w TERNATIONAL HEALTH COSTS r‘ AND EXPENDITURES AND WELFARE ic Health Serv EDUCATION, 1 DEPARTMENT OF HEALTH U.S Ice Pu bl tutes of Health ional lnsti Nat [INTERNATIONAL HEALTH COSTS AND EXPEN DITURES__1 Proceedings of an International Conference on Health Costs and Expenditures sponsored by The John E. Fogarty International Center for Advanced Study in the Health Sciences—National Institutes of Health—and held at the National Institutes of Health, Bethesda, Maryland, USA, June 2-4, 1975 Editor: Teh-wei Hu Professor of Economics The Pennsylvania State University A publication of the GEOGRAPHIC HEALTH STUDIES JOHN E. F OGARTY INTERNATIONAL CENTER FOR ADVANCED STUDY IN THE HEALTH SCIENCES 1976 US. Department of Health, Education, and Welfare Public Health Service National Institutes of Health DHEW Publication No. (NIH) 76-1067 For sale by the Superintendent of Documents, 0.5. Government Printing Office, Washington. D.C. 20402 V (ff/5034620 A1 : f i 3 ‘ i’U BL ?, SOVIET HEALTH STUDIES , 1 Anatomy of East-West Cooperation: U.S.-USSR Public Health Exchange Program, 1958-1967 The Soviet Five-Year Plan for Public Health, 19714975“ Fundamental Principles of Health Legislation of the USSR* Soviet Medical Research Priorities for the Seventies The Soviet Feldsher as a Physician's Assistant Medical Care in the USSR Nutrition Research in the USSR, 1961-1970 Machine Diagnosis and Information Retrieval in Medicine in the USSR’ Soviet Medicine: A Bibliography of Bibliographies Chronic Effects of Mercury on Organisms“ Soviet Biomedical Institutions: A Directory Soviet Personalities in Biomedicine Soviet Research in Pharmacology and Toxicology A Bibliography of Soviet Sources on Medicine and Public Health in the USSR Urban Emergency Medical Service of the City of Leningrad CHINA HEALTH STUDIES Medicine and Public Health in the People's Republic of China Topics of Study Interest in Medicine and Public Health in the People's Republic of China: Report of a Planning Meeting A Bibliography of Chinese Sources on Medicine and Public Health in the People's Republic of China: 1960-1970 Anticancer Agents Recently Developed in the People's Republic of China—A Review Prevention and Treatment of Common Eye Diseases‘ Standard Surgical Techniques, lllustrated“ Neurology - Psychiatry’ China Medicine As We Saw It Chinese Herbal Medicine Respiratory Research in the People's Republic of China An Economic Analysis of the Cooperative Medical Services in the People’s Republic of China SWEDISH HEALTH STUDIES National and Regional Health Planning in Sweden BRITISH HEALTH STUDIES British National Health ServiCe Complaints Procedures Community Medicine in England and Scotland *Translations of Soviet and of Chinese Documents, produced in very limited quantities only. ii INTERNATIONAL HEALTH COSTS AND EXPENDITURES The Government of the United States is reviewing on a continuing basis the national health activities of other countries in order to better serve the escalating medical requirements of the American people. Important elements of all health activities, of course, include biomedical research, medical education, health manpower and health services. An analysis of these foreign health-related activities and programs may provide the US. Government health administrators with new insights in solving some of the complex problems relating to the improvement of health in the United States. Recognizing the historical development of foreign medical systems, no single country or government may have the type of medical care or health system which will provide completely adequate health assistance desired by our citizens. A study of the best features of foreign health systems, however, ultimately may provide a better understanding of the perspective within which health exists in this country. Such a perspective, however, must include an improved comprehension of the political, economic, social, and other cultural aspects of society itself. The Fogarty International Center of the National Institutes of Health, established in 1968, and named in memory of the late Congressman John E. Fogarty of Rhode Island, is an organization envisioned by Mr. Fogarty and called for in his address to the Third National Conference on World Health in September 1963, as “a great international center for research in biology and medicine dedicated to international cooperation and collaboration in the interest of the health of mankind.” With Senator Lister Hill of Alabama, Congressman Fogarty charted the growth of the National Institutes of Health and the nation’s medical research and education for nearly two decades as Chairman of the House of Representatives’ Appropriations Subcommittee on the Departments of labor, and Health, Education, and Welfare. The many-faceted operations of the Fogarty Center have grown and flourished in collaboration with other American, foreign national and international bodies, and by means of bilateral agreements with the govern- ments of several countries including France, Italy, Japan, and the USSR. The Center also has the effective and continuing cooperation of international organizations such as the World Health Organization and the Pan American Health Organization and engages in less formal exchanges involving scientists and physicians from the United States and abroad. Similarly, toward the production of new and valuable medical findings, it shares its resources with other elements of the National Institutes of Health and with the US. Public Health Service. In addition to serving as the communications pulse for scientific informa- tion emanating from abroad, the Center provides American and overseas 184.45 scientists opportunities to deal with complex problems of vital concern in mankind’s well-being. These opportunities and services are inherent in the Center’s International Education Program, in its International Fellowship Program and the Visiting Program. Also being implemented is the Center’s International Research Exchange Program that enables American health professionals to study abroad. Many and varied health-related topics have been investigated by the Center’s Scholars-in-Residence Program, by a continuing program of conferences and seminars, and by its five-year-old Geographic Health Studies. This last mentioned undertaking, a series of studies designed to obtain and disseminate comparative knowledge of the health-care systems of other countries, is this publication’s raison d ’etre. Inquiries about this and other publications of the Geographic Health Studies Program, which are listed elsewhere in this book, should be directed to Dr. Joseph R. Quinn, Head, Geographic Health Studies. Milo D. Leavitt, Jr., MD. Director Fogarty International Center vi TAB LE OF CONTENTS PREFACE ................................ LIST OF CONFERENCE PARTICIPANTS ............... INTRODUCTION ............................ Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6 Chapter 7 Teh-wei Hu, Conference Program Director WELCOMING REMARKS AT THE CONFERENCE Dr. Milo D. Leavitt, Jr. INTERNATIONAL HEALTH COSTS AND EXPEN- DITURES—AN HORS D’OEUVRE .......... Robert Maxwell THE AMERICAN MEDICAL ECONOMY—PROBLEMS AND PERSPECTIVES ................ Dorothy P. Rice and Douglas Wilson DISCUSSANT’S COMMENTS—Rashi Fein ........ HEALTH COSTS AND EXPENDITURES IN CANADA. Robert G. Evans DISCUSSANT’S COMMENTS—Roderick D. Fraser . . . HEALTH COSTS AND EXPENDITURES IN THE UNITED KINGDOM ................. Michael H. Cooper DISCUSSANT’S COMMENTS—A. J. Culyer ....... A COMPARISON OF THE HEALTH-CARE SYSTEMS OF FRANCE AND THE UNITED STATES ..... Georges Rosch and Simone Sandier HEALTH-CARE EXPENDITURES IN FRANCE—A SATELLITE NATIONAL ACCOUNTS SYSTEM. . . Alain Foulon and Georges Rosch DISCUSSANT’S COMMENTS on Chapters 6 and 7~ A. Peter Ruderman Page ix XV 23 55 59 81 93 109 115 161 196 vii Chapter 8 Chapter 9 Chapter 10 Chapter 11 Chapter 12 INDEX .................................. viii HEALTH COSTS AND EXPENDITURES IN SWEDEN—THE PROBLEMS OF A PRIVATE GOOD IN THE PUBLIC SECTOR .......... Ingemar Stahl DISCUSSANT’S COMMENTS—O. W. Anderson ..... HEALTH COSTS AND EXPENDITURES IN BEL- GIUM, DENMARK AND THE NETHERLANDS . . André Prims DISCUSSANT’S COMMENTS—Jan Blanpain ...... HEALTH COSTS AND EXPENDITURES IN THE FEDERAL REPUBLIC OF GERMANY AND THE UNITED STATES ................... Uwe E. Reinhardt DISCUSSANT’S COMMENTS—Romuald K. Schicke . . HEALTH COSTS AND EXPENDITURES IN ROMANIA ...................... Cristian-Adrian Havriliuc COMPARISON OF HEALTH-CARE COSTS AND EXPENDITURES FOR WESTERN EUROPEAN AND NORTH AMERICAN COUNTRIES: REPORT ON PAPERS AND DISCUSSIONS AT CONFERENCE .................... Herbert E. Klarman Page 199 229 233 243 249 291 301 339 363 LIST OF CONFERENCE PARTICIPANTS MR. PAUL AHMAD Office of International Health Room 18-87, Parklawn Building leth Resources Administration 5600 Fishers Lane Rockville, Maryland 20852 DR. STUART H. ALTMAN Deputy Assistant Secretary for Health Planning & Evaluation US. Department of Health, Education, and Welfare Washington, D.C. 20210 PROFESSOR RONALD ANDERSEN Center for Health Administration Studies University of Chicago Chicago, Illinois 60637 PROFESSOR O. W. ANDERSON Center for Health Administration Studies University of Chicago Chicago, Illinois 60637 PROFESSOR RALPH ANDREANO Department of Economics University of Wisconsin Madison, Wisconsin 53705 PROFESSOR ROBIN F. BADGLEY Department of Behavorial Science University of Toronto 121 St. Joseph Street Toronto 5, Canada MR. STANLEY E. BISSEY Assistant to the Director Division of Engineering Services Bldg. 12A, Rm. 4015 National Institutes of Health Bethesda, Maryland 20014 PROFESSOR JAN BLANPAIN Director Department of Hospital Administration Leuven University Icuven, Belgium DR. MARTHA BLAXALL Senior Professional Associate Institute of Medicine National Academy of Sciences 2600 Virginia Avenue, N.W. Washington, D.C. 20418 MR JEFF BROWN Roche Laboratories 340 Kingsland Avenue Nutley, New Jersey 07110 DR. CRAIG D. BURRELL Vice President Sandoz, Inc. Pharmaceutical & Chemistry Co. East Hanover, New Jersey 07936 MR. EROL CAGLARCAN Roche Laboratories 340 Kingsland Avenue Nutley, New Jersey 07110 DR. PHIL CAPER Subcommittee on Health Senate Committee on Labor & Public Welfare Washington, D.C. 20510 MR. DAVID L. CHICCHIRICHI Executive Officer Division of Research Resources Bldg. 31, Rm. 53—03 National Institutes of Health Bethesda, Maryland 20014 DR. HAROLD COHEN Executive Director Health Services Cost Review Commission 2100 Guilford Avenue Baltimore, Maryland 21218 MR. JAY CONSTANTINE Senate Finance Committee Dirksen Bldg, Rm. 2227 Washington, D.C. 20515 PROFESSOR M. H. COOPER Department of Economics University of Exeter Exeter, EX4 4RJ, England DR. WILLIAM COPE, M.D. Special Assistant for International Affairs Parklawn Bldg., Rm. 903 5600 Fishers Lane Rockville, Maryland 20852 DR, ANTONY CULYER Institute of Social and Economic Research York University York, England MR. JOHN DALY Office of International Health Parklawn Bldg., Rm. 18-87 5600 Fishers Lane Rockville, Maryland 20852 DR. JOE DAVIS Office of International Health Parklawn Bldg., Rm. 18—87 5600 Fishers Lane Rockville, Maryland 20852 DR. ALAN DOBSON Special Studies Branch Office of Planning, Evaluation and legislation Parklawn Bldg., Rm. 13A—22 5600 Fishers Lane Rockville, Maryland 20852 DR. S. PAUL EHRLICH, JR. Director Office of International Health Parklawn Bldg., Rm. 18-67 5600 Fishers Lane Rockville, Maryland 20852 MR. JOSEPH EICHENHOLZ Special Assistant to Deputy Assistant Secretary U.S. Department of Health, Education, and Welfare Washington, D.C. 20210 PROFESSOR ROBERT G. EVANS Department of Economics University of British Columbia Vancouver, Canada PROFESSOR RASHI FEIN School of Public Health 643 Huntington Avenue Harvard University Boston, Massachusetts 02015 PROFESSOR PAUL FELDSTEIN School of Public Health University of Michigan 109 South Observatory St. Ann Arbor, Michigam48104 DR. ROBERT FISCHER Director Office of Europe Office of International Health Parklawn Bldg., Rm. 18-75 5600 Fishers lane Rockville, Maryland 20852 DR. PAUL FISHER Chief, International Staff Office of Research & Statistics Social Security Administration 1875 Connecticut Avenue Washington, D.C. 20009 MR. ALAIN FOULON Centre de Recherches et de Documentation sur la Consumation 45 Bd De La 75-6 34 Paris France DR. PETER FOX Director, Division of Health Analysis U.S. Department of Health, Education, and Welfare Washington, D.C. 20210 PROFESSOR R. D. FRASER Department of Economics Queens University Kingston, Ontario Canada PROFESSOR VICTOR FUCHS Prof. of Economics at Stanford University Vice President, Research at National Bureau of Economic Research 204 Junipero Serra Blvd. Stanford, California 94025 MR. WILLIAM FULLERTON House Ways & Means Committee Rm. 102, House Longworth Bldg. Washington, D.C. 20515 DR. EUGENE B. GALLAGHER Dept. of Behavorial Science University of Kentucky Medical Center Lexington, Kentucky 40506 DR. FRANK GAULDFELDT Director, International Organizations Division of Program Analysis Office of International Health Parklawn Bldg, Rm. 18-87 5600 Fishers Lane Rockville, Maryland 20852 DR. CLIFTON GAUS Assistant Director Research & Evaluation North Universal Bldg, Rm. 915 1875 Connecticut Ave., N.W. Washington, D.C. DR. DEREK GILL Dept. of Community Health and Medicine Practice University of Missouri School of Medicine Columbia, Missouri 65201 DR. FREDERICK GOLLADAY World Bank 1818 “H" Street, N.W. Washington, D.C. 20433 MR. JACK L HADLEY National Center of Health Services Research Health Resources Administration Parklawn Bldg, Rm. 15-A06 5600 Fishers Lane Rockville, Maryland 20852 MR. MARK HARRIS Health Economics Group Roche Laboratories 340 Kingsland Avenue Nutley, New Jersey 07110 DR. CRISTIAN-ADRIAN HAVRILIUC Institute of Hygiene and Public Health Bucharest, Romania DR. ROBERT HELMS Director, Center for Health Policy Research American Enterprise Institute 1150 17th Street, N.W. Washington, D.C. 20036 DR. JOHN HOLAHAN Urban Institute 2100 “M” Street, N.W. Washington, D.C. 20037 DR. ERIK HOLST Joint Center for Studies of Health Programs University of Copenhagen Juliane Mariesvej 32 DK2100 Copenhagen Denmark DR. LIEN-F U HUANG Department of Economics Howard University Washington, D.C. 20001 PROFESSOR TEH-WEI HU Visiting Research Professor F ogarty International Center National Institutes of Health Bethesda, Maryland 20014 DR. LEE HYDE Public Health Subcommittee House Committee on Interstate Commerce Rayburn Bldg., Rm. 2125 Washington, D.C. 20515 MR. STAN JONES Professional Staff Associate Subcommittee on Health Senate Committee on Labor and Public Welfare Washington, D.C. 20510 MR. GEORGE J. KELLEY Vice President Blue Cross Association 1700 Pennsylvania Ave., N.W. Washington, D.C. 20006 DR. JAMES C. KING Deputy Director Office of International Health Parklawn Bldg., Rm. 18-67 5600 Fishers Lane Rockville, Maryland 20852 xii PROFESSOR HERBERT KLARMAN Graduate School of Public Administration New York University 4 Washington Square, North New York, New York 10003 DR. MICHAEL J. KREMER Management Analysis Officer Division of Research Services Bldg. 12A. Rm. 4047 National Institutes of Health Bethesda, Maryland 20014 MR. STEVE LAWTON Chief Counsel Public Health Subcommittee House Committee on Interstate Commerce Rayburn Bldg, Rm. 2417 Washington, D.C. 20515 PROFESSOR M. L. LEE Department of Economics University of Missouri Columbia, Missouri 65201 DR. ROBERT LEYTON Office of Program Planning & Evaluation, NHLI Bldg. 31, Rm. 5A-27 National Institutes of Health Bethesda, Maryland 20014 MR. GERALD C. MACKS Office of Clinical & Management Systems Bldg. 10, Rm. 15—229 National Institutes of Health Bethesda, Maryland 20014 DR. HAROLD MARGULIES Deputy Administrator Parklawn Bldg, Rm. 10—05 Health Resources Administration 5600 Fishers Lane Rockville, Maryland MR. ROBERT MAXWELL McKinsey & Company London SW1A 1PS, England MR. JAMES MEAD Capital Blue Cross Association 100 Pine Street Harrisburg, Pennsylvania 17101 MRS. MARJORIE S. MUELLER Social Science Research Analyst North Universal Bldg., Rm. 919-G 1875 Connecticut Ave., N.W. Washington, DC. 20009 DR. JOSEPH NEWHOUSE Rand Corporation 1700 Main Street Santa Monica, California 90406 DR. WILLIAM POLLAK Urban Institute 2100 “M” Street, N.W. Washington, D.C. 20037 MR. MICHAEL R. POLLARD Institute of Medicine National Academy of Sciences 2101 Constitution Ave., N.W. Washington, DC. 20418 / PROFESSOR ANDRE PRIMS Faculty of Medicine University of Leuven Vital Decosterstraat 102 3000 Leuven, Belgium DR. JOHN RAFFERTY Director, Intramural Research Section Bureau of Health Services Research Health Resources Administration 5600 Fishers Lane Rockville, Maryland 20852 MR. ROBERT E. RAMSBURG Operations Research Analyst Bureau of Quality Assurance Parklawn Bldg. Rm 16A—44 5600 Fishers Lane Rockville, Maryland 20852 PROFESSOR UWE REINHARDT Woodrow Wilson School of Public Affairs Princeton University Princeton, New Jersey 08540 MRS. DOROTHY P. RICE Deputy Assistant Commissioner for Research & Statistics Social Security Administration 1875 Connecticut Avenue Washington, D.C. 20009 DR. GERALD ROSENTHAL Director, Bureau of Health Services Research Health Resources Administration 5600 Fishers Lane Rockville, Maryland 20852 DR. PETER RUDERMAN School of Public Health University of Toronto Toronto, Ontario Canada DR. GEORGE F. RUSSELL Director, Division of Management Policy Bldg. 1, Rm. 209 National Institutes of Health Bethesda, Maryland 20014 DR. LOUISE B. RUSSELL Center for Health Policy Studies National Planning Association 1625 Massachusetts Ave., NW. Washington, D.C. 20036 PROFESSOR D. SALKEVER School of Public Health Johns Hopkins University Baltimore, Maryland 21205 MME. S. SANDIER Centre de Recherches et de Documentation sur la Consumation 45 Bd De La 75~634 Paris, France DR. ROMUALD K. SCHICKE Department of Health Care Medical School 3 Hannover Karl Wiechert Allee 9 Federal Republic, Germany DR. SOLOMON SCHNEYER Director, Division of Program Analysis Bldg. 1, Rm. 228 Office of Program Planning and Evaluation National Institutes of Health Bethesda, Maryland 20014 PROFESSOR ROCKWELL SCHULZ Director, Program in Health Services Administration Medical School University of Wisconsin Madison, Wisconsin 53706 PROFESSOR S. O. SCHWEITZER Department of Community Medicine Georgetown University Washington, D.C. 20007 MR. RICHARD'L. SHERBERT Management Analysis Office Office of Administrative Management Bldg. 31, Rm. llA—33 National Institutes of Health Bethesda, Maryland 20014 MR. JOSEPH SIMANIS Office of Research & Statistics Social Security Administration 1875 Connecticut Avenue Washington, D.C. 20009 PROFESSOR HERMAN SOMERS Woodrow Wilson School of Public Affairs Princeton University Princeton, New Jersey 08540 PROFESSOR I. STAHL Institute of Economics University of Lund Lund, Sweden DR. ERNST W. STROMSDORFER Director, Office of Evaluation Department of Labor 200 Constitution Ave., NW. Washington, D.C. 20210 DR. PIERRE TACIER Sandoz Institute for Health & Socio-Economic Studies 5 Route de Florissant Geneva C.H. - 1206 Switzerland MR. HAROLD THOMPSON Office of Special Program Operation Division of Program Operation Indian Health Service Parklawn Bldg, Rm. 6A-55 5600 Fishers Lane Rockville, Maryland 20852 xiii DR. ROBERT VAN HOEK Acting Administrator Health Services Administration Parklawn Bldg., Rm. 14-15 5600 Fishers Lane Rockville, Maryland 20852 PROFESSOR B. WEISBROD Department of Economics University of Wisconsin Madison, Wisconsin 53705 MR. JACK WERNER Medical Research Scientist Department of Health Commonwealth of Pennsylvania Harrisburg. Pennsylvania 17101 MR. JOHN R. WHITE Director, Office of Program Development Bureau of Community Health Services Parklawn Bldg. Rm. 7A-33 5600 Fishers Lane Rockville, Maryland 20852 MRS. NANCY WORTHINGTON Social Science Research Analyst North Universal Bldg., Rm. 919-B 1875 Connecticut Ave., NW. Washington, DC. 20009 DR. ALONZO S. YERBY Visiting Research Professor Fogarty International Center National Institutes of Health Bethesda, Maryland 20014 PROFESSOR DONALD YETT Director, Human Resources Research Center University of Southern California Los Angeles, California 90007 PROFESSOR MICHAEL ZUBKOFF Chairman, Department of Community Medicine Dartmouth Medical School Hanover, New Hampshire 03755 MR. DANIEL ZWICK Associate Administrator Parklawn Bldg., Rm. 10A-53 5600 Fishers Lane Rockville, Maryland 20852 Conference Hogram Director DR. TEH—WEI HU Conference Planning Committee DR. RASHI FEIN DR. MAW LIN LEE DR. UWE REINHARDT DR. A. P. RUDERMAN DR. DAVID SALKEVER DR. STUART SCHWEITZER DR. BURTON WEISBROD Fogarty International Center DR. JOSEPH R. QUINN Chief, International Cooperation & Geographic Studies Branch Bldg. 31, Rm. 2C08 Fogarty International Center National Institutes of Health Bethesda, Maryland 20014 DR. TEH—WEI HU Visiting Research Professor Bldg. 31, Rm. 2C06 Fogarty International Center National Institutes of Health Bethesda, Maryland 20014 MRS. MICHIKO M. COOPER Conference Coordinator Executive Office Fogarty International Center Bldg. 31, Rm. 2C07 National Institutes of Health Bethesda, Maryland 20014 INTRODUCTION TEH-WEI HU CONFERENCE PROGRAM DIRECTOR Visiting Research Professor, F ogarty International Center, National Institutes of Health There have been two major conferences on health economics during the past 12 years. The Proceedings of the first conference were published in 1964 by the University of Michigan under the title, The Economics of Health and Medical Care. The Proceedings of the second conference were published by The Johns Hopkins Press in 1970 and were titled, Empirical Studies in Health Economics. In his Introduction to the second conference Proceedings, Herbert Klarman pointed out that: “Most striking perhaps is the absence of papers on the recent increase in health services costs and expenditures (except tangen- tially) and on the source and mechanism of financing health services.” He went on further to mention that, “In the not too distant future, it may be desirable to consider the advisability of conducting a conference that would concentrate on a single problem or policy issues.” Since the international exchange of knowledge in health sciences is the major function of the Fogarty International Center at the National Institutes of Health, while at the Center I was asked to coordinate a conference on health economics. Evidence of the growing concern over rising health costs and expenditures is seen in the increase to about 5 to 8 percent of the current Gross National Products (GNP) among European and North American countries, compared to 3.5 to 6 percent of the GNP in the early 60’s. In constant 1974 dollars in the United States, the real per capita spending increase was from about $309 in 1965 to $485 in 1974, or an increase of about 57 percent. Furthermore, the price of medical care has been steadily rising at a considerably faster rate than consumer prices in general. For instance, in the United States the Medical Care Services component of the Consumer Price Index in February 1975 was 14 percent above its value a year ago, compared to a 10.9 percent increase for all items except medical care. Therefore, it seems logical and timely that the rapid increase in health costs and expenditures in the US. and abroad should be a relevant and useful conference theme. The increasing costs and expenditures relevant to health care are the result of many factors, such as the increase in labor costs, changes in the nature of the health services provided as to their range, intensity, and sophistication, changes in health-financing mechanism, and the increase in hospital facilities. Not all these causes are known or uniform among European and North American countries. The purpose of this conference was to exchange views on XV what these causes are in these countries and how they can relate to each others’ experiences to contain increases in health cost and expenditure, taking into account the differing socio-economic and political environments. During the course of preparing this international conference on health costs and expenditures, a general outline that included eight questions was sent to each author so that a common basis of data collection and presentation during the conference could be established. Needless to say, the eight questions asked more than any one paper could answer, even if an author had all the required data. The questions were: 1. What are the costs of receiving health-care services (physician fees, hospital charges, and prescription drug charges)? What is the relationship between quality of health care and health costs? What are the major factors in the changes of these costs? 2. What are the costs of providing health-care services (mainly hospital capital costs and hospital current costs)? What is the relationship between the change of technology and hospital costs? 3. Which of these health-care components (hospital, physician, pharma- ceutical) contributes most to the total increase in health costs? Which of these health care costs increases most rapidly? 4. What are the expenditures on health-care services (private expendi- tures and public expenditures)? What are the major factors influencing these expenditures? 5. What is the relationship between reimbursement mechanism (fee-for- service, individual and group insurance, government national health insur- ance programs) and costs and expenditures? 6. Were there any major changes in health-care policy in your country during the past decade? If so, what were the changes? What impact have these changes had on costs, expenditures, and utilization? 7. What are the current trends in health costs and expenditures in your country? Are the government and consumers concerned about the increase in health costs? Have the government, medical profession, or the consumers developed any cost control mechanism? What have the results been? 8. What are the implications of the experience in your country for the future changes in the US. health-care system? In this Proceedings there are a total of eleven papers, covering Belgium, Canada, Denmark, France, the Netherlands, Romania, Sweden, the United Kingdom, the United States, and West Germany. The sequence of these papers is in accordance with the order of presentation during the conference. Discussant’s comments follow each paper (chapter). The first paper (chap- ter 2), prepared by Robert Maxwell, is an overview of the health costs and expenditures among most of the countries covered by the conference. It was hoped that this background information should provide a general picture for those conference participants who were not familiar with costs and expendi- tures in other countries. Herbert Klarman prepared the Proceedings’ last chapter, which also is a survey, but with different perspectives. Essentially, Klarman’s chapter summarizes the major points of each presented paper by xvi incorporating floor discussion and comments provided by the 120 conference participants. Therefore, readers who do not have time to read this entire volume, should read Klarman’s chapter. And, since Klarman has provided this excellent summary, this Introduction will not recite the major conclusions. The consensus of the conference was that the free market mechanism for containing health costs and expenditures in most of these European and North American countries is losing ground. The conference participants suggested that the role of government might be more active in this respect. On the demand side of health care, some governments are using the co-payment mechanism or are rationing health services by waiting time. Omthe supply side, some governments have been controlling increases in hospital beds and physicians; reducing the cost of physician reimbursement; and have introduced the regionalization of health-care services and the concept of fixed budgets for hospitals. Each of these control methods has merit, but none is totally effective. It was suggested that an effective solution might be to ask the medical community to cooperate more actively in the management of health services in an effort to contain costs and expenditures. Conference participants agreed that in view of the United States’ interest in national health insurance legislation, this country can learn much from other countries’ experiences involving reimbursement mechanisms and long-term care. It is believed that this volume provides important data sources and recounts experiences valuable for comparisons of health costs and expenditures in the countries represented. Several possible future conferences may be suggested as a result of this conference. First, an international conference on the definition of costs of health services. Brian Abel-Smith, Robert Maxwell, and many other authorities have mentioned that the definition of “cost of health services” varies from country to country. For example, the definition of health costs used by the International Labor Office is different from the definition used by the US. Social Security Administration. It would therefore be most advantageous if each country and international organization would adopt a common health- cost definition so that international comparisons can be facilitated. Second, a further conference on hospital costs would be most valuable. Hospitals account for the largest share of health expenditures in the health-care sector in each country. Although this conference was called to study health costs and expenditures, participants did not treat these two concepts discretely. In fact, as will be seen, this volume’s chapters indicate an emphasis upon increases in health expenditures rather than increases in health costs. Thus, a conference on hospital costs would be a logical consequence of this conference; and, of course, data on hospital costs are more readily available than data on other types of health costs. Finally, it is mentioned that the study of the costs of health services is but one part of the resource allocation concern. Also, a meaningful analysis would concern the relation of health costs and expenditures to the economic and social benefits received by individuals and nations. An international conference on the cost-benefit or cost-effectiveness of several selected health programs or various biomedical innovations is worthy of consideration. xvii During the past year many people made contributions to the conference and to these Proceedings. Dr. Joseph R. Quinn, Chief, International Cooperation and Geographic Studies Branch of the Fogarty International Center was initially instrumental in organizing the conference. His help and encouragement were invaluable. Valuable advice was also given by the members of the conference planning committee: Professors Rashi Fein, Maw Lin Lee, Uwe Reinhardt, Peter Ruderman, David Salkever, Stuart Schweitzer, and Burton Weisbrod. Thanks also are due to the authors, discussants, and all participants for making the conference successful. The last, but certainly not least of my acknowledgments is of the financial support afforded by the Fogarty International Center for the conference, and for my other activities during my year’s tenure at the National Institutes of Health. Bethesda, Maryland December 1975 xviii Chapter 1 WELCOMING REMARKS AT THE CONFERENCE MILD D. LEAVITT, JR. Director, Fogarty International Center I take great pleasure in welcoming you to this Fogarty International Center Conference to consider health costs and expenditures from an international perspective. The urgent need for action to deal with the rising costs and expenditures for health care is clearly indicated in most Western European and North American countries. The objective of this conference is to exchange views on factors which contribute to increases in health costs and expenditures and to discuss ways to contain these increases. We hope that the information developed by this conference can be shared by concerned individuals and governmental agencies in various countries for the development of possible solutions to this mounting problem. About 5 years ago the Center became interested in health systems which have been established and utilized by governments other than that of the United States. We were interested in the effectiveness of these systems in meeting the particular health needs of their people. This was the nucleus of the Center’s Geographic Health Studies Program. The first two countries studied were the Soviet Union and the People’s Republic of China. During the last 2 years, the Center has expanded its interest in the health systems of several European and North American countries, including those of Sweden, Denmark, the United Kingdom, and Canada. Publications on these and other subjects have been issued or will soon be issued by the Center. More recently, the Fogarty International Center has begun to examine the relationship of the social sciences to medicine and how it might examine current problems and possible solutions, combining the use of various disciplines in the social and behaviorial sciences. We are therefore establishing a Health Policy Studies Project that during the next academic year will bring together a multidisciplinary group of individuals from universities to examine, in greater detail, some important questions relative to the general theme of national health insurance. Dr. Teh-wei Hu, principal organizer of this meeting and chairman of the conference planning group is the Fogarty International Center’s first Visiting Research Professor. As a result of Professor Hu’s participation in the activities of the Fogarty Center during the past year, we have been encouraged to expand this relationship with the academic sector by establishing a multidisciplinary group, and we look forward to closer relationships with representatives of the social and behaviorial sciences in considering important problems in years to come. ' Because the United States is currently considering alternative approaches to health financing, costs and expenditures of health services are particularly relevant to our health planning. We are most fortunate to have participating in this conference 20 distinguished experts from Europe and North America, United States Congress staff members, and representatives of several Federal agencies such as the Social Security Administration, the Office of Assistant Secretary for Health, the Office of International Health, the Health Resources Administration, and the Health Services Administration. We have invited Dr. Stuart Altman, Dr. Alonzo Yerby, Dr.Jan Blanpain, Dr. Erik Holst, Dr. Victor Fuchs, and Dr. Herman Somers to serve as con- ference session chairmen. I wish you a most successful and productive con- ference. Chapter 2 INTERNATIONAL HEALTH COSTS AND EXPENDITURES— AN HORS D’OEUVRE ROBERT MAXWELL McKinsey & Company, London, England My purpose in this opening paper is to complement the main papers. The figures given for each country in the national papers will no doubt be more detailed, more reliable and more up to date than those I can give. And since I am not an economist, but a cost accountant and management adviser, others can explain with much more authority the workings of the economic forces. What I will do, drawing on an international survey of health trends and of the way that resources are used, and on my consulting experience in a number of the involved countries, is to start off the conference with a simplified—but I hope not distorted—overview. We are all familiar with the fact that health expenditures have everywhere been rising faster than national income. I will look first at expenditure trends and patterns, at the forces underlying the trends, and then at the way in which individuals and nations are responding to them. Against that background I will consider the implications for the future. Expenditure Trends To compare health expenditures internationally one has to consider not only public but also private expenditures, and one must try to exclude items like cash benefits to the extent that these fall outside the health field. The 1969 health expenditure figures shown in Figure 1 include those for most of the nine countries with which this conference is concerned, and they show a threefold spread from the less prosperous countries to the wealthier ones. Later figures, where we have been able to obtain them, show substantial further increases (Figure 2) and suggest no lessening of the spread. However, this spread is reduced by considering the number of hours a person must work in each country to “cover” his health expenditure (Figure 3) and by relating health expenditure to GNP (Figures 4a and 4b). Quite clearly, national health expenditures are related to national income— the wealthier the country, the greater its tendency to spend a larger portion of its GNP on health care (Figure 5). This is one explanation of why health expenditures have everywhere risen faster than GNPs. While our figures are not as comprehensive or conclusive as I would like, they also suggest that the rise has been accelerating. Whereas in the 1950’s most countries increased their ANNUAL HEALTH EXPENDITURE PER PERSON (U. S. $) 1969 United States '298 Sweden Canada West Germany Australia*** France ‘Netherlands* United Kingdom Italy Republic of Ireland“ *— 1969 ** — 1971 ***— 1970 Source: McKinsey Research Department (calculations from total expenditure and population statistics) Figure 1. Annual health expenditure per person in 10 countries, comparisons in US. dollars. health spending by about 1 percent of GNP, an increase of 1% percent was more typical in the 1960’s (Figure 6). For the 1970’s, on current indications, this might become 2 percent, despite a reduction in the rate of increase of GNP (Figure 7). So a country that spent 3.5 percent of its GNP in 1950 could be spending about 6 percent of a much larger GNP by 1970, and 8 percent by 1980. Expenditure Patterns Difficult as it is to get crude expenditure totals for international comparison, it is even more difficult to get the totals broken down into standard expenditure headings. Three important generalizations do apply, however, in a range of countries and in different years. The first is that salaries and wages typically account for between 66 and 70 percent of total expenditure (Figure 8). Health services are very labor intensive, 4 and nobody has convincingly demonstrated that expenditure on capital equipment alters this. Indeed, as we shall see, the tendency is the other way: manning has increased relative to facilities as services have become more sophisticated. Today something like 150 people per 10,000 population, or 4 percent of the working population, are employed in health care (excluding closely related activities like the pharmaceutical industry (Figure 9)). In the principal professional categories, however, there are substantial differences between countries relative to population; for example, in the number of doctors and, less reliably, of nurses (Figures 10 and 11). These staffing figures show that contrasts in health expenditure cannot be wholly explained by salary and wage differentials (such as the fact that most professional people in Britain are poorly paid) important as such differentials are. The second generalization is that hospital services typically absorb more than one-half the total health expenditure while dealing with a relatively small ANNUAL HEALTH EXPENDITURE PER PERSON (U. S. $) , 1969 _ 298 ”MMStates 1973 _ 447 _ Sweden 1969 234 1971 _ 322 c d 1969 — 228 “a a 1971 _ 305 F 1969 — 150 ”me 1973 — 288 _ 1969 _ 150 Australia 1973 — 259 1968 _ 116 Netherlands 1972 — 249 1969 - 93 U 'ted K' d "' '"9 °”‘ 1973 _ 158 Source: McKinsey Research Department (calculations from total expenditure and population statistics) Figure 2. Increases in annual health expenditures, per person in seven countries. part—in the main, of course, the gravest part—of the incidence of illness (Figure 12). Typically the hospital share of the total, like the salary and wage proportion, has remained relatively fixed, despite the announced intention of many governments to shift the balance in favor of community-based services. However, the number of hospital beds per 10,000 population has not increased at anything like the same rate as has health manpower. In the 1960’s in the countries we surveyed, overall reductions in hospital beds were at least as common as increases. With the notable exception of Sweden, most countries seemed to be stabilizing around a norm of 100 beds per 10,000 population (Figure 13). Incidentally, new construction costs per hospital bed have risen sharply everywhere, though there is no fixed relationship between capital expenditure and running costs; rather, capital expenditure is easier to control and hence fluctuates more sharply according to government policy and the availability of finance. AVERAGE HOURS WORKED TO PAY FOR ANNUAL HEALTH EXPENDITURE, 1969 France 148 Australia Sweden West Germany Italy Netherlands United States United Kingdom Source: McKinsey Research Department (calculations from total expenditure and average direct hourly wages) Figure 3. Average hours worked to pay for annual health expenditure, 1969 comparisons between nine countries. (a) PROPORTION OF GNP SPENT ON HEALTH SERVICES (%) 7960‘ 1969 V United stem i 6.1 sweden —— 6.4 Canada I 6.4 Netherlands - 5.9 . (1963-68) West Germany - 5.7 < (1961-69) Austria I 5.6 ‘ (1960-68) France _ - 5.4 Italy - 5-2 Finland - 5.1 Australia | 4.9 ‘ (1961-70) United Kingdom I 4.6 Source: McKinsey Research Department from national and IMF statistics (b) PROPORTION OF GNP SPENT ON HEALTH SERVICES (%) 1969 V Unitedswes 1960—- 6.7 1913—11 Sweden 1960 _ 6.4 1971—10 Canada 1960—. 