes pregnant women. the criteria variable were obtained. GoT-1I Table I-64. Serum vitamin E status of persons 20-74 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 T T T Mexican American Cuban Puerto Rican | T T | T T | T TTTTOy TT Mean ; Mean : ; Mean ] y Number of | vitamin E iS tandang | Numper of vitamin E | Standard | Numper of vitamin t | © tandard | examined | level (error of j exam iney | level jerror of { 2xamineq | level jerver of Sex and age jpersans 1/ junai/L 2/ | the mean jpersons 1/ june) /L 2/ | the mean |persens 1/ |uel/L 2/ | the mean al 1 1 1 1 1. 1 1 iced. ce eo Male 20-74 yR&IrS.. .sviumsrmwve em 1,390 24 0.5 372 26 0.6 423 21 0.6 20-74 years, age adjusted... L 25 LL 2 id 25 + ry. 22 20-29 yeArS..uss xm rune cum 421 20 0.6 55 20 1.1 109 17 0.7 30-39 years. ............... 354 25 1.1 53 25 1.7 86 21 1.5 40-49 YCArS. . «sinus suis ams 230 28 1.3 82 26 1.6 84 25 1.7 BO-89 years. . .ciaus envi cna 229 29 1.1 109 28 0.9 87 27 i.5 60-69 years. ............... 121 26 1.2 45 29 1.6 37 +24 1.4 10=73 YeBNS uns vis mms smuis eon 35 x27 2.8 28 *29 1.9 10 ‘ ' Female 3/ 20-74 yeArS. ..cuccvvemmrnme 1,711 25 0.4 463 25 0.5 705 21 0.4 20-74 years, age adjusted.. 2 # of 22 vex : 8 24 sow . 22 20-29 years. . «cvs nrswwe srw 479 20 0.5 62 20 0.9 172 18 0.7 30-39 years. ............... 427 23 0.7 90 21 0.8 160 20 0.8 40-49 years. ... : cus: nmus ins 300 28 1.1 100 23 0.7 161 23 1.0 B0-859 years... ..s.c wees emn 309 30 1.0 112 30 1.0 126 26 1.1 60-69 years. ............... 136 30 1.4 68 30 1.3 72 26 1. 70-74 years. .. .: insti amiss 60 31 1.9 31 +28 2.0 14 + + 1/ Includes persons for whom usable measurements for the criteria variable were obtained. 2/ umol/L = mg/dl1#*23.22. 3/ Excludes pregnant women. €0T-1I Table II-65. Serum vitamin E status of Mexican-American children 4-19 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Below poverty Above poverty I ——— T T T T Mean ; Mean Number of I vitamin E { Standard Number of I vitamin E | Standard examined I level jsrror of examined | level error of Sex and age persons | umol/L 2/ | the mean |persons Vi) umol/L 2/ j tie mean 1 1 1 1 1 - Both sexes 4-5 y@ars. ..... css. 87 18 0.8 133 18 0.5 6-11 years. ...... «x 378 17 0.3 570 17 0.3 Male 12-18 years. « « «usm 115 15 0.5 200 15 0.3 16-19 years......... 80 16 0.6 142 16 0.6 Female 3/ 12-15 years. ..... ax 124 16 0.4 175 16 0.4 16-19 years. ........ 107 17 0.5 160 17 0.6 1/ Includes persons for whom usable measurements for the criteria variable were obtained. 2/ umol/L = mg/d1%23.22. 3/ Excludes pregnant women. YOT1-11 Table II-66. Serum vitamin E status of Mexican-American persons 20-74 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Below poverty Above poverty ————— J ce ce ce: i: se es es wd T T T T | Mean ; Mean ; Number of | vitamin E j Standard Number of | vitamin E j Sxandard examined | level jerror of examined | level jsrror of Sex and age persons 1/jumol1/L 2/ | the mean persons 1V/jumol/L 2/ | the mean 1 1 1 1 Male 20-74 yCArS.:::sv:swmui sunsmuns 307 22 0.7 964 25 0.6 20-74 years, age adjusted..... “ix x 23 «nw - oo 26 20-29 years. .: as: ssn mrinns s 96 19 0.9 287 20 0.8 30-39 years. ......... 64 23 1.3 269 26 1.3 40-49 years. ............c.c.c... 45 27 2.2 164 28 1.7 80-59 years... ..csvsvives ives 48 24 1.3 157 30 1.4 60-69 years... ................ 40 +26 1.9 69 28 1.7 70-74 years. .................. 14 * + 18 * Female 3/ 20-74 years... vices isms mees 520 24 0.6 1,029 25 0.58 20-74 years, age adjusted. .... S 24 LL L 26 20-29 years... sv uns ims nines 141 20 0.9 303 20 0.6 30-39 years. ........c... i .. 118 23 1.3 276 23 0.8 40-49 YEAS... cr smvr inns inns 86 26 1.8 182 29 1.5 B0-59 years. ...cuinei inns sma 94 27 1.4 179 32 1.8 60-69 years. .................. 45 29 1.7 72 32 2.3 70-78 Y8ArS. .: xr covs suns suns 36 +33 2.5 17 4 + 1/ Includes persons for whom usable measurements for the criteria variable were obtained. 2/ umol/L = mg/d1*23.22. 3/ Excludes pregnant women. goT-1I Milligrams U.S. Food Supply Thiamin 5 1 1 1 t 0 1905 1915 1925 1935 1945 1955 Year 1965 1975 1985 Figure 11-23. Thiamin: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Thiamin Thiamin Grain products 42.37% S—Other foods 1.8% Meat, poultry, fish 25.7% Fruits 4.9% Legumes, nuts, and soy 6.4% Dai ducts 8% Vegetables 10.9% airy products Figure 11-24. Thiamin: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include eggs; fats and oils; sugars and sweeteners; and miscellaneous foods) Table 11-67. Thiamin: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 715 90 95 All women 2056 1.05 .02 .48 .56 .76 .98 1.29 1.61 1.81 Age 20-29 years 661 1.09 .02 .48 .56 .78 1.02 1.35 1.69 1.90 30-39 years 812 1.05 .02 .48 «55 .76 .98 1.29 1.60 1.77 40-49 years 583 1.00 .02 .50 = .58 .73 .94 }.21 1.51 1.70 Race 1/ White 1775 1.07 .02 .50 .58 .76 1.00 1.30 1.62 1.81 Black 167 .93 .04 .38 .47 .66 .88 1.14 1.44 1.68 Other 76 1.05 .05 * * .79 .94 1.43 * * Poverty status 1/ < 100 315 .99 .03 .42 .50 .65 .92 1.24 1.51 1.72 = > 100 1575 1.07 .02 .50 .59 17 1.00 1.30 1.64 1.84 pt 7 < 131 414 1.00 .03 .44 .52 .67 .93 1.24 1.51 1.72 > 131 1476 1.07 .02 .50 .59 77 1.00 1.30 1.64 1.87 Education 1/ < High school 305 .93 .03 .37 .48 .60 .86 1.21 1.49 1.76 High school 854 1.03 .02 .48 .58 .76 .96 1.23 1.54 1.74 > High school 891 1.11 .02 .53 .63 .81 1.04 1.38 1.66 1.85 Region Northeast 448 1.03 .03 .49 «55 .73 .95 1.25 1.54 1.79 Midwest 564 1.12 .03 .50 .61 .78 1.05 1.36 1.68 1.87 South 660 1.03 .02 .46 .54 .73 .95 1.24 1.60 1.88 West 384 1.06 .04 .50 .58 .78 .99 1.30 1.57 1.69 Urbanization Central city 499 1.06 .03 .48 .58 .76 .98 1.34 1.63 1.80 Suburban 1039 1.05 .02 .50 .57 .76 .99 1.27 1.56 1.75 Nonmetropolitan 518 1.06 .02 .45 .55 .75 .98 1.28 1.67 1.88 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. LOT-II Table 11-68. Thiamin: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 15 90 95 All children 1/ 647 1.13 02 .64 +71 .90 1.09 1.31 1.59 1.74 Age 1-2 years 224 1.04 +03 .61 .69 .81 .99 1.19 1.39 1.59 3-5 years 423 1.18 .02 .65 .76 .94 1.12 1.36 1.67 1.80 Race 2/ White 559 1.12 .02 «65 «73 .90 1.08 1.31 1.58 1.73 Black 53 1.238 .07 * * .98 1.11 1.47 * * Other 26 1.00 .08 * x x .97 x * * Poverty status 2/ < 100 140 1.11 .05 .58 .65 .87 2.07 1.32 1.68 1.75 > 100 471 1.13 .02 .66 .74 .90 1.08 1.31 1.59 1.74 < 131 192 1.13 .05 .59 .68 .87 1.08 1.33 1.65 1.76 > 131 419 1.12 .02 .66 +13 .90 1.07 1.30 1.59 1.74 Education 2/ < High school 99 1.18 .05 * .65 .91 1.11 1.37 1.74 * High school 252 1.14 .03 .66 71 .88 1.07 1.34 1.70 1.80 > High school 295 1.10 .02 .64 .76 .91 1.08 1.25 1.47 1.65 Region Northeast 111 1.19 +05 x .71 .98 1.17 1.37 1.68 * Midwest 199 1.16 .04 .65 «73 .86 1.08 1:35 1.69 2.05 South 187 1.07 .03 .60 .67 .84 1.03 1.21 1.48 1.74 West 150 1.12 .04 .67 +72 .91 1.11 1.31 1.58 1.65 Urbanization Central city 171 1.21 .05 .66 .76 .94 1.17 1.41 1.70 1.79 Suburban 310 1.09 «02 .63 .71 .84 1.05 1.25 1.45 1.71 Nonmetropolitan 166 1.13 .04 +01 L177 .90 1.06 1.31 1.55 1.80 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 80T-II Table II-69. Thiamin: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES I NFCS NHANES II CSFI1l1 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 0.80 0.01 0.90 0.02 0.92 0.01 1.08 0.04 3-5 0.97 0.01 1.10 0.02 1.14 0.01 1.23 0.03 6-11 1.16 0.02 1.39 0.01 1.40 0.02 - - Male 12-15 1.48 0.04 1.75 0.03 1.77 0.05 - - 16-19, 1.64 0.06 1.78 0.03 1.97 0.07 - - 20-29 1.53 0.05 1.57 0.03 1.73 0.05 1.81 0.08 30-39 1.40 0.05 1.52 0.03 1.60 0.06 1.72 0.09 40-49 1.31 0.06 1.51 0.03 1.56 0.05 1.77 0.12 50-59 1.19 0.04 1.50 0.03 1.41 0.04 - - 60-69 1.16 0.02 1.43 0.03 1.40 0.02 - - 70+ 1.04 0.02 1.39 0.03 1.28 0.03 - - Female 12-15 1.10 0.03 1.31 0.02 1.16 0.03 - - 16-19 0.92 0.04 1.15 0.02 1.05 0.04 - - 20-29 0.94 0.02 1.05 0.02 1.09 0.03 1.17 0.03 30-39 0.87 0.02 1.03 0.02 1.05 0.05 1.13 0.03 40-49 0.87 0.02 1.01 0.02 1.02 0.03 1.07 0.03 50-59 0.85 0.03 1.04 0.02 0.98 0.03 - - 60-69 0.82 0.01 1.04 0.02 1.01 0.02 - - 70+ 0.77 0.01 1.03 0.02 0.95 0.02 - - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. 601-11 Riboflavin U.S. Food Supply Riboflavin Milligrams 5 1 i 1 1 1 L 1 a 1 1 1 0 1905 1915 1925 1935 1945 1955 1965 1975 1885 Year Figure II-25. Riboflavin: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Riboflavin Dairy products 34.77% Meat, poultry, fish 24.37% Eggs 4.57% Grain products 247% Figure II-26. Riboflavin: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include fats and oils; legumes, nuts, and soy; sugars and sweeteners; and miscellaneous foods) OTI-II Table II-70. Riboflavin: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 1.35 .02 .56 .68 .90 1.20 1.63 2.14 2.55 Age 20-29 years 661 1.42 .03 .59 .69 .93 1.27 1.78 2.31 2.79 30-39 years 812 1.32 .03 .53 .65 .90 1.21 1.64 2.07 2.48 40-49 years 583 1.26 .03 .56 .67 .87 1.14 1.50 1.94 2.46 Race 1/ White 1775 1.38 .02 «59 .71 .93 1.25 1.67 2.21 2.62 Black 167 1.15 .04 .44 .53 .74 .97 1.38 1.90 2.22 Other 76 1.13 .08 x * .70 1.04 1.59 * x Poverty status 1/ < 100 315 1.26 .04 .47 .56 .81 1.09 1.57 2.02 2.35 > 100 1575 1.37 .02 .58 .69 .92 1.22 1.65 2.16 2.62 13) 414 1.25 .04 .47 .54 .82 1.12 1.56 2.04 2.37 > 131 1476 1.38 .02 .60 .71 .92 1.24 1.67 2.15 2.65 Education 1/ < High school 305 1.14 .04 .45 .53 .76 .96 1.37 1.85 2.47 High school 854 1.31 .03 .55 .64 .87 1.16 1.59 2.01 2.45 > High school 891 1.45 .03 .65 .79 1.01 1.35 1.74 2.32 2.70 Region Northeast 448 1.32 .05 .56 .68 .90 1.17 1.61 2.06 2.62 Midwest 564 1.43 .04 .60 .71 .94 1.31 1.78 2.29 2.55 South 660 1.28 .04 .50 .64 .85 1.13 1.54 2.04 2.48 West 384 1.39 .05 .59 .73 .94 1.30 1.69 2.22 2.53 Urbanization Central city 499 1.38 .04 +55 .69 .92 1.22 1.74 2.16 2.68 Suburban 1039 1.33 .03 .56 .68 .92 1.23 1.60 2.06 2.49 Nonmetropolitan 518 1.34 .03 «53 .63 .86 1.15 1.63 2.25 2.70 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. i-n Table II-71. Riboflavin: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 1.61 .03 .90 1.05 1.29 1.54 1.88 2.20 2.50 Age 1-2 years 224 1.56 .04 .90 1.02 1.29 1.53 1.78 1.99 2.41 3-5 years 423 1.64 .03 .94 1.06 1.30 1.58 1.92 2.27 2.50 Race 2/ White 559 1.63 .03 .94 1.06 1.30 1.55 1.90 2.23 2.53 Black 53 1.54 .08 * * 1.23 1.50 1.82 * * Other 26 1.43 .09 x * * 1.48 * * * Poverty status 2/ 100 140 1.58 .07 .92 1.07 1.27 1.50 1.79 2.32 2.61 > 100 471 1.61 .03 .90 1.06 1.30 1.57 1.90 2.19 2.44 131 192 1.598 .06 .92 1.08 1.29 1.50 1.82 2.32 2.60 > 131 419 l.61 .03 .90 1.04 1.30 1.57 1.90 2.19 2.44 Education 2/ < High school 99 1.62 .06 * 1.07 1.34 1.48 1.79 2.46 * High school 252 1.58 .04 .90 1.02 1.23 1.50 1.84 2.27 2.53 > High school 295 1.63 .03 .94 1.07 1.34 1.62 1.90 2.12 2.37 Region Northeast 111 1.66 .05 = 1.09 1.36 1.65 1.85 2.10 * Midwest 199 1.73 .05 1.00 1.14 1.37 1.65 1.97 2.47 2.72 South 187 1.46 .04 .78 .91 1.16 1.41 1.71 2.01 2.19 West 150 1.63 .06 .98 1.09 1.34 1.60 1.85 2.27 2.48 | Urbanization Central city 171 1.67 .06 1.07 1.24 1.37 1.60 1.92 2.19 2.42 Suburban 310 1.56 .03 .88 1.00 1.23 1.50 1.83 2.09 2.34 Nonmetropolitan 166 1.66 .05 .90 1.06 1.29 1.62 1.90 2.49 2.56 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. CII-1I Table II-72. Riboflavin: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 1.56 0.02 1.48 0.02 1.54 0.02 1.65 0.05 3-5 1.70 0.02 1.64 0.02 1.74 0.01 i.712 0.04 6-11 2.02 0.03 2.02 0.02 2.18 0.04 - - Male 12-15 2.49 0.07 2.51 0.04 2.63 0.08 - - 16-19, 2.62 0.10 2.51 0.05 3.08 0.13 - - 20-29, 2.39 0.08 2.13 0.04 2.53 0.08 2.29 0.12 30-39, 2.06 0.08 1.93 0.03 2.21 0.10 1.97 0.09 40-49 1.98 0.07 1.92 0.05 2.05 0.08 2.01 0.13 50-59 1.79 0.06 1.92 0.04 2.00 0.07 - - 60-69 1.76 0.04 1.86 0.04 1.90 0.03 - - 70+ 1.57 0.03 1.77 0.04 1.77 0.06 - - Female 12-15 1.78 0.05 1.85 0.03 1.73 0.05 - - 16-19 1.45 0.05 1.59 0.03 1.59 0.06 - - 20-29 1.40 0.02 1.39 0.02 1.49 0.04 1.51 0.04 30-39 1.31 0.02 1.34 0.02 1.44 0.07 1.42 0.04 40-49 1.33 0.04 1.30 0.03 1.44 0.08 1.34 0.04 50-59 1.31 0.05 1.40 0.03 1.35 0.06 - - 60-469 1.32 0.04 1.40 0.03 1.36 0.02 - - 70+ 1.11 0.02 1.35 0.03 1.33 0.04 - - i CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. eIT-1I Niacin U.S. Food Supply Niacin Milligrams 30 25 T 20 5k 1 a 1 i 1 a 1 1 0 i x 1 i 1 A 1 i 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-27. Niacin: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Niacin Meat, poultry, fish 467% Vegetables 11.17% Figure 11-28. Niacin: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include dairy products; eggs; fats and oils; fruits; sugars and sweeteners; and miscellaneous foods) PIT-1I Table II-73. Niacin: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 30 95 All women 2056 16.0 «18 7.8 9.2 11.9 15.4 19.1 23.9 26.4 Age 20-29 years 661 16.0 +31 7:7 9.6 11.7 15.0 19.9 24.3 26.4 30-39 years 812 15.8 21 8.0 9.2 11.8 15.4 19.0 23.1 26.1 40-49 years 583 16.0 .24 7.6 9.2 12.3 15.6 19.0 23.2 26.4 Race 1/ White 1775 16.2 .21 8.2 9.7 12.1 A5.5 19.6 24.0 26.4 Black 167 14.6 37 6.1 7:2 10.5 14.8 17.2 20.9 25.7 Other 76 15.3 . 90 * * 11..8 15.1 18.7 x * Poverty status 1/ < 100 315 15.0 .36 6.5 8.0 10.9 14.5 18.1 23.1 25.6 > 100 1575 16.2 .19 8.2 9.8 12.0 15.6 19.4 23.7 26.6 < 132 414 15.0 «35 6.5 8.0 11.0 14.6 18.1 22.7 25.6 > 131 1476 16.2 «18 8.3 2.9 12.1 15.6 19.6 23.8 26.6 Education 1/ < High school 305 14.3 .41 6.5 7.4 9.6 13.8 17.5 21.6 26.8 High school 854 15.6 .24 8.0 9.5 11.9 14.8 18.7 23.2 25.2 > High school 891 16.9 wel 8.5 10.2 12.17 16.1 20.6 24.4 26.8 Region Northeast 448 16.1 «32 8.1 9.6 12.1 15.5 19.4 24.0 26.2 Midwest 564 16.5 .36 8.1 9.9 12.5 15.5 20.0 24.4 26.7 South 660 15.8 .29 7.4 8.7 11.7 15.2 18.7 24.0 26.9 West 384 15.5 «352 7.8 9.1 11.4 15.3 19.0 23.0 24.8 Urbanization Central city 499 16.0 .38 7.6 8.8 11.7 15.2 19.5 24.4 26.4 Suburban 1039 15.9 .26 8.2 9.6 12.0 15.6 19.2 23.5 26.2 Nonmetropolitan 518 15.9 .24 7.4 9.6 11.9 15.1 18.7 23.2 26.3 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. SIT-1II Table II-74. Niacin: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 13.8 «27 7:1 8.5 10.6 13.1 16.4 19.5 22.7 Age 1-2 years 224 12.5 .44 6.6 7.3 9.7 12.2 14.0 17.6 21.1 3-5 years 423 14.5 .30 7.9 9.5 11.2 13.9 17.1 19.9 23.3 Race 2/ White . 