6.2 1971— 7.1 1963_-5.9* ‘ ”ethe"a““s 1972— 7.3 F 1960 - 5-4 ran“ 1973 — 5.8 1961_I 4.9” A . “"3"“ 1973— 5.6 1960 _,9 _ * United Kingdom I 4 6 — 1968 1973—53 H_ 1970 Figure 4. Proportion of GNP spent on health services (3) increases percent in 11 countries; (b) increases percent in seven countries. PROPORTION OF GNP SPENT ON HEALTH SERVICES (%) GNP Per CapitaI 1969 (U. S. Dollars) United States 1969 4240 Sweden 1969 2920 Canada 1969 2650 Netherlands 1968 1760 West Germany 1969 2190 Austria 1968 1470 France 1969 2460 Italy 1969 1400 Finland 1969 1980 Australia 1970 2300 United Kingdom 1969 1890 Source: McKinsey Research Department from national and IMF statistics Figure 5. Proportion of GNP spent on health services and the 1969 GNP per capita in 11 countries. My third generalization is that not all the resources—manpower, hospitals, money—are rationally used. A rational person would (assuming he could grasp the necessary, complex facts) allocate the resources where most needed and where they could do most good. Priority of need would thus be one pre-eminent criterion, and effectiveness of the existing or proposed service would be the other. Most health systems show some glaring examples of obvious contraventions of one of these criteria, despite everyone’s good intentions within the limited perspective of their own activities, institution or agency. Thus, for example, the Inverse Care Law frequently applies: that not only are the resources unevenly distributed, but that they are more thinly spread where the need is greatest. Services tend to be at their worst in poorer communities, such as among immigrant populations, with their great social and health needs (Figure 14); and they tend to be at their worst for those with chronic complaints, as opposed to episodes of illness (Figure 15). On the 8 criterion of effectiveness, the evidence is often absent—we simply do not know what lasting benefit is achieved for the patient. But it is known that the value of some intensive, sophisticated services is in doubt; that there is overprovision of some of the most expensive facilities; and that the way services are used is often determined more by what is available than by appropriateness to the problem (Figures 16a and 16b). In pointing to irrationalities in the use of resources I do not imply that eliminating these irrationalities would be easy; neither do I underestimate the efforts made to do so. Indeed, all three of the points made about patterns of expenditure emphasize the self-perpetuating tendency, the relative rigidity in the way resources are used. Making any substantial alteration in the patterns is very difficult, requiring determination, patience and opportunism. ESTIMATED INCREASES IN PERCENT OF GNP SPENT ON HEALTH IN THE 19605* Sweden 2. 9 Finland United States Netherlands France Italy In. , In West Germany Canada .1 Austria United Kingdom Australia 0.3 AVEfiAGE 1.3 * — Between 1960 and 1969 Source: McKinsey Research Department from national and IMF statistics Figure 6. Estimated increases in percent of GNP spent on health in the 1960s in 11 countries, and the average increase. ESTIMATED INCREASES IN PERCENT OF GNP SPENT ON HEALTH SINCE 1969 EQUIVALENT NUMBER OF RATE FOR INCREASE YEARS 10 YEARS Netherlands 1.4% 4 3.5 Canada 2 3.5 Sweden 2 3.0 United States 4 2.5 Australia 3 2.0 United Kingdom 4 1.8 France 4 1.0 Source: McKinsey Research Department from national and IMF statistics Figure 7. Estimated increases in percent of GNP spent on health since 1969, in seven countries. THE REVENUE COST OF HEALTH SERVICES IN ENGLAND AND WALES (1971-72) £MILLION Other Costs £672m. Salaries and Wages £1 ,434m. 10 Figure 8. The revenue costs of health services in England and Wales, 1971-72. People employed in health services Survey average per 10,000 population 0.000....0... 0. Doctors15 0.00.00.00.00 :000000000000 NursesZ7 0.00 Dentists4 0.0.0 Pharmacists 5 0......000000 0 . .00 .0. O 0 0.0.0... 0 . 00.0.... : : :::::m 000 Source: World health : : 0"“.... statisticsand Supporting staff 100 McKlnsey estimates Figure 9. Survey average of people employed in health services per 10,000 population. DOCTORS PER 10,000 POPULATION 1972 V USSR — 25-6 Italy — 189* West Germany _ 18.4 Belgium —16.4* Denmark — 16-3 Canada — 15.8 Scotland — 15.6 United States —15.4** Sweden — 14.7 Netherlands _ 13.6 France —13.2*** England and Wales _12.7** Australia - 12.6” * — Includes physicians practicing dentistry Source: World ** — 1971 Health Statistics *** - 1970 Reports Figure 10. Doctors per 10,000 population in 1972 in 13 countries. NURSES* PER 10,000 POPULATION Australia — 61.7 Denmark — 48.7 Canada — 45.9 ussn _ 42.6 Sweden — 40.7 United States _ 35 3 England and Wales — 30. 7 France _ 26.6 West Germany _ 23.1 Netherlands - 19.2 Italy - 6.9 Source: World Health Statistics Annuals Portugal I 4'5 and National Statistics * — Fully qualified Figure 11. Nurses per 10,000 population in 1970 in 12 countries. RUNNING COST OF THE NATIONAL HEALTH SERVICE IN ENGLAND AND WALES* £ MILLION £2,612m. 1,125 43% Source: Health and personal social ser- vices statis- £416m. 57% tics for England Other 183- 44% Hospitals 233- 56% 1951 1973 * -— 1973 — England only 12 Figure 12. Running cost of the National Health Service in England and Wales. HOSPITAL BEDS* BY COUNTRY BEDS PER 10,000 PERCENTAGE POPULATION CHANGE (1971) (1960-69) Sweden — 166 +5 Scotland _ 121 -4 Australia - 121 +1 Republic of Ireland _ 119 -40 Netherlands _ 117 +22 Northern Ireland — 115 -3 Luxembourg — 115 -6 West Germany _ 113 +7 ltaly - 106 +13 France _ 104 ' -31 Canada - 98” -10 Denmark - 96 -10 England and Wales - 92 -9 Belgium - 83” 0 United States - 75 -12 * — All hospitals ** — 1970 Source: World Health Statistics Reports Figure 13. Number of hospital beds per 10,000 population and percentage changes therein in 15 countries. Underlying Forces Why have health expenditures risen? Once again the information is patchy, but there is enough evidence to support the conclusion that population increases, and increases in the quantity of demand, have played a relatively small part. The two big factors have been increases in personnel costs, particularly through higher pay (Figures 17a and 17b), and changes in the nature of the services given, in their range, intensity and sophistication (Figure 18). The first factor is relatively simple. There has been a catching up process in bringing the wages of many supporting health service personnel more into line with other occupations*. Many of the skill groups have raised their standards *While this catching up process has been talked of for a long time, and may have happened prior to 1960 in the United States, it has been relatively recent in many West European countries. 13 of qualification, their status and their salaries. And there is some evidence that when individuals and communities increase their willingness to spend on health without at the same time expanding the resources paid for, a substantial part of the increase goes into higher salaries and wages. The sophistication factor is more complex. It stems partly from the changing nature of disease. With the advance of medicine, declining birth rates and lower perinatal and infant mortality, health services have increasingly had to care for an aging population affected by the chronic, degenerative diseases. It is one of the unalterable paradoxes of health care that you cannot win in the battle against ill-health. You can win battles, but in the long run you cannot win the war because death is inevitable, and so, it would seem, is the physical deterioration associated with aging. Other influences are also at work on the sophistication factor. Expectations change concerning what people can expect from health services; people’s demand threshold has tended to fall and public and professional judgment of what constitutes a satisfactory quality of service has risen drastically. And, finally, science and technology have made possible an enormous extension of the tools and techniques of intervention, and these new tools and techniques are generally expensive. looking ahead, so far as I can see all the underlying forces except one point in the direction of higher expenditures and an accelerating rate of cost inflation. First, people’s health needs and expectations wfll continue to change and, in the main, to rise. The expansion of the over 75’s, with their great health needs, is by no means over; nor is the increase in public and professional expectations, particularly on what constitutes an appropriate level of service. Today, certain areas of existing services in most countries quite obviously need improvement; and there will always be some gap between the standard of services we have and the standard we would like to have. Then, there is the likely influence of medical science. While I am sure that advances in medicine will continue and will provide the opportunity for more sophisticated services, in very few cases will they actually reduce the need for care. Tuberculosis was a case where an advance in medicine did sharply reduce the need for an important service; but that was a very rare case, and now that chronic, degenerative diseases dominate we can expect few such cases in the future. On the personnel side, there are as yet no signs of moves to reduce numbers. And so far as medical manpower goes, which has an influence on expenditure far greater than its direct cost, some substantial increases are already mandated by the expansion of medical schools that has already taken place (Figure 19). Nor is the drive towards higher wages and salaries by any means over. Combining these upward pressures it is very easy to see how the proportion of the GNP spent on health care can escalate. In the United Kingdom, for example, the current expenditure level of 5 percent of the GNP could quickly reach 8 percent (Figure 20). The one factor working the other way is a clash with other priorities. At some point, varying from case to case, nations are bound to say, explicitly or implicitly, “We would like to spend more on health but we cannot do so.” In this discussion of underlying forces I have left out human intervention that would deal with some of the irrationalities in the present pattern of 14 DOCTORS PER 100,000 POPULATION IN APPALACHIAN STATES BY WEALTH OF COUNTY (1967) DOCTORS O 20 40 60 80100120 Alabama Georgia Kentucky Maryland Mississippi North Carolina Ohio Tennessee Virginia West Virginia Counties with median disposable income of - $5,000 or less More than $5,000 Source: Towards a Comprehensive Health Policy for the 19705, Department of Health, Education and Welfare, May 1971 Figure 14. Number of doctors per 100,000 population of the Appalachian States of the U.§.,l)y wealth of county, 1967. Mental handicap institutions are short-staffed and there are large variations in staffing standards. . . . PATIENTS PER MEMBER OF WARD STAFF ON DUTY Children's All Wards Wards Best staffed hospital in sample 10.3 5.0 Average for sample 15.8 11.1 Worst staffed hospital in sample 34.0 19.4 Source: Pauline Morris Put Away, 1969, p. 105-6 (U. K. data) Figure 15. Mental handicap institutions (UK), number of patients per member of ward staff on duty. resource use. This could, and I hope will, be an important influence. It is, however, a force of a different kind from those I have just described, which will be at work whatever health leaders do. Effective intervention, on the other hand, depends on them, and to some extent on conference participants. National Responses I shall give only a brief summary of national responses, since my information is by no means comprehensive, particularly on what nations are proposing to do in the future. As regards the past, there has been a universal tendency for governments to take an increasing share of the financing burden as health care costs have outstripped most individuals’ capacity to meet them, even on an insurance basis. Typically, governments have stepped in directly or via the insurance systems to help the less affluent meet the costs of acute illness, and to help everyone meet the disaster of long-term illness and incapacity. Sweden and Great Britain have gone to the extreme of meeting most health expenditures out of general taxation. Many European countries now have compulsory insurance schemes covering the majority of the population in a manner that is very little different from financing by general taxation. At the other extreme has been the United States with some 60 percent of expenditure covered from private sources, including non-compulsory insurance. But the United States is travelling along the European route. Some countries, most notably France, have deliberately retained some substantial direct charges to ‘ patients across a broad range of services in an attempt to bring home to the public the cost of these services. Naturally enough, the shouldering of a major part of the financial burden has been followed by some government intervention in administration. Typically, this has started with involvement in reimbursement and regulation of the insurance funds. To very varying degrees some intervention in the health services themselves, and the key resources of medical manpower and health facilities has followed, in efforts to ensure quality and to influence resource distribution. The governments of nearly all the developed countries now have some controls over new hospital building, the accreditation of hospitals, the licensing of their own physicians and specialists, the size of medical schools, and the immigration of non-nationals. Relatively few have gone so far as to try to control total health expenditure, to force the closure of beds considered to be surplus, to rationalize medical programs among institutions, or to direct where physicians may and may not practice. But by becoming increasingly involved in the administration of health care, governments are having to consider whether or not they will intervene in these issues. Implications Today, you will often hear it said by thoughtful commentators that the root problemin financing health care is that, while resources are inevitably finite, demand is infinite. Personally, I believe this to be an oversimplification in its 16 (a) DUPLICATION AND UNDERUTILIZATION OF FACILITIES WORKLOAD OF 777 U. S. HOSPITALS EOUIPPED TO DO CLOSED HEART SURGERY One case or more per week No such cases Fewer than _ one case In the year per week studied Source: A National Program to Conquer Heart Disease, Cancer and Stroke. (Washington, U. S. Government Printing Office, 1965) (b) MISUSE OF ACUTE BEDS IN GENERAL HOSPITALS (data from Radcliffe Hospital, Oxford) 0 20 40 60 80 100 I I I I I 7 Percentage of patients who could have been placed under GP care Percentage of days unnecessarily absorbed by these patients Unnecessary Source: The Demand for Hospital Care admission ISL London, 1970 - Delayed discharge Figure ’16. (a) Duplication and underutilization of facilities in 777 US. hospitals equipped to perform closed heart surgery; (1)) misuse of acute beds in general hospitals (U.K.). 17 (a) Higher hospital costs in Canada . . . . AVERAGE COST PER PATIENT DAY $62 $23 I 1961 1971 = + $39 (index increase from 100 to 268) L_____> (b) . . . . Are due mainly to personnel costs. . . . MAIN FACTORS BEHIND THE INCREASE PERCENT OF $39 INCREASE Labor Cost Per 58 Patient Day Labor cost per Paid hours paid hour of work of work per patient Non-Labor Cost Per - 20 day Patient Day Number of Patient 95 22 Days I \ Popu- Patient days lation per capita increase Source: Health and Welfare, Canada Figure 17. (a) Average cost per patient per day in hospital in Canada; (b) factors behind the increase in this cost. CAUSES OF INCREASED HOSPITAL COSTS 1955-68, TOTAL ACUTE HOSPITAL COST GROWTH 265% INCREASES IN: I Input Quantity — 106% o Labor7Patient7Day -3OA’ Supplies and Services/ _ 0 Patient Day 76% Wages Supplies and Services 0 Prices I 91’ Utilization —70‘7_o 0 Population - 32A) Hospitalization Rate - 38% Source: McKinsey Research Department based on U. S. government statistics Figure 18. Causes of increased hospital costs. NUMBER OF DOCTORS IN SWEDEN 000 30 20 1970 1975 1980 1985 1990 Source: S. Ake Lindgren, Head of Planning Department, National Board of Health and Welfare. Health Service Prospects, Lancet/NPHT, 1973 Figure 19. Number of doctors in Sweden, projected from 1970 through 1990. 19 description of demand. It suggests a misleading inevitableness and homoge- neity. In fact health is such a professional business that most of us cannot formulate a clear demand at all; much of the time we turn to the doctors with a want, and the doctor decides for us the urgency of our need and what (if anything) should be done about it. The concept of need, involving an objective and informed assessment of a health or social problem, in the health care context seems to me much more important than the concept of demand. Whether or not need is infinite, it is certainly not homogeneous in type or degree. In its totality it'cannot be defined with precision, and it shades away- from the clear and unambiguous through conditions of less clarity and less urgency. The root problem in health care is that, as indicated earlier, inherited patterns of service and the forces influencing their expansion do not do a good job of placing the available resources where they are most needed, or where they will do most good. ' _ Some of those attending this conference have wisely said that it is not the total health care bill that should worry governments so much as obtaining better value, or greater effectiveness of care, for the money spent. I agree with that. But when health costs are rising much faster than a nation’s income the government also has to worry about the total bill. The British Government has had to do so for years, and even the Swedish Government with its greater national wealth and fine standard of service has been doing so recently. So is the Netherlands Government. Not for long are many governments going to be able to foot a sharply rising bill without concern about the total; and that may even be a good thing. For, in my experience, awareness of a budget limit provides a useful stimulus, both for agencies and for individuals, to think about priority of need and effectiveness of service. There are relatively few services, existing or proposed, that can be said categorically to be harmful or an utter waste of money; but there are many that do not survive comparison with alternative uses within some kind of global limit. The elements I would like to see in a response to the problem we are considering at this conference are: (1) Definition of need. That all the varying health needs of a whole community, be it national or local, should be considered in allocating limited resources: therefore for planning purposes there should be close coordination among all the agencies serving that population, and basic epidemiological information should be available to them all. (2) Assessment of effectiveness. That the effectiveness of services should be much more carefully evaluated, and that new and different information is required for this purpose, concentrating on the end results achieved and how those results can be improved. (3) Awareness of limitations and of individual responsibility. That much more emphasis should be placed on what individuals can do to protect their own health—on responsibilities as well as rights, and that policymakers should be much more skeptical about the value of expanding services. (4) Use of resources. That sensible and sensitive decisions should be taken, involving the community, on what total resources can be made 20 Potential growth of health service expenditure in England and Wales between 1971 and 1980 ' Percentage of GNP 10 P Public demand for all-round higher standards 8 .— / Care for disadvantaged groups 6 — Increased staffing costs \ Reducing regional variations 4 — . 1971 Aging population 2 — 4.9 ' 0 Source: Mc Kinsey estimates Figure 20. Potential growth of health service expenditure in England and Wales, 197 1-1980. available for health purposes, and that these resources should be allocated and used on the basis of priority of need and effectiveness of treatment. ‘ I see very broad scope for choice by each country, and within countries by local communities, on how these elements are pieced together in their specific solution. The choices will show greater variety and Ihope that there will be continuing an exchange of experience so that we can all learn what works and what does not and thereby improve our response to one of the great problems of our time. 21 Chapter 3 THE AMERICAN MEDICAL ECONOMY~PROBLEMS AND PERSPECTIVES DOROTHY P. RICE and DOUGLAS WILSON Office of Research and Statistics, Social Security Administration, Department of Health, Education, and Welfare, Washington, D. C. Introduction The provision and financing of health-care services is complex, comprising an intricate pluralistic system of private, federal, state, and local government sectors. Expenditures for medical care have been increasing at a rapid rate in recent years, reaching $104.2 billion in Fiscal Year 1974. In Fiscal Year 1950, medical-care expenditures amounted to $12 billion dollars and represented 4.6 percent of the Gross National Product; by Fiscal Year 1974, they were 7.7 percent of GNP (Table l and Figure 1). The private share of the American medical market has always been larger than the public one. With the addition in recent years of new public programs of Medicare and Medicaid (Titles XVIII and XIX of the United States Social Security Act, respectively), there has been a significant increase in public financing. From the end of World War II to 1966, private outlays constituted three-fourths of the total. Within the public sector, state and local governments were spending more than the federal government. But by Fiscal Year 1974, the government’s proportion had increased to 40 percent of the total. Federal expenditures were more than twice that of state and local governments. Although the private medical market is a dominant source of expenditures, the mechanism of paying for medical care is predominantly through third.party payers, both public and private. In Fiscal Year 1974, 65 percent of personal medical care services was paid through third-party payers (Table 2). . The American medical economy is also characterized by its entrepreneurial spirit and community support. Community hospitals represent 83 percent of all hospitals, handle over 92 percent of a1] admissions and 77 percent of all outpatient visits, employ 78 percent of the industry’s total labor force, and account for 79 percent of the total cost of hospital care. Of the 900,000 community hospital beds, less than one-fourth are in state and local government hospitals (1). Nursing homes are predominantly operated for profit. The 211,000 physicians in private practice are primarily engaged in fee-for-service care. Unlike physicians in some other countries, it is essential for the livelihood of private-practice American physicians to have staff privileges at local hospitals. Competition among physicians and other medical practitioners 23 and among hospitals is by non-price means. To a large extent, supply determines demand. One of the motivating forces in American medicine is the stress on maintaining and augmenting medicine’s scientific base. This concern is manifested in the increasing specialization among medical personnel and the drive by hospitals to acquire and utilize increasingly sophisticated medical equipment. In 1960, only 10 percent of community hospitals had intensive care units; in 1973 about 57 percent had them (2}. There appears, however, to be a disparity between the increase in training and sophistication of the treatment processes and the improvement of medical-care outcomes. This disparity raises the question of the value being derived from the additional private and public dollars spent for medical care. Despite the infusion of public dollars into the medical-care system and marked improvements in providing medical care to the poor and aged, there are wide differences in the use of services and benefits received by these groups (3). The increased emphasis on medical care as a right rather than as a privilege suggests that all persons have equal access to adequate medical care provided in a manner that respects the rights and dignity of the individual. In summary, the dominance of third-party payers and relatively low out-of-pocket price for treatment, combined with the noncompetitive aspects of American medicine and increasing sophistication of medical-care treatment, have resulted in rising medical-care prices and expenditures. The control of this chronic expenditure inflation in the medical industry and the search for mechanisms to fill the gaps in existing private and public medical-care programs PERCENT OF GNP 1950 1960 1970 197| 1972 1973 1974 Figure 1. National healflr expenditures as a percent of GNP, selected Fiscal Years, 1950-1974. 24 SZ Table l AGGREGATE AND PER CAPITA NATIONAL HEALTH EXPENDITURES, BY SOURCE OF FUNDS, AND PERCENT OF GROSS NATIONAL PRODUCT, SELECTED FISCAL YEARS, 1928- 29 THROUGH 1973-74 Health expenditures Gross Total Private Public Fiscal year National Product Amount P f Amount f Amount P t f (in billions) (in Per capita ercent 0 (in - Per capita Percent o (in Per capita ercen o millions) GNP millions) total millions) total 1928-29. $101.0 $3,589 $29.16 3.6_ $3,112 $25.28 86.7 $477 $3.88 13.3 1934-35 . 68.7 2,846 22.04 4.1 2,303 17.84 ‘ 80.9 543 4.21 19.1 1939-40. 95.1 3,863 28.83 4.1 3,081 22.99 79.8 782 5.84 20.2 1949-50 . 263.4 12,028 78.35 4.6 8,962 58.38 74.5 3,065 19.97 25.5 1954-55 379.7 17,330 103.76 4.6 12,909 77.29 74.5 4,421 26.46 25.5 1959—60 495.6 25,856 141.63 5.2 19,461 106.60 75.3 6,395 35.03 24.7 1964—65. 655.6 38,892 197.75 5.9 29,357 149.27 75.5 9,535 48.48 24.5 1965-66. 718.5 42,109 211.56 5.9 31,279 157.15 74.3 10,830 54.41 25.7 1966-67. 771.4 47,879 237.93 6.2 32,057 159.30 67.0 15,823 78.63 33.0 1967-68 827.0 53,765 264.37 6.5 33,727 165.84 62.7 20,040 98.54 37.3 1968-69 899.0 60,617 295.20 6.7 37,682 183.51 62.2 22,937 111.70 37.8 1969-70. 954.8 69,202 333.57 7.2 43,964 211.92 63.5 25,238 121.65 36.5 1970-71 . 1,013.6 77,162 368.25 7.6 48,558 231.74 62.9 28,604 136.51 37.1 1971- 72. 1,100.6 86,391 408.31 7.8 53,365 252.22 61.8 33,025 156.09 38.2 1972-73.1,225.2 94,235 441.94 7.7 58,415 273.95 62.0 35,819 167.98 38.0 1973- 74* 1,349.8 104,239 485.36 7.7 62,929 293.01 60.4 41,311 192.35 39.6 *Preliminary estimates. Source: Worthington, Nancy L., “National Health Expenditures, Fiscal Year 1974,” R&S Note No. 32, 1974, Office of Research and Statistics, SSA. 93 Table 2 PERSONAL HEALTH-CARE EXPENDITURES AND PERCENT DISTRIBUTION BY SOURCE OF FUNDS AND BY TYPE OF SERVICE, FISCAL YEAR 1974 Third—party payments (percent) Personal health . Direct Type of service care expenditures Private Govern Philanthro payments . . . ‘ PY (in billions) Total . health ment and industry (percent) insurance Total ........................ $90.3 65 26 38 1 _ 35 Hospital care ..................... 40.9 90 35 53 1 10 Physicians’ services ................ 19.0 61 37 24 — 39 Dentists’ services .................. 6.2 14 9 5 — 86 Drugs ......................... 9.7 14 6 8 — 8 6 All other* ....................... 14.5 54 3 46 5 46 *Includes other professional services, eyeglasses and appliances, nursing-home care and other services not elsewhere classified. Source: Worthington, Nancy L., “National Health Expenditures, 1929-1974”, Social Security Bulletin, February 1975. have become the dominant themes in public policy discussion about medical care. This paper, then, focuses on four major problems that are central to the discussion of the perspectives in the American medical economy. They are: (1) The chronic expenditure increase in the industry and the sources of these increases. (2) The explosive increase in federal subsidization of medical care. (3) The disparity between the sophistication of medical care and the outcome in terms of health. (4) Public policy alternatives in providing equal access to all population groups and eliminating financial hardships. The Chronic Inflation of Expenditures The chronic inflation of medical expenditures is of acute concern to federal policymakers who are stewards of large federal subsidies to the medical economy. Not only have there been large increases in demand that have strained the existing supply, but it has also become increasingly clear that suppliers generate demand. As early as 1959, Shain and Roemer found that “a hospital bed built is a hospital bed filled” (4). Subsidization of hospital facilities through the Hill-Burton program (5) and, more recently, the encouragement of increases in the supply of physicians and other medical professionals have resulted in increased use of these resources, thereby contributing to the rise in expenditures. An examination of the problem of chronic expenditure inflation reveals that it is a long-term phenomenon. The annual rate of increase in medical care expenditures since 1929 has been 7.8 percent (Table 3). During the Great Depression years of the 1930’s the rate of increase was less than 1 percent a year. The 1940’s, with World War II, saw an accelerated annual rate increase of 12 percent. During the 1950’s and the first half of the 1960’s, inflation was reduced to 8-8.5 percent a year. With the introduction of Medicare and Medicaid, medical-care use and prices increased significantly, resulting in an all-time increase of 12.9 percent a year for the 5-year period 1966-71. The deceleration to 10.5 percent in the annual rate of increase since 1971 in aggregate health spending reflected, in part, the slowdown in the rate of medical care price inflation resulting from the Economic Stabilization Program (ESP). Instituted on August 15, 1971, the ESP was in effect for the medical economy through April 1974. Despite the continued high rate of inflation in the medical—care market, health expenditures as a percentage of GNP have been relatively stable since 1971—at about 7.7 percent. It appears that during periods of inflation in the general economy, medical-care spending tends to keep pace with the inflationary growth of the nation. With high unemployment and the reduced rate of growth in the GNP, however, it is likely that health spending as a percentage of GNP will advance rapidly once again. 27 8Z Table 3 NATIONAL HEALTH EXPENDITURES, BY TYPE OF EXPENDITURE, SELECTED FISCAL YEARS, AND AVERAGE ANNUAL INCREASES FOR SELECTED INTERVALS, 1929-74 Personal Health Care Fiscal year or H 't'l Nonpersonal Research and interval Total ospx d care Physicians’ Other services"? construction Total . personal Total Community* serv1ces services** Aggregate amount (in millions) 1929 ......... $3,589 $3,] 65 $651 $380 $994 $1,520 $217 $207 1940 . . . .‘ ..... 3,863 3,413 969 551 946 1,498 316 134 1950 ......... 12,028 10,400 3,698 2,234 2,689 4,013 781 847 1960 ......... 25,856 22,729 8,499 5,706 5,580 8,650 1,433 1,694 1966 ......... 42,109 36,216 14,245 10,499 8,865 13,106 2,445 3,448 1970 ......... 69,202 60,113 25,879 19,693 13,443 20,791 3,952 5,137 1971 ......... 77,162 67,228 29,133 22,598 15,098 22,997 4,534 5,400 1974 ......... 104,239 90,281 40,900 30,593 19,000 30,381 6,902 7,056 Average annual increase (percent) 1929-74 ....... ‘ 7.8 7.7 9.6 10.0 6.8 6.9 9.7 8.2 1929-40 ..... 0.7 0.7 3.7 3.4 -0.4 -0.1 3.5 -3.8 1940—50 ..... 12.0 11.8 14.3 15.0 11.0 10.4 9.5 20.2 1950-60 ..... 8.0 8.1 8.7 9.8 7.6 8.0 6.3 7.2 63 Table 3—Continued NATIONAL HEALTH EXPENDITURES, BY TYPE OF EXPENDITURE, SELECTED FISCAL YEARS, AND AVERAGE ANNUAL INCREASES FOR SELECTED INTERVALS, 1929-74 Personal Health Care Fiscal year or Hospital care _ . , Other Nonpersonal Research and interval Total Total Fifi/21:25 personal servicesT construction Total Community* r c services** Average annual increase (percent)—Continued 1960-66 ..... 8.5 8.1 9.0 10.7 8.0 7.2 9.3 12.6 1966-74 ..... 12.0 12.1 14.1 12.9 10.0 11.1 13.9 9.4 1966-71... 12.9 13.2 15.4 16.6 11.2 " 11.9 13.1 9.4 1971-74. . '. 10.5 J 10.3 12.0 10.6 8.0 9.7 15.0 9.3 *Nonfederal short term general and other special hospitals. “Includes dentists’ services, other professional services, dru services. 'l'Includes expenses for prepayment and administration, government public health activities, raising activities. gs and drug sundries, eyeglasses and appliances, nursing home care and other health and expense of private voluntary agencies for fund- Source: Compendium of National Health Expenditures Data. Office of R esearch and Statistics, Social Security Administration (forthcoming 1975) and unpublished data from the Office of Research and Statistics. Sources of Increase in Medical-Care Expenditures What are the factors behind the chronic increase in medical-care spending? Three broad factors can be identified: Price inflation, product change, and the increase in utilization or the quantity of care demanded and supplied. This simple list, however, does not capture the enormous changes that have taken place in the organization and delivery of medical care, about which leading medical economists have become concerned. Additional medical, demographic, and economic changes also affecting the increased demand for medical-care services and the growth in spending include rising per capita incomes, higher education levels, the rising proportion of elderly in the population, a shift from care of acute illnesses to more expensive long-term illnesses, growth of private health insurance and prepayment plans, increasing public support of medical care for the aged and the poor, and finally the growing awareness of the benefits of medical care. Theories of Hospital Inflation Hospital care is the largest expenditure item, accounting for two-fifths of total spending. This category of expenditure has been the fastest growing component, increasing at an average annual rate of 10 percent since 1929 (Table 3). Davis ( 6 ) has summarized the various theories of inflation: (1) Demand-pull inflation resulting from increasing insurance coverage wherein the hospital can greatly increase the cost of hospital care without increasing the direct financial burden on its patients. (2) Labor cost-push inflation resulting from a “catching-up” of hospital wages as well as increasing unionization of hospital employees. (3) Wasteful capital expenditures including the unwarranted expansion in hospital-bed capacity as well as the wasteful duplication of specialized facilities among hospitals in the same area. (4) Cost reimbursement and hospital inefficiency under which hospitals have little incentive to economize and keep costs down. (5) Changing medical technology, much of which is very expensive including intensive care units, radiation-therapy units, renal-dialysis facilities and burn therapy units. (6) Expanding role of community hospitals in providing many new services. These theories are not all mutually exclusive. On the basis of an empirical study of the impact of Medicare on the financial status of hospitals, Davis concludes that rising hospital costs following the introduction of Medicare and Medicaid are consistent with the demand-pull hypothesis of inflation, although the other expanations have also contributed to inflation in the hospital industry. ( 7) Feldstein’s earlier study of the “Rising Cost of Hospital Care” also yielded similar results relevant to the demand-pull hypothesis (8). u 30 Components of Hospital Costs Increases in the average expense per patient-day for short-term community hospitals result from two major facrors: (1) Increases in wage rates and in prices paid by hospitals for nonlabor inputs. This cost rise represents the additional cost necessary to maintain the same level of hospital services. (2) Additional expenditures for improvement in services, including the cost for additional employees and for other expenses, such as additional equipment and supplies. These arise in large part from new medical technology, procedures, and techniques. Table 4 shows the factors contributing to hospital cost increases for selected periods, 1950-73. The data show that, except for the 1965-67 period, each factor has accounted for about half of the total rise in expense per patient-day over the 23-year period. During the 1965-67 span, improvements in services accounted for more than three-fifths of the increase; apparently, hospital administrators anticipated greater and more intensive use of hospital services Table 4 FACTORS CONTRIBUTING TO INCREASES IN EXPENSES PER PATIENT- DAY, COMMUNITY HOSPITALS, SELECTED PERIODS, 1951-73 Average annual percentage increase Item 1951-601960-65 1965-67 1967-691969-71 1971-73 Total increase . . 7.5 6.7 10.3 13.8 14.8 11.5 Increase in wages and prices ........... 3.8 3.5 41 8.0 8.2 5 9 Wages .......... 5.2 4.7 4 7 9.9 10.0 6 6 Prices .......... 1.5 1.3 2 9 4.8 5.1 4 9 Changes in services. . . 3.7 3.2 6.2 5.8 6.6 5 6 Labor ......... 3.1 1.7 3.8 2.8 3.7 2 3 Other .......... 4 6 5.6 9.6 9.8 10.3 10 0 Percent of total increase due to: Wages and prices . . 50.0 51.5 39.7 58.2 55.3 51.3 Change in services. . 50.0 48.5 60.3 41.8 44.7 48.7 Source: US. Department of Health, Education, and Welfare, Social Security Adminis- tration, Office of Research and Statistics, “Background Information on Medical Expenditures, Prices, and Costs,” September 1974 (Preliminary), p. 34. 31 under the newly-created Medicare and Medicaid programs and subsequently expanded labor and nonlabor inputs to meet the increased deman . Increases in wages and prices in.the 1971-73 period were considerably lower than those reported during the previous 2-year period (1 ). Mandatory wage and price controls for all sectors of the economy began on November 1971 and continued through January 1973, while wage and price controls remained in effect for the health-care industry until April 1974(9). Technology and Hospital Costs Hospital costs have grown in response to increased demand as well as improvements in knowledge and medical technology."I'hese improvements and extensions of medical treatment processes have been underwritten by the tremendous increases in the supply of insurance and third-party payment mechanisms. To an important degree, it is the availability of financing, rather than medical efficacy that governs the adoption rate of high-cost technology. In recent years, the tenor of technological advance in health care has changed from one which saw considerable advances through the use of antibiotics to the implementation of new techniques that are usually resource intense, necessitating institutionalization of the patient. These technologies and treatment processes generally have only a small impact on the general health of the population and in some cases on morbidity and mortality related to specific diseases. Examples of this trend would include chemotherapy, radiation therapy in cancer treatment, renal dialysis, open-heart surgery, organ transplant, intensive care units for heart attack, burns, and traumatic shock. Even today, many years after introduction of these new modes of treatment, questions have been raised concerning their appropriate use and, in some cases, their efficacy. Unlike most pharmaceuticals, which are not permitted on the market until after they have been extensively tested, new medical procedures remain on the market until they are found unsound or possibly irrelevant. Since these technologies are often required by an institutionalized patient, the hospital became the focal point of amassing the needed machines and labor. Many treatments that could be done at least part of the time on an ambulatory basis (for example, radiation therapy, renal dialysis) are often concentrated in the inpatient hospital setting. There have been many reasons offered for the relatively rapid diffusion of these costly new technologies. In a regime of full-cost reimbursement, there is little incentive on the part of the patient, hospital administrator, or the physician to create a situation favorable to the adoption of resource-saving technology. Instead, perverse incentives towards the adoption of resource-using technology are present. As previously noted, Feldstein has argued that rising incomes and more intensive health insurance coverage have increased the demand for hospital care { 8, 10).The existence of extensive hospital insurance has resulted in a major change in the character, quality, and style of hospital care as evidenced by major increases in the use of real inputs per patient-day. Supportive of the same phenomenon is a tendency of administrators and hospital staff to maximize their prestige by commandeering certain key inputs to their hospitals (11). It has also been noted that the modern hospital under 32 extensive hospital-based insurance is essentially a physician’s rent~free work- shop, and that the physician staff will press administrators and trustees to add those inputs that enhance staff income and prestige {12). Technological acquisitiveness is reinforced by a strong technological imperative instilled in physicians in their medical training programs, and is tacitly encouraged by the present cost-based financing system. ~ There may be, in fact, a feedback mechanism at work that is leading the health-care system down an explosive growth path. The new high-cost hospital-based technology necessitates specialization, and fosters a narrow professionalism among new physicians. For reasons relating to income, prestige, and the way that modern medicine is practiced, new physicians are drawn away from primary care office-based settings and toward the practice of specialized medicine within an urban institutionalized setting. This trend toward medical specialization, in turn, proliferates physicians’ demands to induce hospitals to adopt still more technology. Technology has its primary impact through direct and indirect operating costs and spillovers that cause alterations in other hospital functions and the increased intensity in the use of health resources. Berry (13) suggests that hospitals grow along a logical path; hospitals do not have surgical suites without blood banks and recovery rooms. Thus, the installation of a particular piece of equipment may involve additional supportive services or alterations in the way previously-existing services are used. With regard to the direct operating costs of capital equipment, Bloom and Peterson (14) estimate that in 1970 the cost of building and equipping all coronary care units in Vermont, New Hampshire, Massachusetts, and Rhode Island was $3.7 million, while the annual cost of operating these units was $6.4 million. Finally, Redisch (15) has shown that a large part of the rise in resource use in hospitals from 1967-1971 was related to more intensive use of a basic set of ancillary services (for example, laboratory tests, radiological procedures, etcetera) rather than to increased utilization of more esoteric technology. Thus, the effects of technology—initial cost, its continuing operating cost, indirect spillover effects on the organization and operation of the hospital, and the cost arising from its availability and intensive use—are reinforcing effects to increase the cost of hospital and medical care. An additional effect of technology on individual hospitals is to attract more complex cases which require more resources. In terms of all hospitals, the increased technology has the tendency of making many cases that were untreatable previously become subjects of complex treatments. Scitovsky (16) suggests that this is true, even with cases that are relatively straightforward. For example, it is relatively unheard-of to deliver a baby outside a hospital or to have a broken arm set in a physician’s office. Some general statistical support is given for the pervasive effect of technology on hospital costs. Worthington (17) finds a strong simple relation- ship between the change in nonlabor input in a hospital and the change in the per capita expenditure for hospital care (I = 0.88). Approximately 77 percent of the variation in the change in annual per-capita expenditure for hospital care is explained by the change in real nonlabor inputs. Although “real nonlabor 33 input” measures such diverse items as food and bedding and surgical suites, it also measures the intensity with which real nonlabor inputs are used. Thus, the change in nonlabor inputs is pervasive, with over half of its effects being indirect effects on other determinants of change in hospital per-capita spending. Technology and Physicians’ Services Physicians in their office practice have not been immune to the same trends that have been affecting hospital care. Physicians are traditionally the central providers of medical care. They not only provide care directly to patients, but they also determine a large part of other services and supplies required by patients throughout the course of treatment. Expenditures for hospital care and outpatient drugs, for example; are primarily made at the direction of a physician. Physicians in recent years have become increasingly specialized. In 1949, 54 percent of all physicians were in general practice. By 1973, this proportion had dropped to 15 percent with the remaining 85 percent made up of medical, surgical, and other specialists. During 1963-73, there was a 55 percent increase in medical specialists, and a 31 percent increase in surgical specialists. Although much primary care formerly performed by GP’s is now performed by internists, pediatricians, obstetricians, and gynecologists, there is little doubt that this increased specialization has had a sizable impact on the nature and quality of physicians’ services, as well as the amount that is spent for them. Fuchs and Kramer (18) find that supply factors—including technology and number of physicians—appear to be of decisive importance in determining the utilization of, and subsequent expenditures for, physicians’ services. This conclusion contradicts those reached by numerous other investigators that trends in price, income, and insurance coverage adequately explain the variation in demand for physicians’ services. Fuchs and Kramer’s study further underlines the importance of the physician in the medical economy. Not only does the physician influence the demand for other medical services such as hospital care and drugs, but he also determines to a considerable extent the demand for his own services. Klarman et al.(19)"'indicate that although price increases account for a substantive portion of the increase in expenditures in recent years, the per capita output of physicians has also increased. Worthington (17) finds, in an analysis of the annual change in per-capita expenditure for physicians’ services, that the growth in real inputs employed by physicians over time may be attributed to increasing specialization. Worthington’s analysis suggests that physicians are substituting inputs, such as paramedical personnel and equipment, for their own labor so they are spending relatively less of their own time per patient. As the price of their own labor is relatively more expensive, the real price per visit declines over time. This interpretation of more care being delivered in the physicians’ offices through paramedical personnel, specialized services or equipment is compatible with the observation that year-to-year movement in real price per visit is negatively 34 correlated with the volume of the patient visits. This is intuitively reasonable, assuming that individual physicians attempt to hold their working hours relatively constant. Worthington also finds that real inputs in physician visits are as important a contributor to changes in the annual expenditures per capita for physician care as are the changes in labor and nonlabor prices paid by the physician. F edéral Subsidization of Medical Care The improvements in technology clearly have contributed significantly to the chronic expenditure inflation in the medical economy. The role of federal subsidization of medical care, however, cannot be overlooked. Federal monies are now an important undergirding of the medical economy. The federal initiative of the 1960’s strongly redistributed the sources of payment for care, enabling two large population groups—the agedland the poor—to obtain needed medical-care services. The last decade saw the initiation of Medicaid and Medicare, community mental-health centers, regional medical programs, comprehensive health- planning agencies, the National Center for Health Services Research and Development, the first direct support of medical students, and the training of other medical professionals. During the same period, some programs were eliminated, like regional medical programs, or transferred, like 0E0 health programs, or reduced in priority, such as NIH-supported biomedical research. Eligibility and services under the Medicaid program were cut back in some states, and additional controls were placed on Medicare in an attempt to control expenditures. From August 1971 to April 1974, an attempt was made at direct control of medical-care prices-under the Economic Stabilization Program. The shift toward heavy federal involvement, however, seems permanent. Since Fiscal Year (FY) 1965, federal medical expenditures for health services and supplies increased from $3.1 billion to $25.3 billion in FY 1974, an annual rate of increase of 26 percent (Figure 2, line C). State and local spending increased to $12 billion from $4.6 billion in the same period (line A) at an annual rate of increase of 11 percent. Assuming the continuation of the trends from 1950-66, and without the influence of federal legislation, it is estimated that state and local medical expenditures would have been approximately $6 billion in Fiscal Year 1974, or about one-half their current level. Moreover, federal expenditures, exclusive of research and construction, would have been approximately $3.5 billion. Thus, without the initiatives in the last decade, public expenditures in FY 1974 would have been approxi- mately $9.5 billion, rather than the FY 1974 level of $37.4 billion. The impact of federal initiatives clearly has been explosive. The real increase in public resources in FY 1974 alone, exclusive of research and construction, is estimated roughly at $17.0 billion. Thus, about 61 percent of the additional $27.9 billion in public expenditures in FY 1974 beyond those expected without federal initiatives was an increase in real resources. With regard to individual medical care components, the emphasis was away from construction and research and toward subsidies for personal 35 9E ‘VL61-0961 salddns pun sooyues qnnaq 10} samupuadxa [n30] pun 9121s Tampeg ': am 31:1 Billions of Dollars 25 20-— (A) -.._._- State and Local (current dollars) / (B) —— State and Local (constant 1967 dollars) (C) — — Federal (current dollars) / (D) ——-— Federal (constant 1967 dollars) 1954 1958 1962 1966 1970 1974 FISCAL YEARS care (20). Construction expenditures in FY 1974 were 4 percent of total public medical expenditures; they were 7 percent in FY 1965. Research outlays were down to 6 percent of public expenditures in FY 1974; they were 13 percent in FY 1965. Not only did research outlays lose their relative place, but in the decade 1965 to 1974 they grew only 8 percent annually compared with 24 percent in the preceding decade. Figure 3 shows the private expenditures for health services and supplies since 1950 in current and constant (1967) dollars. In calendar year 1973, private medical care outlays amounted to $57.5 billion in current dollars (line A) and $41.8 billion in 1967 dollars (line B). Trend line B’ estimates what real private expenditures would have been without Medicare and Medicaid—$44.9 billion, or a difference of $3.1 billion. Figures 2 and 3 imply a simultaneous relationship after 1966 between inflationary aspects of the private market and the surge of real resources added by the public sector. That is, the increase in public resources was a catalyst for the medical market’s expansionary characteristics. The explosion of federal monies is associated with a strong redistribution of medical-care expenditures for the poor and the aged. These trends can be examined in terms of hospital and physician visits. Wilson and White (21) have presented data from the Health Interview Survey showing changes between 1964 and 1973 on morbidity, disability, and utilization differentials between the poor and nonpoor. Summary data are shown in Table 5. The data distinguish between the nonpoor and the poor who are defined as those with family incomes of the lowest quintile in 1964 and 1973—under $3,000 in 1964 and under $6,000 in 1973. Also, the data distinguish the white and nonwhite. (Detailed data for four age categories are presented in the original paper.) Some of the improvements illustrated by the data are continuations of long-term trends, and cannot be ascribed to the intervention of the federal government. Nevertheless, there are a number of indications of increased availability and accessibility of services in the period 1964-1973. It was, of course, the purpose of government programs to increase the utilization and availability of medical care to the poor and aged. Hospital utilization increased substantially for the poor, especially the nonwhite. The detailed data show that the elderly poor white appear to have been assisted considerably by Medicaid and Medicare; the nonwhite elderly poor have been assisted to a smaller degree. For example, discharges from short-stay hospitals per 100 persons per year increased 42 percent for the poor elderly white between 1964 and 1973 but only 4 percent for the poor elderly nonwhite in the same period. There has also been a marked reduction in the proportion of the population who have not seen a doctor in the previous 2 years, especially among the poor. This pattern held for all age groups, indicating a narrowing of the gap between the poor and the nonpoor in terms of financial accessibility of medical care. Another indicator of medical care is the number of physician visits per person per year. Since 1964, there has been an increase in the average number of physician visits per person during the year. Again substantial gains are shown for the poor relative to the nonpoor. The improvements in the relative rates of hospital discharge and physician visits has not been reflected in the relative rates of dental visits. The importance of this point is that there was only a 37 60 A 55 — 50 ~— A Current Dollars B --- Constant 1967 Dollars B1 45 _ B1 ----- Projected Trend (in 1967 Dollars) I Billions of Dollars fillllllllll]? 50 52 54 56 58 60 62 64 66 68 70 72 Calendar Years Figure 3. Private expenditures for health services and supplies, 1950-1973. 38 minor increase in the amount of federal and state resources that went into supplementary private dental-care expenditures. Although substantial gains have been made in meeting the medical care needs of the aged and the poor, Davis (3) has shown that there are substantial gaps. Federal Subsidization of Medical Education In an effort to increase medical manpower and in response to the increased demand for medical care resulting from the increased federal subsidization, the federal government undertook a second front of intervention in the area of medical education. The initial interest in medical education by the executive Table 5 SELECTED UTILIZATION RATES BY POOR AND NONPOOR STATUS AND COLOR, UNITED STATES, 1964 AND 1973 Y Total White All other ear Poor Nonpoor Poor Nonpoor Poor Nonpoor Short-stay hospital discharges per 100 persons per year 1964 ...... 13.8 12.6 15.3 12.9 9.9 9.6 1973 ...... 19.0 12.5 20.2 12.6 15.3 11.6 Percent of population with no doctor visits in past two years 1964 ...... 27.7 17.7 25.7 17.1 33.2 24.7 1973 ...... 17.2 13.4 16.8 13.2 18.5 15.3 Number of doctor visits per person per year 1964 ...... 4.3 4.6 4.7 4.7 3.1 3.6 1973 ...... 5.6 4.9 5 7 5.0 5.0 4.3 Number of dental visits per person per year 1964 ...... 0.8 1.8 0.9 1.8 0.6 1.2 1973 ...... 1.1 1.8 1.2 1.9 0.7 1.1 Note: Definition of poor and nonpoor is based on family income: 1964.... 1973.... Under $3,000 Under $6,000 Poor Nonpoor $3,000 and over $6,000 and over Source: Wilson, Ronald W. and White, Elijah L., “Changes in Morbidity, Disability, and Utilization Differentials Between the Poor and the Nonpoor; Data from the Health Interview Survey: 1964 and 1973.” Presented at the 102nd Annual Meeting of the American Public Health Association, October 21, 1974. 39 and congressional leaders was the result of studies indicating not only shortages in the number of physicians but also financial problems in the professional schools. During the 1960’s, the federal government also heavily subsidized medical schools through support of medical research. Examination of the source of funds for medical schools in FY 1966 and FY 1973 in Table 6 shows that federally-supported research is playing a diminished role. In FY 1966 it accounted for 35 percent of medical school funds, compared with 22 percent in FY 1973. The largest increase in funding during this period came from federally and nonfederally supported multipurpose funds. These funds, which were very minor in FY 1966, now account for almost 15 percent of the total. (It is noted that the National Health Expenditure Series developed by the Social Security Administration does not adequately account for outlays for medical training and education (22).) The funds supplied to medical professional schools have led to a large increase in the number of medical students. The number of students who entered the 113 medical schools in 1972 exceeded 11,600, a 23-percent increase in the 5-year period from 1968-1972, compared to a 7-percent increase over a 5-year period to 1965. The increase in funds for educational Table 6 SOURCES OF FUNDS FOR MEDICAL SCHOOLS, 1965-66 AND 1972-73 1965-1966 1972-1973 Source of Funds (in millions) % (in millions) % TOTAL .......... $882 100.0 $2,111 100.0 Tuition and fees ........ 41 4.6 91 4.3 Sponsored research Federal ............. 307 34.7 473 22.4 Non-Federal ......... 68 7.7 115 5.4 Endowment (unrestricted). . 26 2.9 47 2 2 State and local University transfers ............ 142 16.0 369 17.5 Federal-sponsored multipur— pose and service ........ - — 106 5.0 Other nonfe deral-sponsored multipurpose and service . . 17 1.9 203 9.6 Training grants plus other gifts and grants ........ 136 15.4 300 14.2 Medical service funds ..... 70 7.9 201 9.5 Overhead ............. 59 6.7 147 7.0 Other ................ 16 1.8 59 2.8 Adapted from: Journal of the American Medical Association, Nov. 25, 1968, Vol. 206, p. 2019. Ibid, Jan. 1975, Vol. 231, (Suppl.), 13. 28. programs have provided medical professional schools with greater financial flexibility and stability. \ In addition to the increase in medical students in American medical schools, the pool of foreign-trained physicians (including many United States citizens) has doubled since the early 1960’s to approximately 60,000, or 18.5 percent of all active physicians. The NIH estimate that the current annual net influx of 5,200 foreign medical graduates is roughly equivalent of the yearly output of 34 US. medical schools ( 23 ). Without major alterations, in either present output from domestic schools or immigration patterns, the supply _of physicians in the United States is projected to increase steadily by 1980 to approximately 408,000, an increase of 27 percent during the 1971-1980 period. This projection indicates that the nation will have an aggregate physician-population ratio in 1980 of 180 per 100,000 population. This is the current ratio for the mid-Atlantic region, the highest ratio in the nation ( 23 ). Successful federal efforts to increase the overall physician supply have not significantly altered the imbalance in the geographic and medical-specialty distribution of physicians. The projected increase in supply may bring about little change in this regard. Both US. and foreign~trained physicians have tended to locate in populous, urban settings, and to locate disproportionately in mid-Atlantic and East North Central States. The Impact of Medical Care on Health An assumption underlying the federal subsidization of medical care is that there is a positive probability, if not certainty, that provision of more care will favorably impact on health. The ultimate question in examining the inflation of medicalcare expenditures is what has been the impact of increased amounts of medical care on the nation’s health. If it can be demonstrated that the effect has been salutary, then inflation will be viewed differently than if no discernible effect can be measured. A number of investigators, including Abel-Smith, (24) Deputy, (25) Forbes,{26) Fuchs, (27, 28) and Ginzberg, (29) have commented on the efficacy of medical care to favorably influence health, each concluding there is often lack of a positive correlation between the two. This is not to imply we can do without medical care, for a cessation of all medical care would surely be followed by a worsening of health status. A meaningful question, however, is whether increments or decrements in the amount of medical care result in improvement or deterioration in the level of health. Fuchs concludes that variations in mortality across and within countries do not seem to be related to differences in the availability of physician or other medical-care inputs. He argues that medicine has demonstrated effectiveness in such areas as treatment of trauma and skin cancer and those areas where there have been scientific advances to control disease through immunization or drug therapy, such as diphtheria, typhoid fever, pneumonia, tetanus and poliomy- elitis. To support his claim that a marginal change in the amount of medical care will not change health status, Fuchs relies heavily on mortality statistics. He 41 cites large differences in infant mortality among socioeconomic classes in England and Scotland, although free national health services are available to all, and points out that the Netherlands, where a large proportion of all births occur at home with a midwife rather than in a hospital with attending physicians and elaborate facilities, has one of the lowest infant mortality rates in the world. Although prenatal care appears to reduce the risk of infant mortality, other nonmedical factors such as income, schooling, race and the mother’s physical condition are also important. Between 1950 and 1965, infant mortality in the United States decreased slowly, about 1 percent a year on the average. However, in the second half of the ’sixties and into the early part of this decade of the ’seventies, the rate declined more than 4 percent annually. Fuchs proposes as possible explana- tions improved contraception and more liberal abortion laws with a resulting decrease in unwanted births. There was also greater availability of maternal and infant-care services and intensive-care units for premature babies. However, the relative contributions of more medical care and the other factors are unknown. Fuchs also proposes that economic and social factors are important determinants of adult mortality, which is therefore unresponsive, to a certain extent, to medical care. Accidents, suicide and homicide account for 3 of every 4 male deaths from ages 15 through 24. The rate for violent deaths was 40 percent less 20 years ago, although treatment of trauma improved during that period. Heart diseases and neoplasms are important causes of death for males ages 35 through 64. These diseases are apparently influenced by nonmedical factors such as diet, heredity, smoking, lack of exercise and stress. Fuchs believes “the greatest potential for reducing coronary disease, cancer, and other major killers still lies in altering personal behavior.” For further support, he provides evidence that as female and male life-styles become more similar, so does mortality experience. Between 1949 and 1959 the ratio of male to female deaths from unexpected heart attacks was 12 to l,while for 1967-1971 it was 4 to 1, with 90 percent of the females dying in the latter period being smokers, a majority of them heavy smokers. Dupuy ( 25 ), examining mortality statistics in France, found life expectancy for women increasing by only 1 month a year since 1960 and not at all for men since 1965. And for men ages 15 to 24, the death rate has been increasing 2 percent a year. Health expenditures, on the other hand, have been increasing faster than expenditures for most other consumption in France. Forbes, (26) studying longevity and medical costs through the mid-1960’s concluded, “In the United States there is no longer any significant relation between the money spent on health and the results achieved.” Many factors besides medical care influence the health of a population, including heredity, environment, and behavior. Accidents, especially on the highway, are the leading cause of death among young persons. Heart disease, related apparently to diet, smoking, lack of exercise and stress, is the most important cause of death among adults. Deaths in New York City from respiratory and heart diseases have been found to vary significantly with daily levels of air pollution ( 30). At present, it appears that changes in life-style may do more to prevent these deaths than changes in the quantity of medical care. Detailed statistical analyses of mortality have been performed by Auster, 42 Leveson and Sarachek (31), Brenner (32, 33 ), and Fuchs (34 ). Applying regression analysis to state data for 1960, Auster et al. found environmental conditions, especially income and education, to be more important deter- minants of interstate variation in death rates than the quantity of medical services. Brenner, studying heart disease mortality in the United States between 1900 and 1967, observed increased mortality associated with economic downturns, measured by unemployment, and decreases in heart disease deaths during economic upturns. Brenner suggests various stresses accompanying economic recession might cause increases in heart disease mortality. Brenner also reports significant changes in perinatal, neonatal, and postneonatal mortality associated with economic fluctuations. In a series of randomized controlled trials, Cochrane (35) finds that several therapies that are commonly used today are ineffective or ambiguous. He suggests that tonsillectomies are effective in no more than 20 percent of the cases, result in a small but definite mortality, and should be limited to cases involving obstruction. He also argues that the death rate from carcinoma of the cervix was falling before the use of cervical smears became popular and has continued to fall at about the same rate. Further, it has not been shown that the decrease has been greater in areas of relatively high screening of the female population. The value of anticoagulant therapy for ischemic heart disease is dobutful, except perhaps for men with prolonged angina or previous infarction; and it is not clear that coronary care units save lives. In one randomized controlled trial, Cochrane failed to find any gain for cases admitted to hospitals with coronary care units compared to those treated at home. He also concludes that insulin therapy for mature diabetics provides no advantage over diet, and that the oral drugs tolbutamide and phenformin may be ineffective or even dangerous. Other studies belie the .value of iron for anemia at certain hemoglobin levels in nonpregnant women and ergotamine tartrate for newly-diagnosed cases of migraine. In short, there is a considerable lack of information about the existence of the cause and effect relationship between what is done to the patient and the outcome of the medical-care process as measured by mortality, morbidity, disability, and capacity to carry on their activities ( 36 ). Attempting to Control Chronic Inflation of Expenditures Health program managers and federal policymakers have become increas- ingly concerned about finding ways to curb the chronic inflation of medical expenditures. Numerous kinds of controls have evolved, many of them in the last decade (37, 38). Among these controls are supply limitations (certificate of need for capital construction), financial disincentives to the patient (insurance deductibles and coinsurance rates), authorization requirements (preadmittance screening, surgery authorization), review mechanisms (utilization and claims processing review), legal action (malpractice suits) and rate regulation (state hospital ratesetting commissions). These mechanisms differ with regard to who is being regulated, who is doing the regulation, the degree of medical judgment required, and whether the control occurs before, after, or during medical treatment. 43 The controls have grown more numerous and have been applied more intensely during the last decade, though many of them originated before 1965. The Medicare and Medicaid programs represented the first serious attempt by the government to include controls beyond those associated with closed-end appropriations. These included deductibles and coinsurance (Medicare), limita- tions of benefits, establishment of utilization review committees and require- ments for certification and recertification of inpatient hospital and nursing- home stays. During the 1950’s and 1960’s excess utilization through hospital over- building became so serious that some 36 states have Certificate of Need laws which require approval by the state before new construction of facilities may be initiated by the hospital. It is interesting to note that during this period the disincentives associated with coinsurance and deductibles declined in use. The out-of-pocket price (adjusted for inflation) for hospital care and physicians’ services has declined over the last decade. It appears that it is in response to this decline that the other administrative control methods have grown. For example, the use of prospective reimbursement of hospital costs increased sharply as states were faced with large rate increases by insurers. The success of these controls in the public and private sectors is spotty. In 1970 Medi-Cal required prior authorization from a state-employed physician for all non-emergency hospital admissions. Currently, this authorization has been extended to include the expected length of stay; any extensions of the stay are subject to recertification. Stuart and Stockton (39) showed that although the number of hospital admissions rose after an initial decline, the rate of increase was lower than expected. Despite a 23-percent increase in eligibility, the Medi-Cal program paid for only 3.5 million patient-days in 1970, as opposed to 3.6 million in 1969. The United Mine Workers Union (UMW) provides comprehensive medical care to about 500,000 coal miners and their dependents. To control utilization, the Union requires that each covered family select a plan-approved physician to be its primary provider. This doctor becomes a managing physician through whom all treatment is channeled. Care provided by a nonaffiliated physician or specialist customarily requires written approval of the managing physician. Since its initiation in the mid-1950’s, this program, along with other innovations, has succeeded in reducing utilization in certain areas by up to 25 percent ( 39). Utilization review has been found to be most effective where there is a scarce supply of beds and excess demand. Under these conditions, utilization review committees may have adequate motivation to control 'the use of hospital beds. In an empirical study of the effect of scarcity as a motivation device for utilization review committees, Bonner, Decker, and Kasten (in Stuart and Stockton) (39) found that utilization review was more successful in those institutions with high occupancy, low rate of turnover, long initial length of stay than where facilities were being under-utilized. Finally, prepaid health center care or health maintenance organizations have given ample evidence that they can be quite effective in reducing hospital admission and inpatient stays (40). It is not clear, however, that this result derives from any incentive toward greater emphasis on preventive treatment, 44 nor is there any reason to believe that an HMO can be expected to produce better health outcomes than more traditional modes of delivery {41). Although some success in the control of utilization of medical care through administrative means has been exhibited, there are numerous reasons for the current lack of success in other attempts. Successful results appear difficult to replicate in other places or, at best, difficult to measure. Many times the effects that have been measured have shown negative or unintended effects. In part, these disappointing results occur because medical care, for all its technology and science, is still at the margin an application of judgment applied to the enigmatic healing process. In addition to difficulties of judgment, the incentive to reduce total utilization or expense per day is not strong. The incentives for'self-regulation are weak; it is difficult for rotating parttirne review groups of physicians to criticize their peers. Insurance companies have little incentive to reduce over-utilization. Some forms of regulation have adverse incentives. Hellinger (42) found the Certificate of Need legislation prompted hospitals to initiate capital projects in the months before Certificate of Need agencies became effective. Hellinger (43) also found that in several hospitals, prospective reimbursement schemes, based on a per-diem rate, tended to result in longer lengths of stay. Ginsburg (44) points out several disincentives in the Economic Stabilization Program (ESP). He writes that cost-justification requirements under ESP eliminated all incentives from hospitals not experiencing difficulty in staying under the 6-percent increase in price limit. Since base period data for calculating price increases were each institution’s own prices, many hospitals were not limited by the controls. If they feared that their current performance might become a base for the future, incentives to increase costs may have existed. Thus, one set of reasons for lack of success was the inappropriateness of many features of the regulations themselves. The burgeoning problem of malpractice suits, allegedly a deterrent to unnecessary or medically-dangerous procedures, also may have led to “de- fensive” medicine characterized by medically unnecessary but potentially legally valuable tests and X-rays. In those cases where there are penalties for noncompliance, as under Certificate of Need legislation, they may be ineffectual. Under current law, Medicare and Medicaid cannot reimburse capital construction built without permission of the state’s planning agency; this is a far weaker penalty than not reimbursing for all the care of Medicare and Medicaid patients in institutions using unauthorized capital. Financing control on the consumer may result in adverse incentives if high-cost insured services are substituted for low-cost uninsured ones—e.g., substituting insured hospital care for uninsured ambulatory care. A dilemma is presented, however, since comprehensive coverage of services is many times associated with over-utilization. Financial controls, such as deductibles and coinsurance, have the deficiency of controlling primarily through price, which is simply one variable determining utilization. Also, it is unknown whether the care not undertaken was medically unnecessary or simply unaffordable. The federal tax laws encourage comprehensive first-dollar coverage, and are a 45 disincentive to strong application of deductibles and coinsurance rates. Finally, price does affect, perhaps only slightly, utilization of care that is generated by consumer demand. Estimated price and income elasticities of care appear to be small and, as a result, the effect of the relatively small deductibles and low coinsurance rates are not large on demand generated by the consumer. Public Policy Alternatives in Financing Medical Care Services There has been increasing discussion in the United States concerning the need for a national health insurance program and various proposals have been under consideration by the Congress I 45). A national health insurance program is expected to help meet various objectives. One is to provide basic protection against health-care costs for the entire population and to eliminate financial hardship. Our present system of voluntary health insurance and government medical care programs has left gaps in coverage, and has provided inadequate coverage for substantial segments of the population. A national health insurance program is also viewed as a vehicle for implementing controls over costs and utilization for the entire medical economy ( 46). Opinions vary widely concerning the most desirable methods of financing and administering a national health-insurance program. These can, with some oversirnplification, be divided into three approaches. One approach is to use a mixed public and private mechanism which would build on our present structure of private health insurance and government programs. The private health-insurance mechanism would be retained for the working population, but employers would be required to make available to their employees a plan providing specified health services. The plan would usually be insured through private insurance, and the employer would pay part of the cost. Private insurance would also be retained for the self-employed, farmers, and others not in an employment group under provisions designed to facilitate coverage for the group. A uniform government program for the nonemployed and the poor would also be established, and this would be financed and administered by the federal government or jointly by the federal and state governments. Also, the Medicare program for the aged would be integrated into the national plan and would continue to be financed mainly by the federal government. While there are many differences in details, a plan of this general type has been proposed by the Administration and by representa- tives of the hospital industry, the medical profession, the commercial insurance industry, and certain business groups. The labor movement in the United States generally supports a federal program, based on the social insurance approach, which would cover the entire population under one plan. This proposed program is financed by social insurance taxes on employers, employees and other income receivers, and by contributions from federal general revenues. The program would be adminis- tered by the federal government. A national health budget would be established each year, and the funds would be allocated by type of service to the various regions of the nation. The government would be given considerable administrative discretion in allocating funds, establishing standards for pro- viders of service, establishing reimbursement policy, and conducting reviews of utilization. 