559 13.6 +25 7:1 8.8 10.7 13.0 16.1 19.3 22.5 Black 53 15.9 1.17 x * 11.5 14.9 19.5 * * Other 26 11.8 .81 x * * 11.3 * * * Poverty status 2/ < 100 140 13.7 .61 7.3 8.6 10.2 13.0 15.9 19.5 22.4 > 100 471 13.8 .30 7.1 8.8 10.7 13.1 16.3 19.4 23.3 < 131 192 13.9 .61 7.2 8.2 10.2 12.9 16.5 20.2 23.8 > 131 419 13.7 .26 Te2 9.1 10.7 13.1 16.1 19.3 22.1 Education 2/ < High school 99 14.4 .68 * 8.0 11.2 13.5 16.7 21.3 * High school 252 14.4 .44 7.8 9.2 10.9 13.5 17.6 20.4 23.8 > High school 295 13.1 +33 7.1 7.8 10.3 12.7 15.1 18.2 20.0 Region Northeast 111 14.9 +75 * 8.9 10.8 13.9 19.1 21.1 * Midwest 199 14.0 .51 6.6 8.4 10.3 13.2 16.3 21.6 24.9 South 187 13.3 .39 7.4 8.6 10.3 iz.8 16.0 18.2 19.5 West 150 13.3 «52 7.2 9.2 10.8 12.6 15.1 18.5 21.4 Urbanization Central city 171 14.8 .70 6.9 8.8 11.5 13.8 17.9 22:4 24.4 Suburban 310 13.1 «31 7.1 8.2 10.2 12.7 15.7 17.9 19.5 Nonmetropolitan 166 14.0 .48 7.6 8.8 11.0 12.8 16.7 19.9 23.3 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 9T1-1I Table II-75. Niacin: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANESI), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 8.82 0.15 10.3 0.18 10.82 0.14 12.4 0.53 3-5 11.20 0.14 13.5 0.20 13.92 0.11 15.6 0.52 6-11 14.37 0.24 17.5 0.18 17.52 0.27 - - Male 12-15 18.89 0.51 22.3 0.33 23.00 0.61 - - 16-19, 23.85 0.83 24.4 0.40 29.55 1.02 - - 20-29 25.21 0.70 24.6 0.37 29.42 0.70 28.0 1.09 30-39; 24.94 0.80 23.6 0.39 26.12 0.72 26.2 1.37 40-49 23.52 0.71 23.8 0.42 25.90 0.76 25.3 1.05 50-59 20.52 0.61 23.5 0.35 23.15 0.63 - - 60-69 18.54 0.30 21.6 0.36 21.23 0.26 - - 70+ 15.82 0.28 19.4 0.42 18.78 0.41 - - Female 12-15 14.01 0.38 17.0 0.3) 15.34 0.42 - - 16-19 13.65 0.49 16.4 0.32 15.04 0.50 - - 20-29 14.36 0.20 15.7 0.26 16.17 0.34 17.3 0.43 30-39 15.14 0.25 16.1 0.24 16.81 0.45 17.1 0.32 40-49 15.05 0.30 16.4 0.24 16.50 0.47 16.6 0.36 50-59 + 14.41 0.45 17.0 0.27 15.10 0.40 - - 60-69 13.45 0.24 16.0 0.26 14.83 0.18 - 70+ 11.33 0.18 14.8 0.26 14.14 0.28 - % CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. LIT-1II U.S. Food Supply Vitamin B6 Milligrams 5 Vitamin B6 1 1 1 1 1 1 1 0 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure 11-29. Vitamin B6: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Vitamin B6 Meat, poultry, fish 41.17% Other foods 2.8% Vegetables 21.9% y Lequmes, nuts, and soy 47% Grain products 8.9% Dairy products 10.7% Fruits 10.6% Figure I1-30. Vitamin B6: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include eggs; fats and oils; sugars and sweeteners; and miscellaneous foods) 8TT-II Table 11-76. Vitamin B6: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 15 30 95 All women 2056 1.17 .02 .50 .62 .82 1.07 1.44 1.83 2.12 Age 20-29 years 661 1.20 .02 .54 .64 .81 1.07 1.48 1.91 2.17 30-39 years 812 1.17 .03 .48 .59 .83 1.07 1.45 1.8) 2.13 40-49 years 583 1.13 «+02 7% | .64 .81 1.06 1.39 1.72 1.97 Race 1/ White 1775 1.20 «02 .54 . 64 .84 1.09 1.46 1.86 2.15 Black 167 1.00 +03 .40 .49 .75 .95 1.25 1.59 1.76 Other 76 1.07 .06 * * .76 1.00 1.40 * * Poverty status 1/ < 100 315 1.05 +03 .40 .47 .70 .96 1.28 1.73 1.92 > 100 1575 1.20 .02 +55 .65 .83 1.08 1.46 1.83 2.15 < 131 414 1.05 .03 .40 .49 +72 .98 1.28 1.76 1.92 > 131 1476 1.23 .02 .55 .65 .84 1.11 1.47 1.86 2.18 Education 1/ < High school 305 .98 +03 .40 +47 .67 .91 1.13 1.56 1.88 High school 854 1.12 .02 .49 .61 .79 1.01 1.35 1.75 1.99 > High school 891 1.28 .02 .61 «72 .91 1.19 1.57 1.94 2.26 Region Northeast 448 1.15 .03 «53 +62 .81 1.05 1.36 1.86 2.10 Midwest 564 1.22 .04 «91 .64 .84 1.11 1.54 1.91 2.12 South 660 1.14 «+03 .47 +58 .78 1.03 1.39 1.78 2.21 West 384 1.20 .04 +55 .69 .87 1.12 1.48 1.76 2.04 Urbanization Central city 499 1.20 .03 .49 .63 .81 1.09 1.48 1.92 2.18 Suburban 1039 1.17 .03 «52 .63 .84 1.07 1.42 1.78 2.05 Nonmetropolitan 518 1.15 «03 .50 .59 L717 1.02 1.41 1.84 2.10 do not add to the number of all women. 611-11 Table II-77. Vitamin B6: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 «79 .01 .46 .52 .61 .74 .90 1.12 1.28 Age 20-29 years 661 L717 .01 .46 .51 .60 .73 .87 1.11 1.26 30-39 years 812 «79 .01 .45 .50 .60 . 75 .92 1.12 1.29 40-49 years 583 .81 .01 .49 .53 .63 .75 «92 1.14 1.33 Race 1/ White 1775 .79 .01 .46 .52 .61 .74 .91 1.14 1.30 Black 167 .76 .02 .41 .47 .59 .71 .88 1.09 1.20 Other 76 17 .03 * * .64 .73 .88 * * Poverty status 1/ < 100 315 .76 .02 .44 .48 .59 .71 .86 1.05 1.21 > 100 1575 .79 .01 .46 .52 .61 .74 .92 1.13 1.28 < 131 414 .75 .02 .46 .50 .59 .71 .85 1.02 1.20 > 131 1476 79 .01 .46 .52 .62 .75 .92 1.14 1.29 Education 1/ < High school 305 17 .03 .44 .50 .58 .70 .85 1.09 1.29 High school 854 .75 .01 .45 .51 .59 .71 .87 1.04 1.21 > High school 891 .83 .01 .48 .52 .64 .78 .94 1.18 1.33 Region Northeast 448 .80 .01 .48 .53 .64 77 .92 1.15 1.29 Midwest 564 .78 .02 .45 .51 .61 .73 .91 1.09 1.31 South 660 .78 .02 .42 .49 .59 .73 .90 1.10 1.27 West 384 .79 .01 .49 .54 .62 .74 .90 1.17 1.26 Urbanization Central city 499 .80 .01 .47 .53 .62 .74 .91 1.17 1.30 Suburban 1039 .79 .01 .46 .52 .61 .75 «91 1.13 1.28 Nonmetropolitan 518 17 .02 .45 .50 .60 .7 .90 1.08 1.23 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 021-11 Table II-78. Vitamin B6: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 1.24 .02 .65 .76 .94 1.19 1.45 1.78 2.00 Age 1-2 years 224 1.19 .04 .63 .76 .92 1.13 1.40 1.76 1.86 3-5 years 423 1.26 .03 .66 .76 .96 1.22 1.48 1.80 2.03 Race 2/ White 559 1.24 .02 .68 .78 .95 1.20 1.45 1.76 1.97 Black 53 1.28 .09 * * .88 1.19 1.57 * Other 26 1.10 .09 x * * 1.07 * * Poverty status 2/ 100 140 1.20 .07 .61 .69 .82 1.18 1.46 1.74 1.97 > 100 471 1.25 .02 .70 .80 .98 1.20 1.45 1.79 2.03 < 131 192 1.22 .06 .64 .71 .82 1.20 1.47 1.78 2.07 > 131 419 1.25 .02 .69 .81 .98 1.20 1.45 1.78 1.97 Education 2/ < High school 99 1.25 .06 * .73 .92 1.22 1.45 1.78 * High school 252 1.24 .04 .65 .74 .94 1.16 1.46 1.86 2.12 > High school 295 1.22 .03 .65 .78 .94 1.20 1.44 1.76 1.84 Region Northeast 111 1.28 .06 * .64 .96 1.28 1.58 1.87 * Midwest 199 1.26 .04 .73 .78 .95 i.21 1.46 1.79 2.23 South 187 1.17 .04 .63 .74 .90 1.14 1.38 1.75 1.79 West 150 1.26 .06 .70 .81 .97 1.24 1.47 1.80 1.97 Urbanization Central city 171 1.33 .06 .73 .81 .99 1.30 1.57 1.97 2.18 Suburban 310 1.20 .02 .63 .75 .92 1.14 1.40 1.76 1.88 Nonmetropolitan 166 1.22 .04 .64 .74 .91 1.21 1.45 1.76 1.87 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 121-11 Table II-79. Vitamin B6: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 .88 .01 .55 .61 .72 .85 1.01 1.13 1.29 Age 1-2 years 224 .92 .02 .57 .63 .74 .88 1.04 1.20 1.35 3-5 years 423 .85 .01 .53 .61 .71 .83 .98 1.11 1.25 Race 2/ White 559 .87 .01 +57 .62 71 .84 1.01 1.13 1.29 Black 53 9 .06 * x .73 .92 1.08 * * Other 26 .91 .06 * * * .88 * * * Poverty status 2/ < 100 140 .86 .03 .46 .58 .68 .86 .99 1.14 1.31 > 100 471 .88 .01 .57 .63 .73 .86 1.01 1.13 1.29 < 131 192 .89 .03 .49 .62 .71 .88 1.00 1.14 1.31 > 131 419 .87 .01 .55 .63 .72 .85 1.01 1.13 1.29 Education 2/ < High school 99 .89 .04 * .62 .70 .87 1.04 1.23 * High school 252 .88 .02 .54 .61 .1N .87 1.00 1.13 1.28 > High school 295 .87 .02 .57 .62 L712 .83 1.00 1.21 1.27 Region Northeast 111 .89 .03 * .61 .74 .87 1.03 1.13 * Midwest 199 .89 .02 .57 .62 .713 .86 1.00 1.19 1.30 South 187 .85 .02 .58 .63 .70 .78 .98 1.11 1.25 West 150 .89 .03 .51 .59 .74 .89 1.01 1.14 1.30 Urbanization Central city 171 .90 .02 .61 .64 .17 .87 1.01 1.14 1.30 Suburban 310 .87 .02 .53 .60 .70 .85 1.01 1.12 1.28 Nonmetropolitan 166 .87 .02 .56 .63 .71 .82 .99 1.19 1.30 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. GCI-11 Vitamin B12 U.S. Food Supply Vitamin B12 Micrograms 14 12+ 10+ 8 | 1 i 1 1 1 1 1945 1955 1965 1975 1985 Year 1 1 0 1 1905 1915 1925 1935 Figure 11-31. Vitamin B12: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Vitamin B12 Meat, poultry, fish 75.4% N Other foods .1% Grain products 1.5% ggs 5.47% Dairy products 17.6% Figure 11-32. Vitamin B12: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include fats and oils) 31-11 Table II-80 Vitamin B12: mean intake in micrograms, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 4.6 .15 1.3 1.7 2.4 3.4 4.9 7.0 10.2 Age 20-29 years 661 4.5 .32 1.4 1.8 2.4 3.4 4.9 7.0 8.7 30-39 years 812 4.4 .20 1.2 1.5 2.5 3.4 4.8 6.9 10.3 40-49 years 583 5.1 «32 1.3 1.7 2.3 3.4 4.8 7.1 12.6 Race 1/ White 1775 4.5 .14 1.3 1.7 2.5 3.5 4.9 7.0 10.1 Black 167 6.0 .84 «9 1.4 2.0 3.1 4.7 6.9 12.6 Other 76 4.4 .46 x * 2.1 3.2 5.3 % * Poverty status 1/ < 100 315 5.5 .75 1.0 1.4 2.0 3.1 4.4 7.0 10.9 > 100 1575 4.4 .11 1.3 1.7 2.5% 3.4 4.9 7.1 10.1 < 131 414 5.0 .58 1.0 1.4 2.0 3.2 4.6 6.6 8.7 > 131 1476 4.5 A 1.3 1.7 2.5 3.5 5.0 7.1 10.5 Education 1/ < High school 305 4.6 .46 «9 1.5 2.0 3.1 4.1 6.0 11.8 High school 854 4.5 .31 1.3 1.7 2.4 3.4 4.8 6.8 8.6 > High school 891 4.8 .14 1.4 1.8 2.5 3.5 5.1 7.6 11.2 Region Northeast 448 5.0 .24 1.4 1.8 2.6 3.4 4.9 7.7 11.7 Midwest 564 4.7 .36 1.3 1.7 2.6 3.5 4.9 6.9 8.9 South 660 4.5 .30 1.2 1.5 2.2 3.2 4.7 6.9 10.1 West 384 4.4 .17 1.3 1.7 2.5 3.7 5.0 7.1 8.6 Urbanization Central city 499 5.0 «29 1.3 1.7 2.5 3.7 5.1 7.3 10.3 Suburban 1039 4.3 .13 1.3 1.7 2.5 3.3 4.7 6.9 10.1 Nonmetropolitan 518 4.9 .46 1.2 1.6 2.2 3.3 4.8 6.9 11.2 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. ¥e1-11 Table II-81. Vitamin B12: mean intake in micrograms, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 4.0 .13 1.8 2.2 2.8 3.5 4.4 5.5 6.8 Age 1-2 years 224 4.0 .17 1.8 2.2 2.8 3.5 4.1 5.5 6.8 3-5 years 423 4.0 .15 1.8 2.3 2.8 3.5 4.6 5.6 6.9 Race 2/ White 559 3.9 .11 1.8 2.3 2.8 3.5 4.4 5.5 6.8 Black 53 4.8 .62 * A 2.9 3.6 4.8 * x Other 26 3.4 .18 * * * 3.5 * * Poverty status 2/ < 100 140 4.4 .50 1.9 2.2 2.9 3.5 4.7 6.1 7.9 > 100 471 3.8 .12 1.9 2.3 2.8 3.5 4.3 5.5 6.4 < 131 192 4.3 .38 2.0 2.3 2.9 3.5 4.7 5.9 7.7 > 131 419 3.8 «32 1.8 2.3 2.8 3.5 4.3 5.4 6.5 Education 2/ < High school 99 4.1 .23 * 2.5 3.2 3.7 14.7 5.9 * High school 252 4.2 «27 1.8 2.1 2.8 3.5 4.7 6.3 7.3 > High school 295 3.8 .14 1.8 2.3 2.8 3.5 4.1 5.4 6.0 Region Northeast laa 4.0 .38 * 1.8 2.7 3.6 4.7 5.6 * Midwest 199 4.4 .27 2.2 2.6 3.3 3.9 4.7 6.0 7.2 South 187 3.7 «21 1.6 2.0 2.7 3.2 3.8 5.1 5.9 West 150 3.8 .14 2.1 2.4 2.9 3.5 4.3 5.8 6.5 Urbanization Central city 171 4.2 33 2.1 2.5 2.9 3.7 4.6 5.7 6.7 Suburban 310 3.6 .10 1.7 2.0 2.7 3.3 4.2 5.1 6.8 Nonmetropolitan 166 4.5 .27 1.9 2.3 2.9 3.7 5.1 6.1 7.6 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. AR U.S. Food Supply Vitamin C Milligrams 140 120+ 100 art IN 80 60 40} 20} 0 ! J LL ee XQ 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-33. Vitamin C: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Vitamin C Vitamin C Vegetables 47.97% eat, poultry, fish 2.3% Dairy products 3.17% Other foods 47% Fruits 42.77% Figure II-34. Vitamin C: food sources in the U.S. food supply, 1 Food Supply Series (other foods include legumes, nuts, an and sweeteners; and miscellaneous ft 985: U.S. sugars 921-11 Table II-82. Vitamin C: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 78 2.1 18 25 40 67 105 148 177 Age 20-29 years 661 79 3.6 20 26 40 65 107 i153 187 30-39 years 812 78 2.6 37 23 38 67 105 147 171 40-49 years 583 78 2.6 19 26 43 68 102 141 170 Race 1/ White 1775 78 2.4 20 26 40 67 104 146 176 Black 167 78 6.2 16 22 39 70 109 156 178 Other 76 792 6.9 * * 35 68 109 * * Poverty status 1/ < 100 315 61 3.6 13 17 28 48 79 123 140 > 100 1575 83 2.3 20 26 43 70 109 153 186 < 131 414 64 3.3 14 18 32 51 83 123 143 > 131 1476 83 2.4 20 26 43 70 109 155 188 Education 1/ < High school 305 57 3.3 13 16 27 47 70 109 150 High school 854 72 2.5 17 23 37 59 97 132 165 > High school 891 90 2:5 26 31 50 80 115 162 194 Region Northeast 448 80 3.4 18 25 46 71 108 151 178 Midwest 564 79 545 18 23 40 67 102 146 189 South 660 74 3.0 16 23 37 63 101 144 165 West 384 83 4.7 23 29 43 70 109 149 176 Urbanization Central city 499 81 3.8 18 24 41 68 113 153 170 Suburban 1039 81 3.3 20 26 42 70 106 149 187 Nonmetropolitan 518 69 3.2 15 23 35 55 90 126 156 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. L3T-1I Table II-83. Vitamin C: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 83 2.75 26 32 49 69 107 146 168 Age 1-2 years 224 84 4.04 26 33 49 70 106 150 175 3-5 years 423 82 2.95 26 32 49 69 108 143 164 Race 2/ White 559 84 2.93 26 34 49 69 110 150 175 Black 53 75 6.82 * % 49 69 97 * * Other 26 85 11.23 * * * 79 * * * Poverty status 2/ < 100 140 67 4.76 23 29 41 59 87 114 138 > 100 471 88 2.717 28 36 52 78 116 15% 175 < 131 192 71 4.14 24 30 43 63 91 121 146 > 131 419 88 3.08 28 36 52 79 117 153 176 Education 2/ < High school 99 65 5.39 x 26 36 54 85 119 * High school 252 78 4.02 26 32 49 66 98 132 157 > High school 295 92 3.62 26 39 54 85 124 156 176 Region Northeast ii) 93 7.97 * 29 54 72 125 157 * Midwest 199 83 3.79 29 35 52 76 11a 140 165 South 187 75 5.26 23 29 41 59 100 143 157 West 150 84 4.68 34 40 54 71 105 150 164 Urbanization Central city 171 89 4.92 27 34 52 75 116 155 167 Suburban 310 82 4.24 25 31 48 68 109 146 176 Nonmetropolitan 166 75 4.18 27 33 47 66 93 124 147 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 821-11 Table II-84. Vitamin C: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 71 2.4 66 2.2 88 1.8 81 3.8 3-5 82 2.2 72 2.2 100 1.6 81 3.1 6-11 81 2.5 87 2.0 107 3.3 - - Male 12-15 91 5.2 95 3.1 117 6.3 - - 16-19, 109 7.4 100 3.2 125 7.7 - 20-29 102 6.0 89 2.7 118 5.7 107 8.6 30-39 78 5.3 84 2.3 102 5.8 104 7.1 40-49 84 5.1 86 3.1 98 5.2 124 13.4 50-59 91 5.4 95 2.8 105 5.6 - - 60-69 93 2.4 90 2.8 101 2.0 - - 70+ 91 2.6 89 3.3 102 3.3 - - Female 12-15 79 4.2 79 2.2 82 4.5 - - 16-19 82 5.6 77 2.1 79 4.9 - 20-29 80 2.3 74 2.2 95 4.1 86 4.1 30-39 76 2.2 73 2.0 86 4.7 86 4.1 40-49 79 2.9 78 2.2 91 4.9 81 3.2 50-59 90 4.6 86 2.3 102 5.0 - - 60-69 96 2.5 92 2.4 102 1.8 - 70+ 83 2.3 85 2.2 112 3.2 - i CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. 