46 A third approach would be to establish a catastrophic health-insurance program. A program of this type would be designed to pay benefits only to persons who incur unusually high health-expenses. For example, one of these proposals would pay hospital benefits after the 60th day of hospital care and medical benefits for medical expenses exceeding the first $2,000 of expenses. This approach is based on the concept that the role of a national health-insurance program should be limited to the financing of extraordinary expenses. All of the major proposals include provisions designed to control costs and utilization. Generally, the mixed public-private proposals incorporate the types of measures, described above, which have been added to the Medicare and other government programs, and apply them to all plans under the proposal. For example, the Administration’s proposal includes the Professional Standards Review Organization and provisions for withholding of reimbursement for capital expenditures not approved by the state planning agency. It also provides an option for all persons under the program to enroll in a health maintenance organization. The proposal does not specify methods of reim- bursement to be used, but indicates these will be developed by the states under federal guidelines. As suggested above, the social insurance proposal provides considerable administrative authority which would permit establishing of cost and utiliza- tion controls. It also includes specific provisions which would give preference to HMO’s in providing benefits. Also the reimbursement of institutional providers would be based on a predetermined annual budget and the reimbursement provisions for physicians are designed to encourage compensa- tion on a salary or capitation basis and discourage practice on a fee-for-service basis. Further, institutional providers of service could be ordered to add or reduce services or provide services at a new location. The proposal also establishes a health resources development fund to finance the growth of HMO’s and otherwise develop facilities and train manpower. Conclusion Federal initiatives during the last decade have resulted in considerable changes in the American medical economy. Large sums of money have been infused into the system, substantially improving access to care for the aged and the poor. At the same time, improvements in knowledge and medical technology have occurred, largely underwritten by the tremendous increases in the third-party payment mechanism. The rate of increase in medical care expenditures continues at a very high level, but there are still gaps in access to care for certain population groups. Many analysts are questioning the efficacy of medical care for improving health status and whether additional dollars should be put into the health-care system. Successful control of medical-care use and costs through administrative controls continues to be elusive. Some analysts view the expenditure inflation that has been witnessed in this country in the past decade as a temporary social perturbation. It is the thrust of this paper,however, that the experience of the 47 last decade is part of a long-term trend that will continue at least as strongly in the future, presenting fundamental problems for policymakers in the next decade. The future role of medical care technology may be to effect massive changes on the entire social fabric of society, not merely the health-care system. We are currently entering into the initial stages of a biological revolution that will have social consequences at least as important as those related to the industrial revolution (47/. Breakthroughs will eventually emerge in research related to genetic surgery, artificial wombs, cloning (making genetic copies of individuals via asexual reproductions), artificial organs, and other far-reaching develop- ments. The rate of innovation diffusion, and genera} acceptance of these procedures-and concepts, will be directly related to our policies for biomedical .research funding and for regulating and financing the diffusion of new health-care technology (48). The policy issues of the last decade, which opened with the concern‘ about equity, ended with new questions being raised about efficiency in the delivery of care, the control of utilization and expenditures, and an increasing number of questions raised about the efficacy of that care. As the decade came to a close, it became increasingly clear that the events of the previous decade were an overture to the next. The technological imperative will increasingly be underwritten by insurance with the possibility that the impact of the biological revolution in the future will affect not only medical-care processes but the entire fabric of our society. The future passage of national health insurance compounds the problem of utilization control which has had uneven success. National health insurance, depending upon which particular version is favored, will increase the reliance upon administrative control while lowering the out-of-pocket price for additional beneficiaries. For the most part, national health insurance has been looked upon as the vehicle for eliminating the inequities remaining from legislation passed in the mid-1960’s. Although these needs surely exist and are not trivial, it may be preferable to look upon the national health insurance debate as the opening of a discussion about how best to organize American medical care to guide the rational use of its resources. The trend in the mid-1960’s was to subsidize the private medical market with legislative prohibitions against interfering with the delivery of care. The experience of the late 1960’s and early 1970’s has shown the deficiencies of this approach. The challenge of the next decade is to begin the creation of unique social mechanisms that will maintain the improvements in equity that have been gained and increase the efficacy of medical care by concentrating on the outcome of treatments rather than the process itself and improving the social efficiency with which care is delivered. 48 10. ll. 12. 13. 14. REFERENCES . U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE (1974), Social Security Administration, ORS, “Background Information of Medical Expenditures, Prices, and Costs,” September (preliminary). AMERICAN HOSPITAL ASSOCIATION (1974), Hospital Statistics 1974 Edition: “1973 Data from the AHA Annual Survey.” DAVIS, KAREN (1974), “National Health Insurance,” In Setting National Priorities: The 1975 Budget, Washington, DC: The Brookings Institu- tion, pp. 207-246. SHAIN, MAX and ROEMER, MILTON I. (1959), “Hospital Costs Relate to the Supply of Beds,” Modern Hospital, Vol. 92, No. 4, pp. 71-73, 168. LAVE, JUDITH R., and LAVE, LESTER B. (1974), The Hospital Construction Act: An Evaluation of the Hill-Burton Program, 1948-1973. Washington, DC: American Enterprise Institute for Public Policy Re— search. . DAVIS, KAREN (1972), “Rising Hospital Costs: Possible Causes and Cures,” Bulletin of the New York Academy ofMedicine, Vol. 48, No. 11. . DAVIS, KAREN (1973), “Hospital Costs and the Medicare Program,” U.S. Social Security Bulletin, Vol. 36, No. 8, August. . FELDSTEIN, MARTIN (1971), The Rising Cost of Hospital Care, Washington, DC: Information Resources Press. . ALTMAN, STUART, and EICHENHOLTZ, JOSEPH (1974) “Control of Hospital Costs Under the Economic Stabilization Program,” Federal Register, Vol. 39, No. 16, pp. 2693-2701. FELDSTEIN, MARTIN S. (1973), “Econometric Approaches to Health Economics,” In M. lntrilligator and D. Kendrick (eds), Hontiers of Quantitative Economics, Vol. 11, Amsterdam: North Holland. LEE, MAW LIN (1971), “A Conspicuous Production Theory of Hospital Behavior,” Southern Economic Journal, Vol. 38, No. 1. PAULY, MARK, and REDISCH, MICHAEL (1973), “The Not-for-Profit Hospital as 3 Physicians’ Cooperative,” American Economic Review, Vol. LXIII, No. 1. BERRY, RALPH (1973), “On Grouping Hospitals for Economic Anal- ysis,”Inquirjy, December. BLOOM, BERNARD and PETERSON, OSLER (1973), “End Results, Cost, and Productivity of Coronary Care Units,” New Engl. J. Med, January 11. 49 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 50 REDISCH, MICHAEL (1974), Hospital Inflationary Mechanisms, Paper presented at Western Economic Association Meetings, Las Vegas, Nevada, June 1974 (revised October 1974). SCITOVSKY, ANNE A. (1967), “Changes in the Costs of Treatment of Selected Illnesses,” American Economic Review, Vol. 57, No. 5. WORTHINGTON, NANCY L. (1975), “Factors Contributing to Annual Changes in Expenditures for Hospital Care and Physicians’ Services,” Office of Research and Statistics, Social Security Administration (forth- coming—1975). FUCHS, VICTOR and KRAMER, MARCIA (1972), Determinants of Expenditures for Physicians’ Services in the United States, 1948-1968, DHEW Publication No. (HSM) 73-3013, December. KLARMAN, HERBERT E., RICE, DOROTHY P., COOPER, BARBARA S., and STETTLER, H. LOUIS (1970), Sources of Increase in Selected Medical Care Expenditures, 1929—1969, DHEW Social Security Admin- istration, Office of Research and Statistics, Staff Paper No. 4, April. RUSSELL, LOUISE B., BOURQUE, BLAIR B., BOURQUE, DANIEL P., and BURKE, CAROL S. (1974). Federal Health Spending, 1969-1974, Washington, DC: National Planning Association, August. WILSON, RONALD W., and WHITE, ELIJAH L. (1974), Changes in Morbidity, Disability, and Utilization Differentials Between the Poor and the Nonpoor; Data from the Health Interview Survey: 1964 and 1973. Presented at the 102nd Annual Meeting of the American Public Health Association, New Orleans, Louisiana, October 21, 1974. WORTHINGTON, NANCY L. (1975), “National Health Expenditures, 1929-1974,” US. Social Security Bulletin, Vol. 38, No. 2. UNITED STATES GOVERNMENT (1974), Special Analyses, “Federal Health Programs,” Budget, Fiscal Year 1974, pp. 135-163. ABEL-SMITH, BRIAN (1974), “Value for Money in Health Services,” US. Social Security Bulletin, Vol. 37, No. 7. DUPUY, JEAN-PIERRE (1974), “On the Social Rationality of Health Policies,” In Mark Perlman (ed.), The Economics of Health and Medical Care, New York: John Wiley. FORBES, WILLIAM H. (1967), “Longevity and Medical Costs,” New Engl. J. Med., July 13. FUCHS, VICTOR R. (1966), “The Contribution of Health Services to the American Economy,” Milbank Memorial Fund Quarterly, Vol. 44, No. 4, part 2. FUCHS, VICTOR R. (1974), Who Shall Live?, New York: Basic Books. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. GINZBERG, ELI with OSTAW, MIRIAM (1969), Man, Money, and Medicine, New York: Columbia University Press. HODGSON, THOMAS (1970), “Short-Term Effects of Air Pollution on Mortality in New York City,” Environmental Science and Technology, Vol. 4, No. 7. AUSTER, RICHARD; LEVESON, IRVING, and SARACHEK, DEBORAH (1969), “The Production of Health, An Exploratory Study,” Journal of Human Resources, Fall. BRENNER, M. HARVEY (1971), “Economic Changes and Heart Disease Mortality,” Amer. J. Pub. Hlth, Vol. 61, No. 3. BRENNER, M. HARVEY (1973), “Fetal, Infant and Maternal Mortality During Periods of Economic Instability,” International Journal of Health Services, Spring. FUCHS, VICTOR R. (1974), “Some Economic Aspects of Mortality in Developed Countries,” In Perlman (ed.), The Economics of Health and Medical Care, New York: John Wiley. COCHRANE, A. L. (1974), Effectiveness and Efficiency —Random Reflec- tions on Health Services, London: The Nuffield Provincial Hospitals Trust. LALONDE, MARC (1974), A New Perspective on the Health of Canadians, Government of Canada, April. U.S. INSTITUTE OF MEDICINE (1974), “Controls on Health Care,” Papers of the Conference on Regulation in the Health Industry, Jan- uary 7-9, 1974. KLARMAN, HERBERT E. (1973), “Reducing the Supply of Hospital Beds,” National Health Insurance~Implications, Hearings before the Subcommittee on Public Health and Environment of the Committee on Interstate and Foreign Commerce, House of Representatives, 93rd Congress, December 12, 1973, pp. 220-232. STUART, BRUCE, and STOCKTON, RONALD (1973), “Control Over the Utilization of Medical Services,” Milbank Memorial Fund Quarterly, Summer. U.S. INSTITUTE OF MEDICINE (1974), HMO’s: Toward a Fair Market Test, IOM Publication 74-03, May. ROEMER, MILTON I., and SCHONICK, WILLIAM (1973), “HMO Performance: The Recent Evidence,” Milbank Memorial Fund Quarterly, Vol. 51, No. 3, Summer. 51 42. 43. 44. 45. 46. 47. 48. 52 HELLINGER, FRED J. (1975), “The Effect of Certificate of Need Legislation on Hospital Investment and Operating Costs,” unpublished manuscript. HELLINGER, FRED J. (1975), “An Empirical Analysis of Prospective Reimbursement,” unpublished manuscript. GINSBURG, PAUL B. (1975), “Inflation and the Economic Stabilization Program,” unpublished manuscript. WALDMAN, SAUL (1974), “National Health Insurance Proposals, Pro- visions of Bills Introduced in the 93rd Congress as of July 1974,” DHEW, Social Security Administration, ,Office of Research and Statistics, DHEW Publication No. (SSA) 75-11920. KLARMAN, HERBERT E. (1974), “What Kind of Health Insurance Should the United States Choose?” In the US. Medical Care Indus- try: The Economist’s Point of View, Joseph C. Morreales, Editor, Michigan Papers, No. 60, University of Michigan, pp. 93-108. CHASE, ALLAN (1971), The Biological Imperatives: Health, Politics, and Human Survival, New York: Holt, Rinehart, and Winston. ETZIONI, AMITAI (1975), “Public Policy Issues Raised by a Medical Breakthrough,” Policy Analysis, Vol. 1, No. 1, Winter. BIBLIOGRAPHY AMERICAN MEDICAL ASSOCIATION (1974), Profile of Medical Practice, 1974 edition. BERLINER, R. W. and KENNEDY, T. J. (1970), “National Expenditures for Biomedical Research,” J. Med. Ed., 45:666. DAVIS, KAREN (1973), “Lessons of Medicare and Medicaid for National Health Insurance,” National Health Insurance—Implications, Hearings before the Subcommittee on Public Health and Environment of the Committee on Interstate and Foreign Commerce, House of Representatives, 93rd Congress, December 12, 1973, pp. 203-215. DAVIS, KAREN (1973), “Theories of Hospital Inflation: Some Empirical Evidence,” Journal ofHuman Resources, Spring. DAVIS, KAREN (1974), “The Role of Technology, Demand, and Labor Markets in the Determination of Hospital Costs,” In Perlman (ed.), The Economics of Health and Medicare Care, New York: John Wiley. FEIN, RASHI (1967), The Doctor Shortage: An Economic Diagnosis, Wash- ington, D.C.: Brookings Institution. FELDSTEIN, MARTIN (1971), “Hospital Cost Inflation: A Study of Non— profit Price Dynamics,” American Economic Review, December. GINSBURG, PAUL (1970), “Resource Allocation in the Hospital Indus~ try: The Role of Capital Financing,” U.S. Social Security Bulletin, Vol. 33, No. 10, October. U.S. INSTITUTE OF MEDICINE (1974), Costs of Education in the Health Professions, Summary, January. MUELLER, MARJORIE SMITH (1975), “Private Health Insurance in 1973: A Review of Coverage, Enrollment, and Financial Experience.” U.S. Social Security Bulletin, Vol. 38, No. 2, February. NATIONAL CENTER FOR HEALTH STATISTICS (1975), Health Resources Statistics, 1974, DHEW Publication No. (HRA) 75-1509, June. SALKEVER, DAVIS (1972), “A Micro-econometric Study of Hospital Cost Inflation,” Journal ofPolitical Economy, November/De cember. SHANNON, J. A. (1967), “The Advancement of Medical Research: A Twenty-year View of the Role of the NIH,” J. Med. Ed., 42:97. U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE (1974), Estimated Health Expenditures Under Selected National Health Insurance Bills: A Report to the Congress, July. 53 DISCUSSANT’S COMMENTS RASHI FEIN Harvard University School of Public Health, Boston, Massachusetts I shall attempt to be very, very brief in my comments. We are already substantially behind schedule, and I fear that if I did take the time originally allotted, there would be no opportunity for general discussion. First, let me indicate that, in my view, Dorothy Rice’s paper is excellent, the very kind of paper that one ought to have at an international conference. It sets a lot of data before us and is organized in a manner that will permit us, as the days go by, to compare the performance of the various nations here represented. Those of you who have had an opportunity to read the paper will also recognize that it is far more than a collection of numbers, that it is organized to tell a story and to point up issues and problems. To do justice to all the issues raised would, of course, be impossible. I will, therefore, in the few minutes that I have limit my remarks to the one matter that seems to me to be of priority importance As Dorothy Rice correctly describes the situation in the United States, that which we have known becomes even clearer and is brought forcefully to our attention: There does not exist a mechanism that might provide for a purposeful social allocation of resources to or within the health sector. There is no overall mechanism or set of inCentives that would yield such an allocation. We have little that can be described as effective planning, or alternatively, competitive market constraints. We have developed financing mechanisms that encourage and enable everyone to “do their own thing” (or, perhaps, more correctly “to try and do everything”). So it is difficult to get hospitals to close down or, indeed, even to cooperate. I have served on a committee that attempted to bring four institutions into a cooperative relationship that would improve patient care, rationalize program mix and reduce both capital needs and operating costs. It was a difficult task. There was no shortage of good will, but there was little incentive for individual institutions to sublimate their own goals in favor of a social imperative. Before one gets very irritated, one ought to perhaps remember that there are a large number of people in this room who come from universities. It is not clear that we in universities do a particularly better job of cooperating with other universities located in the same city, or indeed on occasion a better job of cooperating within the same university as between departments or schools, and so on. We, too, are influenced by prestige factors. So, it becomes, I think, relatively easy to stand by and criticize others without asking whether there are forces at work that cause our behavior, on 55 occasion, to be similar to that which we criticize in others. I think there are some forces, and I think Dorothy has mentioned a number of them, of which the basic force, at least as I see it, is that there is no feeling on the part of key micro decisionmakers that there really are scarce resources that have to be allocated. The health budget is not determined in advance; indeed, it is merely the product that one ascertains by adding up (after the event) what has been spent. This suggests an additional point that can be added to the prestige and other arguments that Dorothy has cited as being among the reasons that we invest in technology and that we build more and ever more. The problem, I think, is that we have put the burden of decision making on the wrong set of actors. We say, on occasion, that there aren’t sufficient incentives for hospital administra- tors or for physicians to allocate resources wisely. That is correct, but I would suggest that the problem really isn’t with the physicians and the hospital directors. If major allocative decisions are simply the summation of individual micro decisions, there may be no realistic incentives which are likely to lead to the favorable results that we would want. We ought to start by making macro decisions and constrain the micro decisions in that manner. If one spends time with physicians, I think one senses that, as a product, perhaps, of their training or self-selection process, they seldom have a population perspective. We can sit in this room and talk about benefits and costs, and about social responsibility, but physicians are trained to think of the patient at the particular moment. As I have spent 7 years in a medical environment and have become 7 years older and, therefore, the probability of my being a patient has increased, I find some sympathy, I must say, with the position that physicians articulate. I don’t think that I would feel very comfortable lying in a hospital bed and having the physician come in and say, “Rashi, you know how it is, we can take care of you, but the same resources will, in fact, take care of 17 extra cases of infant mortality in the area, and I have read what you have written about social responsibility and so I will deny you the care that you thought you might want.” (Laughter.) I don’t think that the physician or the patient looks with favor upon that view and this suggests that, at that particular moment, you cannot ask the physician or the patient to make those allocative decisions. If one is concerned about allocative decisions, the macro ones must set the dimensions for the micro ones, and the macro ones are to be made by a different set of actors, with a different perspective and at a different moment in time. They are to be made, it seems to me, by setting bottom-line budgets and having budget controls that constrain certain kinds of behavior. Even this will be extremely difficult since, to some significant degree, the public is not particularly on our side on this. Perhaps the public is wise enough to say, “Well, if we allocated medical resources in a more socially responsible manner, we would have to ask what would we use those newly available resources for? Is there some evidence that they would be used in a more socially responsible manner?” Perhaps the public is also wrapped up with emotion at the moment in time, and perhaps that says that, important as it is not to waste resources, important as it is not to let the percent of GNP going to health rise and continually rise 56 without deriving much benefit from it, important as it is to educate the public to the ideas that we may not be getting our money’s worth in this field, and that there may be other more socially useful things to do with the funds, perhaps, in the largest measure though, the issues that\the U.S. faces remain the issues of the ’sixties. That is, in largest measure, the battle may not be really over what'percent of GNP is going to the health-care sector, but more the "battle that is alluded to or spoken about in the last paragraph in the volume that Titmuss’s wife and Brian Abel-Smith edited after Titmuss’s death, the little volume, Social Policy, which contains his lectures to his class. The postscript to the volume is his first lecture of the quarter in which he tries to explain to his class why he is teaching the course that year in a different quarter than had been announced. The reason was that he was, in the first quarter, quite ill with the cancer that ultimately led to his death. In the last paragraph of the postscript he writes that one of the things that most impressed him as he was going through the British National Health Service (not being cured but being cared for because no cure was possible— care, a dimension that we economists tend to undervalue) was the fact that he showed up every morning at 10 o’clock in the waiting room together with a 26-year-old West Indian who had cancer of the rectum, and both of them were scheduled at 10 am. to be there in order to go in for radiotherapy. As Titmuss recounts, “Sometimes I went in first, sometimes he went in first, and what determined who waited was not race, color, occupation, or class, but the vagaries of London traffic.” That issue is ultimately what the American agenda still remains about. Perhaps the American public recognizes that far better than do we. Perhaps one of the reasons that the public is not entirely taken with the concept of social priorities and system rationalization is that it senses that the distribu- tional issues involving equity of access are often lost in the discussions about production efficiency. Perhaps it is suspicious of the benefit-cost calculus, feeling that the emphasis on efficacy is a way of avoiding providing to all what some already have. There are a host of other issues that are alluded to in Dorothy Rice’s paper—specific issues involving cost control in the context of the health sector and its behavior and involving the search for equity and its relationship to expenditures and cost control. I promised to be brief, however, and to go into other matters would negate that promise. Let me close, therefore, with the comment that US. policy may be setting the wrong priorities. It seems to me that first one must address the question of equity and only then the problem of cost control. In part this is because the solutions to the equity question will bring into place the very kinds of structures that will enable us to achieve cost control within socially optimal resource allocations; that is, cost control without national health insurance may be unattainable (or if attainable, may be attained only at the expense of equity considerations). In large measure, however, my view—I confess—may derive from the observation that our 57 energies are limited, and if, as I believe, one must choose which problem should be addressed now, it is the question of equity in access that cries for attention. It ill behooves those of us who already have access—and that is all of us in this room—to argue that others should wait for moral solutions while we straighten out the economic issues. Economics once was called moral philosophy. Perhaps we should remember that. 58 Chapter 4 HEALTH COSTS AND EXPENDITURES IN CANADA ROBERT G. EVANS Department of Economics, The University of British Columbia, Canada The Canadian Health-Care System, 1975 In Canada, most health care is free. To the average Canadian citizen point-of-service charges for the services of physicians or access to hospitals are trivial or non-existent and the costs of providing such care are buried in a complex web of budgets at various levels and jurisdictions of government. (1) Furthermore, it appears generally agreed that access to health care unimpeded by (direct) economic barriers is the right of every Canadian citizen, or even resident, and consequently in the areas where such barriers do still exist, prescription drug purchases and dental care, it is expected that universal public “insurance” programs will be established nationwide in due course/2) Few Canadians know or care very much what the costs of such programs are: taxes are, of course, too high and government spending is in general wasteful but these views are rarely translated into concern over the appropriate costs of particular programs or services. Public concern over health-care questions focuses on issues of access, or quality, or the rate of addition of new insured services; but only bureaucrats and academics worry about costs. The most comprehensive estimate of health-care costs in Canada, a new series issued in 1974 by the Department of National Health and Welfare, places the total of national health expenditures at $6,611 million in 1971 , or $306.11 per capita spread over a population of 21.6 million. This represented 7.1 percent of Gross National Expenditure, which in 1971 was $93,094 million. (3) Somewhat less comprehensive than the N.H.E. series is the personal health care estimate, which is restricted to payments to hospitals, physicians, and dentists, and for prescription drugs. It excludes costs of research, capital expenditures, public health and preventive care, education of health personnel, administration of public or private payment plans, and a variety of consumer expenditures which are not directed by the health “establishment”—eyeglasses and appliances, nonprescription drugs, services of health professionals (other than physicians or dentists) outside hospitals. For 1971, expenditures for “personal” and “national” health care are shown in Table 1. Unfortunately, at time of writing (May, 1975) more recent data in these series have yet to be released. A number of partial returns are available from the annual reports of hospitals and of provincial medical insurance agencies, but these tend to be sufficiently diverse in concept and coverage as to make 59 Table l CANADIAN HEALTH CARE EXPENDITURES, 1971, $MILLION “Personal” Concept “National” Concept Total Per Capita Total Per Capita General and Allied Special Hospitals ............ 2,594.6 120.15 2,595 120.15 Other Hospitals ......... 557.4 25.82 557 25.82 Nursing Homes ......... — — 186 8.62 Physicians ............ 1,236.2 57.24 1,236 57.24 Dentists .............. 298.8 13.84 299 13.84 Other Health Professionals. . — _ ~ 120 5.55 Prescribed Drugs ........ 422.5 19.56 423 19.56 OTC Drugs ............ — -— 372 17.14 Eyeglasses and Appliances . . , — 82 3.81 Administration ......... ~ g — 120 5.54 Capital Expense ......... ~ — 338 15.63 Research, Public Health and Other ............... — — 286 13.22 Total ................ 5,109.5 236.61 6,611 306.11 Sources: Canada, Department of National Health and Welfare, National Health Ex- penditures in Canada, 19601971, Ottawa; October 1973 (Le. 1974) and Expenditure on Personal Health Care in Canada, Ottawa, n.d. (1974). their linkage with the above series hazardous for purposes more substantial than seeing which way the wind appears to be blowing. These national expenditures pass through a number of different budgets, but in general one can distinguish three main classes of expenditure in the Canadian health-care system. First and largest are the two federal-provincial cost-shared health “insurance” programs, general hospital and medical care. In 1971 these were about three-quarters of personal health care expenditures and nearly 60 percent of national health care expenditure. (4) Secondly, private expenditure still predominates in drugs and dental care, although as is noted below, several individual provinces are currently putting in place payment programs for children’s dental care and prescription drugs for the elderly. “Welfare” patients have always received such services free of charge or at subsidized rates. Finally a small part of health-care expenditure flows through direct public sector provision; these activities include traditional public health activities at the provincial and municipal level as well as small systems of 60 federal and provincial hospitals for special populations such as the mentally ill, native peoples, the military and veterans. Educational, research, and capital expenditures are also almost all financed by direct public provision or grant. The federal-provincial cost-sharing health “insurance” programs presently form the backbone of health-care delivery in Canada; they have been planned and implemented over the last 30 years and reached comfiletion in 1971 when the last provinces joined the medical insurance system. structure they are provincial programs, since the Canadian constitution assigns to the provinces jurisdiction over health-care matters. Thus each province establishes its own administrative agencies, conditions of eligibility for participation by both patients and providers, and modes and rates of payment. The federal legislation, however, lays down common structural features which each provincial plan must meet in order to be eligible for federal cost-sharing; and since this cost-sharing amounts to roughly 50 percent of total program costs (the percentage varies by formula from province to province) it would be impossible for any province to fmance a plan outside the federal framework. In order to receive federal funding, each provincial plan must have four basic characteristics: universal access “on equal terms and conditions,” portability across provinces, comprehensive coverage, and non-profit administration. These in general imply that the plans must be compulsory (in order to be universal) and provincially administered rather than contracted to private insurers. The resulting hospital and medical insurance programs have a relatively complex structure. In concept, they were originally intended only to remove “financial barriers” to health care, not to affect the institutions and processes of the “supply side.” Hospitals and physicians were to remain independent entities at arm’s length from the public paying agencies, providing care without interference, and submitting the bills to the government. Patients would be able to get all the care they “needed” without the burden of direct payment. The only restraints were that physicians would have to accept direct reimbursement on the basis of a uniform provincial fee schedule (promulgated by each provincial medical association) and hospitals would have to submit their own budgets for auditing and approval. In concept, then, “the state” would not intervene in the relationship between provider and patient. Physicians and hospitals would continue to be “free and self-governing.” The only roles of the provincial government would be to handle bookkeeping and payment, establish the validity of claims, and check for fraud. The federal government’s role was to be even more limited—remittance to the province of a formula share of annual audited costs, whatever these turned out to be, subject only to assurance that the costs were in fact incurred for designated (comprehensive) “shareable” services, and that the provincial plans met the federal structural requirements. In retrospect, it is hard to imagine how such a blank-check approach could ever have been expected to work. The public programs avoided in principle the responsibility for managing the delivery of health care or rationing its availability. Prior to national health “insurance,” the marketplace had imposed some constraints on pricing and costs of providers and utilization by consumers; the results, of course, were quite unsatisfactory but at least no 61 agency or individual had responsibflity for such decisions. In the absence of the marketplace, it seems to have been the hope that the health-care industry (being composed after all of reasonable people) would somehow constrain and manage itself. This philosophy appears rooted in a medico-technical model of health services, which assumes a causal sequence of the form: population -> “need” for care -* specific procedures or services —> manpower and equipment -> costs. Each step'in this sequence is treated as if it were technologically determinate, and in particular as if it were independent of the mode of organization of and payment for health care. Instead, of course, every step depends on the institutional context. The “need” for care expressed by patients depends on the costs (monetary or otherwise) of seeking care. The choice of technology for meeting “needs” made by providers depends both on the availability of facilities and the relative economic or other payoffs to themselves. Provincial governments are more likely to finance services which are federally shareable, although these tend to be the more expensive ways of resolving any given problem. And the last step, from manpower to costs, passes by way of provider incomes which now become primary determinants of total system expendi- tures. Subject neither to market nor to administrative constraint, a health-care system becomes overutilized, provides excessively costly and unnecessary services, uses overtrained personnel and goldplated equipment, and generates ever-higher incomes for its providers. Of course, such absence of all constraint could not last, and regardless of underlying philosophy the provincial governments immediately found it necessary to try to impose all sorts of administrative restraints on health expenditures. Such restraints, however, have been relatively ineffective to date because the basic structure of the programs, with providers at arm’s length, has made management of health delivery almost impossible. The crudcr techniques of turning off financial taps or disapproving new projects, at best tend to be short-run in an industry as politically sensitive as health. And finally the federal cost-sharing structure weakens the incentives of provincial governments to take politically difficult decisions in the health field—health expenditures are after all only “50¢ dollars.” A serious effort by a provincial government to improve the management of health services by, e.g., providing alternatives to hospitali- zation, might well move into “non-shareable” areas of expenditure and cost $1 dollars. The federal government, on the other hand, has no constitutional authority to manage health services; it can only advise—so expenditures continue to climb. It is, of course, difficult, in a complex world, to determine the extent to which Canada’s national health insurance programs caused these expenditure increases, to some degree the programs were responses to more general worldwide forces. But there is no doubt whatever that the structure of Canada’s programs, with their deliberate bias against direct public intervention in the process of provision of health care, has made it difficult or impossible to control or modify these developments. Almost every major study or report for the past 5 years (except, of course, those financed by medical associations) has called for increased public intervention in such forms as direct provincial 62 takeover of hospital systems, establishment of salaried group practices, elimination of fee-for-service, etc. (5) The most recent report summarizes a common critical theme: ( 6 ) “In our judgment there is no area in the private or public sector where the decision-making process is so completely unrelated to any incentive to economize on the use of scarce resources. . . “What is especially disturbing is that the essentiafcharacteristics of the‘ decision-making process, in conjunction with the payment system, have created a health sector that is inherently unstable, by which we mean that the increase in total health costs may result largely from cost escalating characteristics of the health-care system and not from the provision of substantially more and better health care.” Yet the very real success of Canada’s health programs in relieving its citizens from the direct burdens of illness and equally important the anxieties associated with such burdens, makes it difficult to muster a political constituency for change. The cost inflation‘goes on, apparently out of control, but apparently also out of sight of the average Canadian. Historical and Statistical Perspective So we have achieved, in the early 1970’s, an unsatisfactory health care system with which most Canadians are satisfied. How did we get here? The major developments of the past 20 to 25 years can be traced through changes in the legislative and insurance structure, shifts in the availability of health resources, and trends in utilization, costs, and prices. The major milestones of this period are the federal Hospital Insurance and Diagnostic Services Act of 1957 and the Medical Care Act of 1968. These established the federal terms for provincial programs, in each case building on pre-existing programs in some provinces and stimulating action in others. In each case there was some delay before all provinces could put in place qualifying programs, but participation became universal in hospital insurance in 1961 and in medical insurance in 1971. Thus 1971 was both the first year of complete national health insurance (hospital and medical) in Canada and the last year of complete or consistent system data. Of course, prior to the federal legislation both hospital and medical insurance were widespread through a mixture of private insurance companies, non-profit agencies, and provincial plans without federal assistance, but the level of coverage is difficult to establish. As late as 1967, only about 30 percent of the population had medical coverage through a non-profit plan with the first dollar, full coverage features of the public program. The private carriers fielded an array of different contracts whose total effect on coverage is difficult to calculate. {7) In a sense this is a significant disadvantage in any attempt to analyze the impact of the public programs; it is clear, for example, that the effects of increasing health insurance coverage establish trends which predate the public plans. On the other hand, if one recognizes that from the perspective of policy the interesting questions relate public health insurance not to some hypothetical uninsured alternative, which has not existed for a generation, but to the health system as 63 it really was, warts and all, prior to 1957 or 1968, then it is legitimate to use these dates as reference points in studying the impact of the new programs. However, national health insurance was also accompanied by major shifts on the supply side which tended to obscure the direct identification of impact. Advance planning for demand shifts created certain “self-fulfilling prophecies” which may have been independent of changes in the payment mechanism per se. In the case of hospitals, a federal building subsidy program was begun in 1948 which combined with the spread of private insurance and rising incomes to support relatively rapid growth in the stock of hospital beds per capita. This measure is reported in Table 2. (8) Thus the insurance program was predated by a planned expansion in hospital capacity to meet the expected demand, and what is interesting is that the demand expansion also appears to have predated public insurance. Thus the common assertion among Canadian health-care observers that, by introducing hospital insurance first Canada biased its health-care system towards excessive'and costly use of hospitals, must be interpreted as meaning the intent to introduce insurance led to the over- building of hospitals which in turn increased utilization and drove up expenditures. This in turn necessitated the public program. As noted below, ambulatory medical insurance seems to push up hospitalization slightly. A similar development parallels the introduction of medical insurance; the same Royal Commission which recommended a national medical insurance plan simultaneously recommended a major increase in the capacity of medical schools. (9) This was done, although lead-times tend to be longer for producing new physicians than for building hospital beds. Unfortunately, this new capacity was put in place simultaneously with a drop in birth rates and a surge in in-migration of foreign physicians, the results were very rapid increases in the physician/population ratio in the last few years. Thus it is difficult to Table 2 CANADIAN PUBLIC HOSPITALS*, BEDS AND UTILIZATION (per 1,000 population) Beds Patient Days Admissions 1953 ..................... 5.13 1,473.1 130.2 % change 1953-59 ................. 1l.5% 12.0% 11.2% 1959-65 ................. 4.2% 7.8% 5.1% 1965-71 ................. 7.6% 6.7% 8.3% 1971 ..................... 6.41 1,896.6 164.9 *Includes general and allied special, chronic and convalescent. Source: R. G. Evans “Beyond the Medical Marketplace“, (See Note 1). 64 determine whether changes in delivery system behavior result from universal insurance or from a growing physician surplus. Consistent data on the physician stock are harder to find than for hospitals, but the federal Health Manpower Inventory (1974) (10) reports 36,095 physicians or 1.618 per thousand population in 1973 compared with 28,209 and 1.362 in 1968. A longer series focusing on physicians in active fee practice shows an average growth of 1.1 percent per year from .780 per thousand in 1957 to .879 in 1968, followed by an average of 4.5 percent per year to 1.004 in 1971. (11) More recent data are not yet released, but the Inventory data show the all-physicians ratio growing at just under 3 percent per year from 1971 to 1973. The trend is disturbing. The data in Table 3 provide a general summary of historical rates of change in the main components of Canadian health expenditure; the leading sector is clearly general and allied special hospital expenditure which has grown from $280.4 million in 1953 to $25946 million in 1971, and makes up an ever-growing proportion of the total. Expenditures on physicians’ services have also shown rapid growth, as has related prescription drug expense. All other expense components have tended to increase their share of national income, Table 3 GROWTH OF CANADIAN HEALTH CARE EXPENDITURES, SELECTED COMPONENTS, ANNUAL PERCENTAGE CHANGE 1953-71 1960-71 General and Allied Special Hospitals ........... 13.2 14.1 Other Hospitals ......................... 8.7 9.5 Nursing Homes ......................... —— 14.8 Physicians ............................. 1 1.4 13.0 Dentists .............................. 9.3 10.0 Other Health Professionals .................. — 7.5 Prescribed Drugs ........................ 10.6” 1 1.1 OTC Drugs ............................ — 9.5 Eyeglasses and Appliances .................. ~ 8.0 Administration, Research, Construction, etc. ..... —— 7.5 Total (Included Items) .................... 12.0* 11.8 *Because of a conceptual break in the prescription drug series, these rates are computed for the period 195 7-71 only. Sources: Canada, Department of National Health and Welfare, Expenditures on Per- sonal Health Care in Canada, 1953-61, Ottawa, March 196 3, and National Health Expenditures in Canada, op. cit. 65 but except for nursing homes have tended to grow significantly less rapidly than hospitals or physicians. It is clear, therefore, that over the past 25 years health-care spending has increased dramatically in Canada. So what? There are a number of more specific questions which one can address to these data. In particular it is important to know to what extent the numbers reflect general growth patterns and inflationary trends in the Canadian economy. Insofar as these exogenous effects can be accounted for, can the “endogenous” trends in the health care sector be related to the introduction of the national health insurance programs? Can the “endogenous” trends be broken into price and quantity effects of various sorts, and can these components be related to public programs or to other supply side effects? If all this could be achieved, can one sketch in what may be happening in the period more recent than 1971? Ideally, to answer these questions one would probably wish to have a large and sophisticated structural model of the health care sector—assuming one could ever agree on what the structural equations should look like! But some answers can be sketched in by studying the available data directly, and these may permit one to draw some conclusions about future developments. (12) To begin with, one can recast the expenditure categories in Table 3 into per capita amounts or into shares of personal income. Population growth makes little difference, growth rates averaged 2.1 percent in the period 1960-71 and 1.7 percent so the pattern in per capita expenditures looks very much like Table 3 with each figure reduced about 2 percent and the more recent period growth rates relatively somewhat higher. Personal income growth is much more important, having averaged 7.6 percent per year in 1953-71, and 8.6 percent during the shorter period 1960-71. Comparing these rates with Table 3 reveals an interesting ranking in expenditures. The services insured during the period 1953-71, hospital care and physician care, both rose substantially faster than personal income and raised their shares at about 5 percent and 3-1/2 percent per year. Dentists’ services and prescribed drugs, both uninsured but high “professional content” services, increased their share of personal income at between 1 percent and 2 percent per year. Other services in which the expenditure decision is more likely to be made either by the individual independently or by a public agency, rose at about the same rate as personal income, give or take a percent. Only nursing homes fall outside this hierarchy. There is, of course, no magic significance to an income elasticity of unity, and any association between rapid growth of expenditure and insured status says nothing about direction of causality (if any). Nevertheless it turns out that the largest jumps in personal income share devoted to hospital care occur precisely during the period of introduction of insurance. From 1958 to 1961, growth in income share devoted to general and allied special hospitals rose on average at 11.5 percent per year—double the long-term rate. Part of this effect was due to the recession conditions of 1961, personal income grew at an average rate of only 4.1 percent during that rather soft patch. But this effect is not nearly enough to account for the large jump, particularly when one notes the stability of the rate of growth of personal income share devoted to hospitals outside that period—3.9 percent annually from 1953 to 1958 and 4 66 percent from 1961 to 1971. There appears to be a built-in expansion factor in hospital care, but the insurance programs introduced between 1958 and 1961 significantly increased that share by about half a percentage pomt in 3 years. The same phenomenon appears in the physician price data, from 1969 to 1971 “the annual growth in personal income share rose 7.3 percent compared with a 1953-71 rate of 3.5 percent. No other year-over-year increase comes close to that amount. In this case, however, interpretation is confounded by the rapid increase in the total numbers of physicians over the same period. The data must be examined on a province-by-province basis; and while some provinces show very large increases in the average receipts of physicians associated with the introduction of insurance, others show no such effect. There is some suggestion that in fact the long-rim upward trend in the relative income status of physicians may have been very little affected by insur- ance, (13) and that recent expenditure surges may be the result of new physicians generating more activity. Of course, in the absence of insurance this generation of activity might not be possible. Until the federal government begins to catch up on its release of more recent data, it is difficult to know what is going on. This leads into the issue of price and quantity effects. One of the most striking aspects of public hospital insurance in Canada is the effect it did not have on utilization. Over the period 1958-62, (14) when most provinces were initiating plans and expenditures were soaring, utilization rates were relatively flat. Patient-days per thousand rose 1.46 percent per year in that period; admission rates rose 1.15 percent. As noted above, utilization growth was much faster in the years prior to public insurance—2 percent per year growth in patient days per thousand from 1953 to 1958—than it was either during introduction or after. From 1962 to 1969 patient-day utilization rates rose 1.08 percent per year. A more significant effect occurred when medical insurance plans were introduced; from 1969 to 1971 admission rates rose 2.68 percent each year and patient-day rates rose 2.25 percent—the fastest of the period. This tends to confirm the “paradox” noted by Lewis and Keairnes: that insurance for ambulatory medical care (in a fee-for-service environment) increases hospital utilization. {15) Whether this is a backlog effect or a permanent increase will have to be answered by future data. (16) The major increases in hospital costs are clearly identifiable as increases in costs per patient-day—insurance did not trigger a demand explosion. These patient-day costs in turn can be traced primarily to wage and salary costs, the share of total hospital budgets devoted to wages and salaries rose from 57.7 percent in 1953 to 67.9 percent in 1971. In the earlier part of the period, and particularly over the years of introduction of insurance, much of this increase represented staff increase. Total paid hours worked per patient-day rose 15.2 percent from 1953 to 1969 and 22.6 percent from 1959 to 1965. The earnings of hospital workers also rose relative to the average weekly wage at a rate of 2-3 percent per year in these years. Since 1965, however, the increase in staff has been minimal and the main inflationary force has been wage rates. Wages in the hospital sector have risen over 5 percent per year relative to general wage levels from 1965 to 1971. So it appears that the hospitals’ reaction to public 67 insurance was both to raise wages and to increase staff. ,This led by the mid-sixties to increased administrative pressure limiting staff increases; but at the same time hospital unions discovered the absence of effective wage controls in a non-strikeable industry. At present, hospital costs seem driven by wage inflation, and it is not clear how the process can be controlled. In physician services, processes are less clearcut due to the absence of any uniform quantity index even as unsatisfactory as the admission or the patient-day. Data on physician expenditures province-by-province generally show a jump in the year of introduction of insurance or immediately after. Data on list prices (fee schedules) certainly show no corresponding increase, although an adjustment upward of actual fees of unknown amount was associated with insurance. Available indices of physician fees generally show them rising more rapidly than consumer prices generally prior to public insurance, more slowly thereafter. Combining these indices with reported expenditures, however, leads to quantity estimates per physician or per capita which increase at an implausibly rapid rate. Clearly shifts have taken place both in the relation of actual list fees and in the labeling of procedures for billing purposes, and the size of such effects is indeterminate. Such sketchy data as exist do not support the view that any significant shift in patient-determined utilization behavior has taken place. It is of interest that the first published analysis of service patterns in the immediate post-Medicare years shows a high correlation between increases in physician stock by specialty and increases in total billings. But it is not in any way conclusive. Similarly suggestive is the Enterline et al. finding in Montreal that pre- and post-Medicare physician contacts per capita were equal, but that they had been redistributed somewhat from higher to lower income groups. (17) The evidence is difficult to read, but it appears that to some extent public insurance has increased the share of physician care received by lower income people,has not triggered a significant demand increase by patients; has influenced physician behavior with respect to billing patterns and perhaps hospitalization, has not generated list-fee increases but if anything may have moderated them; and has had all its effects seriously confounded by the effects of the physician stock increase which, of course, also may influence these variables. One important point which does emerge from the Canadian data, however, is that list fees are singularly unhelpful indicators of market conditions. Prior to universal insurance they deviate from actual fees by an indeterminate amount, subsequently their effect on incomes of physicians and expenditures is more determinate but is modified by shifts in billing practices. Moreover, even in the absence of public insurance, physician fees in Canada have tended to be unrelated or perversely related to market conditions, being consistently highest in the province with the largest relative supply of physicians. (18) It was stressed earlier that data subsequent to 1971 are available but are incomplete and not always consistent with the longer-term series on which the comments thus far have been based; hence these data must be treated as indicative only. Preliminary releases, for example, showing small increases or reductions in average income per physician since 1971 can be generated by shifting the manner of counting physicians in the denominator, or by changing 68 the concept of what is counted as income in the numerator, while the relatively up-to—date Quarterly Hospital Indicator data are not based on the full universe of institutions and thus are not entirely consistent with the longer-term series. It is thus rather unfortunate that federal statistics should be falling even farther behind at a time when sound data are needed more than ever to evaluate our programs. Nevertheless, recent data suggest that in spite of the inherently uncontrolled nature of both hospital and medical insurance programs, expenditure growth relative to the general economy was mitigated somewhat in the early 1970’s. Whether this was a permanent development or an accident is at present an open question. Hospital data show a drop in patient-days per thousand population from 1971 to 1973, from 1905.9 to 1871.0. On the other hand, admissions rose slightly, 167.7 to 170.4, beds available rose 4-1/2 percent, which may bring utilization up again as occupancy rates have moved down slightly (82.5 to 79.4 percent). (19) Hospital operating expenses per patient-day rose 9.9 percent per year from 1971 to 1973, and total patient-days were almost static. Personal income rose substantially faster, an average of 13.1 percent per year, so that if the initial data are confirmed, the share of hospital expenditure in the Canadian economy will have fallen for the first time in 20 years. This view is supported by recently released data on hospital salaries and wages which show very large gains for the major classes of hospital employees in the years 1969-70 and 1970-71, but much slower increases in the years 1971-72 and 1972-73. (20) The danger sign, however, is that much of this change may be the result of the redistributional effects of inflation. Hospital workers’ earnings continued to grow very slightly faster than average wages generally from 1971 to 1973, but over this period and continuing into 1974 labor incomes have tended to run behind rising food costs, increasing effective tax rates due to inflation, rising corporate profits, and energy costs. The result has been explosive increases in labor contract awards in 1974, and hospital workers have apparently been very successful in securing spectacular increases in salary rates. It is impossible at this stage to point to specific data on these expenditure effects, but it is extremely likely that when the national data finally emerge they will show a relative flat spot in 1972-73 followed by large increases in 1974. Whether this will restore the trends of the period to 1971 is at this point impossible to say. The same development appears to have occurred in physician services. The large gains associated with Medicare seem almost to have put provincial medical associations off guard—reported list fee increases from 1971 to 1973 were very small. Between January 1971 and January 1974 the National Health and Welfare composite fee-schedule index rose only 4.9 percent. (21) During this time, of course, physician incomes were hit by the forces of accelerating inflation and rapid increases in physician numbers, so it would not be surprising if when consistent data become available they show a drop in the real income of the average Canadian physician in 1972-73. Precisely because data are delayed, and negotiations between physicians and provincial payment agencies are based on income data, there is a response lag built into the fee-setting process. 69 By January 1975, however, average fee schedules rose an additional 6.3 percent and negotiations may lead to larger increases in 1975. There is talk of strikes or other militant action in several provinces, although physician strategy now appears to focus on the use of salaried government or hospital physicians as front-line troops in any confrontation. How the process will resolve itself is something which may not be known for several years, but it is clear that tensions are currently building in this area. Pressure is also being exerted by medical associations to curtail the inflow of immigrant physicians, as might be expected. Thus the historical record can be summarized as: steady increase in the relative costs of hospital and medical care, exacerbated but not necessarily caused by public insurance, primarily associated with increases in health workers’ earnings, but not in general driven by independent demand shifts resulting from insurance. The more recent data since 1971 may indicate that health costs are no longer taking an ever-increasing share, but whether this is due to a topping-out of the once-for-all effects of insurance or merely a result of health-care workers having been on the short end of a period of unanticipated inflation is a crucial and unsettled question. The writer’s own view is that the latter is the case, and that a reaction which will push expenditures up yet again is inevitable. This view is primarily based on an inability to detect any institutional or market force which would restrain such a reaction, combined with a professional pessimism about the willingness of individuals, health-care workers or others, to passively accept income reductions. Future Developments The evolution of health services delivery in Canada in the next quarter- century will probably exhibit two main themes: modification of the existing structure of hospital and medical care insurance and the addition of elements to that structure by the extension of coverage to dental care and pharmaceuti- cals. Obviously the two are related, since the way in which coverage is extended will depend on political judgments as to how satisfactory is the functioning of the existing insurance programs. In particular, if one takes the view that the expenditure increases associated with national health insurance were self-limiting and are now under control, one might wish to extend the same insurance mechanism with minimal intervention on the supply side—and if not, of course not. The argument for significant change in the hospital and medical-care programs stresses the absence of any incentives in the present structure to encourage efficiency by providers as well as the absence of any long-run mechanisms for income limitation. In fact present incentives encourage the excessive use of relatively high cost “technologies,” the over-use of hospital beds and physicians relative to less-costly ambulatory or home care or nurse practitioners. The problem has two aspects—sheer waste in the sense of hospitalized days or physician services which should never have been supplied, and inappropriate production technology in the sense of services which were 70 required, but which were provided by an inappropriately expensive person or institution. Current modes of organization of and payment for hospital and medical services create powerful economic incentives for both these forms of expenditure inflation in ways which are now familiar. The general effects of fee-for-service reimbursement on patterns of medical practice by independent entrepreneurial physicians, or of cost reimbursement on the budgetary behavior of independent hospitals do not really need further comment. The essence of the problem is to reduce the use of high-cost inputs, expand the use of low-cost inputs, and reduce “unnecessary” output. But at present the key decisionmaker in each sphere, the hospital administrator or physician, is also the supplier of the high-cost input. Hence physicians will use more auxiliaries only if they can be used to “add-on” additional output (and costs), not to reduce costs of existing output levels. Hospitals react the same way—an ambulatory surgery program which reduces inpatient utilization and hence reduces overall rates of reimbursement is unlikely to be popular. An increase in system efficiency through cost-saving input substitutions which are already well-known and well proven cannot take place unless the locus of effective management shifts from the providers who have a direct interest in preserving present modes of production. Hence we find recommendations for the “provincialization” of all hospitals and health-related facilities in the hope that a unified provincial management will be able to achieve a better balance in the use of facilities as well as being better able to discourage unnecessary use by physicians. In the same spirit variations on the theme of the Community Health Center (C.L.C., H.S.O., HMO, P.P.G.P., and so forth) are proposed as a method of moving away from fee-for-service and thereby curtailing excessive hospital use. (22) In addition it is hoped that such institutions may change the mix of health personnel to make less use of physicians and more use of nurses or other health professionals, and improve the quality of care through consultation, better coordination, more preventive care, etcetera. The asserted benefits of such alternatives are as diverse as the various descriptions, but all share the common objective of relieving the individual physician qua physician of the management role and at the same time removing the economic incentives which encourage him to seek that role and to use it to build inefficient distortions into the process of delivery. Such an approach does not, of course, provide any answer to the problem of placing upper limits on the income status of health workers, but at least it may permit income inflation to occur among a less costly average mix of workers. In the short-term, however, prospects for such changes do not appear bright for a number of reasons. As mentioned above, the economic incentives to change faced by a provincial government (which is, constitutionally, the level at which change must occur) are diluted by the present cost-sharing agreements which require half of all savings to be shared with Ottawa. The political fall-out, however is not shareable and is certain to be major, given the vested interests and political muscle of medical associations. Even highly sympathetic governments are understandably reluctant to act in this area. Moreover, the very satisfaction of the average citizen with “free” care severely weakens any 71 political constituency for change—there is no strong massxsupport for radical action in a population which see the benefits but not the costs of the present system. Proposals have been made to close the open-ended nature of cost-sharing, but to date no agreement has been reached. In the absence of major institutional change in the management structure of the health-care sector, interest has recently focused on administrative measures to limit the total supply of both hospital beds and physicians. Such a policy certainly seems to be a precondition for the success of any more-far-reaching policy to shift the input mix in provision of health care away from reliance on the highest-cost inputs. The present concern with a physician surplus in Canada makes study and development of the nurse practitioner as a substitute for the physician rather absurd, and leads to suspicions that expanded-role auxiliaries may be just one more “add-on” to raise health-care costs yet further. Certainly, chronic and convalescent care hospital beds have turned out to be such an “add-on,” because acute care beds have not been correspondingly withdrawn. But if such restraints are all that are attempted, the result will be a perceived (and actual) reduction in services with no reason to believe that the “unnecessary” services are being eliminated. Such a policy is unlikely to be supportable, and in fact there is no evidence yet that direct restraint is succeeding. Given the open-ended cost commitments which they have undertaken in hospital and medical insurance, and the fact that the federal government has not yet been willing to share the costs of major extensions, provincial governments have been moving carefully in the dental and pharmaceutical field. Most provinces have programs of some sort underway or projected, but these tend to be partial rather than universal. Most provinces are now moving into children’s dental care, although age limits vary and coverage is being phased in. Families eligible for welfare receive subsidized dental care and prescription drugs. Free prescription drugs for the aged (over 65) are also available in several provinces; one province has gone universal but with a deductible for those under 65 which effectively restricts coverage to the chronically ill or other high volume prescription users. The experience with “insurance” plans has also encouraged provinces to take a much greater interest in the operation of the supply side of these serviCes before launching payment plans. Saskatchewan (23) has introduced its children’s dental care plan built around school-based dental nurses—public employees who provide the bulk of both preventive and restorative care— backed up by salaried dentists. This approach not only holds out the promise of lower ecst, since it is now well' established that preparing and placing amalgam restorations is not a task requiring the skills of a dentist; but even more important, the school-based service will reach a significantly higher proportion of the child population and should therefore yield a greater increase in overall dental health than could a private-practice-based system. Other provinces are studying this system program with interest. (24) Manitoba’s prescription-drug plan includes requirements that pharmacists supply generic drugs when available, and involves the provincial government in ensuring that supplies of such drugs are available. In most provinces the 72 situation is still very much in flux, but it is clear that provincial govermnents will certainly experiment with a variety of forms of direct intervention as dental care and prescription drugs are shifted out of the marketplace, in an effort to gain some control over the resulting expenditure behavior. This is particularly necessary since the problems of negotiating with independent providers become even more acute in partial-coverage insurance plans. If negotiations break down between a province and a medical association, a physician strike is a political disaster for the government but it also cuts off almost all income of the physicians involved. But if dentists, for example, withdraw their services from a children’s dental plan they may lose about a quarter of their total billings. Therefore there is a significant asymmetry of bargaining power. This extends to the individual level, for participation in a public insurance plan is not compulsory. Refusal to participate in a partial plan is likely to be feasible for a significantly higher proportion of practitioners than if the plan is total. Concern for the functioning of the supply side in extending public programs " to dental care and drugs can be justified on grounds other than expenditure control, however. It is important to recall that hospital and medical “insurance” programs developed from the traditional insurance concerns with risk-spreading. Dental and pharmaceutical insurance have always been slow to develop because they are “uninsurable” in the traditional sense. Levels of family expenditure are relatively low and regularly recurring, except for certain segments of the population who are well-identified and who would thus self-select into any voluntary plan. Therefore the introduction of public plans in these areas cannot be justified as improvement in markets for risk-bearing in the classic Arrow justification for health insurance. (25 } Instead, public intervention must be justified on grounds of merit goods (children’s dental care is intended to raise utilization), income redistribution G’harmacare is intended to relieve the elderly and chronically ill of high drug-expense burdens) and remedy of market failure (both dental care and prescription drug markets are asserted to have significant monopolistic features and large quasi-rents (26)). The correction of market failure clearly requires supply-side intervention; the promotion of dental utilization appears to require change in present modes of delivery, not merely insurance plus posters; and the process of income redistribution by subsidizing the users of large quantities of drugs is likely to accentuate market failure problems by eliminating the price sensitivity of the most price-conscious drug buyers. ( 2 7) The likely outcome, then, is that within a decade dental and prescription drug coverage will be universal in Canada, probably with some federal participation but (it is hoped) not the present open-ended sharing form. There will be a mixture of direct public provision of services and contracting with private providers, using perhaps Community Health Centers or school services, but there will undoubtedly be direct regulation of private providers in some form. The system will probably not be uniform across provinces. As experience accumulates in these areas, it is even possible that it may also serve as a basis for change in the medical and hospital sectors. 