621-11 Folacin U.S. Food Supply Folacin Micrograms 400 300 “oT Nl 200+ 100 1 i 1 i Xk i 1 & 1 i 1 i 1 i 1 0 i 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year : Fi, 11-35. Folacin: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Dy rir perc PRY Folacin Legumes, nuts, and soy 19.57% . Vegetables 24.87% Grain products 12.7% S—Other foods 1.8% Eggs 7.7% Meat, poultry, fish 12.6% ’ Dairy products 8.5% Fruits 12.4% Figure 11-36. Folacin: food sources in the U.S. food supply, 1985: U.S. Rots Supply Series (other foods include fats and oils; and miscellaneous 0eT-1I Table 11-85. Folacin: mean intake in micrograms, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 193 3.1 78 93 127 176 236 310 359 Age 20-29 years 661 197 4.9 79 93 130 182 239 325 379 30-39 years 812 193 4.6 75 91 126 179 242 311 355 40-49 years 583 183 4.3 78 99 127 166 218 287 337 Race 1/ White 1775 196 3.7 82 99 132 179 239 311 363 Black 167 165 6.5 56 70 106 153 199 292 330 Other 76 179 12.7 * * 104 170 237 * * Poverty status 1/ < 100 315 166 6.1 59 15 100 139 199 302 334 > 100 1575 198 3.3 83 100 134 182 238 308 367 < 131 414 168 6.1 59 78 105 148 201 294 331 > 131 1476 199 3.3 83 101 136 183 239 311 370 Education 1/ < High school 305 157 6.0 58 69 95 137 190 269 332 High school 854 181 4.0 75 90 120 161 222 292 328 > High school 891 214 3.5 96 116 147 196 260 328 387 Region Northeast 448 188 6.1 75 98 126 ina 224 309 363 Midwest 564 193 7.7 75 89 127 176 240 310 348 South 660 190 5.0 76 89 124 168 233 300 374 West 384 200 6.1 89 100 138 189 245 313 346 Urbanization Central city 499 197 4.9 78 97 126 182 244 319 388 Suburban 1039 193 5.1 81 96 134 179 233 306 343 Nonmetropolitan 518 186 5.1 74 85 118 161 228 305 351 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 181-11 Table II-86. Folacin: mean intake in micrograms per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 130 1.4 67 75 94 121 153 197 231 Age 20-29 years 661 129 2.3 64 73 92 118 150 196 217 30-39 years 812 131 2.4 66 73 95 121 155 199 241 40-49 years 583 131 2.0 70 17 95 121 153 199 232 Race 1/ White 1775 1131 1.6 69 16 96 121 153 198 231 Black 167 126 4.0 56 66 87 112 158 197 220 Other 76 128 8.2 * % 86 120 161 * * Poverty status 1/ < 100 315 119 2.8 58 67 80 104 147 183 205 > 100 1575 132 1.7 69 78 96 122 153 198 231 < 131 414 120 2.4 61 69 83 110 147 182 205 > 131 1476 132 1.8 69 78 96 122 154 198 233 Education 1/ < High school 305 123 4.1 62 67 85 110 145 198 213 High school 854 123 1.9 66 73 88 113 145 184 210 > High school 891 139 2.0 72 84 102 126 164 209 241 Region Northeast 448 132 3.0 67 76 96 124 161 196 214 Midwest 564 125 3.4 64 72 89 119 146 187 220 South 660 132 2.4 66 72 92 118 152 199 243 West 384 133 1.9 74 81 98 122 159 197 231 Urbanization Central city 499 134 2.1 67 75 94 119 156 204 241 Suburban 1039 130 2.1 67 75 96 122 155 197 222 Nonmetropolitan 518 126 3.2 68 72 89 114 145 190 231 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. e1-11 Table 11-87. Folacin: mean intake in micrograms, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 191 4.8 92 108 137 177 229 287 350 Age 1-2 years 224 180 6.6 90 104 120 164 216 274 335 3-5 years 423 197 5.3 93 111 146 185 242 295 352 Race 2/ White 559 189 4.6 95 109 137 177 228 276 340 Black 53 221 22.1 * * 148 226 295 * * Other 26 177 18.6 * * * 158 * * * Poverty status 2/ < 100 140 188 12.8 91 103 120 164 231 295 363 > 100 471 192 4.8 97 111 141 180 230 286 345 < 131 192 191 10.8 92 103 124 173 231 318 363 > 131 419 192 4.5 95 111 141 181 229 280 340 Education 2/ < High school 99 196 11.0 x 110 143 174 232 321 * High school 252 194 8.3 93 103 130 174 230 318 372 > High school 295 187 4.9 93 110 138 180 229 271 313 Region Northeast lia 205 13.6 * 104 141 195 247 335 * Midwest 199 193 9.1 93 108 128 177 230 313 376 South 187 178 7.0 92 103 124 164 228 268 307 West 150 196 9.6 103 124 149 184 230 283 318 Urbanization Central city 171 206 10.6 107 116 146 187 257 325 372 Suburban 310 184 6.2 91 103 127 175 228 272 328 Nonmetropolitan 166 188 8.9 89 109 136 169 226 286 356 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. eeT-1I Table 11-88. Folacin: mean intake in micrograms per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 135 2.17 79 86 102 127 159 193 217 Age 1-2 years 224 138 3.9 77 87 102 134 161 203 221 3-5 years 423 134 2.8 79 84 100 124 159 190 212 Race 2/ White 559 132 2.5 79 86 101 123 155 187 212 Black 53 160 15.8 * * 106 155 192 * * Other 26 145 12.7 * * * 126 * * * Poverty status 2/ < 100 140 136 7.0 70 82 94 121 160 210 230 > 100 471 135 2.6 79 87 103 128 159 188 212 < 131 192 140 5.8 71 82 100 125 168 208 230 > 131 419 133 2.6 79 87 102 127 155 187 212 Education 2/ < High school 99 142 7.9 * 77 103 130 176 217 * High school 252 137 4.5 79 83 100 130 161 207 213 > High school 295 131 2.4 82 87 102 123 152 185 206 Region Northeast 111 142 7.0 * 87 106 137 174 201 * Midwest 199 135 4.4 74 84 100 125 162 193 213 South 187 130 5.4 79 86 100 117 149 186 221 West 150 138 4.0 80 85 109 134 160 188 210 Urbanization Central city 171 141 4.6 82 88 104 134 163 202 228 Suburban 310 133 3.9 77 85 102 123 157 187 212 Nonmetropolitan 166 135 5.0 75 82 99 123 159 196 221 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. Ve1-11 U.S. Food Supply Iron Milligrams 20 10} 5k 0 i s i 1 4 1 i 1 i 1 i 1 i 1 A 1 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-37. Iron: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Iron Iron Grain products 41% Meat, poultry, fish 23.8% ggs 4.4% y Legumes, nuts, and soy 6.67% Vegetables 12.6% Other foods 11.6% Figure II-38. Iron: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include dairy products; fats and oils; fruits; sugars and sweeteners; and miscellaneous foods) Table I-89. Iron: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 10.1 .14 4.9 5.9 7.5 9.6 11.9 15.2 17.5 Age 20-29 years 661 10.2 .23 4.9 5.8 7.3 9.6 12.1 16.2 18.7 30-39 years 812 10.2 .19 4.9 5.9 7.5 9.8 12.0 15.2 16.9 40-49 years 583 9.9 «7 5.4 6.1 7.7 9.3 11.4 14.5 16.0 Race 1/ White 1775 10.2 .15 5.2 6.1 7.6 9.6 12.0 15.1 17.1 Black 167 9.1 .37 4.0 4.5 6.2 8.8 11.0 14.1 16.4 Other 76 10.7 .71 x * 7.3 10.3 13.6 * * Poverty status 1/ < 100 315 9.5 .24 4.3 4.9 6.8 8.9 11.3 14.8 16.5 = > 100 1575 10.3 .14 5.3 6.2 7.7 9.7 2.1 15.2 17.1 — 8 < 131 414 9.6 .26 4.3 5.1 6.9 9.1 11.4 14.8 17.5 > 131 1476 10.3 .13 5.3 6.2 7.17 9.7 12.1 15.2 16.9 Education 1/ < High school 305 8.9 .27 4.0 4.7 6.1 8.4 10.9 14.5 16.3 High school 854 9.8 .18 4.9 5.8 7.3 9.3 11.4 14.5 16.5 > High school 891 10.8 .16 5.9 6.7 8.1 10.3 12.7 16.0 18.7 Region Northeast 448 10.1 .31 5.1 5.9 7.4 9.8 11.5 14.6 18.4 Midwest 564 10.4 .28 5.1 6.1 7.7 9.6 12.7 15.6 18.4 South 660 9.8 .20 4.6 5.8 7.2 9.2 11.5 14.6 16.9 West 384 10.3 .33 5.3 6.4 7.9 9.9 12.5 15.3 16.6 Urbanization Central city 499 10.3 .29 5.2 6.0 7.5 9.6 12.3 16.2 18.3 Suburban 1039 10.0 .20 5.0 6.0 7.6 9.7 11.9 14.4 16.5 Nonmetropolitan 518 10.1 .20 4.9 5.8 7.4 9.3 11.8 15.5 17.8 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 9eT-1I Table II-90. Iron: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 6.8 .06 4.5 5.0 5.6 6.5 7.7 9.1 10.0 Age 20-29 years 661 6.6 .10 4.3 4.8 5.4 6.3 75 8.8 9.8 30-39 years 812 6.9 .08 4.6 5.0 5.7 6.6 7.8 9.1 9.9 40-49 years 583 7.1 .07 4.8 5.1 5.8 6.8 7.9 9.6 10.3 Race 1/ White 1775 6.8 .06 4.6 4.9 5.6 6.5 7.6 9.0 10.0 Black 167 6.8 .19 4.5 4.9 5.4 6.5 8.0 8.1 10.0 Other 76 7.6 «37 * x 6.4 7.9 8.6 * * Poverty status 1/ < 100 315 6.8 «12 4.6 5.1 5.7 6.6 8.0 8.9 9.4 > 100 1575 6.8 .06 4.6 5.0 5.6 6.5 7.6 9.2 10.0 < 131 414 6.9 .12 4.6 5.1 5.6 6.7 7.9 9.0 10.0 > 131 1476 6.8 .07 4.6 4.9 5.6 6.4 7.6 9.1 9.9 Education 1/ < High school 305 6.9 «15 4.5 4.8 5.8 6.7 7.9 9.3 10.3 High school 854 6.6 .09 4.4 4.8 5.5% 6.4 7:5 8.9 9.8 > High school 891 6.9 .07 4.7 5.1 5.7 6.6 7.9 9.2 10.2 Region Northeast 448 7.1 .13 4.7 5.2 5.9 6.8 8.0 9.3 10.0 Midwest 564 6.7 .12 4.6 4.9 5.5 6.3 7.5 9.1 10.0 South 660 6.7 .10 4.5 4.8 5.4 6.4 7.5 8.8 9.9 West 384 6.8 «11 4.4 5.0 5.7 6.6 7.8 9.2 10.2 Urbanization Central city 499 6.9 .09 4.5 4.9 5.7 6.6 7.8 9.3 10.2 Suburban 1039 6.7 .08 4.6 5.0 5.6 6.5 7.6 9.0 9.9 Nonmetropolitan 518 6.9 .15 4.5 5.0 5.6 6.6 7.8 9.2 10.5 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. LET-II Table 11-91. Iron: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 9.8 «21 5.7 6.3 7.5 9.3 11.3 14.0 15.9 Age 1-2 years 224 9.2 .29 5.4 5.8 6.9 8.6 10.4 13.3 15.6 3-5 years 423 10.2 «23 6.1 6.7 8.0 9.7 12.5 14.3 16.1 Race 2/ White 559 9.7 .17 5.8 6.4 75 9.2 11.0 13.7 15.4 Black 53 11.3 1.10 * * 8.2 10.1 13.1 * * Other 26 9.2 .80 * * * 9.2 * * Poverty status 2/ < 100 140 10.0 ,58 5.1 5.8 7.1 9.2 11.7 15.1 17.5 > 100 471 9.8 «21 5.9 6.4 7.7 9.3 11.0 13.6 15.4 < 131 192 10.1 .56 5.2 5.8 7.1 9.2 12.0 15.3 18.3 > 131 419 9.7 «15 5.9 6.5 7.8 9.3 10.9 13.4 15.3 Education 2/ < High school 99 10.4 .52 * 5.8 7.5 10.2 12.3 14.9 * High school 252 10.1 .37 5.6 6.4 7.6 9.4 11.5 15.3 16.8 > High school 295 9.5 .20 5.9 6.4 7.5 9.2 10.6 12.7 14.5 Region Northeast 111 10.9 .64 * 6.2 8.0 10.4 12.6 15.4 * Midwest 199 10.2 «39 5.9 6.4 7.5 9.2 11.6 15.3 18.2 South 187 9.0 «23 5.2 5.9 7.1 8.9 10.2 12.0 13.7 West 150 9.7 .36 5.8 6.5 8.0 9.2 10.6 13.7 16.4 Urbanization Central city 171 10.5 .54 5.8 6.5 8.0 9.6 11.5 15.6 17.5 Suburban 310 9.5 .24 5.6 6.2 7.5 9.2 10.7 13.4 15.3 Nonmetropolitan 166 9.7 .28 5.4 6.3 7.2 9.2 1pr.2 14.1 16.3 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. Table II-92. Iron: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 7.0 .10 4.7 5.1 5.6 6.5 7.9 9.2 10.5 Age 1-2 years 224 7.1 .17 4.7 5.1 5.6 6.5 8.2 9.8 11.5 3-5 years 423 6.9 «12 4.8 5.1 5:7 6.5 7.7 8.9 10.1 Race 2/ White 559 6.8 .10 4.7 5.1 5.6 6.4 7.8 8.9 10.0 Black 53 8.1 .58 * * 6.4 7.1 8.9 * * Other 26 7.5 .49 * * * 7.5 * * Poverty status 2/ < 100 140 7.1 .30 5.0 5.1 5.7 6.5 8.0 9.9 11.6 = > 100 471 6.9 «11 4.6 5.0 5.6 6.5 7.9 9.0 10.2 re < 132 192 7.3 vad 4.9 5.0 5.6 6.5 8.2 9.9 11.8 8 > 131 419 6.9 .10 a.6 5.1 5.6 6.5 7.8 8.8 10.0 Education 2/ < High school 99 7.4 .36 * 5.1 5.8 6.7 8.5 10.5 High school 252 A wl? 5.0 5.2 5:7 6.5 8.2 9.3 10.5 > High school 295 6.7 +12 4.6 5.0 5.6 6.2 7.6 8.7 10.2 Region Northeast 111 7.6 «23 * 5.8 6.4 7.1 8.4 9.9 * Midwest 199 7.2 .20 4.7 5.3 5.6 6.7 8.0 9.3 11.6 South 187 6.6 .18 4.7 5.0 5.3 6.0 7.6 8.8 9.8 West 150 6.9 +15 4.6 5.2 5.7 6.4 7.7 8.9 11.1 Urbanization Central city 171 1.2 .25 4.8 5.2 5.6 6.5 8.0 10.0 11.6 Suburban 310 6.9 +12 4.7 5.1 5.7 6.5 7.9 8.9 10.0 Nonmetropolitan 166 6.9 17 4.7 5.0 5.6 6.5 7.7 8.1 10.4 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 6€1-11 Table II-93. Iron: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 7.38% 0.16 8.1 0.16 8.57 0.13 10.2 0.51 3-5 8.58 0.11 9.5% 0.12 10.02 0.09 11.0 0.34 6-11 10.81 0.17 12.2 0.12 12.34 0.21 - - Male 12-15 14.13 0.42 15.6 0.20 16.01 0.45 - - 16-19 16.70 0.51 16.9 0.26 18.15 0.60 - - 20-29 16.55 0.40 16.2 0.23 17.59 0.45 16.7 0.79 30-39 16.54 0.48 18.9 0.21 16.48 0.49 15.6 0.70 40-49 15.26 0.42 16.1 0.24 16.58 0.53 15.6 0.81 50-59 13.85 0.38 15.9 0.23 15.21 0.43 - - 60-69 13.08 0.18 14.9 0.25 14.73 0.20 - - 70+ 11.68 0.16 14.2 0.28 13.24 0.29 - - Female 12-15 10.44 0.28 11.9 0.21 10.71 0.32 - - 16-19 9.54 0.30 11.2 0.20 10.04 0.34 - - 20-29 10.06 0.13 10.7 0.17 10.67 0.23 131.1 0.27 30-39 10.36 0.14 11.1 0.15 11.08 0.31 11.2 0.26 40-49 10.40 0.18 11.0 0.14 11.10 0.34 10.6 0.22 50-59 10.15 0.28 11.5 0.17 10.30 0.30 - - 60-69 9.53 0.14 11.0 0.15 10.53 0.13 - - 70+ 8.63 0.13 10.4 0.16 10.18 0.22 - - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. 01-1 Table 11-94. Iron deficiency determined by the MCV model of children 4-19 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 T T T | Mexican American Cuban Puerto Rican | | | i | Standard | Standard | Standard {Number of jPercent j error of Number of Percent jerror of Number of jpercent j Error of j examined ith iron | the jexanineg itn iron | the | examined juith iron | the Sex and age jpersons 1/ deficiency 2/ |percent jpersons 1/|deficiency 2/ | percent jPersons 1/ deficiency 2/ percent 1 1 1 1 1 1 1 1 1 Both sexes 4-5 years. .......... 225 3.9 1.1 14 * * 60 7.0 83.2 6-1Y years. .....«... 1,012 3.4 0.5 97 3.1 1.7 313 2.5 0.8 Male 12-15 years. ....co:ux 339 3.5 1.0 50 6.7 3.3 134 2.6 1.4 16-19 years... ...... 242 0.4 0.4 51 0.0 0.0 133 1.2 0.9 Female 3/ 12-15 years......... 307 5.9 1.3 39 *0.0 0.0 134 6.5 2.1 16-19 years... ...... 277 7.9 1.6 42 *0.0 4.6 117 7.9 2.4 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The Criteria variable is discussed in the table notes. 2/ Iron deficiency assessed by MCV model. 3/ Excludes pregnant women. 91-11 Table 11-95. Iron deficiency determined by the MCV model of persons 20-74 years of age by sex, age, and specified Hispanic origin: Hispanic Health and Nutrition Examination Survey, 1982-84 T T T Mexican American ; Cuban Puerto Rican | | | Standard i | Standard | | | Standard Hunper of jrercent jereor of | Number of jFercent j error of jhlunber of jPercent jerror of | &xamined julien iron | the j SXanined jaiib iron the j examined juith iron the Sex and age Persons 1/ deficiency 2/ percent jPsrsons 1/ deficiency 2/ percent jpersons 1/|deficiency 2/ |percent 1 1 1 1 1 1 1. 1 1 Male 20-74 years... ..cuviveeenn 1,344 0.9 0.3 363 0.5 0.4 415 0.4 0.4 20-74 years, age adjusted 0.9 0.6 0.3 20-29 years. ..... cc vue own 413 0.7 0.5 54 1.5 2.1 107 0.0 0.0 30-39 years. ............. 343 1.3 0.8 52 0.0 0.0 83 1.1 1.5 40-49 years... xxx ssms 226 1.0 0.7 80 0.0 0.0 83 0.6 1.0 B0-50 years... ccvcvwninn 216 0.9 0.6 104 0.0 0.0 95 0.0 0.0 60-69 years. ............. 111 0.0 0.0 45 0.0 0.0 37 *0.0 0.0 70-74 years. .....«.xsiine 35 2.9 3.2 28 +3.8 3.6 10 * + Female 3/ 20-74 years. .... cc ccvvena 1,653 9.0 0.8 449 5.8 1.2 677 6.9 1.1 20-74 years, age adjusted 8.5 5.2 6.2 20-29 years. ............. 468 8.2 1.5 60 2.8 2.5 162 7.2 2.5 BO-39 YOANrS. wv uvvvn ween 407 11.7 1.8 87 7.2 3.3 155 7.9 2.7 40-49 YOAPrS. ..c.nnnsinmen 288 12.9 2.0 100 10.2 3.3 155 8.3 2.4 50-59 years. ............. 301 4.0 1.1 109 4.0 1.9 120 3.9 1.6 60-689 years. . ox vexmas 132 4.2 1.8 66 1.7 1.5 72 3.8 2.0 70-74 years. ............. 