73 Lessons for the United States? (1) In a sense, the effects of national health insurance on the demand side of the health-care market may be the least interesting aspect of the problem. Excess demand resulting from “free care” did not appear to be quantitatively significant in Canada, either because health care was price-inelastic or because of the pre-existence of private insurance; the same may be expected in the United States. (2) The non-quantitative effects of health insurance on the population are very important. Universal first-dollar full coverage provides great security and relief of anxiety for all citizens. The average Canadian regards the United States as a medical jungle. In this regard, the absence of the various direct charges proposed for some United States insurance plans may be significant, as these expose the patient to additional risk and anxiety. This is particularly true if the provider is permitted to use direct charges to the patient as a vehicle for raising his charges above the program reimbursement rate and hence unilaterally reducing the patient’s coverage. This is not permitted in Canada. (3) On the other hand, general satisfaction and freedom from anxiety makes major institutional change very difficult. A program set up to handle the demand side now and the supply side later may find that later it is politically impossible to generate support for the changes in management structures necessary to control health care expenditures. (4) The supply side is the crucial sector, and the government (state? federal? in Canada provincial) must take responsibility for some degree of management of the supply side. (5) This management should be aimed both at limiting unnecessary services provided (and promoting necessary services) and at altering the mix of inputs used in provider institutions (or mix of institutions used); existing incentive patterns clearly discourage efficient production of both hospital and medical care. “Limiting unnecessary services” should include determination of the therapeutic effectiveness of procedures and products, and conscious effort to steer providers away from ineffectual activities by financial incentive or regulation. (6) It is doubtful that “management at arm’s length” is possible. Independ- ent provider organizations subjected only to financial restraint and direct regulation can usually modify the information flow to the regulator and/or operate in the political arena so as to frustrate regulation. Public management of health systems may be unavoidable, if unattractive, in the long run. “Management at two arms’ length,” that is independent providers reimbursed by independent insurance carriers, is, of course, doubly difficult. Initially, some Canadian provinces attempted this system in medical care—it was rapidly abandoned as being impractical and unworkable. It was urged by some provider organizations, perhaps because it rendered effective management impossible. (7) Canadian experience tells little about appropriate mechanisms for determining the relative income status of health workers, which becomes an administrative necessity when health care is removed from the market. Depending on how one reads the evidence, either we have been fortunate in 74 that income claims have stabilized, or we are still on the escalator—in any case we have no answers. (8) National health insurance per se seems to have some desirable effects on the distribution of access to care by income groups, but it does not have much impact on access problems more broadly defined (for example, geographic or social-class barriers). Nor has it had any effect on quality of care that we know of (since we have not looked). Presumably the ability to set conditions of participation creates levers to encourage quality upgrading, but this has not yet been tried. What Canadian experience may show is only that access improves if you get many more doctors, for eventually some get pushed to less desirable areas; and this is an expensive process. (9) There is really no end to the good advice you can get for asking. 75 NOTES AND REFERENCES l. The best single description of the Canadian system is M. LéCLAIR “The 76 Canadian Health Insurance Program”, In S. Andreopoulos (ed.) National Health Insurance: Can We Learn from Canada? New York: John Wiley, 1975. The specific features of the individual provincial plans are described in CANADA, DEPARTMENT OF NATIONAL HEALTH and WELFARE, Social Security in Canada, Ottawa, 1974. An alternative recent description and analysis of the Canadian health insurance system focusing more on statistics and less on institutions than the LeClair paper is R. G. EVANS, “Beyond the Medical Marketplace: Expenditure, Utilization, and Pricing of Insured Health Care in Canada,” In R. Rosett, (ed.), The Role ofHealth Insurance in the Health Services Sector, Universities-N.B.E.R. Conference Volume, New York; Columbia University Press, 1975, (reprinted in S. Andreopoulos, National Health Insurance, op. cit.), which includes documentation of data sources and a partial list of additional descriptive material. The list is perhaps more exhausting than exhaustive. . In addition, the funding of such plans is slowly evolving away from the use of premiums and occasional minor point-of-service charges (e.g. dollar-a- day hospital bed charges) to complete general revenue funding, but the pattern varies from province to province. In an inflationary world, failure to raise premiums leads to phase-out, although some provinces are specifically committed to elimination. Social Security in Canada, op. cit. details the variety of different provincial arrangements as of 1974, but the trend is the same in all provinces. . Population and G.N.E. data are from CANADA, DOMINION BUREAU OF STATISTICS, Canadian Statistical Review, Historical Summary 1970, Ottawa, August 1972. All expenditure data are in $ Canadian, which has generally moved close to par with the $US in recent years. Over the period from 1953 to 1971 it moved by less than 10 percent in either direction from the US. $ and is currently about 4 percent below. . The “insurance” label for such programs is rather dubious since premiums and charges to patients are compulsory and are in no way related to risk. They represent a tax system component which appears to be transitional. “Insurance” merely distinguishes these public programs from direct public services. . The original is, of Acourse, the Rapport Du Commission d’Enquete sur la Sante et le Bien-Etre (Castonguay Commission) Quebec, 1970 which brought the community health center concept to the forefront of national attention. The Report of the federal Community Health Center Project (Hastings Report) Ottawa, 1972 strongly endorsed the concept of community-based salaries group practices, while emphasizing that the details of such structures would vary across provinces and regions. The Report of the British Columbia Health Security Program Project, (Foulkes Report) Victoria, 1973 reached similar conclusions, as have studies in Manitoba and Saskatchewan, and most recently the Report of the Nova 10. ll. l2. l3. 14. 15. Scotia Royal Commission on Education, Public Services, and Provincial- Municipal Relations (Graham Report) Halifax, 1974 calls for provincial takeover of ownership and management of all hospitals and health-related facilities and the development of alternative payment modes to fee-for- service for physicians. . Nova Scotia, Graham Report, op. cit., Vol. II, Chapter 9, pp. 97 and 98. . Some analysis of the pre—Medicare physician insurance data has been carried out in R. G. EVANS, Price Formation in the Market for Physicians’ Services, 195 7-69, Ottawa, Prices and Incomes Commission, 1973, based on the data from the affiliated non-profit provincial plans making up Trans Canada Medical Plans. Particularly interesting is the finding that per capita expenditures for persons with comprehensive first dollar coverage in such plans was little different from national average rates, though, of course, populations are not standardized. Total medical insurance coverage was, of course, much higher, an earlier study by the Department of National Health and Welfare shows coverage at nearly 50 percent by 1961. But this coverage was partial as well as comprehensive and included for-profit company plans with a variety of deterrent features. (DEPARTMENT OF NATIONAL HEALTH AND WELFARE (1963), Voluntary Medical Insurance in Canada, 1953-61, Ottawa.) . The choice of 1959 as a cut—point is justified by the fact that the two largest provinces, Ontario and Quebec, entered in 1959 and 1961. Four other provinces had plans predating the federal legislation; only one initiated a plan in 1958. . CANADA, Report of the Royal Commission on Health Services, (1964) (Hall Commission), Ottawa. CANADA, DEPARTMENT OF NATIONAL HEALTH AND WELFARE (1974), Canada Health Manpower Inventory, 1974, Ottawa, n.d. CANADA, DEPARTMENT OF NATIONAL HEALTH AND WELFARE, Earnings of Physicians in Canada (annual) various issues, Ottawa. Data assembled in R. G. Evans, “Beyond the Medical Marketplace,” op. cit. Data used in subsequent discussions are drawn from a variety of sources, assembled and documented more extensively in R. G. EVANS, “Beyond the Medical Marketplace,” op. cit. See for example, R. G. EVANS, Price Formation, op. cit. The period 1958-62 rather than 1958-61 is reported because the official data seem to embody a mis-classification error in 1961, see R. G. EVANS, “Beyond the Medical Marketplace,” op. cit., and are suspect. LEWIS, C. E. and KEAIRNES, H. W. (1970), “Controlling Costs of Medical Care by Expanding Insurance Coverage: Study of a Paradox,” in New Eng. J. Med, Vol. 282, No.25. 77 16. 17. 18. 19. 20. 21. 22. 23. 78 Some Nova Scotia data presented in the Graham Report (op. cit.) Chapter 9, p. 72 suggest a backlog effect—rising utilization in the year of Medicare followed by a drop—but these are not conclusive. ENTERLINE, P. E. et al. (1973), “The Distribution of Medical Services Be- fore and After ‘Free’ Medical Care—the Quebec Experience,” in New Eng- land Journal of Medicine, Vol. 289, No. 22. Data on rates of performance of specific services are reported in CANADA (1974), DEPARTMENT OF NATIONAL HEALTH AND WELFARE, Physicians’ Services Provided under Medical Care Insurance Programs of Five Canadian Provinces 1970-71 and 1971-72, Ottawa, n.d. Unpublished fee schedule index data are prepared by the Department of National Health and Welfare, Health Economics and Statistics Division. Some analysis of this data is reported in both EVANS, Price Formation, op. cit. and EVANS, “Beyond the Medical Marketplace,” op. cit. Hospital statistics here and subsequently are based on CANADA, DOMINION BUREAU OF STATISTICS, Hospital Indicators January- December 1971 and 1973, Ottawa, 1972 and 1974. These are based on a partial quarterly survey of hospitals, and their reported data are thus estimates of the data which will be reported subsequently in the Hospital Statistics series. The 1971 data do not exactly match those reported elsewhere in this paper, but 1971—73 changes are reported from the quarterly indicators data to maintain internal consistency. Recent popula- tion and personal income data are drawn from the DES. Canadian Statistical Review (Monthly) Ottawa, April 1975. CANADA (I975), DEPARTMENT OF NATIONAL HEALTH AND WELFARE, Health Economics and Statistics Division, Salaries and Wages in Canadian Hospitals 1969 to 1973, Ottawa. CANADA, DEPARTMENT OF NATIONAL HEALTH AND WELFARE, Health Economics and Statistics Division, unpublished data. Canadian studies tend to bear out US. experience, that non-fee-for—service medical practitioners in groups generate significantly lower rates of hospitalization from similar populations. Studies in Saskatchewan by D. O. ANDERSON, What Price Group Practice?, University of British Columbia, Vancouver, 1973 and J. L. McPHEE, Community Health Association Clinics: an Evaluation Covering the Period April I, 1972 to March 31, 1973, Saskatchewan Department of Public Health, Regina, 1973, and in Ontario by J. E. F. HASTINGS et al., “Prepaid Group Practice in Sault Ste. Marie, Ontario: Part I”, in Medical Care, March-April 1973, show reductions in patient-day rates of the order of 20-25 percent. The initial plan of this program is described in SASKATCHEWAN, DEPARTMENT OF PUBLIC HEALTH (1972), A Proposal for a Dental Program for Children in Saskatchewan, Regina. Now that the program is in the field, of course, changes have been made, but these are not yet documented. 24. 25. 26. 27. A recent evaluation of a number of alternative children’s dental care programs ranging from a pure private-practice-based model to a system somewhat similar to that of Saskatchewan has been conducted by a representative committee funded by the Government of British Columbia and the College of Dental Surgeons of B.C., Children’s Dental Health Research Project, Report, Victoria; the Queen’s Printer, 1975. ARROW, K. J. (1973), “Uncertainty and the Welfare Economics of Medical Care,” in American Economic Review, Vol. 53, No. 5. References abound, but in particular P. J. FELDSTEIN (1973), Financing Dental Care: An Economic Analysis,‘Lexington, Mass., D.C. Health; and H. D. WALKER, (1971) Market Power and Price Levels in the Ethical Drug Industry, Bloomington, Indiana University Press. The very recent rapid growth of private dental insurance plans on a group basis is likely to make the market failure problem yet more acute by reducing the price-responsiveness of demand even further. Whether optimum short-run strategy by supplier organizations is to respond to such plans by moderating fee increases to defer government intervention or to move fees up as fast as possible in order to establish a high base for public plan reimbursement remains to be seen. 79 DISCUSSANT’S COMMENTS RODERICK D. FRASER Queens University, Kingston, Ontario, Canada I think Bob has touched on most of the points that should be touched on. Where I disagree principally is in the interpretation of what has happened and as to where we should go from here. Before dealing with the concerns that I have with the conclusions that he draws—either explicitly or otherwise—I would like to supplement what he has said in four ways. There is a set of statistics handed out under the title Predicted, Actual and Residual Values of Infant Mortality for 25 Countries, I950, 1960, and 1965 (Table 4). It is with respect to this and other tables that I comment. First of all, with respect to physician incomes and Tables 5 and 6, there is a presentation of relative income data taken from our taxation statistics. These data thus have the warts of taxation statistics data. In spite of the warts, the tables do, I believe, provide an opportunity to compare various professional groups with physicians and surgeons over the time period that we have been dealing with, namely from the introduction of hospital insurance plans around the year 1960 and then the medical insurance programs after 1965. I think in spite of the warts of the data there are some pretty clear conclusions one can draw about the significant jumps in the relative incomes of physicians. The second set of supplemental data deals with the impact of the hospitalization plans. Bob mentioned in this paper that, as far as one could see, there was very little impact on the utilization of hospitals by the introduction of the plans themselves. Supportive of this conclusion are the results presented in Tables 7 and 8. These are the results of a fairly simple statistical analysis of the impact of utilization as seen either in patient-days of care or number of admissions per 100 people in Canada and each province in the period 1946-1964.* The introduction of the insurance plan is controlled for by the introduction of a dummy variable. I think you can see from these results that, with respect to the Canadian data, the estimated regression coefficient for the dummy variable is negative though insignificant. With respect to the individual provinces, there are some 16 positive coefficients with respect to the dummy variable; however, only four of those are significant. These results are thus consistent with the conclusion that the hospital plans per se have had little impact on utilization. However, it should be stressed that these are only two aspects of utilization. Two other important elements are the *These are preliminary results of my ongoing study of the impact of the hospital and medical care programs. 81 Z8 Table 4 PREDICTED, ACTUAL AND RESIDUAL VALUES OF INFANT MORTALITY FOR 25 COUNTRIES, 1950, 1960, AND 1965 Count 1955 1960 1965 ry Actual Predicted Residual Rank Actual Predicted Residual Rank Actual Predicted Residual Rank Bulgaria ........... 80.2 57.5 22.7 23 46.3 52.4 -6.1 9 32.0 48.8 -16.8 1 Finland ........... 28.7 44.2 —15.5 2 21.8 38.0 -16.2 2 16.5 31.0 -14.5 2 Norway ........... 21.1 39.8 -18.7 1 18.5 36.3 -17.8 1 16.8 30.7 -13.9 3 Denmark .......... 25.7 36.6 -10.9 4 21.9 33.4 -11.5 3 18.1 28.9 -10.8 4 Japan ............. 41.7 44.0 -2.3 9 31.0 36.9 -5.9 10 19.4 29.2 —9.8 5 Israel ............. 38.8 44.1 -5.3 7 30.0 41.3 -11.3 4 27.0 34.8 -7.8 6 New Zealand ........ 23.9 34.4 -10.5 5 23.1 30.3 -7.2 6 18.8 26.6 -7.8 7 Belgium ........... 40.5 41.6 -1.1 10 29.9 38.4 -8.5 5 25.0 32.1 -7.1 8 Netherlands ......... 21.5 28.2 -6.7 6 17.7 24.6 -6.9 7 14.6 21.3 -6.7 9 U.S.S.R ............ 58.0 48.9 9.1 20 35.8 40.9 -5.1 11 27.5 34.2 -6.7 10 Romania .......... 82.8 56.2 26.6 24 74.7 53.5 21.2 25 46.4 50.2 —3.8 11 Poland ............ 78.2 51.1 27.1 25 60.9 49.5 11.4 24 41.8 44.4 -2.6 12 Italy ............. 50.9 47.9 3.0 11 43.3 43.1 0.2 13 35.5 38.0 -2.5 13 Sweden ............ 17.8 31.4 -13.6 3 16.3 22.5 -6.2 8 13.4 15.6 -2.2 14 G.F.R ............. 41.1 32.7 8.4 16 33.3 28.2 5.1 19 24.2 24.8 -0.6 15 Hungary ........... 59.8 48.1 11.7 21 48.0 44.3 3.7 17 38.8 39.3 -0.5 16 France ............ 38.5 29.4 9.1 19 27.5 29.3 -1.8 12 22.3 20.2 2.1 17 Australia .......... 22.1 24.5 -2.4 8 20.4 18.8 1.6 15 18.6 15.7 2.9 18 Ireland ............ 36.7 27.7 9.0 18 30.6 25.1 5.5 20 25.6 22.3 3.3 19 Austria ............ 45.7 31.2 14.5 22 36.7 27.7 9.0 22 28.5 24.4 4.1 20 Czechoslovakia ...... 34.4 29.6 4.8 13 24.0 22.8 1.2 14 23.5 19.0 4.5 21 Switzerland ........ 26.5 22.3 4.2 12 21.4 19.5 1.9 16 17.8 12.6 5.2 22 Canada ............ 31.7 25.4 6.3 15 27.6 22.8 4.8 18 23.8 17.2 6.6 23 United Kingdom ...... 24.7 19.8 4.9 14 21.9 16.1 5.8 21 19.3 8.8 10.5 24 United States ....... 26.3 17.6 8.7 17. 25.9 16.1 9.8 23 24.4 9.6 14,778 25 Table 5 AVERAGE INCOME* OF SELECTED PROFESSIONAL GROUPS RELATIVE TO THAT OF ALL TAXPAYERS, ONTARIO, 19501972 Occupations 1950 1955 1960 1965 1970 1972 (1) Physicians and surgeons. . . 3.7 3.9 4.0 4.9 5.8 5.2 (2) Dentists ............. 2.2 2.5 3.0 3.4 3.7 3.8 (3) Lawyers ............. 3.7 4.4 4.0. 4.3 4.8 4.2 (4) Consulting engineers and architects ........... 4.3 3.6 3.6 3.7 3.3 2.6 (5) Total professionals ...... 2.6 2.8 2.9 3.2 3.6 3.2 *Average total income assessed for tax purposes. Source: Table 11-1 and in turn Canada, Department of National Revenue, Taxation Statistics, 1952 through 1974. Table 6 AVERAGE INCOME* 0F SELECTED PROFESSIONAL GROUPS RELATIVE TO THAT OF ALL TAXPAYERS, CANADA, 1950-I972 Occupations 1950 1955 1960 1965 1970 1972 (1) Physicians and surgeons . . 3.3 3.4 3.9 4.7 5.4 5.3 (2) Dentists ............. 2.1 2.4 2.9 3.2 3.5 3.6 (3) Lawyers ............. 3.3 3.5 3.5 3.9 4.2 3.9 (4) Consulting engineers and architects ........... 3.7 4.0 3.7 3.9 3.5 3.3 (5) Total professionals ...... 2.4 2.6 2.9 3.2 3.4 3.4 *Average total income assessed for tax purposes. Source: Table 11-2 and in turn Canada, Department of National Revenue, Taxation Statistics, 1952 through 1974. switching that probably has occurred from low cost-low technology hospitals to high cost-high technology hospitals, and within hospitals the probable increased bundle of resources allocated to the care of a patient in a given patient day. In effect, as yet we know little about the overall impact that the hospitalization and medical care plans have had on “utilization.” I have looked at the geographical differences in the levels of different rates of mortality, infant mortality, standardized age-adjusted death rate, and so on. There is no readily discernible impact on the geographical variation in 83 mortality as one comes forward in time from the introduction of the hospitalization and medical care plans through the ’sixties and up until about 1972. With respect to the distribution of health-care resources, some evidence produced for eastern Ontario suggests that there has been some “shift” of physicians from rural areas and into urban centers. A full analysis awaits the detailed results of our 1971 census. The hypothesis that I would entertain is that with the introduction of hospital insurance and, in particular, medical insurance plans, the physician is no longer faced with the constraint of having to generate consumer dollars to back the demand of those consumers, and thus is able to make his decision on location with respect to other variables in his preference function. As a result my expectation would be that you would find physicians setting up practices in those socioeconomic areas in which they want to, on the assumption that they are likely to be able to generate enough demand on the part of their patients to provide them with the income that they would like to have. Such areas, with the removal of the constraint of demand backed by consumer dollars, are likely to be the better geographical, socioeconomic areas. My final reference is to a study that I did on international systems of financing health care. (1 ) Briefly, it was a study of 25 well-developed countries. A very simple framework was used in an attempt‘to relate the inputs of the health sector to its outputs and, at the same time, to control for some of the demographic, geographic, and socioeconomic factors that one expects to affect levels of health status. The prime measure of health status is infant mortality. The results are thus clearly circumscribed by having only that one measure of mortality and by not having measures of other aspects of the output of the health-care sectors. Taking this prime limitation into consideration, we can note the residuals obtained through the simple statistical analysis. The chart (Figure 1) describes the same results. In this figure Canada’s position is illustrated about mid-page; the dashed lines are the levels of infant mortality that one would predict given the information about the input of health care resources and the socio- economic demographic yariables that were controlled for in this study; the solid line gives the actual levels. It is interesting to note that the middle dots refer to 1960 and the final to 1965. Over that period the position of Canada actually appeared to get worse. The first of the two principal concerns I have is with the impression thatI think Bob Evans has left you with, namely, that we should all forget about doing anything with the demand side and should concentrate on further intervention or in setting up more detailed regulations. There is nothing like the objective base of knowledge of the impact of the plans that would justify such a conclusion. I am not at all convinced that we have exhausted what we can do in having consumers play some role in influencing the level of resource use. ‘ Though the utilization data that I have just referred to are as they are, they do in fact simply refer to patient-days and admissions. The demand by consumers for some of the high-cost technologies—be they in terms of the 84 specialist vis-a-vis the GP, or in terms of hospital care vis-‘a-vis their local physician’s office—cannot be ignored. To my mind, those demands are real and have had quite an influence on the generation of costs in the health-care sector in Canada. Infant Mortality Per 1,000 Live Births 10 40 30 20— 10 Switzerland ~\ Belgium Israel \- Ireland Czechoslovakia France \ \ New Zealand \\ Denmark >— \\ Australia Norway "\ Finland 0. ‘\\ \\ \\ Netherlands \\ Sweden 19515 1960 19651955 1960 19651955 1960 19651955 1960 196519551960 1965 Actual """ Predicted *""""' U.S. White Population Only Figure 1. Actual and predicted values of infant mortality per 1,000 live births in 1955, 1960 and 1965. 85 The second concern is with respect to a reference that Bob made in his written paper—with respect to the fee-for-service versus salary debate, he seems to believe that we have said all we can with respect to that debate. To the contrary, it seems to me that we have not. We have in Canada, certainly in the last couple of years, witnessed tremendous power (on the part of different occupational groups in the health-care sector) to raise wages significantly above those of other occupational groups irrespective of whether they are fee-for- service or salaried. The power of the strike threat in the health-care sector isI believe, especially strong. In addition, I think as long as one has to live with physicians practicing alone, the costs of setting up a monitoring system to make judgments with respect to salary changes, and so on, would be enormous, and the ensuing problems at least as big, if not greater, than those of trying to control the claimed over-provision of services under the fee-for-service schedule. However, the same conclusion does not hold with such force when one moves to an institutional setting such as the hospital. But then, we might develop the analogy that was put forward this morning with respect to the university system, and how we behave in universities, and how might we expect physicians to behave in the hospital sector. It seems to me that moving to a salary system in the hospital sector is likely to give rise to the same use of discretionary power on the part of individuals to allocate their time to things other than the provision of services, in the same way that professors are said to allocate their time to things other than teaching classes. Furthermore, the demands of some of the salaried physicians for sophisticated equipment is, I think, a demand that places extremely high costs on the health-care system that we have today. Lastly, I would like to underline two things that Bob mentioned in his conclusions; very firmly agree with them. Dorothy Rice in her paper this morning talked about the objectives of the United States in going into health-care insurance, and one of them was that there was the expectation that there might be greater leverage thereby to change the supply side. I would firmly agree with Bob that you lose a lot of your leverage once you move into that political arena where it just becomes, in my mind, ever more difficult for politicians to make decisions that go against established patterns of practice. The other thing I would like to underline is the statement that we really do not have a lot of information on the therapeutic effectiveness of the bulk of the things we do in the health-care sector, and it seems to me that this is one of the most important things we need to know before we start committing ourselves, either in Canada to our Pharmacare programs or Dentacare programs, or in a country such, as the United States with the Medicare program. Bob has detailed, and there are others who have detailed, the problems of the Canadian system, and they are horrendous problems. Ultimately you have to judge whether the problems that you have with the present system are worse than the ones you are likely to get if you go quickly into a national health insurance program. 86 L8 Table 7 SIMPLE CORRELATION COEFFICIENTS AMONG UTILIZATION VARIABLES AND POSTULATED DETERMINANTS, CANADA, 1946-1966 Number of Observations: 21 .................. Y1 Y2 X1 X2 X3 X4 X5 X6 X7 Y1 Patient Days Per Capita ...................... — .955 .813 .834 .420 -.795 .935 .723 .648 Y; Admissions Per 100 Persons ................... — .824 .778 .547 -.701 .922 .661 .771 X1 Ratio of Female to Male Population ............. — .891 .351 -.783 .847 .792 .491 X2 Personal Disposable Income Per Capita (Constant 1957 Dollars) ........................... -— .013 -.885 .904 .850 .372 -X3 Percent of Population Aged 0 to 4 or Over 65 ....... — -.029 .259 .046 .629 X4 Other Hospital Beds Per 10,000 Persons ........... — -.761 —.740 -. 144 X5 Physicians Per 10,000 Persons .................. — .817 .647 X6 Provincial Hospitalization Insurance Dummy ........ — .313 X7 Total Hospital Beds Per 10,000 Persons Significant values ofr are: 0.423 (P = 0.05) 0.492 (P = 0.02) 0.537 (P = 0.01) Source: See text. 88 Table 8 MULTIPLE REGRESSION OF UTILIZATION VARIABLES AND POSTULATED DETERMINANTS: CANADA AND THE PROVINCES, 1946-1966 . -2 Durbin- Equatlon R F-value Watson Canada. . .. Yl =12.904— 16.598xl + .001x2 +13.824x3 — .188X4 +.202x5 — .010x.5 .948 61.947 2.898 (~2.319) (1.377) (3.598) (—2.836) (3.080) (—.217) Y2 =10.118— 38.997xl + .003x2 +100.115x3 — .704x4 + 2.044x5 — .552x6 .957 75.046 2.034 (—.762) (.666) (3.696) (—1.487) (4.372) (— 1.615) Prince Y1 = 0.696— 2.671xl + .ooox2 +13.647x3 + .030x4 — .017x5 +.186x6 .778 3.160 1.813 Edward (— 1.498) (.845) (2.935) (.651) (.225) (1.140) [Sland' ~ - - Y2 =—17.273 + 1.602xl + .oosx2 + 85.356x3 + 575x4 — .428x5 +1.363x.5 .730 10.016 1.695 (.102) (1.176) (2.078) (1.421) (.645) (.948) Nova Scotia Y1 =-9.618 +6.141x1 + .001x2 +16.507x3 + .039x4 + .0175 + .101x6 .923 41.189 1.695 (1.875) (3.367) (3.207) (1.610) (.363) (1.815) Y2 =—88.966 + 39.331xl + .018x2 + 249.205x3 + .210x4 — .769x5 — .009x6 .911 35.008 1.606 (1.361) (5.544) (5.484) (.987) (-1.849) (—.o19) New Y1 = 3.910— 3.234x1 + .ooox2 — 2.621X3 — .040x4 +.171x5 + .276x6 .972 117.808 2.336 Brunswick (— 1.949) (.890) (—.822) (— 1.362) (4.543) (6.361) Y2 = 28.028 — 12.937x1 — .004x2 — 92.840x3 — .786X4 + 3.503x5 + .492x6 .920 39.338 1.340 (-.535) (-l.042) (—1.998) (71.835) (6.400) (.777) Bracketed figures are t-values 68 Table 8—Continued MULTIPLE REGRESSION 0F UTILIZATION VARIABLES AND POSTULATED DETERMINANTS: CANADA AND THE PROVINCES, 1946-1966 . -2 _ Durbin- Equatlon R F value Watson Quebec. . .. Y1=11.900- 15.7O4X1-.002X2 +16.959X3 - .021X4 + .001X5 - .055X6 .719 9.519 1.852 (-1.014) (2.428) (1.882) (-.189) (.009) (-.317) Y; =-5.567 — 21.681X1 + .010X2 +112.983X3 - .264X4 + .678X5 + .095X6 .894 29.218 1.777 (-.312) (2.827) (2.791) (.535) (1.000) (.122) Ontario . . . Y1 =4.921- 7.618X1+.001X2 + 6.849X3 +.016X4 +.085X5 +.059X6 .926 42.440 2.576 (-.958) (4.104) (1.677) (.249) (1.292) (1.051) Y2 = 9.456- 38.596X1+.008X2 +104.955X3 +.326X4 + .713X5 - .173X5 .971 112.920 1.606 (- 1.245) (6.469) (6.591) (1.348) (2.785) (-.796) I Manitoba .. Y1= 5.603 - 4.685X1+.001X2 + 2.863X3 - .101X4 + .027X5 + .125X6 .838 18.181 1.749 (-l.124) (.819) (1.383) (-l.795) (.343) (1.156) - , Y2 =-35.647 +13.919X1 + .007X2 + 88.143X3 + .860X4 + .533X5 + .380X6 .956 73.217 2.085 (.747) (4.416) (9.522) (3.424) (1.526) (.786) Saskatchewan Y1 =-4.563 + 6.639X1 +.000X2 + 5.094X3 - .026X4 — .055X5 +.363X6 .830 17.267 1.716 (1.861) (-.050) (1.972) (-.863) (-l.254) (3.465) Y2 = 43.483 + 54.043X1 - .001X2 + 66.255X3 - .088X4 - .109X5 + 3.013X5 .857 20.988 1.720 (1.456) (-.589) (2.466) (-.279) (-.237) (2.765) 06 Table 8—Continued MULTIPLE REGRESSION 0F UTILIZATION VARIABLES AND POSTULATED DETERMINANTS: CANADA AND THE PROVINCES, 1946—1966 . —2 Durbin— Equat1on R F-value Watson Alberta. . . . Y1 = 0.581 - .638Xx + .001X2 +1.166X3 + .053X4 + .094X5 + .208X6 .934 48.549 1.265 (-.264) (2.981) (.924) (1.992) (1.444) (3.876) Y3 =— 15.486 - 4.009X1 + .002X2 + 92.115X3 + .246X4 +1.689X5 - 1.907X5 .777 12.611 1.133 (—.106) (1.160) (4.642) (.586) (1.302) (-2.265) British Y1 = 4.146 — 3.666X1 + .003X2 - .958X3 + .026X4 + .054X5 +.104X6 .275 2.264 2.390 Columbia (-1.611) (1.281) (-.564) (1.117) (1.362) (1.097) Y2 =-11.638 +19.036X1+.000X; + 4.511X3 + .005X4 + .672X5 .132X6 .868 22.960 2.189 (1.743) (.058) (.553) (.048) (3.564) (.290) Newfound- Y1 = 5.273 - 7.017X1 + .002X2 + 6.941X3 + .027X4 — .OOOXS + .006X5 .813 10.447 1.564 1and*. . . . (*1.139) (2.589) (1.092) (.581) (—.005) (.124) Y; =-72.664+101.327X1- .002X2 - 64.772X3 .561X4 + .690X5 + .292X6 .958 50.180 V 2.844 (2.703) (-.491) (- 1.674) (—1.957) (1.685) (.965) *Newfoundland Data Covers Only The Period 1953 to 1966. Source: See text. REFERENCE 1. FRASER, R. D. (1973), “Health and general systems of financing health care revisited,” International Journal ofHealth Services, 3, 3, pp. 369-97. 91 Chapter 5 HEALTH COSTS AND EXPENDITURES IN THE UNITED KINGDOM MICHAEL H. COOPER Donald Reid Professor of Economics, University of Otago, New Zealand The original 'objective of the United Kingdom’s National Health Service (NHS) was to establish access to health care for all those in genuine need as a “human right.” This right was to become a reality by the nationalization of all health-care resources and their subsequent provision at zero or near zero prices at the point of consumption. The political assumption underlying this policy was that the need for health care was finite and that it was possible for society to divert sufficient resources from private goods and services to ensure adequacy of provision. In practice, 27 years’ experience of working a near zero priced system has shown that the need for medical treatment is virtually limitless, and that no society can afford to provide all the services that are technologically possible at any given moment in time. Indeed, assessment of need by the professions tends to gravitate towards whatever level of provision is being currently made available. The basic insatiability of need has led to continuous demands for more finance and manpower and to almost annual forecasts of the system’s imminent collapse. As one ex-minister has observed, “One of the most striking features of the NHS is the continual, deafening chorus of complaint which rises day and night from every part of it.” (I) The basic misunderstanding as to the true nature of need has produced a system which is extremely popular with its customers but reviled by its suppliers. Doctors, nurses and other professional groups have found themselves in the front line of a system which could not deliver what it had seemingly promised. Having set out to provide the impossible, namely the elimination of unmet need, the professions have increasingly found themselves fulfilling the role of assessing relative needs and rationing scarce health-care resources amongst them. Many health professionals still believe that all those in “need” of help should get it. They feel that their conception of adequacy is the only legitimate one (even if they differ amongst themselves as to what it is) and that political decision-makers are blind as to the real shortfall between current provision and real needs. It is basically this conflict which has produced a history of professional discontent. (2) Incremental Planning A careful investigation of health-service spending reveals that by far the best explanation of the pattern of expenditure is the service’s original inheritance. 93 The priorities implied by the prenationalization distribution of manpower, finance and capital resources have for the most part been preserved intact. On the appointed day (July 5, 1948) resources were allocated to reflect the old order and it has since proved extremely difficult to radically change them. As Enoch Powell once remarked, “Once the butter is in the dog’s mouth it can only be scraped out at the risk of getting bitten.” As late as 1969, Richard Crossrnan (3) still regarded geographical and other disparities as the single biggest problem the system faced. To put right such disparities imone region or speciality, would have involved the “political odium of being seen to reduce expenditure” in another. (1) Clearly, once established, priorities cannot be changed quickly in any service industry. Manpower has specific skills, a long gestation period and is . geographically immobile. Moreover, the objectives of the NHS have never been stated in a form readily translatable into policy, and criteria by which to attest their attainment have consequently remained obscure. Added to this, all previous expenditure has a momentum of its own (yesterday’s capital spending commits tomorrow’s current expenditure, etcetera) and so it is not surprising that the Treasury’s approach has been almost entirely incremental—a process described by Klein (3) as “fiscal blind man’s buf .” A clear indication of the extent to which expenditure reflects and responds to external forces rather than to deliberate planning is given by comparing the 1973-74 estimating error for public spending of £2,356M with the planned changes of only £1 ,227M. As Sir William Armstrong has pointed out, the scope for such changes in direction is extremely limited and probably at no time is more than 2% percent of current‘spending. In 1968 Bierman (5) and his colleagues from the U.S. National Institutes of Health concluded that “despite 20 years of public control the NHS has failed to evolve an effective planning mechanism.” One very distinguished Chief Medical Officer has admitted that NHS planning amounts to “the use of last year’s budget with a bit added here and a bit taken off there—we never ask ourselves the big questions.” ( 6 ) Basically, planning has amounted to last year’s expenditure plus N percent plus the square of recent public scandals. Supply In practice, real expenditure upon the NHS increased 130 percent between 1953 and 1972. Total public spending over the same time period grew only 83 percent. Health care, although growing faster than total public spending, fared rather worse than education (240 percent), personal social services (399 percent) and social security benefits (160 percent). The relatively favorable treatment of all social spending was largely made possible by a decline in defense spending and in subsidies of one kind and another. Defense was 25 percent of public expenditure in 1953 but only 11 percent by 1972. Health care grew from 7.8 to 9.7 percent, while education grew from 6.9 to 12.9 percent (Table l). The bulk of the NHS expenditure goes on the hospital system (Table 3). Its share of current spending increased from 55 percent in 1950 to 66 percent by 1973. The general medical services bill has declined from 12 to 7 percent, and 94 ophthalmic services from 5 to 1 percent. The labor bill dominates hospital costs, accounting for 58 percent in 1949 and 71 percent in 1973 (Table 4). In contrast, pharmaceuticals formed only 3.4 percent of hospital costs in 1949 and this had declined to 2.6 percent by 1973. In terms of the average weekly costs of keeping a patient in hospital, nursing is the biggest single item at 28 percent. The non-treatment departments, however, in aggregate account for 40 percent in contrast to only 5.8 percent for doctors’ services. It is difficult to imagine any private free-for-item-of-service system in which the average cost of medical supervision in hospital came to less than the NHS figure of £3.98 a week; which is less in fact than the combined cost of hospital portering and laundry (Table 5). Table 1 ANALYSIS OF PUBLIC EXPENDITURE 1953-72 Relative share of Real growth public spending 1953-72 % % 1953 1972 National Health Service ............. 