57 8.4 3.8 27 +7.0 4.9 13 * + 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Iron deficiency assessed 3/ Excludes pregnant women. by MCV model. [44811 Table II-96. Iron deficiency determined by the MCV model of Mexican-American children 4-19 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Below poverty Above poverty ——————— | Standard | | Standard Number of jrercent jsrror of Number of |Percent jerror of examined with iron | the examined jrith iron the Sex and age | persons 1/ deficiency 2/ percent |persons 1/ deficiency 2/ percent 1 1 i 1 1 1 Both sexes 4-5 years. .......... 84 3.3 1.7 128 3.8 1.6 6-11 years. ......... 372 4.1 0.9 559 2.8 0.6 Male 12-15 years. ........ 111 5.0 1.9 193 2.0 1.0 16-19 years. ........ 76 $2 1.3 136 0.0 0.0 Female 3/ 12-15 years......... 116 4.3 1.8 163 5.9 1.8 16-19 years. ........ 103 9.2 2.8 151 6.3 1.9 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Iron deficiency assessed by MCV model. 3/ Excludes pregnant women. EVI-11 Table II-97. Iron deficiency determined by the MCV model of Mexican-American persons 20-74 years of age by sex, age, and poverty status: Hispanic Health and Nutrition Examination Survey, 1982-84 Below poverty Above poverty T umber of Percent jerror of Number of [Percent with iron the examined Ni examined [with iron p ersons 1/ T | | | Standard ; | Sex and age | 1 T | | | | | | 1 De ——————) deficiency 2/ percent 1 Male 20-74 YRAIS. ... sinus smms insisis 298 0.2 0.3 930 1.2 20-74 years, age adjusted. .... 0.3 .2 20-29 Y@APS. ... sr rennin usivsss 95 0.0 0.0 280 1.0 BO-39 ¥YRaIrS.«: suns neimweswaus 64 0.0 0.0 259 1.6 40-49 years. .........oa.an 45 0.0 0.0 161 1.1 50-59 years. ............cnn.nn 46 1.8 1.9 147 0.8 BO=BY YRBArS, =n wstimms smnsicmmn 34 *0.0 0.0 65 0.0 70-74 Years. .... cicurvwursnvs 14 * * 18 * Female 3/ 20-748 YEAPS. i: :ussummssmmivns 505 10.6 1.4 994 7.9 20-74 years, age adjusted..... we 10.1 7.2 20-29 years... asin imnsrnn 140 9.6 2.9 296 7.7 30-39 yRAIrS. ...ionvnrmmnrmns smn 112 14.6 3.6 265 9.8 40-49 years... .... cnn 84 19.2 4.2 172 9.8 BOBO YRAPrS. «ix nz s 5 vice mvoimwm ws mn 92 3.1 1.7 174 4.1 60-69 years. ............o.... 42 *2.6 2.5 71 5.8 TO~-784 YRAPrS. . +: cs asammsvmwewe 35 *7.6 4.6 16 * | Standard jerror of | the persons 1/ deficiency 2/ percent : 1. lo] on 000-0 » ONVOO® N= NON = » Nb WO 1/ Includes persons for whom usable measurements for the criteria variable were obtained. variable is discussed in the table notes. 2/ Iron deficiency assessed by MCV model. 3/ Excludes pregnant women. The criteria Yi-11 Table 11-98. Iron deficiency determined by the MCV model of children 4-19 years of age by sex, age, and race: Second National Health and Nutrition Examination Survey, 1976-80 Non-Hispanic white Non-Hispanic black T | | Standard i | Standard Number of jFercent jsrror of Number of jPercent jerror of examined jritn iron the examined jHith iron | the Sex and age persons 1) 1 1 deficiency 2/|percent joersons 1/ 1 deficiency 2/ percent 1 1 Both sexes 4-8 y@ArS. enw imus 771 3.8 0.5 180 9.3 1.4 6-11 years. ......... 1,085 3.2 0.6 219 4.0 1.5 Male 12-15 years. ........ 492 1.8 0.7 87 2.2 1.8 16-19 years..... x. 480 0.8 0.5 91 0.9 1.1 Female 3/ 12-15 years... ...... 418 2.1 0.8 101 8.0 2.9 16-19 years.....sv:u 456 3.8 1.1 83 13.8 4.4 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the tabla notes. 2/ Iron deficiency assessed by MCV model. 3/ Excludes pregnant women. SyI-1I Table 11-99. Iron deficiency determined by the MCV model of persons 20-74 years of age by sex, age, and race: Second National Health and Nutrition Examination Survey, 1976-80 Non-Hispanic white Non-Hispanic black T T T | standard | standard Number of [Percent jerror of Number of jC Ercent jerror of examined with iron | the examined juith iron the P SE Sex and age jpersens 1/ deficiency 2/ percent ersons 1/| det iciency 27 eercent 1 1 1 1 Male 20-74 YOArS. .: wn: nnermms amvames 4,464 .3 0.2 565 6 0.7 20-74 years, age adjusted..... AP .3 2.7 . 20-29 years... ... scr wis oven 977 0.8 0.4 146 0.0 0.0 30-39 years. ..............n 677 0.7 0.4 88 3.1 2.3 40-49 years. ...... vans 551 1.4 0.7 60 2.4 2.7 BO-89 YOAIrS.: wx: cunsnmwsmme nie 551 1.1 0.6 72 5.9 3.4 60-69 years. ..........nn 1,301 2.4 0.3 144 3.0 1.0 70-74 years. ........ououuueenonn. 407 3.5 0.7 S55 6.1 2.3 Female 3/ 20-74 YyO&rS. ..:uvvrsvwrmmnirmunn 4,832 4.2 0.3 666 6. 1.0 20-74 years, age adjusted... .. CEE 4.2 6.1 20-29 years. .... aaa. 956 2.4 0.6 163 3.5 1.8 B0-39 years. .......sssacwssnns 741 6.5 1.2 104 5.9 2.9 40-89 YRArS...: sus ivwwrmns rns 587 6.5 1.4 90 14.0 4.4 50-59 years. ..........oenn- 623 3.5 1.0 96 2.4 1.9 60-69 years. ............n 1,424 3.1 0.4 153 7.9 1.6 70-74 years... ..:cxscccanarsvna 501 2.7 0.6 60 4.2 1.9 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Iron deficiency assessed by MCV model. 3/ Excludes pregnant women. 9YI1-1I Table 11-100. Iron deficiency determined by the MCV model of children 4-19 years of age by sex, age, and poverty status: Second National Health and Nutrition Examination Survey, 1976-80 Below poverty Above poverty T | | | | | 1 | "Standard i | Standard Number of jPercent jsrror of Number of jrercent j error of examined juin iron | the examined with iron | the Sex and age |persons 1/ deficiency 2/ percent |persons 1/ deficiency 2/ percent 1 1 1 1 1 Both sexes 4-5 years........... 192 6.4 1.1 737 4.2 0.5 6-11 years. ......... 235 3.8 1.3 1,034 3.3 0.7 Male 12-15 years... ...... 89 1.1 1.2 466 1.9 0.8 16-19 years. ........ 101 0.9 1.1 445 0.9 0.5 Female 3/ 12-15 years. ........ 102 5.1 2.4 397 2.5 0.9 16-19 years. ........ 132 8.7 2.8 384 3.6 1.2 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Iron deficiency assessed by MCV model. 3/ Excludes pregnant women. LYT-11 Table 11-101. Iron deficiency determined by the MCV model of persons 20-74 years of age by sex, age, and poverty status: Second National Health and Nutrition Examination Survey, 1976-80 Below poverty be ee ee ce en ef | | Standard Number of jrencent jSrror of examined grish iron I the Sex and age persons 1/ deficiency 2/ percent ersons 1/ deficiency 2/ percent 1 1 1 1 Male 20-74 CAPS. ss: sms ni inns NE ns 505 3.1 0.8 4,343 «3 0.2 20-74 years, age adjusted..... 8.2 +2 20-29 years... .. iii 139 0.8 0.9 a952 0.7 0.3 BO-39 yOaArs. .. i «snsnnrsunssnms 59 4.1 3.1 684 0.6 0.4 40-89 YRAPS. . vt cvveccvmammn nimns 32 *2.1 3.2 551 1.6 0.7 50-59 years... ..... on 46 4.0 3.4 544 1.6 0.7 B06 YEBrS: au: mvs cmmspmatnns 147 6.0 1.4 1,249 2.1 0.3 TO~74 YEAPS. .coivavimmrrmwsmnn 82 5.6 1.8 363 3.4 0.8 Female 3/ 20-74 YOArS. . rs: ives numssmancins 804 6.3 0.9 4,476 4.2 0.3 20-74 years, age adjusted..... 53 6.6 4.2 20-29 YRAPrS. uns wwe cvmsmmmeinn 189 3.7 1.7 901 2.5 0.7 30-39 years. ...... ian 111 10.7 3.6 714 5.8 1.1 40-49 YyCArS. i curv: sms swims amas ¢ 3 72 10.1 4.4 585 7.1 1.4 BOBBY yLAIrS.. vx cnwosmmniniwnsnm 76 5.3 3.14 609 3.4 1.0 60-69 years. .............n 240 4.4 1.0 1.251 3.4 0.4 TFO-74 YGArsS... .«::xxsssvswvein 116 4.8 1.5 416 2.5 0.6 1/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 2/ Iron deficiency assessed by MCV model. 3/ Excludes pregnant women. Above poverty | | Standard Number of jrercent jerrer of examined jrith iron | the Pp 81-11 Calcium U.S. Food Supply Calcium Milligrams 1500 1250 } 1000 } 750 } TT 500 } 250 — 1 1 1 1 i 1 i 1 A 1 i 1 1915 1925 1935 1945 1955 1965 1975 1985 Year 0 13805 Figure I1-39. Calcium: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Calcium Dairy products 76.8% \ egumes, nuts, and soy 3.1% Grain products 3.6% Meat, poultry, fish 3.6% Vegetables 6.2% Other foods 6.7% << Figure 11-40. Calcium: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include eggs; fats and oils; fruits; sugars and sweeteners; and miscellaneous foods) 6v1-11 Table 11-102. Calcium: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 630 12 213 273 387 558 783 1062 1280 Age 20-29 years 661 678 17 232 300 408 596 841 1239 1480 30-39 years 812 617 17 199 262 383 568 799 1032 1197 40-49 years 583 562 18 213 256 354 495 685 955 1144 Race 1/ White 1775 661 12 241 303 409 583 808 1102 1314 Black 167 429 26 145 185 260 384 564 682 810 Other 76 494 36 x x 346 446 594 x * Poverty status 1/ < 100 315 532 22 155 208 318 486 695 881 1001 > 100 1575 652 13 238 293 404 573 798 1105 1307 < 131 414 538 23 158 206 333 487 706 890 1032 > 131 1476 659 14 244 297 409 578 802 1116 1311 Education 1/ < High school 305 478 18 151 183 298 419 603 823 946 High school 854 606 17 208 256 367 525 765 1053 1244 > High school 891 699 15 273 337 461 640 857 1128 1349 Region Northeast 448 603 24 217 273 362 529 765 1035 1235 Midwest 564 665 20 231 300 426 598 821 1102 1273 South 660 571 23 192 242 346 489 714 931 1196 West 384 708 19 220 318 443 622 883 1261 1444 Urbanization Central city 499 648 19 215 278 393 567 812 1105 1404 Suburban 1039 631 19 226 279 396 564 779 1063 1268 Nonmetropolitan 518 598 20 197 249 362 515 751 1008 1252 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 0ST-1I Table 11-103. Calcium: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 412 5 202 233 298 384 495 618 710 Age 20-29 years 661 426 9 214 244 311 398 517 650 773 30-39 years 812 410 7 199 224 294 390 506 619 692 40-49 years 583 388 9 206 233 285 362 456 572 666 Race 1/ White 1775 426 6 214 246 309 398 514 623 728 Black 167 316 14 168 193 229 308 385 456 513 Other 76 354 23 * * 259 334 424 * * Poverty status 1/ < 100 315 376 13 174 212 262 354 465 591 647 > 100 1575 419 6 211 244 305 387 501 620 728 < 131 414 373 12 183 214 270 355 453 575 639 > 131 1476 423 7 213 244 306 391 507 625 731 Education 1/ < High school 305 364 11 174 205 259 332 448 557 653 High school 854 398 9 199 220 285 368 480 601 718 > High school 891 439 7 237 265 326 412 524 644 733 Region Northeast 448 413 10 202 235 296 387 501 620 692 Midwest 564 419 9 216 247 320 399 500 613 684 South 660 378 11 199 215 271 340 443 588 690 West 384 455 12 226 255 332 416 549 691 789 Urbanization Central city 499 417 7 202 237 297 394 495 637 747 Suburban 1039 415 9 207 237 305 384 501 613 705 Nonmetropolitan 518 395 11 175 214 279 366 480 612 687 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 181-11 Table 11-104. Calcium: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 804 15 410 490 614 769 966 1129 1299 Age 1-2 years 224 798 23 387 488 616 789 977 1129 1203 3-5 years 423 807 18 418 492 614 764 966 1155 1324 Race 2/ White 559 826 15 453 505 621 795 990 1160 1333 Black 53 639 31 * * 565 657 724 * * Other 26 725 56 * * * 665 * * * Poverty status 2/ < 100 140 764 27 396 492 620 724 894 1084 1200 > 100 471 814 18 410 488 4598 794 979 1151 1333 < 131 192 770 25 396 488 618 723 896 1085 1299 > 131 419 818 19 408 490 598 799 989 1149 1314 Education 2/ < High school 99 749 26 * 441 584 738 848 1075 * High school 252 765 19 417 488 578 715 914 1100 1192 > High school 295 852 21 434 494 662 832 1031 1175 1376 Region Northeast 111 812 26 x 530 631 784 991 1141 x Midwest 199 851 23 467 506 663 799 1013 1212 1404 South 187 722 25 370 414 556 710 872 1062 1129 West 150 855 36 460 537 674 841 990 1157 1376 Urbanization Central city 171 818 38 483 537 674 793 266 1098 1253 Suburban 310 799 18 394 486 578 770 987 1129 1229 Nonmetropolitan 166 798 29 418 482 605 752 925 1186 1437 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. SaT-1I Table 11-105. Calcium: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 572 9 345 391 459 554 657 779 881 Age 1-2 years 224 622 18 372 404 492 602 721 869 967 3-5 years 423 545 8 329 383 448 528 623 716 796 Race 2/ White 559 582 9 347 393 476 567 667 789 872 Black 53 460 15 x x 407 456 500 * * Other 26 606 53 * x x 522 * * * Poverty status 2/ < 100 140 555 18 375 393 433 529 651 748 848 > 100 471 574 10 329 385 464 556 663 784 889 < 131 192 564 17 372 391 439 534 649 770 910 > 131 419 572 11 323 380 465 556 665 774 860 Education 2/ < High school 99 538 17 * 391 432 495 593 719 * High school 252 544 12 309 380 438 527 626 748 821 > High school 295 605 12 329 401 506 597 698 850 910 Region Northeast 111 577 18 * 393 458 570 653 850 * Midwest 199 602 14 351 403 510 599 680 796 883 South 187 520 14 303 366 425 505 608 702 749 West 150 604 21 380 420 490 569 697 849 957 Urbanization Central city 173 567 25 376 401 456 534 644 742 910 Suburban 310 579 12 307 385 459 572 674 813 889 Nonmetropolitan 166 561 12 354 383 462 538 653 757 834 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. €a1-11 Table II-106. Calcium: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 885 13 755 16 770 11 843 23 3-5 921 12 759 13 818 8 828 18 6-11 1,093 18 936 12 1,029 19 - - Male 12-15 1,309 37 1,146 22 1,202 41 - - 16-19 1,310 55 1,144 26 1,370 56 - - 20-29 1,115 46 909 18 1,096 40 1,067 65 30-39, 917 43 819 19 889 35 853 42 40-49 887 42 749 19 830 36 847 59 50-59 784 31 757 16 832 38 - - 60-49 763 15 708 16 755 12 - - 70+ 693 12 708 20 664 19 - - Female 12-15 940 31 849 15 854 32 - - 16-19 744 33 716 16 725 35 - - 20-29 685 12 628 11 662 21 691 20 30-39 605 12 567 11 595 24 656 20 40-49 604 16 532 10 596 27 600 19 50-59 582 22 555 10 569 28 - - 60-59 559 11 558 9 552 9 - 70+ 537 11 555 11 546 14 - i CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. ¥a1-11 U.S. Food Supply Phosphorus Milligrams 2000 1200} 800 400} 1 1 d 1 0 1 - 1905 1915 1925 1935 1945 1955 Year 1 1 i 1965 1975 1985 Figure II-41. Phosphorus: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Phosphorus Phosphorus Dairy products 35.7% Meat, poultry, fish 29.2% Legumes, nuts, and soy 6.1% Grain products 13.2% Vegetables 7.5% Figure 11-42. Phosphorus: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include fats and oils; fruits; sugars and sweeteners; and miscellaneous foods) Ga1-1I Table 11-107. Phosphorus: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 975 13 469 548 722 909 1172 1437 1634 Age 20-29 years 661 1019 18 485 590 745 939 1248 1575 1854 30-39 years 812 963 18 457 534 716 908 1178 1420 1600 40-49 years 583 915 