129.4 7.8 9.7 Defense ........................ -14.8 24.5 11.4 External relations ................. 86.1 1.3 1.3 Road, transport & communications ..... 196.1 4.4 7.2 Industry and trade ................. 215.1 5.0 8.5 Research ....................... 33 3.3 0.3 0.8 Agriculture and fishery .............. -34.2 4.7 1.7 Housing ........................ 15.1 8.5 5.3 Environmental Services .............. 164.3 3.4 4.9 Libraries, etcetera ................. 31 1.1 0.2 0.5 Law and order .................... 92.7 3.0 3.2 Education ...................... 242.4 6.9 12. 9 Personal social services .............. 398.5 0.5 1.4 Welfare food ..................... -5.4 1.2 0.6 Social security benefits .............. 159.2 13.3 18.9 Other services .................... 63. 2 3.0 2.7 Debt interest ..................... 36.2 12.0 9.0 Total ......................... 82.9 100.0 100.0 Source: Based upon R. Klein (1974), Social Policy and Public Expenditure, Tables 3 and 4. Centre for Studies in Social Policy, 1974.(4) 95 Table 2 GROSS COST OF THE NHS AND THE COST AS A PROPORTION OF" NATIONAL INCOME, 1950-73 (UK)1 Year Cost of NHS NHS as a % of f. million national income 1950 ...................... 477 4.42 1951 ...................... 500 4.22 1952 ...................... 523 4.09 1953 ...................... 548 3.98 1954 ...................... 567 3.89 1955 ...................... 607 3.91 1956 ...................... 662 3.93 1957 ...................... 721 4.04 1958 ...................... 764 4.11 1959 ...................... 828 4.24 1960 ...................... 902 4.33 1961 ...................... 981 4.40 1962 ...................... 1,025 4.41 1963 ...................... 1,092 4.42 1964 ...................... 1,186 4.46 1965 ...................... 1,308 4.62 1966 ...................... 1,434 4.85 1967 ...................... 1,594 5.12 1968 ...................... 1,741 5.23 1969* ..................... 1,886 5.33 1970 ...................... 2,083 5.38 1971 ...................... 2,369 5.48 1972 ...................... 2,732 5.68 1973 ...................... 3,l79 5.73 1 Includes current and capital expenditure by central and local government and NHS patient payments. *Change in definition of NHS from 1969 onwards. Certain local authority services transferred from NHS to Social Services. Manpower, over the period 1949-73; also showed marked increases in numbers (Table 6). Hospital doctors, nurses and administrative staff more than doubled, and general practitioners and dentists increased by 1 percent and 20 percent respectively. At the same time the population grew only 12 percent. Since 1971, hospital doctors have increased by 10 percent, nurses 8 percent and administrative staff by 15 percent. General practitioners and dentists are again less favored, with increases of 31/2 percent. Again, the increase in population is far more modest at 0.7 percent. The major growth in both time periods was in professional and technical staff—193 percent from 1949-73 and 11 percent from 1971-73. The number of beds provided by the NHS has declined since 1949 from 453,000 to 450,000 and from 10.3 to 9.2 per 1,000 population. All the 96 indicators of the work load have, however, increased considerably. The number of inpatients treated has doubled and the number of outpatients has grown by ' about 76 percent. Despite all these large increases in expenditure and manpower and the consequent increase in cases treated, cries of shortage have continued unabated, and the waiting list for admission to hospital remains relatively stable at half a million people (Table 7). This is rather more, in fact, than the total number of beds available. Collectively, since 1949 the medical profession appears to have reassessed its conception of need in line with changes in the levels of provision, and it will be interesting to see whether the impact of the Table 3 HEALTH SERVICES AS A PROPORTION OF TOTAL COST OF NHS 1950 TO 1972, UK . Pharma- General General General Local Year Hospital ceutical medical dental ophthalmic health Other Total servrces . . . . authority servrces servrces servrces servrces . servrces % % % % % % % % 1950 54.9 8.4 11.7 9.9 5.2 7.8 2.1 100 1951 56.0 9.8 11.0 7.8 2.8 8.4 4.2 100 1952 56.0 9.8 11.1 5.9 2.1 8.4 6.7 100 1953 55.3 9.5 10.8 5.5 2.2 8.9 7.8 100 1954 56.4 9.3 10.6 5.8 2.3 9.2 6.4 100 1955 57.3 9.6 10.2 6.3 2.5 8.7 5.4 100 1956 57.6 9.8 10.0 6.3 2.3 8.6 5.4 100 1957 57.0 9.7 10.3 6.4 2.2 8.7 5.7 100 1958 58.0 10.0 10.3 6.5 2.1 8.9 4.2 100 1959 57.4 10.1 9.7 6.5 2.1 9.3 4.9 100 1960 56.4 10.1 9.8 6.3 1.9 9.0 6.5 100 1961 56.8 9.8 9.0 6.2 1.8 9.3 7.1 100 1962 59.0 9.7 8.5 6.0 1.7 9.7 5.4 100 1963 60.1 10.1 8.3 5.7 1.6 9.9 4.3 100 1964 60.5 10.2 7.9 5.6 1.7 10.0 4.1 100 1965 60.5 11.1 7.8 5.1 1.6 10.2 3.7 100 1966 60.9 11.2 7.5 5.2 1.5 10.2 3.5 100 1967 59.9 10.6 7.9 5.0 1.4 10.7 4.5 100 1968 60.0 10.2 7.9 4.7 1.4 10.6 5.2 100 1969* 63.1 10.4 8.0 4.8 1.5 7.4 4.8 100 1970 64.2 10.0 8.3 4.9 1.4 7.0 4.4 100 1971 65.5 9.8 8.1 4.8 1.3 6.9 3.6 100 1972 66.0 9.7 7.9 4.5 1.7 6.8 3.9 100 1973 66.2 9.4 7.4 4.4 1.1 6.9 4.6 100 *Change in definition of NHS. Certain local authority services transferred from NHS to Social Services. 97 Table 4 REVENUE EXPENDITURE 0F NHS HOSPITALS BY BROAD CATEGORY 1949-50 AND 1972-73 % Total % Total 1949-50 1972-73 Salaries and wages ...................... 58.1 71.4 Provisions ............................ 10. 7 4.2 Uniforms and clothing ................... 1.0 0.6 Drugs and dressings ..................... 3.4 2.6 Appliances and equipment ................. 2.9 4.4 General services (laundry, power, etcetera) ...... 6.3 4.2 Maintenance of plant and grounds ............ I 3.4 2.5 Domestic repairs and renewals .............. 3.4 1.3 Central administration .................. ‘. . 2.4 2.9 Other ............................... 7.8 4.8 Total ............................. 99.4 98.9 Sources: Annual Report of the Ministry of Health for 1951, 1953, Cmnd. 8655, p. 8, and DHSS. consultants’ recent “work to rule” will produce a proportionate increase in the waiting times, or whether correspondingly fewer people will be referred by their general practitioners and fewer of those actually referred added to the waiting list by consultants. Priorities The political, as against professional, realization of the insatiability of need came very early. When the original bill proved to be more than double the Beveridge prediction and still showed no signs of meeting all legitimate needs, panic ensued and urgent measures were sought to apply the brakes. In consequence, the growth in expenditure in real terms actually declined in 1953. This rapid political awareness of the problem, however, established patterns of spending and priorities which have persisted throughout the life of the NHS. Four such patterns of spending are clearly discernible. First, curing has been favored rather than caring. Indeed the caring sections of the service have often appeared to be an unwanted burden which had to be tolerated but were in effect an interference in the real work of the service, namely curing the acutely sick. Translated into practice this has meant 98 Table 5 AVERAGE WEEKLY COST OF MAINTAINING AN INPATIENT IN AN ACUTE NONTEACHING HOSPITAL WITH OVER 100 BEDS IN 1969-70 AND 1970-71 1969-70 1970—71 £ % £ % Inpatients wards 1 Pay: medical ............ 3.10 5.4 3.98 5.8 nursing ............ 13.67 23.8 17.54 25.5 domestic ........... 2.77 4.8 3.09 4.5 other ............. 0.39 0.7 0.45 0.7 2 Drugs ................ 1.62 2.8 1.83 2.7 3 Dressings: prepacked ...... 0.47 0.8 0.18 0.3 other ......... 0.37 0.5 4 Patients’ appliances ........ 0.05 0.1 0.06 0.1 5 Equipment: major ........ 0.09 0.2 0.12 0.2 other — traditional 1.27 2.2 0.99 1.4 disposable 0. 50 0.1 6 Contract services ......... 0.08 0.1 0.09 0.1 7 Ward Total ............. (23.51) (40.92) (29.20) (42.51) Other treatment departments 1 Operating theaters ......... 5.70 9.9 7.03 10.2 2 Radiography ............ 0.14 0.2 0.19 0.3 3 Diagnostic X-rays ......... 1.01 1.8 1.15 1.7 4 Pathology .............. 1.93 3.4 2.34 3.4 5 Physiotherapy ........... 0.36 0.6 0.46 0.7 6 Pharmacy .............. 0.50 0.9 0.60 0.9 7 Ancillary medical service . . . . 0.51 0.9 0.66 1.0 8 Total ................. (10.15) (17.66) (12.43) (18.09) Nontreatment departments 1 Nurses in training ......... 0.81 1.4 1.05 1.5 2 Catering ............... 6.07 10.6 6.46 9.4 3 Staff residences .......... 2.15 3.7 2.30 3.3 4 Laundry ............... 1.20 2.1 1.28 1.9 5 Power, light and heat ....... 1.88 3.3 2.01 2.9 6 Building maintenance ...... 1.93 3.4 2.62 3.8 7 Medical records .......... 0.50l 0.9 0.59 0.9 8 General administration ...... 2.54; 4.4 2.92 4.3 9 General portering ......... 1.38? 2.3 1.59 2.3 10 General cleaning .......... ' 0.55 1.0 0.65 0.9 11 Maintenance of grounds ..... 0.23 0.4 0.26 0.4 12 Transport .............. 0.23 0.4 0.28 0.4 13 Other ................. 4.35 7.6 5.04 7.3 14 Total ................. (23.79) (41.40) (27.05) (39.38) Total .......... 57.45 68.68 99 Table 6 001 MANPOWER CHANGES 1949-73 (ENGLAND AND WALES) % Change 1949 1971 1973 1949-73 1971—73 Hospital Service Ancillary staff (porters, etcetera) ................ No. 157,112 239,112 238,883 52.0 -0.4 Professional and technical ..................... Wte. 13,940 36,817 40,858 193.0 11.0 Medical staff ............................. Wte. 11,735 23,806 26,152 122.9 9.9 (Consultants) ........................... (3,488) (8,655) (9,496) (172.2) (9.7) Dental staff .............................. Wte.’ 206 753 782 279.6 3.9 Nursing staff .............................. No. 137,636 288,065 310,760 125.8 7.9 Administrative and clerical ..................... No. 23,797 47,690 54,764 130.1 14.8 Regional Board staff ........................ No. 1,320 7,243 8,359 533.3 15.4 Executive Councils General practitioners ........................ No. 20,400 21,910 22,686 11.2 3.5 Dentists ................................. No. 9,495 10,962 11,374 19.8 3.8 Ophthalmic medical practitioners ................ No. 996 920 938 -6.2 2.0 Population ................................. (000) 43,785 48,815 49,174 12.3 0.7 Sources: Various reports of the Ministry and Department of Health 1950-72; Health and Personal Social Services Statistics for England, 1974, and Health and Social Statistics for Wales, 1974. Wte. = whole time equivalents. Table 7 WAITING LISTS PER 1,000 POPULATION AND WAITING TIMES— ENGLAND AND WALES . . . . . . Mean waiting Year Total'wartrng Tlptal wajrtrng 111st per times (weekS), list t ousan popu ation All causes“ 1949 ......... 497,700 11.37 1950 ......... 530,500 12.11 1951 ......... 503,600 11.49 1952 ......... 500,300 11.38 1953 ......... 525,900 11.92 1954 ......... 474,300 10.71 1955 ......... 454,900 . 10.24 1956 ......... 430,800 9.64 1957 ......... 440,300 9.80 1958 ......... 442,803 9.82 1959 ......... 475,626 10.48 1960 ......... 465,539 10.17 1961 ......... 474,177 ~ 10.26 1962 ......... 469,091 10.05 1963 ......... 475,834 10.13 1964 ......... 498,834 10.54 14.4 1965 ......... 517,142 10.84 14.5 1966 ......... 536,447 11.18 14.8 1967 ......... 537,005 11.12 14.2 1968 ......... 534,890 11.01 13.7 1969 ......... 561,365 11.50 14.0 1970 ......... 555,883 11.35 14.7 1971 ......... 525,892 10.77 13.9 *Excluding maternity in 1967-70. Source: Health andPersonalSocial StatisticsforEngland, HMSO 1973 and A. J. Culyer and J. G. Cullis, New Society, 1973. open-heart surgery rather than new central heating for the geriatric block. As Sir Keith Joseph (7) has put it, “Take my advice and do not be old or frail or mentally ill here.” Of course the caring are the least vocal and the least in the public eye. Ironically, a good deal of the so-called curing proves upon closer examination to actually be caring. John Bunker (8) has argued that only some 10 percent of surgery is in any sense life-saving or unavoidable, the other 90 percent being intervention in the attempt to improve the quality of life. Similarly, some 30 percent of all general practitioners’ consultations have little to do with medical services as such. (9) Again, 60 percent of us will die from a chronic disorder rather than from an acute illness. Second, preventive medicine remains largely untouched by the system. It has suffered from being statistical rather than personal, from being long-run 101 rather than immediate, and therefore from being imminently postponable. Health education amounts to less than 0.01 percent of NHS expenditure despite the fact that increased exercise, better dietary habits, less consumption of alcohol and no smoking would probably have more impact upon health than even a doubling of NHS expenditure. Third, current expenditure has been increased at the expense of the capital account. During the first 13 years the NHS built only one new hospital; indeed, proportionately less was spent on building than in the 1930‘s. The NHS “inherited" an obsolete capital stock. About 30 percent were former local government institutions often sited away from population concentrations or built for the sick poor. The rest were largely the result of random philanthropic accidents. Consequently, 50 percent of the bed stock is in buildings erected prior to 1900. Only 18 percent of beds are in new or replacement buildings provided under the NHS. Despite this, capital spending has in recent years increased. Less than 4 percent of current spending in 1950, it now exceeds 12 percent. Further, it could be argued that had expenditure been more adequate in the 1950’s we should have by now been over-supplied with beds, as the constraints have made professional conservatism come to terms with shorter lengths of stay and increased throughput. Finally, overwhelmed by the size of manifest demand, until recently the state has avoided, as far as is possible, any attempt to seek out and quantify the full extent and nature of need. There is a persistent fear of discovering new areas of need for which no resources exist. The Deputy Secretary of the DHSS infomied the Expenditure Committee in 1972 that, “The fact is we are only just beginning to explore unmet need and its implication for national policy—the more we go into the problem the more we uncover, as it were, need that we have not seen before.” The implications of this confession for rational planning and for the determination of priorities are plain. The Growing Provision Gap That there is a gap between what is possible and what is currently provided is becoming increasingly obvious to the public. The publicity given to the lack of facilities for renal dialysis and the emergence of costly treatments such as Factor VIII for hemophiliacs have made rationing decisions quite explicit for the first time. Further, the often appalling conditions in which the mentally handicapped, the geriatric and the mentally ill are treated, are increasingly coming before the public eye. In 1972 it was still possible to find wards in which patients slept, ate, excreted, lived and died in one large room, and in which one consultant was expected to treat 660 patients. Again two-fold variations in manpower and resources between regions imply either that some are deprived or others are being wasteful. (10) The problem for the future is that this gap is bound to grow. The NHS is to a large extent an “old-age service.” The retired represent 16 percent of the population but 48 percent of the bed use (excluding maternity and psychiatric beds), 28 percent of the expenditure, 25 percent of the prescriptions, and have an average consultation rate of 7.5 compared with that of3.6 for those aged 15 102 to 44. While the proportion of those aged 65-75 years is likely to remain fairly stable to the end of the century, those over 75 years will increase some 30 percent. Further, the expectations and the propensity to complain of the young are far higher than the current retired. Certainly they have lower thresholds of tolerance as measured by certificates of sickness. To these factors must be added the impact of structural changes in delivery (e.g., the growth of health centers), the influence of changing life styles and the advance in medical science. (2} Finance The uniqueness of the NHS is one of degree rather than of kind. There is, for example, no national health system which is entirely free of state finance. The essence of the British system lies in the smallness of its direct charges and in its universal coverage (although some 4 percent of the population consume private medicine they remain eligible to use the NHS at any time they may wish—it is not possible to “opt out”). The continuous pressure on the NHS to provide more finance has led to active political consideration of alternative sources of revenue—principally a return to some form of insurance scheme with reimbursable (either in part or whole) prices at the point of consumption. However, any attempt to fundamentally change the system away from taxation would completely undermine the system as at present conceived. The main case for dismantling the present tax-financed health system rests on the so-called “tax illusion.” The public tends to see little or no connection between the payment of taxes and the consumption of public goods and services. Indeed, the argument runs, they regard the payment of tax as little more than a confiscation, and therefore are very reluctant to pay. The supply of health care is currently dependent upon the public willingness to substitute the payment of tax for direct command over private goods and services. Therefore, it is argued, if health were placed in the private rather than the public sector, people would be more willing to direct resources to it. It may well be that people are more willing to spend their money when they can see an immediate and obvious return in the form of benefits. The present NHS, however, demands a pooling of risks and a transfer of purchasing power from the fit to the sick, from the rich to the poor, from the wage earner to the young, and the retired, and so forth. Increasing the non-tax element decreases all these other features of the system. Interestingly enough, the Government Social Survey found that 60 percent of the public considered that the entire cost of the NHS was met, not from general taxes, but from the weekly national insurance contribution (the insurance stamp). In practice this amounted in 1972 to only 24.1p per week per adult male worker, one-third of which was paid by the employer on the employee’s behalf. This represents only 81/2 percent of the total cost of the NHS (Table 8). It is theretbre small wonder that the public think the NHS a good bargain. ' 103 Table 8 NATIONAL HEALTH AND WELFARE SERVICES—SOURCES OF FINANCE (GREAT BRITAIN) Source Unit 1958-59 1972-73 All Services ................. £2 million 803 3,109 Percent 100 100 Central Government Services ...... £ million 707 2,526 Percent 88.0 81.2 Consolidated Fund ............. £1 million 568 2,198 Percent 70.7 70.7 Insurance Stamp contributions ..... £ million 102 228 Percent 12.7 7.3 Charges to recipients ............ £ million 35 91 Percent 4.4 2.9 Miscellaneous ................ £2 million 2 9 Percent 0.2 0.3 Local Authority Services. . . . . ; . . . £ million 96 583 Percent 12.0 18. 8 Rates and Consolidated Fund Grants . £ million 83 525 Percent 10.4 16.9 Charges to recipients ............ £2 million 13 58 Percent 1.6 1.9 Source: Health and Personal Social Services Statistics for England, 1 974. DHSS 1974. There are four basic reasons why large-scale fundamental changes in the system of finance and the introduction of an insurance element are undesirable and hopefully unlikely to take place in the foreseeable future: (1) The degree of medical risk varies very widely between individuals. The risks are of a much greater order than those associated with theft or fire. It can be more catastrophic when it strikes. Those persons with the higher health risks would have to be compulsorily admitted by all insurance companies equally; otherwise a company which accepted bad risks on the same terms as its other business would find its premiums forced up, driving its better risks to other companies, finally driving its own premiums up still higher. (2) The present state monopsony would break up. The NHS as the sole purchaser of medical talent and supplies has used its power to hold costs down. The service is probably one of the cheapest in the world. We have, for example, amongst the lowest drug prices in Europe. (11) Insurance companies on the other hand, would become little more than bill payers, automatically adjusting premiums to risks and leaving patients to increase their coverage in line with inflationary price and 104 wage increases. Thus, the state’s leverage would be missing under an insurance system. The danger is that an insurance system would result in more expenditure but fewer real resources. It is true that the tax illusion would vanish but there may well be an insurance illusion to match it! In the United States, for'example, health professionals frequently despair at the failure of Americans to adequately invest in their own health- insurance protection. People notoriously discount the future at absurdly high rates. Any insurance broker can provide ample evidence of inadequate fire cover, and so forth. As with tax, no immediate connection between payment and benefit is present and indeed, once the premium has been paid, the same moral hazards are present (up to the individual’s limit of cover). It is often overlooked by the advocates of insurance-based schemes that insurance and tax are merely two pipelines running off the same well (GNP). The overall constraint is the same in both cases (i.e., national income and wealth). (3) The argument that an insurance system would promote greater efficiency is also far from convincing. The efficiency of the service is largely producer-determined. Numbers and types of hospital admission and length of stay are not determined by patients but by doctors. Nor is there evidence that insurance, even with only partial reimbursement, would discourage “trivial” visits to the doctor. Indeed the very act of payment might make patients less guilty about it, actually increasing the incidence of such calls. In any case, what is a “trivial” ailment? A face blemish on a model, a stomach upset 2 hours before a scholastic test, or desperate loneliness in a bereaved widow?—what is trivial to the doctor may be a crisis to the patient. If such cases were diverted from the surgery, where would they go? (4) Finally, an additional problem encountered by partial types of scheme is that they would remove the articulate and informed from the state scheme, leaving those—like the mentally ill and aged—to fend for themselves. Much more likely than any such major change is the gradual expansion of less and less “nominal” user charges (Table 9). To date, however, charges form a more-or-less steady percentage of total revenues since 1958 (around 5 percent), although currently they are increasing slightly. In fact, quite substantial charges are needed to raise even very modest revenues because of the very large number of unavoidable exemptions. Prescription charges are currently 20p per item, which represents roughly 25 percent of the average prescription cost. In 1972 there were 256 million prescriptions written in England at a total cost of £211 million. Of these, 150 million prescriptions were exempt from the charge. In the period 1972-73, dental charges amounted to £30 million which was raised by a charge of 50 percent of the actual cost, up to a limit of £10. Ophthalmic charges consist, in the main, of a full-cost charge on lenses, up to a maximum of £3.50. Both charges are subject to the usual list of exemptions. There seems no very good reason for these rather odd variations between the serv1ces. 105 901 CHARGES T0 PERSONS USING THE SERVICES Table 9 Services . . . Pharmu- General General Welfare Local Local Personal social Per1od All servrces HOSP 1tal ceutical dental ophthalmic foods health welfare services £m £m £m £m £m £m £m £m £m 1958-9 ......... 48 6 12 9 6 2 3 10 ~ 1963-4 ......... 75 8 24 12 8 2 5 16 — 1964-5 ......... 78 9 24 12 8 2 6 17 — 196 5-6 ......... 58 8 — 12 9 2 7 20 4 1966-7 ......... 60 8 — 13 9 2 7 2] , 1967-8 ......... 63 9 « 13 9 2 8 22 — 1968-9 ......... 80 10 1 l 15 9 2 7 26 - 1969-70 ........ 93 12 18 17 10 2 7 27 — 1970-1 ......... 107 12 18 19 12 l 3 4 42 1971—2 ......... 131 15 24 25 14 l 3 ~ 49 1972-3 ......... 149 17 27 30 16 l 3 — 55 Source: Health and Personal Social Service Statistics for England 1 974, DHSS 1974. Since 1969 charges for visits to general practitioners, “hotel” charges for stays in hospital, encouragement for private insurance schemes in the way of tax rebates on premiums, and prescription charges graduated up to a maximum amount to reflect actual costs, have all been considered by the Conservative Party as candidates for official policy. None of them has come close to implementation because of the cost of collection, the list of unavoidable exemptions and the existence of a substantial “anti” lobby. More than 45 percent of all occupied beds are filled by the mentally ill or mentally handicapped and 40 percent of the remainder by patients over 65 years old. Further, as doctors are not paid on a fee-per-item basis there is no obvious mechanism by which user charges could be collected. A restricted list of “free” drugs plus a more extensive one with modest user charges may be on the horizon. Certainly, it is probable that pharmaceuticals will cease being a relatively stable share of the NHS bill and will begin to become an increasing one. As the relative importance of the drug bill increases, the almost complete blank check for prescribing enjoyed by doctors since 1948 (save for a very modest check on doctors who persistently prescribe out of line with the average for their area) is bound to be increasingly challenged. Barriers of convenience and user charges—plus the exclusion of forms of treatment from some patients altogether—do not seem any more compatible with the kind of clinical freedom sought by the medical profession than limited access by means of explicit and universally-applied lists of free drugs. Restriction of service to those treatments (surgical and medical) which have the most important and well substantiated effects seems preferable to user charges, which are likely to be a burden only to those least able to bear them. Certainly it is likely that to become significant sources of revenue, these charges would have to be at levels where they were also significant deterrents to demanding care, and there exists no evidence that it is the more trivial demands that are deterred first. Conclusions A much more realistic appraisal of the curing aspect of the NHS, and a re-examination of priorities are long overdue. In the early 1950’s Ferguson and Macphail (12) found that 36 percent of all patients medically discharged from hospital were dead 2 years later. Allowing that most medical intervention is to improve the quality of life, we need to know the outcome of such intervention. The increasing literature on the subject of diseases of medical practice (D.O. MP) bears further testimony to the urgent need for more critical appraisal (13) as do the startling variations in medical practice when con- fronted with similar medical states, which suggest the existence of considerable economic waste. (2) Further, the observable impact of the NHS upon morbidity, absence from work, and life expectancy, leaves no room for complacency. The public, for its part, must learn that medical science does not offer a cure for everything, while the medical profession must be less reluctant to advertise its own limitations. 107 Finally, it must be said that recent years have seen a great increase in awareness of these problems, not only in terms of cost effectiveness but also in the introduction of overall program analysis and review. The NHS does not need to be dismantled, rather it requires a close scrutiny of its problems and zealous protection of its many virtues. 10. ll. 12. 13. 108 REFERENCES . POWELL, E. (1966), Medicine and Politics, London: Pitman. COOPER, M. H. (1975), Rationing Health Care, New York: Halstead Press, Wiley. . CROSSMAN, R. (1969), The Guardian, November 27. KLEIN, R., et a1. (1974), Social Policy and Public Expenditure, London: Centre for Studies in Social Policy. . BIERMAN, P. (1968), Milbank Memorial Fund Quarterly Review, 46, 77. . BROTHERSTON, J. M. H. (1970), In W. Lathem and A. Newberry (eds.), Community Medicine: Teaching, Research and Health Care, London: Butterworths. . JOSEPH, K. (1973), Daily Telegraph (London), June 30. BUNKER, J. (1974), “Risks and benefits of surgery,” In Benefits and Risks in Medical Care, London: OHE. . THOMAS, K. B. (1974), Brit. Med. J., March 30. COOPER, M. H. and CULYER, A.J. (1972), “Equality in the NHS: Intentions, performance and problems in evaluation,” In M. Hauser (ed.), The Economics ofMedical Care, London: Allen & Unwin. COOPER, M. H. (1975a), European Pharmaceutical Prices 1964-74, London: Croom Helm. FERGUSON, T. and MacPHAIL, A. N. (l954), Hospital and Community, Oxford: Nuffield Provincial Hospitals Trust. MALLESON, A. (1973), Need Your Doctor Be So Useless?, London: Allen & Unwin. DISCUSSANT’S COMMENTS A. J. CULYER York University, England Treasury grope and medical gripe (but not consumer) may indeed be broad characteristics of the NHS. Also, the monopsony power of the state has, doubtless, had a substantial effect—though a once-and-for-all effect—on the health bill in Britain. So far, I find little to disagree with in Mike’s paper. Because he covers so much ground I shall be selective in what I comment upon. In particular, I should like to focus on the question of demand and need; the question of the determinants of the distribution of expenditure; and the question of priorities. Needology In essence, I agree with Mike’s View here: namely that the elimination of unmet “need” is an impossible objective for a health system and that, unless such an objective is supplemented by more realistic specification of what is both meaningful and possible, then the system is going to lumber along—as indeed it largely has—in a rather directionless manner providing endless opportunities for discontent all round. On the other hand, I do think Mike could have gone a bit further (I) in the “need” discussion. I would like to throw two pebbles into the pool, two sets of assertions, the first of which refers back to an earlier discussion today: ‘ (l) Distinguish two concepts of need: the supply concept and the demand concept. (a) In the supply concept the word is used in its normal logical sense of “something being a necessary condition to attain an objective.” In health, this requires outcome measurement and input-output/ production function studies—all of which are clearly areas for relevant economic expertise. (b) In the demand concept we are concerned with “who should get what” and, with all due respect to Peter Ruderrnan and Bob Maxwell, I don’t know why they imagine their value judgments on these matters should be of particular interest to us. Anyway, this is clearly not the economist’s business and I’ll not tell you my value judgments (2) in this field. (2) The second set is reasonably inferred, I think, from the objectives of Britain’s socialized medical-care system. 109 (a) Only those conditions for which acceptable and clinically effective treatments exist (including curing as well as caring treatments) should be given; other treatments, including ineffective ones, should be provided by the private sector (including the Church of England—or any close substitute other than the NHS doctor) as well as the local Chemist’s shop and private medical practitioners. (b) Only those acceptable and effective treatments or regimes of care be given that can, at the general level, be justified on cost-effective grounds (interpreted broadly to include humanitarian consid- erations). ( 3) Clearly, both these tasks amount to a very considerable research program but it is a program that has, to some extent at least, been got underway in the more enlightened British research centers and it is a program whose sense is seen also by the more enlightened managers in the service. Expenditure Patterns Once again, I believe Mike is quite right in saying that the NHS has never fully shaken off its prewar inheritance in the matter of financial (or real) resource distribution. Nevertheless, there has been some success in the redeployment of GP’s; more recently there has been some success in getting a shift away from institutional towards community care, reductions in inpatient lengths of stay, hospital bed/population ratios, an increase in hospital throughput potential and, even more recently, there is an attempt being made to secure greater equity in the distribution of financial resources both territorially (vis—‘a-vis regions) and within health sectors (e.g., vis-a-vis teaching and non-teaching hospitals). The point is not so much that any of these changes are desirable in themselves as that they can happen! I think there are two principal difficulties which help to explain why it still remains that relatively little has been done to arrange expenditure on a more explicit and formal basis. The first is that productivity growth in the health sector has, in real terms, been slow—which is perhaps a general reflection of the conservatism of the medical profession and its resistance to radical measures to reform. Non-growth industries are never the easiest ones to “rationalize,” , especially when they are in the public sector. Some preliminary work by my colleagues Bob Lavers and Dave Whynes revealed, for example, the following indicators of hospital productivity. Taking 1960 = 100, by 1970 doctors’ price index 204.0 doctors’ real input index 117.5 nurses’ price index 194.0 nurses’ real input index 146.1 total real input index 133.6 (a) crude output index 133.2 (discharges and deaths) (b) social output index 135.6 ((a) by diagnostic category weighted by reciprocal of waiting time) (c) economic output index 113.5 ((a) by age group weighted by (1964—72) discounted future earnings) 110 While these data may tell us nothing about the absolute productivity of the health sector, they do suggest, I think, that it has not, on average, been rising very fast over the decade 1960-70. Indeed, output on all crude indicators (save the crude social index) has risen at a slower rate than inputs; thus it is scarcely surprising that hospital current spending has taken an increasing percentage of all health spending. The second and more fundamental reason for the absence of more explicit and rational allocation criteria, or of effective controls on spending, is ignorance. We have no very clear idea of what regional demands or “needs” are, or will be, in the future; we have no very clear idea of what procedures are effective in either clinical terms (for many conditions) or in cost-effective terms (for many more conditions). Consequently, there is very little idea of precisely what NHS money is buying, let alone of the value (social or private) of what it is buying. Until we have a much better idea of these things, the conservative policy of incremental budgeting seems to me to be a natural thing for central policymakers to be undertaking in confronting the powerful and articulate medical pressure groups. But, again, there is light at the end of the tunnel. The DHSS is developing a client-orientated program budget; there is work on the substitutability of medical inputs and cost-effectiveness of medical procedures; there is work— some of it well-advanced—on output measurement. But until much more of this kind of work has been done, I don’t think that very much can be inferred from the data that Mike has presented in the central part of his paper (nor would he, I am sure, want too much to be inferred from them). Priorities Mike identifies four clear patterns of expenditure. Of these, the first two (a bias pro curing rather than caring and an unfair bias against preventive care) seem far less clear to me than to him. One of the difficulties in interpreting the highly aggregated form of Mike’s statistics lies in their very aggregation. In Table D1 I show some indexes, derived by multiple regression analysis from the hospital costing returns by Alan Hamlin, of current expenditure by hospital specialties. These seem to indicate that, at least since 1966, most of the relatively fast-growing specialties in financial terms (pediatrics, special baby care, chronic and geriatric sickness, traumatic and orthopedic surgery, dentistry, gynecology, mental handicap and “unallocated”-—which includes hotel facilities, as well as residuals) have a relatively high “care” as opposed to “cure” content. Mike may well be right that there isn’t enough of caring around, but the trend seems to be one of which he should approve. His second pattern relates to the Cinderella status of preventive medicine— yet it is far from clear to me that this charge is really warranted. While I would not dispute that “Trirnm Dich” courses and abstention from smoking, alcohol, fast cars and fast women may put tens of units on your health-status index we really don’t know the parameters of a fully specified demand-for-health function, or what the real costs of “going without” are, compared with trying lll 311 Table 10 CURRENT EXPENDITURES BY SPECIALITIES FOR ALL NHS HOSPITALS (AT CURRENT PRICES) 1966-67 = 100 . 