19 457 556 712 873 1096 1331 1461 Race 1/ White 1775 1002 14 497 570 745 932 1197 1459 1679 Black 167 794 28 350 416 579 774 956 1251 1333 Other 76 894 45 * * 689 822 1109 * * Poverty status 1/ < 100 315 879 23 380 453 625 823 1105 1290 1425 > 100 1575 995 13 492 576 744 927 1189 1449 1635 < 131 414 882 24 371 452 636 842 1107 1313 1455 > 131 1476 1003 13 506 583 751 229 1194 1453 1651 Education 1/ < High school 305 813 23 350 418 582 765 986 1255 1367 High school 854 955 16 469 547 710 874 1146 1420 1561 > High school 891 1045 16 534 632 798 980 1232 1553 1738 Region Northeast 448 932 26 475 547 692 874 1113 1355 1582 Midwest 564 1022 25 509 582 747 939 1250 1545 1718 South 660 931 23 424 505 695 875 1104 1393 1558 West 384 1030 24 498 590 782 969 1249 1497 1692 Urbanization Central city 499 1000 25 448 565 734 917 1215 1497 1738 Suburban 1039 968 19 480 547 729 910 1166 1435 1611 Nonmetropolitan 518 954 20 441 529 702 890 1133 1396 1558 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 9g1-11 Table 11-108. Phosphorus: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 1010 16 616 684 814 980 1162 1392 1498 Age 1-2 years 224 963 21 577 664 799 947 1119 1315 1376 3-5 years 423 1036 20 633 695 827 1006 1192 1436 1547 Race 2/ White 559 1025 16 631 695 820 998 1189 1394 1520 Black 53 930 51 * * 763 954 1066 * * Other 26 915 42 x % * 912 * * * Poverty status 2/ < 100 140 992 34 568 667 835 954 1128 1384 1475 > 100 471 1016 18 616 695 806 988 1189 1394 1507 < 131 192 988 30 592 667 814 945 1132 1392 1416 > 131 419 1020 19 622 695 811 998 1188 1394 1524 Education 2/ < High school 99 986 33 * 648 817 971 1114 1376 * High school 252 994 23 621 671 763 941 1147 1394 1475 > High school 295 1030 21 627 697 870 1020 1188 1373 1507 Region Northeast 111 1020 33 * 669 795 1016 1191 1394 * Midwest 199 1044 26 665 720 841 1006 1186 1407 1524 South 187 947 28 568 607 771 926 1076 1281 1465 West 150 1048 36 657 726 8717 1017 1188 1402 1507 Urbanization Central city 171 1046 36 657 783 870 989 1212 1416 1487 Suburban 310 990 18 596 665 783 980 1141 1318 1464 Nonmetropolitan 166 1013 34 587 648 811 967 1187 1384 1541 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. LGT-1I Table II-109. Phosphorus: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANESI), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES I NFCS NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 1,005 12 889 14 1,014 13 987 25 3-5 1,100 12 976 13 1,164 10 1,083 21 6-11 1,337 17 1,230 12 1,412 23 - - Male 12-15 1,643 37 1,560 23 1,757 51 - - 16-19, 1,873 58 1,663 28 2,060 70 - - 20-29 1,775 47 1,534 22 1,911 50 1,703 71 30-39 1,605 47 1,441 22 1,609 44 1,472 59 40-49 1,498 42 1,385 24 1,520 46 1,446 58 50-59 1,336 36 1,358 19 1,435 41 - - 60-69 1,244 17 1,259 23 1,290 14 - - 70+ 1,096 14 1,178 22 1,133 21 - - Female 12-15 1,204 29 1,178 16 1,246 40 - - 16-19 1,046 33 1,074 18 1,125 51 - - 20-29 1,029 13 1,010 12 1,117 38 1,065 21 30-39 970 13 976 13 1,024 28 1,049 22 40-49 974 17 949 13 994 28 972 21 50-59 939 25 965 13 932 30 - - 60-69 880 12 932 12 894 10 - - 70+ 793 11 890 13 876 16 o- - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. 8ST-1I U.S. Food Supply Magnesium Milligrams 500 400 + 300 + MAAN 200 + 100 + he 1 i 1 1 1 1 1 —~—k 1 & 1 0 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure II-43. Magnesium: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Magnesium Magnesium Grain products 17.7% Vegetables 15.87% Meat, poultry, fish 15.47% Legumes, nuts, and soy 137% Figure I-44. Magnesium: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include eggs; fats and oils; sugars and sweeteners; and miscellaneous foods) 6GT-1I Table 11-110. Magnesium: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 207 2.8 99 119 154 199 254 310 346 Age 20-29 years 661 204 4.0 103 117 150 192 250 314 356 30-39 years 812 210 3.7 96 118 155 200 259 309 346 40-49 years 583 207 4.1 99 123 156 201 244 303 339 Race 1/ White 1775 213 2.9 106 124 159 205 258 314 351 Black 167 158 6.8 64 85 112 152 185 264 304 Other 76 202 11.9 * * 139 197 251 * x Poverty status 1/ < 100 315 175 5.4 76 95 125 159 216 268 309 > 100 1575 213 2.17 105 124 160 206 260 314 352 < 131 414 177 5.6 76 95 126 166 220 276 313 > 131 1476 215 2.6 108 126 161 208 260 315 353 Education 1/ < High school 305 170 4.5 74 96 124 158 205 265 299 High school 854 197 3.2 95 112 146 191 238 296 331 > High school 891 227 3.2 117 137 171 218 275 335 361 Region Northeast 448 199 5.4 105 120 147 185 239 300 330 Midwest 564 212 6.4 101 117 156 203 260 315 356 South 660 196 4.7 86 112 145 182 235 301 346 West 384 225 4.7 112 133 171 222 275 318 351 Urbanization Central city 499 211 6.2 98 117 153 199 260 318 356 Suburban 1039 208 4.0 105 123 156 201 254 308 345 Nonmetropolitan 518 198 3.6 91 107 144 193 240 297 333 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 091-11 Table II-111. Magnesium: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 139 1.0 91 99 115 134 159 189 206 Age 20-29 years 661 132 1.6 87 95 109 128 147 173 188 30-39 years 812 142 1.6 91 100 115 136 161 196 209 40-49 years 583 147 1.6 97 106 121 140 167 195 220 Race 1/ White 1775 141 1.1 94 102 117 136 160 190 210 Black 167 119 3.5 78 84 96 110 137 169 177 Other 76 146 6.1 * * 121 141 172 x * Poverty status 1/ < 100 315 126 2.4 85 90 104 123 144 170 195 > 100 1575 141 1.1 92 101 117 136 162 190 206 < 131 414 126 2.1 84 90 104 123 143 169 189 > 131 1476 142 1.2 93 102 118 137 163 191 208 Education 1/ < High school 305 132 2.5 86 92 108 132 151 184 198 High school 854 133 1.3 89 96 109 129 153 183 202 > High school 891 146 1.3 97 108 122 142 165 195 218 Region Northeast 448 139 1.9 92 101 116 137 158 184 199 Midwest 564 137 2.7 90 97 113 132 156 190 202 South 660 134 1.7 85 94 110 129 152 183 205 West 384 148 2.1 102 107 124 145 168 202 220 Urbanization Central city 499 140 1.9 91 99 116 135 158 182 209 Suburban 1039 140 1.5 91 101 116 136 163 190 205 Nonmetropolitan 518 135 2.2 89 96 110 130 152 188 214 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 191-11 Table 11-112. Magnesium: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 194 3.5 108 130 156 190 229 267 298 Age 1-2 years 224 186 4.5 107 130 154 185 213 245 264 3-5 years 423 199 4.4 109 130 158 193 237 276 310 Race 2/ White 559 198 3.2 113 132 158 193 233 270 298 Black 53 177 12.8 * * 143 167 202 * * Other 26 176 11.4 * * * 181 * * * Poverty status 2/ < 100 140 185 11.4 95 118 147 180 212 267 302 > 100 471 198 3.4 114 135 158 194 231 267 297 < 131 192 185 9.1 105 118 142 177 216 266 314 > 131 419 199 3.3 114 141 160 197 231 268 296 Education 2/ < High school 99 178 6.4 * 124 146 172 202 247 * High school 252 189 5.6 112 125 150 179 214 271 309 > High school 295 204 4.4 107 144 165 203 237 268 293 Region Northeast 111 196 7.6 * 130 163 193 214 273 x Midwest 199 193 4.8 110 134 157 189 228 259 274 South 187 181 5.9 101 121 147 175 217 244 274 West 150 211 8.5 130 142 173 203 249 281 322 Urbanization Central city 171 209 9.2 124 141 164 202 250 302 323 Suburban 310 191 3.3 107 131 155 189 226 250 272 Nonmetropolitan 166 184 7.1 107 121 149 179 210 263 282 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. C9T-1I Table II-113. Magnesium: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 137 1.6 98 106 120 135 154 171 182 Age 1-2 years 224 143 2.1 102 114 126 141 162 179 192 3-5 years 423 134 1.7 97 104 115 133 149 167 178 Race 2/ White 559 139 1.5 102 107 122 137 154 171 180 Black 53 126 6.3 * * 108 127 140 * * Other 26 145 7.5 x * * 152 * * * Poverty status 2/ < 100 140 133 4.5 87 98 114 128 154 168 178 > 100 471 139 1.4 101 108 122 137 154 172 184 < 131 192 133 3.4 89 104 114 131 154 167 173 > 131 419 139 1.6 101 108 123 138 154 175 184 Education 2/ < High school 99 127 3.4 * 93 113 126 146 160 * High school 252 133 2.2 97 105 116 128 149 167 176 > High school 295 144 2.1 102 110 127 141 156 179 191 Region Northeast 111 136 3.3 * 106 122 137 153 166 * Midwest 199 136 2.8 103 108 118 134 154 168 176 South 187 iz 2.6 93 101 113 128 146 166 1373 West 150 148 3.0 103 114 127 149 164 184 193 Urbanization Central city 171 141 3.1 103 110 126 139 157 i 184 Suburban 310 138 2.2 97 105 121 136 154 176 187 Nonmetropolitan 166 129 2.8 96 104 113 124 144 163 170 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 91-11 Table II-114. Sodium: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 2372 25 1061 1338 1748 2252 2862 3518 3971 Age 20-29 years 661 2483 41 1083 1361 1827 2407 3022 3742 4228 30-39 years 812 2293 34 1051 1303 1716 2190 2811 3429 3850 40-49 years 583 2283 49 1034 1354 1740 2178 2738 3381 3710 Race 1/ White 1775 2386 26 1119 1366 1791 2263 2862 3504 3931 Black 167 2209 82 914 1102 1542 2073 2708 3236 3804 Other 76 2502 175 * x 1717 2425 3326 * * Poverty status 1/ < 100 315 2291 66 938 1129 1576 2124 2908 3564 4138 > 100 1575 2396 26 1121 1379 1803 2279 2872 3501 3907 < 131 414 2301 61 938 1131 1602 2218 2914 3564 3947 > 131 1476 2400 28 1165 1391 1806 2275 2860 3502 3925 Education 1/ < High school 305 2078 64 800 1033 1377 1969 2696 3296 3704 High school 854 2376 38 1099 1361 1805 2263 2862 3511 3846 > High school 891 2459 38 1237 1456 1814 2321 2922 3632 4180 Region Northeast 448 2257 52 1107 1312 1723 2140 2700 3239 3641 Midwest 564 2526 46 1118 1453 1855 2384 3079 3714 4208 South 660 2304 41 1021 1294 1706 2236 2767 3464 3839 West 384 2414 59 1029 1356 1781 2229 2901 3571 4242 Urbanization Central city 499 2418 49 1102 1390 1740 2235 2962 3632 4242 Suburban 1039 2345 38 1098 1320 1740 2257 2818 3467 3865 Nonmetropolitan 518 2367 38 1008 1286 1788 2263 2853 3532 3909 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. yI1-11 Table II-115. Sodium: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 2036 34 1192 1289 1608 1945 2406 2883 3102 Age 1-2 years 224 1858 44 1096 1201 1519 1792 2153 2643 2943 3-5 years 423 2133 40 1233 1391 1682 2021 2511 2990 3362 Race 2/ White 559 2037 37 1192 1258 1604 1936 2410 2890 3120 Black 53 2135 79 * * 1688 2016 2450 * * Other 26 1885 86 * * * 1880 * * * Poverty status 2/ 100 140 2077 87 1184 1312 1576 1968 2426 2933 3096 > 100 471 2026 37 1188 1305 1609 1928 2393 2883 3102 131 192 2042 68 1221 1329 1590 1963 2387 2922 3067 > 131 419 2035 39 1158 1258 1611 1932 2412 2890 3112 Education 2/ < High school 99 2101 84 * 1305 1590 2028 2471 2999 * High school 252 2067 48 1201 1278 1631 1968 2424 3007 3174 > High school 295 1992 47 1147 1261 1598 1926 2357 2746 2989 Region Northeast 111 2018 66 * 1233 1519 1979 2412 2970 * Midwest 199 2130 76 - 1238 1325 1618 2038 2511 2990 3406 South 187 1982 59 1127 1270 1555 1926 2333 2806 3362 West 150 2023 69 1209 1423 1644 1919 2334 2836 3008 Urbanization Central city 171 2101 60 1231 1508 1771 2012 2424 2922 3044 Suburban 310 1968 51 1127 1233 1555 1868 2328 2814 3074 Nonmetropolitan 166 2126 59 1235 1312 1604 2005 2465 3007 3528 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. e91-11 Table II-116. Sodium: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES I NFCs NHANES II CSFII 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 1,631 26 - - 1,828 21 1,873 65 3-5 1,925 24 - - 2,173 18 2,169 51 6-11 2,393 37 - - 2.718 44 - - Male 12-15 2,923 83 - - 3,405 96 - - 16-19 3,219 117 - - 4,030 135 - - 20-29 3,123 96 - - 3,916 107 4,021 209 30-39, 2,928 116 - - 3,550 112 3,604 160 40-49 2,839 103 - - 3,542 127 3,330 127 50-59 2,515 88 - - 3,278 105 - - 60-69 2,381 39 - - 2,975 38 - - 70+ 2,114 31 - - 2,804 61 - - Female 12-15 2,094 59 - - 2,567 77 - - 16-19 1,812 69 - - 2,336 80 - - 20-29 1,928 30 - - 2,404 58 2,593 50 30-39 1,822 30 - - 2,354 67 2,491 52 40-49 1,793 39 - - 2,327 83 2,486 77 50-59 1,713 64 - - 2,186 71 - - 60-69 1,548 25 - - 2,108 27 - - 70+ 1,473 27 - - 1,903 36 - - 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. 991-11 Table 11-117. Sodium/potassium ratios, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean ration Ratios at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 1.20 .01 .67 .76 .92 1.14 1.40 1.69 1.90 Age 20-29 years 661 1.26 .02 .73 .81 .99 1.22 1.49 1.79 1.95 30-39 years 812 1.15 .02 .67 .76 .90 1.09 1.34 1.62 1.81 40-49 years 583 1.14 .02 .60 .73 .90 1.09 1.32 1.67 1.90 Race 1/ White 1775 1.16 .01 .64 .75 .91 1.11 1.35 1.64 1.83 Black 167 1.44 .05 .84 .90 1.19 1.41 1.64 2.00 2.23 Other 76 1.33 .06 * * 1.00 1.26 1.68 ® * Poverty status 1/ < 100 315 1.35 .04 .70 .80 1.01 1.31 1.64 1.90 2.09 > 100 1575 1.17 .01 .66 .76 .91 1.10 1.36 1.63 1.82 < 131 414 1.33 .03 .70 .80 1.01 1.29 1.59 1.90 2.06 > 131 1476 1.16 .01 .65 .76 .91 1.09 1.34 1.62 1.81 Education 1/ < High school 305 1.28 .04 .56 .76 .94 1.24 1.57 1.88 2.04 High school 854 1.23 .02 .70 .79 .95 i.17 1.44 1.74 1.90 > High school 891 1.14 .01 .65 .75 .89 1.09 1.32 1.54 1.77 Region Northeast 448 1.17 .02 .67 .76 .93 1.13 1.36 1.62 1.77 Midwest 564 1.22 .03 .67 77 .93 1.18 1.41 1.74 1.91 South 660 1.24 .02 .70 .79 .95 1.17 1.45 1.80 1.94 West 384 1.12 .02 .60 .72 .89 1.05 1.29 1.57 1.74 Urbanization Central city 499 1.21 .03 .65 .79 .94 1.17 1.43 1.71 1.90 Suburban 1039 1.17 .02 .64 .75 .91 1.11 1.36 1.65 1.86 Nonmetropolitan 518 1.24 .03 «71 .78 .93 1.16 1.46 1.80 1.93 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. LI9T-11 Table II-118. Sodium/potassium ratios, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean ratio Ratios at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 1.10 .02 .66 .72 .86 1.07 1.28 1.54 1.69 Age 1-2 years 224 1.03 .03 .58 .66 .76 .98 1.24 1.54 1.61 3-5 years 423 1.14 .02 .70 .78 .91 1.11 1.29 1.54 1.70 Race 2/ White 559 1.07 .02 .65 .71 .84 1.05 1.25 1.46 1.61 Black 53 1.36 .06 x * 1.06 1.32 1.53 * * Other 26 1.15 .08 * * x 1.14 * * * Poverty status 2/ < 100 140 1.20 .04 .71 .82 1.02 1.16 1.36 1.69 1.80 > 100 471 1.06 .02 .65 .70 .82 1.03 1.25 1.52 1.63 < 131 192 1.18 .04 .68 .80 1.00 1.17 1.37 1.58 1.77 > 131 419 1.05 .02 .65 .70 .82 1.00 1.24 1.49 1.64 Education 2/ < High school 99 1.25 .05 * .82 .97 1.24 1.41 1.77 * High school 252 1.13 .02 .73 .76 .95 1.10 1.28 1.53 1.61 > High school 295 1.02 .03 .60 .68 .80 .98 1.20 1.45 1.59 Region Northeast 111 1.09 .05 * .71 .87 1.11 1.32 1.47 * Midwest 199 1.12 .03 .67 .76 .86 1.08 1.28 1.56 1.77 South 187 1.14 .04 .65 .72 .89 1.10 1.32 1.58 1.70 West 150 1.03 .03 .67 .70 .81 .98 1.22 1.46 1.61 Urbanization Central city 171 1.08 .02 .73 .75 .87 1.05 1.25 1.43 1.58 Suburban 310 1.08 .03 .60 .67 .82 1.07 1.28 1.55 1.70 Nonmetropolitan 166 1.17 .03 .76 .82 .92 1.15 1.33 1.55 1.77 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 891-11 Table 11-119. Hypertension among persons 20-74 years of age by sex, specified Hispanic origin, and age: Hispanic Health and Nutrition Examination Survey, 1982-84 T T Male | Female 1/ | T T | T T I | | | | | Number of | Percent Standard Number of Percent Standard | | | I | Sex, Hispanic origin, | examined with | error of | examined | with | error of and age | persons 2/ hypertension jhe percent) persons 2/ hypertension j the percent 1 1 1 x. 1 Mexican American 20-74 years... ...... i... 1,426 17.6 1.1 1,754 15.5 0.8 20-74 years, age adjusted.... C. 23.9 20.4 . 20-29 years... 431 6.9 1.5 502 2.0 0.6 30-39 years. ... ns ns iimmenmn 364 12.2 2.0 429 6.6 1.2 40-489 yOArsS. ... sins imninms 240 18.8 2.5 307 14.3 1.8 50-59 years. ................. 231 38.5 2.8 313 34.1 2.3 60-69 years. .. .:-:. sss unswns 121 53.8 4.3 141 55.8 3.8 70-74 years. ................. 39 *65.2 7.6 62 70.4 5.4 Cuban 20-74 YOArS. «:n:vinisinnisans 370 23.9 2.4 478 16.0 1.8 20-74 years, age adjusted.... ain 20.7 14.4 20-29 years. ..on: nm: innidmme 55 2.0 2.4 65 .0 0.0 30-39 years. ... qo, uns snnsmmas 56 7.5 4.4 94 2.2 1.8 40-49 years. ............i... 82 28.1 5.4 102 10.1 3.2 BOBS years... us: srs ssmssmmne 107 36.8 4.6 113 23.0 3.9 60-69 years... .....oc cons nmus 42 *43.0 7.6 72 45.7 5.7 70-74 years. ................. 28 *53.0 9.4 32 *55.9 8.6 Puerto Rican 20-74 years... .........ui... 436 17.1 2.0 749 12.5 1.3 20-74 years, age adjusted.... fa 21.4 os 19.2 20-29 years. ...........i... 112 3.7 2.2 191 5 0.6 30-89 years: . ::snsuaninmess 90 5.5 3.0 167 2.4 1.4 40-49 years. .......... i... 87 24.9 5.2 170 15.7 2.9 50-59 years.................. 99 38.9 4.4 131 37.1 3.7 BO-69 years... ...vvssvwssmmsinn 37 *853.0 7.9 76 48.2 4.9 70-74 years. ................. 11 * * 14 * * 1/ Excludes pregnant women. 2/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. 691-11 Table 11-120. Hypertension among Mexican-American persons 20-74 years of age by sex, poverty status, and age: Hispanic Health and Nutrition Examination Survey, 1982-84 I Male Female 1/ | I | 1 I | | | | I Number of Percent I Standard Number of Percent ; Standard examined | with error of examined I with error of Poverty status and age persons 2/| hypertension the percent persons 2/| hypertension j the percent 1 1 1 1 Below poverty 20-78 YOOrS. ....cowvimwoswn own 319 19.6 2.5 534 19.7 1.7 20-74 years, age adjusted.... is 24.5 22.3 20-29 years. ..........iiann 99 5.3 2.9 146 2.0 1.2 30-30 Years. nu: invincusmmrnn 69 13.4 4.8 120 6.6 2.3 40-49 years... ccrsn vine 47 12.5 4.7 89 18.2 3.7 50-59 years. ...........oiia... 50 46.9 6.2 96 41.1 4.4 60-69 years. ..... scavssnnven 40 *51.3 7.8 46 56.9 6.9 70-74 YOAPS cov vs nic nmuen men 14 * * 37 *70.7 7.2 Above poverty 20-74 years. ...........i..n 988 16.5 1.3 1,057 13.9 1.0 20-74 years, age adjusted.... #5 23.4 _ . % 20.7 20-29 years. .........iiiaannn 294 6.8 1.8 320 2.3 0.9 30-39 years... . viii vrsannen 275 11.4 2.2 277 6.9 1.5 40-40 yeBAIrS. .. «inv ivveinwsn 171 19.6 3.0 184 13.9 2.3 50-59 years. .........c.i.. 158 35.8 3.4 181 33.5 2.9 60-69 years. ............c.0... 69 56.3 5.8 77 53.9 5.1 * * 18 * * 70-74 years. viv ner nnsn answers 21 1/ Excludes pregnant women. 2/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. OLI-II Table 11-121. Hypertension among persons 20-74 years of age by sex, race, and age: Second National Health and Nutrition Examination Survey, 1976-80 T T Male ; Female 1/ | I T | T T Number of ! Percent Standard Number of J Percent ! Standard | examined | with | error of I examined | with I error of Race and age | persons 2/ hypertension j the percent persons 2/ hypertension jine percent 1 1 1 1 1 1 Non-Hispanic white 20-74 years.................. 4,575 34.2 0.9 4,973 27.4 0.8 20-74 years, age adjusted... .. " 33.8 25.2 20-29 years.................. 998 17.2 1.7 994 2.9 0.8 30-39 years. ................. 699 25.6 2.4 764 10.6 1.6 40-49 years... ............... 564 34.3 3.0 598 25.9 2.6 80-89 YRArsS..... -xcinsvcumnsse 563 49.3 3.1 635 44.6 2.9 60-69 years. ................. 1,331 54.6 1.2 1,463 57.6 1.2 70-78 yOBIS. .ov-xviiciomnmena 420 59.9 2.1 519 68.1 1.9 20-74 years. ................. 580 40.3 2.5 692 41.8 2.4 20-74 years, age adjusted.... “ wi 41.6 43.8 20-29 years. ................. 154 17.0 4.1 169 13.8 3.9 30-39 years. ................. 91 31.7 7-1 107 24.7 6.1 40-40 yOAPS. ..ovvincninnig ams 60 51.6 10.2 87 57.5 7.4 SO0-59 years. ................. 77 58.8 7:9 101 69.8 6.4 60-69 years. ................. 144 69.3 3.2 166 79.4 2.6 70-74 years. .....covsnveiisnn 54 66.7 5.3 62 78.3 4.4 1/ Excludes pregnant women. 2/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. TLI-1I Table 11-122. Hypertension among non-Hispanic persons 20-74 years of age by sex, poverty status, and age: Second National Health and Nutrition Examination Survey, 1976-80 T T } Male Female 1/ | T T | T T Number of Percent Standard } Number of } Percent Standard | examined I with | error of I examined | with I error of Poverty status and age | persons 2/ (hypertension | the percent; persons 2/ hypertension jthe percent 1 1 1 1 A 1 Below poverty 20-74 Y@ArS. ...: cnt svssamnsnse 522 36.3 2.6 834 32.5 2.1 20-74 years, age adjusted.... C 38.5 Ce 32.5 20-29 years. .............an- 142 12.7 4.1 204 6.3 2.5 30-39 years. x sxcicurennson 65 34.2 8.9 119 20.1 5.9 40-49 years. .......... «cc u.n 34 *40.0 13.4 68 32.5 B.S 50-59 years. ........... co... 49 66.1 9.8 79 54.7 8.3 60-69 years. ................- 148 57.8 3.6 245 67.5 2.7 70-74 years. ............cuunun 84 60.8 4.8 119 73.2 3.7 20-74 years. .........ceeenn.. 4,527 34.4 0.9 4,696 28.2 0.8 20-74 years, age adjusted.... IE 34.2 26.5 20-29 YOArS. sus inmermurnnns 996 17.5 1.7 959 4.5 0.9 30-39 years........c a nua nn 719 25.3 2.4 743 11.2 1.6 40-49 years. .........ceann- 572 36.0 3.0 606 29.0 2.7 BO-59 YyRArS. i rss ccnin van mn 565 48.8 3.1 632 45.2 2.9 60-69 years. ........... cue... 1,296 55.9 1.2 1,320 58.8 1.2 70-74 years. ..........ccuoeenn- 379 60.5 2.2 436 68.6 2.0 1/ Excludes pregnant women. 2/ Includes persons for whom usable measurements for the criteria variable were obtained. The criteria variable is discussed in the table notes. CLI-1I U.S. Food Supply Potassium Milligrams 5000 3000 2000 + 1000 + 1 1 1 1 1 1 0 i 1 3 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure 11-45. Potassium: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Potassium Potassium Dairy products 21.1% Vegetables 27.1% XZ H—~Legumes, nuts, and soy 6.3% VX... products 6.8% Other foods 8.1% Meat, poultry, fish 19.3% Fruits 11.3% Figure 11-46. Potassium: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include eggs; fats and oils; sugars and sweeteners; and miscellaneous foods) eLI-1I Table 11-123. Potassium: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 2073 25 996 1205 1542 1998 2515 3055 3457 Age 20-29 years 661 2045 38 1031 1154 1448 1955 2462 3094 3586 30-39 years 812 2086 36 971 1224 1579 2012 2553 2998 3306 40-49 years 583 2105 40 995 1259 1600 2056 2491 3073 3388 Race 1/ White 1775 2139 27 1070 1294 1623 2081 2553 3094 3477 Black 167 1589 69 678 807 1170 1523 1917 2553 2807 Other 76 1903 82 * * 1378 1931 2388 * * Poverty status 1/ < 100 315 1782 56 813 962 1234 1646 2234 2834 3082 > 100 1575 2138 25 1089 1294 1630 2082 2553 3094 3471 < 131 414 1805 57 759 965 1276 1676 2260 2792 3123 > 131 1476 2157 25 1103 1315 1643 2103 2569 3103 3477 Education 1/ < High school 305 1685 46 734 926 1223 1581 2058 2621 2988 High school 854 2019 35 970 1190 1518 1940 2443 2978 3281 > High school 891 2241 31 1219 1358 1728 2163 2654 3243 3620 Region Northeast 448 2008 47 1085 3235 1464 1943 2402 2975 3303 Midwest 564 2170 62 1007 1211 1614 2125 2663 3249 3606 South 660 1963 43 853 1097 1448 1872 2361 2912 3353 West 384 2196 43 1152 1358 1668 2207 2632 3082 3428 Urbanization Central city 499 2096 54 995 1171 1503 1982 2537 3094 3606 Suburban 1039 2078 37 1049 1267 1589 2033 2489 3030 3406 Nonmetropolitan 518 2015 37 956 1124 1472 1943 2505 2944 3366 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. bLI-II Table 11-124. Potassium: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 1936 32 1133 1265 1522 1863 2305 2728 2947 Age 1-2 years 224 1913 52 1074 1258 1522 1835 2344 2648 2899 3-5 years 423 1949 42 1141 1274 1526 1881 2295 2738 3059 Race 2/ White 559 1983 32 1176 1291 1537 1933 2364 27177 2992 Black 53 1642 85 * * 1295 1587 1917 * * Other 26 1735 120 * * * 1663 * * * Poverty status 2/ < 100 140 1800 87 1032 1229 1354 1713 2173 2638 2793 > 100 471 1989 35 1194 1314 1566 1933 2364 2777 2976 < 131 192 1798 70 1032 1244 1360 1716 2142 2631 2793 > 131 419 2013 36 1194 1352 1587 1973 2377 2834 3059 Education 2/ < High school 99 1754 62 * 1195 1426 1678 2061 2442 * High school 252 1882 48 1141 1248 1474 1720 2290 2721 2964 > High school 295 2033 40 1194 1308 1644 2031 2370 2834 2947 Region Northeast 111 1925 75 * 1229 1482 1859 2292 2738 * Midwest 199 1958 44 1256 1335 1605 1915 2256 2590 3010 South 187 1834 62 1032 1210 1362 1653 2268 2743 2932 West 150 2051 78 1223 1426 1617 2061 2436 2770 2854 Urbanization Central city an 2040 75 1223 1417 1602 1940 2436 2793 3135 Suburban 310 1898 37 1058 1275 1488 1845 2272 2668 2899 Nonmetropolitan 166 1895 72 1124 1246 1452 1863 2221 2728 3090 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. QLI-II Table II-125. Potassium: mean intake in milligrams, by sex and age, 1 day: National Health and Nutrition Examination Survey (NHANES I), 1971-74; Nationwide Food Consumption Survey (NFCS), 1977-78; Second National Health and Nutrition Examination Survey (NHANES II), 1976-80; and Continuing Survey of Food Intakes by Individuals (CSFII), 1985-86 NHANES 1 NFCS NHANES II CSFI1 1971-74 1977-78 1976-80 1985-86 Sex and age (years) Mean SEM Mean SEM Mean SEM Mean SEM Both sexes 1-2 1,720 21 - - 1,732 18 1.914 51 3-5 1,937 21 - - 1,912 14 1,991 41 6-11 2,334 32 - - 2,365 35 - - Male 12-15 2,915 72 - - 2,940 79 - - 16-19 3,287 108 - - 3,510 114 - - 20-29 3,131 83 - - 3,340 81 3,308 138 30-39 2,897 90 - - 3,026 74 3,209 113 40-49 2,778 80 - - 2,963 81 3,287 136 50-59 2,557 65 - - 2,765 74 - - 60-69 2,398 3 - - 2,560 27 - - 70+ 2,146 29 - - 2,291 43 - - Female 12-15 2,181 53 - - 2,121 55 - - 16-19 1,956 64 - - 1,952 64 - - 20-29 1,956 25 - - 2,055 44 2,143 43 30-39 1,929 26 - - 2,076 51 2,260 46 40-49 1,957 34 - - 2,096 57 2,210 47 50-59 1,980 50 - - 1,993 58 - - 60-69 1,870 25 - - 1,998 20 - 70+ 1,683 22 - - 1,973 34 1 CSFII data for 1985 only. 2 Ages 70-74 years only for NHANES I and NHANES II. 9LT-II U.S. Food Supply Copper Milligrams 5 2t ~~ 1 1 1 A 1 A A 1 1 XL 1 i 1 1925 1935 1945 1955 1965 1975 1985 Year le 0 1 1 1905 1915 Figure 11-47. Copper: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Copper Copper Vegetables 20.3% Meat, poultry, fish 20.6% Figure I-48. Copper: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include dairy products; eggs; fats and oils; sugars and sweeteners; and miscellaneous foods) LLT-1I Table 11-126. Copper: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 1.0 .01 «5 .6 7 .9 1.2 1.5 i.7 Age 20-29 years 661 1.0 .02 5 .6 7 .9 1.1 1.5 1.7 30-39 years 812 1.0 .02 .5 .6 7 .9 1.2 1.5 1.7 40-49 years 583 1.0 .03 .5 .6 .8 .9 1.2 1.4 1.6 Race 1/ White 1775 1.0 .02 .5 .6 .8 .9 1.2 1.5 1.6 Black 167 .9 .03 .3 .4 .6 .8 1.0 1.4 1.9 Other 76 1.0 .07 * * .7 .9 1.3 * * Poverty status 1/ < 100 315 .9 .03 .4 .4 .6 .8 1.0 1.4 1.7 > 100 1575 1.0 .01 «5 .6 .8 1.0 1.2 1.5 1.6 < 131 414 .9 .03 .4 .5 .6 .8 1.1 1.4 1.6 > 131 1476 1.0 .01 i: .6 .8 1.0 1.2 1.5 1.6 Education 1/ < High school 305 .8 .03 .4 .4 .6 .8 1.0 1.3 1.5 High school 854 1.0 .02 .5 .6 7 .9 1.1 1.4 1.6 > High school 891 1.2 .02 .6 I .8 1.0 1.3 1.6 1.8 Region Northeast 448 1.0 .03 .5 .6 7 .9 1.1 1.4 1.6 Midwest 564 1.0 .03 «5 .6 .8 1.0 1.2 1.5 1.8 South 660 1.0 .03 .4 5 .7 .9 1.1 1.5 1.6 West 384 1.0 .02 «5 .6 .8 1.0 1.2 1.5 1.7 Urbanization Central city 499 1.0 .03 .5 .6 7 .9 1.2 1.6 1.9 Suburban 1039 1.0 .02 .5 .6 .7 .9 1.2 1.4 1.6 Nonmetropolitan 518 1.0 .03 .5 .5 7 9 1.1 1.4 1.7 ee = = = ——— — ———————— — — ————— ~~ = = = = = = = 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 8LI-II Table 11-127. Copper: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 .8 .02 .4 «5 “8 .8 .9 1.1 1.2 Age 1-2 years 224 «7 .02 .4 .5 .6 «7 .8 1.0 1.1 3-5 years 423 .8 .02 .5 «5 .7 .5 .9 1.1 1.3 Race 2/ White 559 .8 .01 .4 .5 .6 .8 .9 1.1 1.2 Black 53 .9 .05 * * .6 .8 1.0 * * Other 26 .8 .05 * * * .7 * * * Poverty status 2/ < 100 140 .8 .04 .4 .5 .6 «7 1.0 1.2 1.3 > 100 471 .8 .02 .4 .5 “7 .8 .9 1.1 1.2 < 131 192 .8 .04 .4 .4 .6 «7 .9 1.2 1.3 > 131 419 .8 .02 .5 .5 «7 .8 .9 1.1 1.2 Education 2/ < High school 99 .8 .03 * .5 .6 .7 .9 1.1 * High school 252 .8 .02 .4 .5 .6 .8 .9 1.1 1.3 > High school 295 .8 .02 .5 .5 .7 .8 .9 1.1 1.2 Region Northeast 111 .8 .03 x .5 .6 .8 .9 1.1 x Midwest 199 .8 .02 .4 .5 .6 .7 .9 1.0 1.2 South 187 .8 .03 .4 .5 .6 .7 .9 1.1 1.3 West 150 .8 .03 «5 .6 .7 .8 .9 1.1 1.2 Urbanization Central city 171 .8 .03 .5 .5 .7 .8 1.0 1.2 1.3 Suburban 310 .8 .02 .4 .5 .6 .8 .9 1.0 1.1 Nonmetropolitan 166 .8 .03 .4 .5 .6 «7 .9 1.1 1.3 1/ Excludes two breastfed children. 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 6LT-1I U.S. Food Supply Zinc Milligrams 20 Sa eee a 1 1 1 1 1 1 1 1 0 wk 1905 1915 1925 1935 1945 1955 1965 1975 1985 Year Figure 11-49. Zinc: per capita amount per day in the U.S. food supply, 1909-85: U.S. Food Supply Series Zinc Dairy products 197% Zinc Nest poultry, fish 48.77% Grain products 12.67% Figure II-50. Zinc: food sources in the U.S. food supply, 1985: U.S. Food Supply Series (other foods include eggs; fats and oils; fruits; sugars and sweeteners; and miscellaneous foods) 08T-1I Table 11-128. Zinc: mean intake in milligrams, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 8.7 .10 4.0 4.8 6.4 8.2 10.4 13.0 14.5 Age 20-29 years 661 8.8 .14 4.0 4.7 6.4 8.4 10.4 13.2 15.1 30-39 years 812 8.8 .18 3.8 4.7 6.3 8.2 10.5 12.8 14.5 40-49 years 583 8.4 .17 4.4 5.0 6.3 7.9 10.1 12.2 13.9 Race 1/ White 1775 8.8 .11 4.2 5.0 6.5 8.3 10.4 13.1 14.5 Black 167 7.6 .38 2.9 3.9 5.0 6.8 9.6 12.6 14.6 Other 76 8.8 .43 * % 5.9 8.3 112.2 * * Poverty status 1/ < 100 315 8.1 .22 3.2 4.2 5.5 7.4 10.3 12.2 13.6 > 100 1575 8.8 .11 4.2 5.0 6.6 8.3 10.4 13.1 14.5 < 131 414 8.2 .22 3.2 4.2 5.8 7.7 10.3 12.3 13.6 > 131 1476 8.8 .11 4.2 5.1 6.6 8.3 10.5 13.1 14.6 Education 1/ < High school 305 7.7 .23 3.4 4.1 5.3 6.9 9.4 11.9 13.8 High school 854 8.5 .14 3.9 5.0 6.4 8.0 10.2 12.7 14.5 > High school 891 9.1 .14 4.4 5.1 6.8 8.6 10.8 13.4 15.3 Region Northeast 448 8.4 .21 4.0 4.9 6.2 7.8 10.0 12.5 14.0 Midwest 564 8.9 .22 4.3 5.3 6.7 8.3 10.8 13.4 14.5 South 660 8.4 .17 3.6 4.4 5.8 8.0 10.1 12.8 14.5 West 384 9.0 .21 4.3 5.1 6.7 8.5 10.6 13.2 15.3 Urbanization Central city 499 8.8 .21 4.0 4.7 6.1 8.2 10.6 13.8 15.3 Suburban 1039 8.6 .15 4.0 4.9 6.5 8.2 10.2 12.5 14.4 Nonmetropolitan 518 8.8 19 4.0 4.8 6.2 8.2 10.5 12.6 13.9 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 181-11 Table 11-129. Zinc: mean intake in milligrams per 1000 kilocalories, women aged 20-49 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All women 2056 5.8 .04 3.6 4.0 4.7 5.6 6.7 7.8 8. Age 20-29 years 661 5.6 .06 3.6 4.0 4.6 5.4 6.4 7.5 8. 30-39 years 812 5.9 .08 3.5 3.9 4.6 5.5 6.8 8.0 9. 40-49 years 583 6.0 .09 3.8 4.2 4.9 S+7 6.9 8.0 8. Race 1/ White 1775 5.8 .05 3.6 4.0 4.7 5.6 6.6 7.8 8. Black 167 5.6 .15 3.4 3.7 4.6 5.4 6.5 7.6 8. Other 76 6.3 +23 * * 5.0 5.8 7.1 * Poverty status 1/ < 100 315 5.9 .10 3.6 4.0 4.8 5.5 6.7 7.9 8. > 100 1575 5.8 .05 3.6 4.0 4.7 5.6 6.7 7.8 8. 131 414 5.8 .09 3.7 4.1 4.8 5.5 6.6 7.8 8. > 131 1476 5.8 .05 3.6 4.0 4.7 5.6 6.7 7.8 8. Education 1/ < High school 305 6.0 «33 3.6 4.2 4.7 5.8 6.9 8.0 9. High school 854 5.7 .06 3.6 4.0 4.7 5.6 6.7 7.8 8. > High school 891 5.9 .07 3.6 4.0 4.7 5+5 6.6 7.9 9. Region Northeast 448 5.9 .08 3.7 4.1 4.9 5.7 6.7 7.8 8. Midwest 564 5.8 .09 3.6 4.0 4.7 5.6 6.6 7.7 8. South 660 5.7 .08 3.4 3.8 4.4 5.4 6.7 8.0 9. West 384 5.9 .08 3.8 4.1 4.9 5.7 6.8 7.8 8. Urbanization Central city 499 5.8 .06 3.6 4.0 4.8 5.5 6.6 7.8 8. Suburban 1039 5.8 .06 3.6 4.0 4.6 5.6 6.7 7.8 8. Nonmetropolitan 518 6.0 12 3.7 4.1 4.8 5.6 6.8 7.9 8. 1/ Some women did not report race, poverty status, or education. Therefore, the numbers of women in each category do not add to the number of all women. 81-11 Table 11-130. Zinc: mean intake in milligrams, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 7.6 .14 4.5 5.1 6.1 7.3 8.6 10.6 12.0 Age 1-2 years 224 6.9 .16 4.2 4.8 5.9 6.6 7.7 8.9 10.2 3-5 years 423 8.0 «17 4.6 5.3 6.3 7.6 9.1 11.2 12.5 Race 2/ White 559 7.6 .13 4.4 5.1 6.0 7.3 8.6 10.5 12.0 Black 53 7.9 .54 * * 6.3 7.8 9.5 * * Other 26 7.2 .36 * * * 6.8 * * Poverty status 2/ 100 140 7.9 .34 4.6 5.1 6.0 7.1 8.9 12.5 13.8 > 100 471 7.5 .14 4.6 5.2 6.0 7.3 8.4 10.2 11.4 < 131 192 7.7 .28 4.6 5.1 5.9 6.9 8.9 11.3 13.0 > 131 419 7.6 .14 4.5 5.2 6.1 7.4 8.5 10.2 11.4 Education 2/ < High school 99 8.1 «35 * 5.2 6.1 7.6 9.7 12.5 * High school 252 7.6 .19 4.2 5.0 6.1 7.3 8.9 10.6 13.0 > High school 295 7.4 17 4.5 5.2 6.0 71.2 8.3 9.7 11.1 Region Northeast 111 7.6 .42 * 4.6 5.8 7.2 8.7 11.2 * Midwest 199 7.8 .19 5.0 5.3 6.3 7.6 8.7 11.3 12.0 South 187 7.3 .22 4.3 4.8 5.7 6.8 8.3 9.8 11.2 West 150 7.7 .29 4.8 5.9 6.3 7.2 8.8 10.2 13.0 Urbanization Central city 171 7.7 .30 4.9 5.4 6.2 7.2 8.8 11.0 11.7 Suburban 310 7.4 .19 4.3 4.8 6.0 7.1 8.4 9.7 12.0 Nonmetropolitan 166 7.9 .19 4.6 5.1 6.1 7.6 9.5 10.8 12.3 2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. 81-11 Table 11-131. Zinc: mean intake in milligrams per 1000 kilocalories, children aged 1-5 years, 4 nonconsecutive days: Continuing Survey of Food Intakes by Individuals, 1985-86 Mean intake Intakes at selected percentiles Characteristics n Mean SEM 5 10 25 50 75 90 95 All children 1/ 647 5.4 .07 3.7 3.9 4.6 5.2 6.0 6.8 7.4 Age 1-2 years 224 5.4 .12 3.8 3.9 4.6 5.3 5.9 6.7 7.3 3-5 years 423 5.4 .08 3.7 4.0 4.5 5.2 6.0 6.9 7.4 Race 2/ White 559 5.3 .08 3.7 3.9 4.5 5.1 5.9 6.7 7.3 Black 53 5.7 .22 * * 5.0 5.6 6.2 * * Other 26 6.2 .47 * * * 5:5 * * * Poverty status 2/ 100 140 5.6 .14 3.8 4.2 4.8 5.5 6.5 7.4 7.6 > 100 471 5.3 .08 3.7 3.9 4.5 5.1 5.9 6.7 7.3 < 131 192 5.5 .12 3.8 4.2 4.7 5.4 6.1 7.2 7.6 > 131 419 5.3 .09 3.7 3.9 4.5 5.1 5.9 6.7 7.3 Education 2/ < High school 99 5.7 .18 * 4.0 4.7 5.7 6.4 7.5 * High school 252 5.4 .11 3.5 3.9 4.6 5.4 6.3 6.9 7.3 > High school 295 5.2 .10 3.8 4.0 4.5 5.0 5.7 6.4 6.8 Region Northeast 111 5.3 .21 * 3.7 4.5 5.2 6.0 6.6 * Midwest 199 5.5 .14 3.8 4.1 4.7 5.4 6.2 6.9 7.4 South 187 5.3 .11 3.6 4.0 4.4 5.0 5.8 6.7 7.2 West 150 5.5 .13 3.9 4.0 4.6 5.2 6.1 7.3 7.4 Urbanization Central city 171 5.3 .14 3.7 3.9 4.4 5.1 5.9 6.7 7.4 Suburban 310 5.4 .11 3.7 3.9 4.6 5.1 6.0 6.8 7.3 Nonmetropolitan 166 5.6 .13 4.0 4.3 4.8 5.3 6.3 7.1 7.5 1/ Excludes two breastfed children. ~2/ Race, poverty status, and education were not reported for all children. Therefore, the numbers of children in each category do not add to the number of all children. ¥81-11 Table II-132. Percent® of supplement users consuming specific nutrients in 1980 (Stewart et al., 1985) Men Women Total Supplement Rank population? users’ 16-24 yr 25-64 yr 65 yr & over 16-24 yr 25-64yr 65 yr & over order Nutrient (n=2,751) (n = 1,090) (n = 150) (n=101) (n =103) (n=192) (n= 274) (n = 270) - percent 1 Vitamin C 35.1 90.6 92.0 89.1 94.24 91.2 91.2 88.2° 2 Thiaminf 30.5 78.3 76.7 75.2 73.8° 78.1 83.94 74.1 3 Riboflavin 30.1 77.4 74.7 73.3 71.8° 78.1 83.94 73.7 4 Vitamin B12 30.1 71.3 76.7 73.3 70.9 80.2 82.84 72.2 5 Vitamin B6f 29.9 76.6 76.7 71.3 70.9° 71.6 83.24 73.7 6 Niacin 29.8 76.7 76.7 73.3 71.8° 76.6 82.14 72.2 7 Vitamin E 28.3 72.6 71.3 68.3 79.64 70.8 76.6 71.5 8 Vitamin A 25.1 64.5 66.0 60.4° 63.1 64.6 68.64 63.0 9 Vitamin D 24.4 62.4 62.0 57.4° 58.2 63.5 67.94 61.5 10 Iron’ 21.9 56.1 54.0 51.5 38.8° 65.14 63.9 47.0 11 Pantothenic acid] 21.3 55.1 53.3 57.4 49.5 45.3° 59.54 49.3 12 Folic acid 20.3 52.3 54.0 49.5 37.9 63.04 55.5 44.4 13 Magnesium’ 13.7 35.5 36.7 37.6 35.9 22.9 38.74 30.4 14 Zinc! 13.5 35.1 42.04 35.6 35.9 27.1 36.1 30.0 15 Calcium 135 34.9 30.0 37.6 32.0 27.1 38.34 31.1 16 Iodine 12.9 33.8 40.74 37.6 29.1 24.0° 33.6 28.1 17 Copper’ 12.2 31.8 36.74 35.6 30.1 22.4° 31.0 28.1 18 Manganese! 9.3 24.5 28.74 28.74 22.3 14.6° 23.0 23.7 19 Biotin! 8.5 22.4 24.0 26.74 11.7 18.8 23.0 14.8 20 Phosphorus 8.4 21.8 21.3 23.84 19.4 1.7° 23.0 18.2 21 Potassium 45 11.8 12.0 12.9 11.6 8.8 13.24 6.3 2 The percent is based upon total number of supplement users within each sex—age group. Sex-age groups are weighted proportional to census data and projected to the total U.S. population. Sex-age groups are weighted proportional to census data. Highest group percentage for a nutrient. Lowest group percentage for a nutrient. Statistically significant differences among groups. 0 Aa 0 oo a81-11 Table 11-133. Median and 95th percentile levels? of intake from supplements (measured in percent of RDAP) for individual nutrients in 1980 (Stewart et al., 1985) Men Women Rank All specific order Nutrient nutrient users’ 16-24 yr 25-64 yr 65 yr & over 16-24 yr 25-64 yr 65 yr & over 95th 95th 95th 95th 95th 95th 95th Median Percentile Median Percentile Median Percentile Median Percentile Median Percentile Median Percentile Median Percentile 1 Thiamin 550 6,000 629 6,786 714 7,983 833f 6,250 1508 4,614 700 5,857 636 2,500 2 Riboflavin 420 5,000 345 5,882 475 5,572 714f 5,357 1428 2,876 548 4,654 476 1,688 3 Vitamin C 333 2,810 333 3,445 af 3,542 arf 2,704 1678 1,798 333 2,500 333 1,837 4 Pantothenic acidd 257 1,929 257 2,143 257 3,071 257 1,429 1438 747 257 1,907 257 1,038 5 Vitamin E 200 6,000 1008 3,668 200 6,367 200 5,333 125 3,358 250f 5,094 250f 7,250 6 Vitamin B12 200 3,338 200 6,667 200 3,333 200 2,500 200 1,905 200 3,875 200 1,937 7 Vitamin D 200 375 1148 256 200 400 200 500 1148 200 200 275 200 400 8 Niacin 190 1,200 167 1,067 159 917 316f 1,250 1438 714 231 1,518 231 837 9 Vitamin B§ 143 5,000 200 4,659 136 5,482 221 2,789 1008 2,628 200 5,168 200 2,500 10 Vitamin A 125 430 1008 500 114 395 160 500 125 367 125 365 125 313 11 Iron 119 500 120 539 180f 500 154 500 1008 592 1008 514 180f 578 12 Folic agid 100 200 100 200 100 200 100 196 100 184 100 300 100 150 13 Iodine 100 150 100 149 100 200 100 200 718 100 100 150 100 100 14 Copper® 67 100 67 100 57 120 67 335 488 100 67 100 67 100 15 Zinc 50 286 43 178 50 288 67 313 71 206 64 296 338 265 16 Biotin® 38 300 sof 163 43h 183 27h 5180 25 154 38 450 208 150 17 Magnesium® 22 97 168 7 19 79 19 96 17 67 20f 97 2of 83 18 Manganese® 20 150 20 343 2ofh 1370 200 ol 200 1470 20 172 20 150 19 Calcium® . 16 113 Bn 103 98 n 10 20 13 11 75 24] 111 16 94 20 Phosphorus 14 73 1 67 108" 1 7 1 4 16 71 1 56, 21 Potassium® 1 3 1h 1b & *h 1 2h hy th 1 15 ih 21h 8 The median and 95th percentile are calculated upon the basis of users of specific nutrients only. b Pantothenic acid, copper, manganese, biotin, and potassium are reported as percent of the upper ESADDI level. © Sex-age groups are weighted proportional to census data. d All group medians 2100 percent and <200 percent RDA. € All group medians <100 percent RDA. f Highest percent RDA for a particular nutrient. 2 Lowest percent RDA for a particular nutrient. Cell n <30. Appendix III Glossary of Terms and Acronyms AEDS: Alcohol Epidemiologic Data System Anemia: refers a hemoglobin level below the normal reference range for individuals of the same sex and age. This assessment corresponds roughly, but not exactly, to the third stage of iron deficiency because anemia may result from causes other than iron deficiency, including infection, chronic disease, and deficiencies of folacin or vitamin B12. ARS: Agricultural Research Service, USDA Atherosclerosis: a progressive process that begins in childhood with the appearance of lesions in the form of fatty streaks in the lining of the coronary arteries or aorta. The fatty streaks may eventually progress to fatty and fibrous plaques or even larger, more complicated lesions. As the lesions develop, the progressive narrowing of the vessels reduces blood flow to the tissues supplied by the affected vessels, resulting in angina pectoris (chest pain), myocardial infarction (heart attack), or sudden death. These are the most common manifestations of coronary heart disease. BMI: body mass index BRFSS: Behavioral Risk Factors Surveillance System Calcium: an essential mineral. Carbohydrate: source of energy in the diet; contributes approximately 4 kilocalories/gram. Cardiovascular disease: includes a variety of pathological processes pertaining to the heart and blood vessels. Carotenes: beta-carotene and other provitamin A carotenoids. CDC: Centers for Disease Control Cerebrovascular disease (ICD-9 430-439): includes a group of disorders characterized by ischemic stroke, a serious and sudden decrease of blood supply to the brain, resulting from atherosclerosis. Stroke may also be caused by cerebral hemorrhage. Cholesterol: fatty substance required for the synthesis of sex hormones, bile acids and vitamin D; found in animal foods and synthesized in the body; carried by lipoproteins in the blood. Coefficient of variation: (standard deviation divided by the mean) x 100. Copper: essential mineral. Coronary (or ischemic) heart disease (ICD-9 410-414): several cardiovascular disorders resulting from inade- quate circulation of blood to local areas of the heart muscle, almost always as a consequence of focal narrowing of the coronary arteries by atherosclerosis. CSFII: Continuing Survey of Food Intakes by Individuals Depleted iron stores: the condition in which the iron reserves of the body are decreased to a critically low level and are a marker for the first stage in the development of iron deficiency. DHHS: U.S. Department of Health and Human Services Dietary fiber: heterogeneous group of plant components that are resistant to digestion by the enzymes of the human gastrointestinal tract; soluble fiber include gums, mucilages, some pectins and hemicelluloses; insoluble fiber include cellulose, lignin, and other pectins and hemicelluloses. nt-1 Dietary status: the condition of a population's or an individual's intake of foods and food components, especially nutrients. dl: deciliter EPONM: Expert Panel on Nutrition Monitoring ESADDI: Estimated Safe and Adequate Daily Dietary Intake ERS: Economics Research Service, USDA FASEB: Federation of American Societies for Experimental Biology Fat: energy source in the diet; provides approximately 9 kilocalories per gram. FDA: Food and Drug Administration Ferritin model: a model for assessing the prevalence of iron deficiency that requires abnormal values for at least two of the following measurements—-serum ferritin level, transferrin saturation, or erythrocyte protoporphyrin level. fl: femtoliter FLAPS: Food Label and Package Survey Fluoride: essential mineral. Food components: as discussed in this report, they include nutrients (macronutrients, vitamins, and minerals) and non-nutrients that may affect health (such as dietary fiber). Food energy: chemical energy obtained from foods that supports body functions. Folacin: water-soluble vitamin. Four-variable model: a model for assessing the prevalence of iron deficiency that requires abnormal values for either serum ferritin level and/or mean corpuscular volume together with abnormal values for transferrin saturation and/or erythrocyte protoporphyrin level. g gram GRAS: Generally Recognized as Safe Growth retardation: impairment in linear growth that may result from undernutrition. HANES: Health and Nutrition Examination Surveys HHANES: Hispanic Health and Nutrition Examination Survey HDL: high density lipoprotein Health status: as used in this report, refers to a population's or an individual's status with respect to physical state or disease condition. Hg: mercury bz LA: histocompatibility locus jas! NIS: Human Nutrition Information Service, USDA Hypertension (ICD-9 401): persistently elevated arterial blood pressure. Hypertensive heart disease (ICD-9 402): includes hypertensive cardiomegaly, cardiopathy, cardiovascular disease, and heart failure. 1-2 ICD: International Classification of Diseases Incidence: refers generally to the number of new events (for example, new cases of disease) in a defined population, within a specified time period. Iron: essential mineral. Iron deficiency: refers to the presence of two or more abnormal values for iron metabolism indicators and corresponds to the second and third stages in the development of iron deficiency. Iron deficiency anemia: denotes a low hemoglobin value found in association with evidence of iron deficiency. Theoretically, it corresponds to the third stage of iron deficiency. Iron overload: occurs in the case of excessive accumulation of storage iron in tissues. JNMEC: Joint Nutrition Monitoring Evaluation Committee kg: kilogram L: liter LDL: low density lipoprotein LSRO: Life Sciences Research Office m: meter Magnesium: essential mineral. Marginal nutritional status: a condition in which nutrient stores may be low, the activity of some enzymes may be lower or higher than normal, or growth may be slightly impaired, but impairment of performance, health, or survival may not be evident. Persons with marginal nutritional status are considered at risk of nutritional deficiency, especially when subjected to stress. MCV: mean corpuscular (red blood cell) volume MCV model: a model for assessing the prevalence of iron deficiency that requires abnormal values for at least two of the following measurements——mean corpuscular volume, transferrin saturation, or erythrocyte protoporphyrin level. Mean: measure of central tendency calculated by adding all individual values and dividing by the number of values. Median (or 50th percentile): value that divides a distribution of values into two equal parts, with 50 percent of the values above and 50 percent of the values below this point. mg: milligram mm: millimeter mmol: millimole Monounsaturated fat: fatty acids with one double bond. J: micro (for example, pg = microgram, pmol = micromole) NCEP: National Cholesterol Education Program NCHS: National Center for Health Statistics NCI: National Cancer Institute NFCS: Nationwide Food Consumption Survey 111-3 NHANES I: National Health and Nutrition Examination Survey NHANES II: Second National Health and Nutrition Examination Survey NHEFS: NHANES I Epidemiologic Followup Study NHES: National Health Examination Survey NHIS: National Health Interview Survey NHLBI: National Heart, Lung, and Blood Institute Niacin: water-soluble vitamin; obtained from food preformed or may be synthesized from tryptophan. NIH: National Institutes of Health NNMS: National Nutrition Monitoring System 1 Nutrient deficiency: a condition associated with adverse health consequences arising from inadequate intake or utilization of a nutrient. Nutrient excess and/or toxicity: a condition associated with adverse health consequences arising from excessive intake or utilization of a nutrient. Nutrition assessment: the measurement of indicators of dietary status and nutrition-related health status to identify the possible occurrence, nature, and extent of impaired nutritional status (ranging from deficiency to toxicity). Nutrition monitoring: the assessment of dietary or nutritional status at intermittent times with the aim of detecting changes in the dietary or nutritional status of a population. Nutritional status: the condition of a population's or an individual's health as influenced by the intake and utilization of nutrients and non—nutrients. It reflects, directly or inferentially, the processes of food ingestion and digestion; absorption, transport, and metabolism of food components; and excretion of food components and their metabolic products. As noted in the JNMEC report (1986), indicators of nutritional status include: (1) levels of specific food components in diets; (2) clinical, anthropometric, hematological, and biochemical measurements related to specific food components; and (3) health conditions or diseases that may be associated Nutrition surveillance: continuous assessment of nutritional status for the purpose of detecting changes in trend or distribution in order to initiate corrective measures. Nutritional imbalance: a condition associated with adverse health consequences arising from insufficient or excessive intake of one nutrient or food component relative to another. Obesity: excessive accumulation of body fat. OCA: oral contraceptive agent Overnutrition: the condition resulting from the excessive intake of foods in general or particular food components. Overweight: body weight in excess of standards; taken as an estimate of obesity (see chapters 3 and 4 for discussion of standards for overweight used in this report). PedNSS: Pediatric Nutrition Surveillance System Percentiles: divisions of a distribution into equal, ordered subgroups of hundredths. Phosphorus: essential mineral. PIR: poverty income ratio 111-4 PNSS: Pregnancy Nutrition Surveillance System Potassium: essential mineral. Polyunsaturated fat: fatty acids with two or more double bonds. Poverty index: consists of a set of cash income cutoffs that vary by the size and number of children in a family. Since 1964, various cutoffs for the definition of poverty index have been used for a variety of analytical and policy applications. (See appendices I and II for details on the poverty indices in specific NNMS surveys.) Prevalence: the number of instances of a given disease or other condition in a given population at a designated time. Protein: formed from various combinations of amino acids; energy source in the diet; provides approximately 4 kilocalories per gram. RBC: red blood cell RDA: Recommended Dietary Allowance(s) Relative standard error: see coefficient of variation. Riboflavin: water-soluble vitamin. Saturated fat: fatty acids with no double bonds. SE: standard error SI: Systéme International Sodium: essential mineral Standard deviation: the square root of the sum of the squares of the deviations from the mean divided by n—1. Standard error: the standard deviation (measure of dispersion or variation) of a statistic (mean or percent). Thiamin: water-soluble vitamin. Undernutrition: the condition resulting from the inadequate intake of foods in general or particular food components. USDA: United States Department of Agriculture Vitamin A: fat-soluble vitamin; activity is derived from both preformed vitamin A (retinol) and provitamin A carotenoids. Vitamin B6: water-soluble vitamin. Vitamin B12: water-soluble vitamin. Vitamin C: water-soluble vitamin. Vitamin E: fat-soluble vitamin. Vit/Min: Vitamin Mineral Supplement Intake Survey VLDL: very low density lipoprotein WIC: Special Supplemental Food Program for Women, Infants, and Children Zinc: essential mineral ITI-5 Appendix IV Recommended Dietary Allowances (RDA) The RDA and Estimated Safe and Adequate Daily Dietary Intakes are recorded here from the National Research Council. 1980. Recommended Dietary Allowances. 9th revised edition. Washington: National Academy Press. Table IV-1. Estimated safe and adequate daily dietary intakes of selected vitamins and minerals® Vitamins Age Vitamin K Biotin Pantothenic Acid (years) (pg) (pg) (mg) Infants 0-0.5 12 35 2 0.5-1 10-20 50 3 Children and 1-3 15-30 65 3 adolescents 4-6 20-40 85 3-4 7-10 30-60 120 4-5 11+ 50-100 100-200 4-7 Adults 70-140 100-200 4-7 b Trace Elements Age Copper Manganese Fluoride Chromium Selenium Molybdenum (years) (mg) (mg) (mg) (mg) (mg) (mg) Infants 0-0.5 0.5=0.7 0.5«0.7 0.1-0.5 0.01-0.04 0.01-0.04 0.03-0.06 0.5-1 0.7-1.0 0.7-1.0 0.2-1.0 0.02-0.06 0.02-0.06 0.04-0.08 Children and 1-3 1.0-1.5 1.0-1.5 0.5~1.5 0.02-0.08 0.02-0.08 0.05-0.1 adolescents 4-6 1.5-2.0 1.5-2.0 31.0-2.5 0.03-0.12 0.03-0.12 0.06-0.15 7-10 2.0-2.5 2.0-3.0 1.5-2.5 0.05-0.2 0.05-0.2 0.10-0.3 11+ 2.0-3,0 2.5-5.0 1.5-2.5 0.05-0.2 0.05-0.2 0.15-0.5 Adults 2.0-3.0 2.5-5.0 1.5-4.0 0.05-0.2 0.05-0.2 0.15-0.5 Electrolytes Age Sodium Potassium Chloride (years) (mg) (mg) (mg) Infants 0-0.5 115-350 350-925 275-700 0.5-1 250-750 425-1275 400-1200 Children and 1-3 325-975 550-1650 500-1500 adolescents 4-6 450-1350 775-2325 700-2100 7-10 600-1800 1000-3000 925-2775 11+ 900-2700 1525-1575 1400-4200 Adults 1100-3300 1875-5625 1700-5100 Because there is less information on which to base allowances, these figures are not given in the main table of RDA and are provided here in the form of ranges of recommended intakes. Since the toxic levels for many trace elements may be only several times usual intakes, the upper levels for the trace elements given in this table should not be habitually exceeded. G-Al Table IV-2. Food and Nutrition Board, National Academy of Sciences--National Research Council, Recommended Daily Dietary Allowances®, Revised 1980. Designed for the maintenance of good nutrition of practically all healthy people in the United States Fat-soluble vitamins Age Protein Vitamin 2 Vitamin D Vitamin E (years) (g) (4g RE) (ug) (mg a-TE) Infants } 0.0 - 0.5 kg X 2.2 420 10 3 0.5 - 1.0 kg X 2.0 400 10 4 Children 1-3 23 400 10 5 4-6 30 500 10 6 7-10 34 700 10 7 Males 11-14 45 1000 10 8 15-18 56 1000 10 10 19-22 56 1000 7.5 10 23-50 56 1000 5 10 51+ 56 1000 5 10 Females 11-14 46 800 10 8 15-18 46 800 10 8 19-22 44 800 7.5 8 23-50 44 800 5 8 51+ 44 800 5 8 Pregnant +30 +200 +5 +2 Lactating +20 +400 +5 +3 See footnotes at end of table. €-Al Table IV-2. Food and Nutrition Board, National Academy of Sciences--National Research Council, Recommended Daily Dietary Allowances®, Revised 1980. Designed for the maintenance of good nutrition of practically all healthy people in the United States—— = continued Water-soluble vitamins Age Vitamin C Thiamin Riboflavin Niacin Vitamin B6 Folacin® Vitamin B12 (years) (mg) (mg) (mg) (mg NE) (mg) (ug) (ug) Infants 0.0 - 0 35 0.3 0.4 6 0.3 30 0.5% 0.5 - 1 35 0.5 0.6 8 0.6 45 1.8 Children 1-3 45 0.7 0.8 9 0.9 100 2.0 4-6 45 0.9 1.0 11 1.3 200 2.5 7-10 45 3.2 1.4 16 1.6 300 3.0 Males 11-14 50 1.4 1.6 18 1.8 400 3.0 15-18 60 1.4 1.7 18 2.0 400 3.0 19-22 60 1.5 1.7 19 2.2 400 3.0 23-50 60 1.4 1.6 18 2.2 400 3.0 51+ 60 1.2 1.4 16 2.2 400 3.0 Females 11-14 50 1.1 1.3 15 1.8 400 3.0 15-18 60 1.1 1.3 14 2.0 400 3.0 19-22 60 1.1 1.3 14 2.0 400 3.0 23-50 60 1.0 1.2 13 2.0 400 3.0 51+ 60 1.0 1.2 i3 2.0 400 3.0 Pregnant +20 +0.4 +0.3 +2 +0.6 +400 +1.0 ’ Lactating +40 +0.5 +0.5 +5 +0.5 +100 +1.0 See footnotes at end of table. ¥-Al 8126, Table IV-2. Food and Nutrition Board, National Academy of Sciences--National Research Council, Recommended Daily Dietary Allowances®, Revised 1980. Designed for the maintenance of good nutrition of practically all healthy people in the United States—- continued Minerals Age Calcium Phosphorus Magnesium Iron Zinc Iodine (years) (mg) (mg) (mg) (mg) (mg) (pg) Infants 0.0 - 0.5 360 240 50 10 3 40 0.5 - 1.0 540 360 70 15 5 50 Children 1-3 800 800 150 15 10 70 4-6 800 800 200 10 10 90 7-10 800 800 250 10 10 120 Males 11-14 1200 1200 350 18 - 18 150 15-18 1200 1200 400 18 15 150 19-22 800 800 350 10 15 150 23-50 800 800 350 10 15 150 51+ 800 800 350 10 15 150 Females 11-14 1200 1200 300 18 15 150 15-18 1200 1200 300 18 15 150 19-22 800 800 300 18 15 150 23-50 800 800 300 18 15 150 51+ 800 800 300 10 15 150 Pregnant +400 +400 +150 hE +5 +25 Lactating +400 +400 +150 LL +10 +50 The allowances are intended to provide for individual variations among most normal persons as they live in the United States under usual environmental stresses. Diets should be based on a variety of common foods in order to provide other nutrients for which human requirements have been less well defined. Retinol equivalents. 1 retinol equivalent (RE) = 1 pg retinol or 6 ug g-carotene. As cholecalciferol. 10 pg cholecalciferol = 400 IU of vitamin D. a-tocopherol equivalents. 1 mg d- a-tocopherol =1 a-TE. 1 NE (niacin equivalent) is equal to 1 mg of niacin or 60 mg of dietary tryptophan. The folacin allowances refer to dietary sources as determined by Lactobacillus casei assay after treatment with enzymes (conjugases) to make polyglutamyl forms of the vitamin available to the test organism. The recommended dietary allowance for vitamin B12 in infants is based on average concentration of the vitamin in human milk. The allowances after weaning are based on energy intake (as recommended by the American Academy of Pediatrics) and consideration of other factors, such as intestinal absorption. The increased requirement during pregnancy cannot be met by the iron content of habitual American diets nor by the existing iron stores of many women; therefore the use of 30-60 mg of supplemental iron is recommended. Iron needs during lactation are not substantially different from those of nonpregnant women, but continued supplementation of the mother for 2-3 months after parturition is advisable in order to replenish stores depleted by pregnancy. ¥¥ U.S. GOVERNMENT PRINTING OFFICE: 1989— 2 4 7 - 662 INNEE 3 U. C. BERKELEY LIBRARIES AERA C049979200 RETURN PUBLIC HEALTH LIBRARY TO =p 42 Warren Hall 642-2511 LOAN PERIOD 1 SEMESTER 2 3 4 5 6 ALL BOOKS MAY BE RECALLED AFTER 7 DAY ALL JOURNALS ARE NON-RENEWABLE Return to desk from which borrowed DUE AS STAMPED BELOW SEMEST™R LOAN MAY 2 4 1997 SUBJEL + RECAL| CDPUBI aN 21797 SEMESTER LOAN MAY 7 31998 SUBJECT TO RECALL L CDPUBL HAY 2 2 93 ONE MONTH IO NET TOT TTT Al 967199 SUBJECT TO REC] LL FORM NO. DD26-7, Om, 11/94 UNIVERSITY OF CALFORNIA, BERKELEY BERKELEY, CA 94720