1967-68 1968-69 1969-70 1970-71 1971—72 Accident & emergency .......................... 108.4 115.9 130.8 150.5 171.0 General medicine. . . . .......................... 108.9 117.0 124.9 150.5 170.9 Pediatrics* .................................. 111.7 122.3 135.1 194.7 231.9 Special care of babies* .......................... 115.6 137.5 156.3 212.5 262.5 Chest'l' ..................................... 104.9 111.3 117.2 138.7 152.9 Other medicine ............................... 108.6 116.5 126.3 156.9 181.6 Geriatric & chronic sick“ ......................... 105.9 118.4 131.6 159.4 186.1 General surgery'l‘ .............................. 108.6 114.4 120.3 145.3 166.9 Ear, nose & throat'l‘ ............................ 108.3 115.8 120.8 139.2 164.2 Traumatic & orthopedic surgery* ................... 1 1 1.8 122.5 132.3 160.8 191.4 Other surgery* ............................... 112.2 121.4 130.7 156.4 192.5 Dentistry* .................................. 112.5 125.0 152.5 187.5 197.5 Gynecology* ................................. 110.6 122.3 131.2 166.1 191.4 Obstetrics .................................. 107.3 115.5 125.5 143.2 173.3 Mental illness'l‘ ................................ 104.9 114.8 122.5 144.1 166.3 Mental handicap* ............................. 106.3 117.7 129.9 160.8 197.6 Unallocated* ................................. 160.4 191.8 264.9 291.8 350.0 Total ..................................... 109.4 119.2 131.6 168.4 182.3 *Indicates higher than average growth over period. Tlndicates “sizeably” lower than average growth over period. Source: Data compiled from data included in hospital costing returns. to put the pieces back together afterwards. My guess is that the cost of pursuading people to go without are astronomical. But, in some cases we do have very hard evidence. Britain’s mass-radiog- raphy program for tuberculosis prevention is now almost entirely withdrawn since, with modern chemotherapy it had become plainly cost-ineffective. The advantages of early diagnosis for many serious conditions are equally seriously in doubt. Preventive medicine has a good ring to it but the evidence to date (pending, perhaps, a comprehensive flu vaccine) does not seem to me to suggest that it deserves in general to be anything more than a Cinderella. Let me end by agreeing with Mike once again: None of the evidence suggests that the British NHS should be dismantled but we could do with a great deal more evidence to help the policymakers make it a more cost-effective service. I happen to believe that, in the long-run, the NHS-type structure of direct state management offers the most promising context for a fully nationalized system of health-care delivery. Ideally, it would function best alongside an active and competitive private sector, for the NHS structure offers the potentiality for effective control of the medical professions and industries in the interests of a more clearly specified collective concern for the health of the nation, while the private sector will be there not only to provide the ineffective treatments but also always to compel the public sector’s attention towards the not altogether unimportant point that patients are also people. There used to be a lot of argument about whether consumers could make rational decisions regarding health care (4/, and the NHS was often put forward as the logical institutional corollary for those who believed they could not. For me, the main point is that it offers leverage on the problem of how to control producer choice, not consumer choice. While the NHS can claim some success in this to date, in 10 years’ time I hope the record will look a lot better. NOTES AND REFERENCES 1. As he does in his book: Rationing Health (1975), London, Croom Helm. 2. An excellent survey of “need” notion is: Williams,A. (1974), “Need as a Demand Concept,” In CULYER, A. 1., Economic Policies and Social Goals: Aspects of Public Choice, London: Martin Robertson. For a Paretian Interpretation of “need” see CULYER, A. J. (1973), The Economics of Social Policy, London: Martin Robertson. 3. For a development of this theme: CULYER, A. J. (1975), “The Social Costs of Doctors’ Discretion,” New Society. 4. See: CULYER, A. J. (1971), “The Nature of the Commodity ‘Health Care’ and its Efficient Allocation”, Oxford Economic Papers. 113 Chapter 6 A COMPARISON OF THE HEALTH-CARE SYSTEMS OF FRANCE AND THE UNITED STATES GEORGES ROSCH and SIMONE SANDIER Division d ’Economie Medicale (CREDOC), Paris, France 1. GENERAL OUTLINE OF THE FRENCH AND AMERICAN HEALTH-CARE SYSTEMS Introduction This paper attempts to compare the health-care systems of the United States and France; the statistics available for both countries make such a comparison possible, and these include: (I, 2) Statistics on the health sector compiled in accordance with international practice for National Accounts and which exist in the case of the US. for every year since 1928, and in the case of France since 1950 (3, 4). Data on the “factors of production” or “inputs” which are available for most countries; e.g., number of hospitals, beds, staff, etcetera (5, 6, 7, 8). Figures on the number of medical acts—their supply as well as their utilization ( 9, 10). Information derived from household surveys (which is much less frequently available) (11, 12, I3). Surveys carried out on care “establishments;” e.g., doctors’ offices, hospitals (14, 15). The difficulties of making comparisons It was inevitable that a number of difficulties would be encountered in making these comparisons and these need to be made clear at the outset. These difficulties are of various kinds, such as: The basic differences which exist between countries which are dissimilar in terms of size, geographic, demographic, political and economic character- istics, and the structure of their health-delivery systems and social policies. Basic statistical data which often are not based on the same concepts (area covered, definitions, nomenclature, units of measurement) making it difficult to draw valid conclusions. Lastly, an obstacle which economic theory recognizes as being insur- mountable, i.e. the comparison of amounts, equipment, salaries, outputs, and so forth, which are all calculated in terms of money values which are 115 not stable or comparable in terms of time or space. Only by an iterative process of successive approximations is it possible to get close to a valid comparison. Methods So as to enable data to be compared, ratios have been used, e.g. the density of manpower or hospital beds per 1,000 inhabitants, per capita consumption, output per physician, staff per hospital bed, etcetera. Expenditure and price data Figures are given in both currencies ($ US. and French francs) using the average exchange rate for 1973. In order to compare trends in expenditure and prices, eliminating the effects of fluctuations in the exchange rate and different rates of population growth and inflation in each country, these trends have been expressed in terms of (a) per capita expenditures at 1973 current prices, i.e. in constant money of the country*; and (b) per capita expenditures at constant 1973 prices—or volume per capita'f'. Periods covered Trends are compared over fairly long periods, i.e. 1950-1973 or 1960-1973, so that the picture is not unduly distorted by isolated events in either country, such as, for example, the introduction of Medicare in 1966. Without any doubt, an analysis covering shorter and more specific time periods (which is outside the scope of this paper) would certame provide a wealth of information and would enable one to assess whether these trends in expenditure on medical care have been affected in recent years by the slowdown in economic activity and the growth of inflation. Plan This report is made up as follows: First, a brief outline of the salient features of each country. Second, a description of the two systems of medical care; i.e. present situation, basic organization, structure of the health-delivery system, provision and utilization of medical care, trends. For each type of service: *VNt = expenditures at current prices for the year t CPIt = consumer price index for the year t (1973 = 100) POt = population for the year t VRPt= expenditures in constant money of the country per capita VRt = (VNt / CPIt) / Pot TIPNt = index of price for the considered service in the year t (1973 = 100) VCPt= expenditures per capita at constant prices 1973 VCPt= (VNt / IPNt) / Pot . 116 Finally, an attempt to discuss, interpret and explain as far as possible the similarities, differences and unique aspects of each system via an in-depth study of hospital care and physicians’ services. The Salient Features of the United States and France Physical geography The United States covers an area of 9,363,488 sq. km.; France covers 551,602 sq. km., or one-seventeenth of the area of the United States. More than anything else, this difference in size affects the variety in climate. France is much more northerly (between the 51st and 4lst parallels), whereas the US. (between the 49th and 24th parallels) is, in the south, only 33 minutes from the tropics. On the whole, France has a mild climate, whereas the American climate is much more varied and rigorous—cold in the north, in the south semi-tropical, i.e., arid or humid. Human geography The population of the US. in 1973 was 213,940,000; four times larger than that of France, with its 52,045,000 inhabitants. In this regard, two aspects in particular need to be borne in mind: (a) With 94.4 inhabitants per sq.km., France can be regarded as having a medium population density, i.e. between 50 and 150. With 22.8 inhabitants per sq.km., the US. has to be ranked amongst the countries with a low population density, i.e. less than 50. This factor has an influence on the ease with which a medical-care system can cover the country and be easily accessible to the local population. (b) From its age structure, the French population would seem slightly older on average: 1970 ’ Ufi. France less than 15 years 28.5 percent 24.8 percent 65 years and over 9.9 percent 12.8 percent All other things being equal, medical care requirements in France should therefore be somewhat higher since it is well known that morbidity increases with age, practically as an exponential. Political and administrative organization The systems of government in each country are fairly similar, but important differences can affect the health-care system. In the US, individual states still retain a large measure of autonomy and, as a result, situations vary widely. In France, power is extremely centralized and regulations, organizations, tariffs and suchlike are uniform throughout the country. 117 4 1 c cc cc c c... d....o€o€o.od4.o4 4¢<9&4fififiodéiécococoooqocococo! / #&&&a®&%&%luw*9Qwaawfifiéfifikfififibfifififlfifififififiéwfifi / oooooooooooooocooooooooooooooooooooooo§oooog ooooooooooooo.coocoooooooooooooooooooo.¢ooo ooooo99909.9.:99cooooooooooooooooooooooooo; OOQOCOQQ0.009....‘0....QQOQOOOOOOQOQOOOQQ COO...OQOOO¢OOOQQO.QOOQQOOOOOOOO 0.0.9.6 OQ....OOOOQ.»«OOQQ.§OO§.Q.COOQQOOO OOOOOA %§hb3§€522?¢¢§§3¥&€%§33¢Q§2§€ tfiidg O ’00,.0’QO’OQ’QOOO;vQO00...”....09....0QQOOOOOOQQQ’... ’O‘OOOOQOA 3332?. acipfixflhhuxxxdttddiépnk% ominxnkfih 90 oo o 990 00909000 00%.. o o .. cocoooooooooooooooooooiooooo o oooooo 9009000909009 » ;ooooo¢¢¢:¢¢¢¢oooo¢¢¢o ooooooooooooooooo... .693bbbu333233?83?$ §3¢992999000¢Q000 ;$%%%kkfififix€€€§§vw ,Qooo$oooooooooo 35kmufibmfimflfifiw z » »»koo¢o O O % O O o oo 3°20 a. o '3 $ % % o. O. O. 0%. .‘v oo oo § 6 % Q Q Q 020 9 0 g C O O O % .020 ,é awn oo. .00 .00 000 009 00¢ .09 coo 09% oz 90' 9.0 900 f?” $9? 0 I '2 % % $ 2 o ’3 o ooooooooo Afi£‘&& 0: O O '3 O O O O D O O 2020‘ . Gold Coast 3 / ,’ '_.- Tropic of Cancer ~__ \\ 50° 0 0 3 :m Figure 1. Comparative geography. A large portion of the French economy is socialized; e.g. electricity, gas, coal-mining, atomic energy, transportation, telecommunications, and educa- tion, together with a sizeable part of the automobile and aerospace industries. The economy In 1950, the economic level of the US. was much higher than in France, but this gap has been reduced as a result of the higher average growth rate in the French Gross National Product (GNP). In 1973, per capita GNP in the US. was only 1.2 times the figlre for France, whereas in 1950 it was 2.6 times as large. This is not wholly explained by the fact that France has had a higher rate of inflation over the period 1950-1973. Per capita GNP in the US. and in France (at current prices and average exchange rate for the year) Q. France (3%) ($) (Frs.) 1950 ....... 1,841 707 2,445 1973 ....... 6,053 4,934 21,977 The Health-Care Systems in the United States and France The scope of the study The limits of the “medical care sector of activity” will be based on both technical and economic criteria. Medical care activities (both their provision and utilization) divide into two distinct categories: , (a) “Standard” activities corresponding to “standard” annual expendi- tures. , (b) “investment” activities corresponding to expenditure “amortized” over several years. The “medical-care sector” will be limited to activities concerned with the prevention, diagnosis and treatment of pathological conditions amongst individuals, carried out by professionally-qualified personnel and specialized establishments. Figure 2 shows the area covered by this study. In practice, there is no major problem in choosing limits to this area which are comparable for the two countries. Both in the US. and in France, in fact, in the majority of cases medical activity is carried out by qualified people (physicians, nurses, etcetera) or in establishments such as hospitals and dispensaries, whose legal status is clearly defined and subject to strict laws and regulations. A precise inventory can thus be made of such activities. However, difficulties of definition do exist in three areas: There is a merging and overlapping of definition with regard to hospitals, nursing homes, geriatric establishments (“hospices”), old people’s homes, as 119 HEALTH SECTOR MEDICAL SECTOR IVi in: \ l ndustries® General Economic Act Industries Ecmm Industries++—————— oooo¢ ”3......“ “a...“ ...’oooo nuanaJd angmauoa . Imaua .; . ‘00 0cooooooo‘ ............ . . :.:,:,.,...,,.....,.,. . 5... o o . A‘ . In. saulsnpul IIIIHHTTITIITI Social Securi llllllllllllllll lIHIHIIIHUHI Adm nistra on and General Policy IllIH HHIH 5 o 3 § '3 o E E o 5 'U 5 § 0 8 § ca 2 o E5 hi i in well as establishments for handicapped children of a more-of-less medical character. It is difficult to make a distinction between private or public preventive measures of a purely health nature (immunization, for example), those which are only partly of a health nature (water supply) and those which are applied in all areas and sectors of economic activity (including control of food, hygiene in buildings, and safety measures for public transport). Lastly, it is usually very difficult to obtain separate data for expenditure in the area of health care or its financing which is incurred in the course of the various political, administrative and operational activities of ministries, local authorities and insurance organizations. Such expenditure has thus to be estimated on a very approximate basis. “Inputs ”in the health sector As in any sector of the economy, the production capacity of the health care system depends on the quantity of inputs, such as manpower, buildings and equipment, and their quality and the way they are organized. In the health field, as in most “service” activities, manpower is the most important factor and the biggest cost element (around 80 percent). Manpower density is higher in all areas in the U.S.: 21 percent higher for physicians, 69 percent higher for nurses, 27 percent higher for dentists and 97 percent higher for pharmacists (Table 1). On the other hand, France is better equipped in terms of hospitals. Table 1 HEALTH MANPOWER 1973 —— RATIO PER 100,000 POPULATION US. France Physicians: Total ............................ l 71 142 Private practice .................... 94 97 Nurses: Registered ........................ 361 219 Other ............................ M £6 Total ............................ 5 6 7 33 5 Dentists: .......................... 5 7 4 5 Pharmacists: ........................ 63 35 121 SOURCE OF FUNDS 3: HEALTH CARES \‘ 1800 400 1500 1.5 _ Eyeglasses 300 and Appliances 1000 25.2 Insurance 200 Per Capita 500 100 .1 Public 100 FRANCE U.S. 1973 Figure 3. The United States and France; structures of health care and funding capabilities. The statistical tables showing the provision and utilization of health-care facilities As with Figure 2, two categories are analyzed: (1) The broad category of the nation’s health-care expenditure, admin- istration expenses, investments, education, etcetera. (2) The more restricted but more detailed category of medical care expenditures*. Note that this second table, like the first, covers all *Classified in the U.S. under the heading of “personal health care expenditures.” 122 mideg lad expenditure including investment expenses which are included in the form of amortizationj‘ These data for 1973 are shown in Table 2 (U.S.A.) and Table 3 (France), showing both the different types of health care services provided and utilized and the different sources of funds. A comparison can thus be made between the degrees of utilization of medical services both in total and for each type of service, together with a breakdown for the different sources of funds. A comparison of the level and structure of medical care in 1973 Comparing the level of medical care, three points stand out: (a) Medical care in the U.S. represents a higher percentage of GNP (6.6 percent) than in France (6 percent). Per capita expenditure on medical care is 39 percent higher in the U.S. (b) This difference is primarily due to the higher expenditure in the U.S. on hospitalization (+ 77 percent) and on physicians’ services (+ 83 percent). Expenditure on dentists appears about equal. (c) Consumption of drugs and drug sundries is considerably higher in France (+ 63 percent). These facts obviously influence the structure of medical care. If the share of expenditure devoted to physicians and dentists is not very different in the two countries, hospitalization accounts for a much larger share in the U.S. (53 percent as against 42 percent), whereas drugs and drug sundries account for a greater share in France (24 percent as against 11 percent). Trends An analysis of trends between 1950 and 1973 shows that: Per capita medical expenditures throughout the period have been higher in the U.S. (Figure 3). The rate of growth in the per capita volume has been more rapid in France (7.6 percent per year) than in the U.S. (3.8 percent). The gap between the two countries is therefore tending to diminish. In 1950 expenditures in the U.S. were 250 percent greater; by 1973 they were only 33 percent greater. Since GNP increased at a slower rate in the U.S., the growth in the proportion of GNP accounted for by medical expenditures has been at a similar rate for both countries (Table 6). 7 During the course of these 23 years, the breakdown of expenditures in current money by type of care has slightly changed. In both countries, over the period 1950-1973, one notices: A greater increase in hospitalization expenditure as a share of the total in the U.S. (from 37 to 53 percent) than in France (from 37 to 42 percent). TAlthough the procedure has not yet been fully adopted by the systems of National Accounts, expenditures on education (human investment) and research (investment in knowledge) is considered as part of the general category of investments and has to be considered as a “capital account.” 123 ’VZI PERSONAL HEALTH EXPENDITURES BY TYPE OF EXPENDITURE AND BY SOURCE OF FUNDS UNITED STATES 1973 — MILLIONS 5 OR FRS. Table 2 Private Public Consumer and State and TOTAL insurance Other Total Federal local Total benefits $ Frs. Frs. $ Frs. $ Frs. $ Frs. $ Frs. $ Frs. . Hospital care ............. 17,717 492 18,209 13,605 6,456 20,061 38,270 78,912 2,191 81,103 60,597 28,755 89,352 170,455 . Nursing-home care --------- 3,430 29 3,459 2,040 1,551 3,591 7,050 15,277 129 15,406 9,086 6,908 15,994 31,400 21,147 521 21,668 15,645 8,007 23,652 45,320 Total .................. 94,189 2,320 96,509 69,683 35,663 105,346 201,855 . Physicians’ services ......... 13,978 13 13,991 3,054 1,154 4,208 18,199 62,258 58 62,316 13,603 5,139 18,742 81,058 . Laboratory test ........... - - - _ _ _ _ . Other professional services 1,523 36 1,559 217 124 341 1,900 6,783 160 6,943 967 552 1,519 8,462 . Dentists’ services ----------- 5,639 - 5,639 213 118 331 5,970 25,116 - 25,116 949 525 1,474 26,590 931 PERSONAL HEALTH EXPENDITURES BY TYPE OF EXPENDITURE AND BY SOURCE OF FUNDS Table 2—Continued UNITED STATES 1973 — MILLIONS $ 0R FRS. Private Public Consumer TOTAL . and Other Total Federal State and Total insurance local benefits $ Frs. $ Frs. $ Frs. Frs. $ Frs. $ Frs. S Frs. . Drugs and drug sundries ..... 8,564 - 8,564 386 350 736 9,300 38,144 - 38,144 1,719 1,559 3,278 41,422 . Eyeglasses and appliances ----- 2,006 - 2,006 48 37 85 2,091 8,935 - 8,935 214 164 378 9,313 Total ambulatory care ....... 31,710 49 31,759 3,918 1,783 5,701 37,460 141,236 218 141,454 17,452 7,939 25,391 166,845 Other ................. - 610 610 - - 2,506 3,116 - 2,717 2,717 - - 11,162 13,879 Total personal health expen— 52,857 1,180 54,037 - - 31,859 85,896 ditures ................ 235,425 5,255 240,680 - - 141,899 382,579 Source: Worthington (3) A reduction in the share of drugs and drug sundries which is more marked in the US. (from 16 to 11 percent) than in France (from 26 to 24 percent). A substantial decrease in the share for physicians’ services in both countries; from 25 to 21 percent in the US. and from 20 to 16 percent in France. Sources of fimds The sources of funds for medical care are very different in the US. and France. In the U.S., a great proportion of these are private funds, i.e. individuals or insurance companies. However, in 1966, Medicare and Medicaid instituted the financing out of public funds of health expenditures for persons over the age of 65 or with low incomes. In France, the Social Security scheme, which has been gradually extended beyond the wageeaming class to cover, first farmers and then all self-employed persons, covered 98 percent of the population in 1973. The effect of these measures is reflected in the breakdown of the sources of funds (Table 7). The improvement in insurance cover for health expenditure between 1950 and 1973 is obvious in both countries, from: (a) A drop in the individual’s direct contribution to the financing of health care; nonetheless, in the United States in 1973, this still represented a fairly high proportion, i.e. 36 percent as against 24 percent in France. (b) The growth in the proportion covered by Social Security in France (68 percent in 1973 as against 45 percent in 1950). (c) The increase in the part covered by private insurance in the US. up to 1966 (9 percent in 1950 and 25 percent in 1973); and the part covered out of public funds, particularly since 1966 (22 percent in 1950 and 37 percent in 1973). 1 Demographic and socioeconomic breakdown of the population The influence of the various demographic or socioeconomic factors on health-care expenditures is similar in both countries (11, 12, 13). The inhabitants of rural areas use the health care services to a lesser degree than town-dwellers. The level of income has only a slight influence on overall expenditure on health care, but the type of expenditure is different, i.e. the more wealthy classes use specialist care to a greater extent and are hospitalized less frequently. Age and sex are the two factors that have the most marked effect on the levels and trends of health-care expenditure. Generally speaking, in France as well as in the U.S., the population groups accounting for the highest medical expenditures are babies and the aged, and to a lesser degree women of childbearing age. Women spend more on ambulatory care than men but, except during the time they are of childbearing age, they are less frequently hospitalized and the length of stay is shorter. 126 Despite the marked effect of age on the level and rapid growth in health expenditures, it has been possible to demonstrate that the increasing top-heaviness of the age pyramid, which is a gradual process, has not been and is unlikely to be a significant factor in the increase in average health-care expenditures (14). For example, in France, over the 20 years between 1950 and 1970, the increase in per capita health care expenditure due solely to the increasing top-heaviness of the age pyramid was less than 5 percent on the total health-care expenditure, whereas the volume of health-care expenditure per person increased by 300 percent over the same period. These factors do not seem therefore to have any real influence, in any area of health care, on the growth in demand and no further reference will be made to them in this report. The population ’s health in the United States and in Ftance One would have liked to have finished this presentation of data by giving information on the respective health status of the two countries’ citizens. However, there are two obstacles to this: Although surveys on morbidity exist in both countries, the task of interpreting them is so delicate as to make valid comparisons practically impossible. Thus, mortality statistics are the only indicators available, although indirect and partial. The health-care system certainly has an influence on the population’s health and mortality, but it is itself dependent on a basic incidental morbidity which can vary greatly from country to country and from region to region within the same country. A few of the most significant figures are, however, given in the following table: DATA ON MORTALITY FOR THE U.S. AND FRANCE U.S. France Infant mortality (< 1 year) 1973 ........................ 21.8% 16.4% Life expectancy Male 1972 ........................ 66.6 68.4 Female 1972 ........................ 75.5 76.1 127 8Z1 PERSONAL HEALTH EXPENDITURES BY TYPE OF EXPENDITURE AND BY SOURCE OF FUNDS Table 3 FRANCE 1973 - MILLIONS $ OR FRS. Private Public Consumer Private . direct insurance Total State and Socral Total Total local Security pay (mutuelles) $ Frs. $ Frs. $ Frs. $ Frs. $ Frs. $ Frs. $ Frs. . Hospital care .......... 342.8 128.1 470.9 463.3 5,281.1 5,744.4 6,215.3 1,527 571 2,098 2,064 23,524 25,588 27,686 . Physicians’ services ........ 744.4 131.3 875.7 49.3 1,496.8 1,546.1 2,421.8 3,316 585 3,901 220 6,667 6,887 10,788 . Laboratory tests .......... 54.5 6.0 60.5 0.8 283.6 284.4 344.9 243 27 270 4 1,263 1,267 1,537 . Other professional services 94.3 24.2 118.5 2.4 491.2 493.6 612.1 420 108 528 11 2,188 2,199 2,727 . Dentists’ services 898.4 60.8 959.2 2.9 524.4 527.3 1,486.5 4,002 271 4,273 13 2,336 2,349 6,622 . Drug and drug sundries ...... 1,307.9 217.5 1,525.4 81.7 2,071.6 2,153.3 3,678.7 5,826 969 6,795 364 9,228 9,592 16,387 6Z1 Table 3—Continued PERSONAL HEALTH EXPENDITURES BY TYPE OF EXPENDITURE AND BY SOURCE OF FUNDS FRANCE 1973 — MILLIONS $ OR FRS. Private Public Consumer Private . direct insurance Total Stage and SSeZleilt Total Total pay (mutuelles) ca u y 1% Frs. $ Frs. $ Frs. $ Frs. $ Frs. $ Frs. $ Frs._ 8. Eyeglasses and appliances . . . . 105.5 15.7 121.2 - 98.1 98.1 219.3 470 70 540 - 437 437 977 Total ambulatory care ...... 3,205.0 455.5 3,660.5 137.1 4,965.7 5,103.8 8,763.3 14,277 2,030 16,307 612 22,119 22,731 39,038 Total personal health expendi- tures ................ 3,547.8 583.6 4,131.4 600.4 10,246.8 10,847.2 14,978.6 15,804 2,601 18,405 2,676 45,643 48,319 66,724 Source: CREDOC (4} 400 1800 1500 300 250 1000 200 700 500 100 400 300 250 4.7 5“ 4-9 3.9 3'8 4‘0 % OF P.N.B. Figure 4. Evolution of medical can. 130 6.:— uo =ou83axm .w 233m wm N125“. N on . b \\ \ I on 8. W «W a. > \\ . no. I 8. I x . \ — #6. I \\ \ \ I 2 8. I \I‘ ~ we. I \ um. I | m5 8. I me. I g ow IFIL on 00 cm \\I\\\‘ m4<0m no wGZ coo ‘ c‘c OQ’O‘Q’O‘Q’c‘Ooo‘c‘ 9 O o’¢.“:¢.o,..%:.:,o ‘ Z 9.9.9 DENTISTS 9.5% Figure 2. National health consumption by suppliers of services and types of goods, percent, 1970. 170 Social Benefits in Kind // / PERSONAL HEALTH / CONSUMPTION 29.3% SPITAL “‘ H0 0485 \ 25.0% ‘ 99 ‘9? / "2’ 19.0% / 14.8% Trans. 0.5% 24.7% / .b / Eyeglasses 1.6% S / c /14.5% 5 Physicians NATIONAL '0.9% 8 17.2% HEALTH cows. '33 1.1% a ‘ \\\\\\\ a 7 Drugs $6 5 25.5% ' o - o 15.1% 00 0‘0?» 26.1% “E 22.5% Figure 3. National health consumption breakdown by categories of services, percent, 1970. National Health Consumption by categories of care Table 3 and Figure 3 show health consumption in 1970 analyzed according to whether the treatment was given either to hospital inpatients, ambulatory patients or patients treated at home, plus medical supplies, viz.: hospital care: 39 percent; ambulatory care: 32 percent; medical supplies: 28 percent; transport of patients: 1 percent. 171 National Health Consumption analyzed by establishments Table 4 shows that National Health Consumption was divided among the various health establishments as follows: (a) Public and private hospital establishments ......... 37.1 percent including: Public hospitals* ....................... (25.5 percent) Including: Inpatient care: 24.5 percent Outpatient services: 1.0 percent Private hospitals ........................ (1 1.6 percent) Including: Inpatient care: 11.3 percent Outpatient services: 0.3 percent (b) Private practice .......................... 29.2 percent including: Medical practices ....................... (17.3 percent) Including: Ambulatory and home care: 14.3 percent Treatment in private clinics: 3.0 percent Dental practices ........................ (9.1 percent) Including: Ambulatory and home care: 9.1 percent Treatment in private clinics: (Not available) Medical auxiliaries’ practices ................ (2.8 percent) Including: Ambulatory and home care: 2.7 percent Treatment in private clinics: 0.1 percent (c) Dispensaries ............................. 1.1 percent including: Public dispensaries: 0.2 percent Private dispensaries: 0.9 percent (d) Spas (without inpatients) .................... 0.6 percent (e) Laboratories (outside of hospitals and dispensaries) . . . 2.0 percent (f) Retail trade ............................. 27.1 percent including: Drugs and drug sundries: 25.5 percent Eyeglasses: 1.4 percent Orthopedic appliances: 0.2 percent (g) Ambulance services (and other forms of public or private transport) .............................. 0.6 percent (h) Inter-company in-plant health services ........... 1.0 percent (i) Health-care supply units run by government departments 0.7 percent Including: Maternity and child-care centers: 0.1 percent State, local administration, Social Security: 0.6 percent (j) Private, non-profit-making organizations .......... (not available) (k) Other health care supply units ................. 0.6 percent Total National Health Consumption ................ 100.0 percent *Including military and prison hospitals. 172 SL1 Table 3 MEDICAL CONSUMPTION BREAKDOWN BY CATEGORIES OF CARE IN 1970 Health Care Services _ _ _ Transport Inpatient hospital care Ambulatory care and home visits Medical goods of Sick Total Eyegl People Public Private Total Physic Lab-tests Auxil. Dentist Therm. Rehabilit. Total Drugs Ortho Total | 2 3= 1+2 4 5 6 7 8 9 10=4‘a9 11 12 |3=ll+lZ I4 ”gm“ Social benefits in kind (Soc. Sec.) . 29.3 19.0 48,3 15.1 2.6 3.6 5.2 0.5 012 2712 22.5 1.1 23.6 0.9 100.0 Pe_rsonal health consumgtion . 4 . . 25.0 14.8 39.8 16.2 2.3 2.9 9.7 0.7 0.] 31.9 26.1 1.1 27.7 0.6 100.0 Collective health consumption . . . 34.8 34.8 37.9 0.3 21.0 59.2 6.0 6.0 100.0 “In—plant” health consumption 6616 33.4 100.0 1000 National Health Consumption. . 24.7 14.5 39.2 17,2 2 2 3.5 9.5 0.6 0.1 33.1 25.5 1.6 27.1 016 100.0 Table4 NATIONAL HEALTH ACHVIHES Inpatient hospllul care (Moms Public Privalc ., . Total “I “” Fucs Other costs Free: Olller mm Production unirs 1 2 3 4 5 = l + 4 General and speciallted hospitals , .............................. 1 1.183 2 111.4. 5,069.8 16,370 4 Physicians and prime practice ........................... . . . . . . 1.397 4 1,3974 Demms In private practice ..................................... 1.7 1.7 Auxilmries ||'| prime praclice ............................... 41.8 42.8 Dispensaries ........................................ Thermal Spnng: Establishment ,.. . , , , , , ,. . . . . . ...... Laboratories ..................... , Pharmacies. . . , , , , , . . . .......................... Opticians ................................. Orthopedlsls .................................. Ambulance and mm transport, . . . . . . . . ................ . Inter-branches implant heallh survrccs .......................... . Maternity and children cenlers. . . , , . ...................... State (army, prisons) ............. ,. . .. . . . . .......... 105.4" 105.4 Locui authentic: ....................................... Socialsccurily. . . ...................................... Private administrations ......... . . . . . . . r . . , . . ........ p m pm p m p m omer enterprises ............ l l PERSONAL HEALTH CONSUMPTION ............................ . . |l,l52 2 0,624.! 17:106.} . l NA l'lONAl, HEALTH CONSUMPTION .............. , ,.. ,. ..... 11.2916 6.6241 17317 7 i . 'lAb-llrfli “Army and pnsonx hoxpmils ”'Hmllll van/inn In: salaried 0! public and privulr‘ uauuuiururmu [\llllllifl m Implnnl hullh )tnlnul p m: negligible In fact, a little more than 95 percent of health care was supplied by four major types of establishment, i.e. hospitals, private practices, laboratories and retail outlets for medical supplies. Dispensaries accounted for under 1 percent of the total. The remaining 4 percent was divided amongst the various establishments and services of government departments (in-plant health services, maternity and child care centers, etcetera). Certainly, some items of “collective” health consumption are underestimated, and the absence of any assessment for private organizations reduces the share for this type of production unit, but it is unlikely that a more accurate estimate would increase their share to any great extent. The source of funds for National Health Consumption As is shown in Table 5, health care funds were supplied from the following sources: 65.2 percent from the Social Security, made up of 50.1 percent for the basic and allied regimes, 7.3 percent for the agricultural regime and 7.8 percent for other regimes, e.g. railways, mines, non-salaried non- agricultural workers, etcetera. 5.9 percent from the state and local authorities, made up of 5.2 percent for medical aid and 0.7 percent for collective public services, e.g. school and university health services, armed forces health services, etcetera. 174 Table 4—Continued CONSUMPTION BREAKDOWN BY ENTERPRISE Health-Care Services Total Ambulatory car: and home mus Medical goods T311121“ Physicians Lab-tests Auxll. Dentists Thermal “fig” Total Drugs 53;?) Tom Pwpk Values 6 7 s 9 10 11 12:5211 13 14 ls=13+14 16 ”15531;“ 507.7 49.4 14.6 7.9 579.6 15.950.” 37.1 5,531.5 (5,534.5 7,9319 17,; 4.1664 4,1664 4,163.1 9.1 1,212.1 1,222.1 1,360.9 2.8 214.6 44.3 78.9 1814 519.2 519.2 1.1 294.0 294.0 294.0 0.5 914.2 914,: 914.: 2.0 [1,657.4 11.6574 11.15574 25.5 6413 641.3 (141.3 1.4 76.0 76.0 76.0 0.2 (269 11 (269.1) 0.5 309.8 152.7 462.5 462.5 1.0 29.0 29.6 59 2 59.1 0.1 74.4 0.6 29.1 0 3 104.4 13.9 18.9 234.7 17 01 9.1 1 26.0 26.0} g 0.6 ... ... u. ... pm pm pm pm pm 11111 pm pm pm pm pm pnl pm 146.2 75.3 (49.91 271.4 271.4 0.5 7,256.8 1,077.9 1,315.5 «.3557 294.0 (4991 14279.9 11.6574 717.3 12.374 7 (259.11 44130.0 (97.8) 7,833.3 1,008.5 1.6113 4,356.0 294.0 (49.9) [5.1.53.5 [1.676.] 717.3 [2,393.6 (169.1) 45,7333 100.0 4.4 percent from insurance companies and mutual benefit societies. 1.5 percent from companies (covering in-plant health services). 23.0 percent from households, i.e. the private individual is left to pay a little less than one-quarter out of his own pocket. The amount supplied from the various sources for each category of health care varies widely, due to the regulations governing Social Security which differ according to the type of health care and whether the establishment is public or private, together with the concentration of “collective” consumption and in-plant medical services into certain categories, e.g.: Inpatient hospital care: 80.5 percent for the Social Security, 10.5 percent for the State and local authorities, 3.3 percent for insurance companies and 5.7 percent for households. Ambulatory care: respectively 53.4, 3.2, 5.2, and 33.7 percent plus 4.5 percent for companies. Medical goods and supplies: Social Security 57.0 percent, state and local authorities 2.6 percent, insurance companies 5.3 percent and 35.1 percent for households. Transport of patients: 100 percent for Social Security, due to the method of assessment adopted. The variations within the sub-categories of care can be even greater, e.g. financing by public bodies (the state, local authorities and Social Security) was highest in the case of inpatient care in public hospitals (92.6 percent) and private hospitals (88.3 percent) and lowest in the case of ambulatory dental care (35.8 percent). 175 National Current Health Expenditure National Current Health Expenditure was estimated at Frs. 52,042 million in 1970 of which F. 45,734 million represented the National Health Con- sumption. The remaining “collective consumption” calculated in accordance with the strict definition of these terms, has been estimated at F.6,308 million; teaching and research F. 1,067 million; prevention F. 484 million; general administration F. 4,757 million. Calculated on a per capita basis, these expenditures work out as follows: National Current Health Expenditure per capita ...... F. 1,027 ($186)* made up of: National health consumption per capita ......... F. 902 ($163) (personal health consumption: F. 883) (public collective health consumption: F. 6) (“in-plant” health consumption: F. 13) Expenditure on teaching and research per capita . . . F. 21 ($4) Expenditure on prevention per capita ........... F. 10 ($2) Administrative expenses per capita ............ F. 94 ($17) A comparison of National Current Health Expenditure and the national aggregates National Current Health Expenditure amounted to 6.4 percent and National Health Consumption to 5.7 percent of Gross National Product. Moreover, Personal Health Consumption amounted to 7.2 percent of National Income. In the case of social benefits, it will be seen that the total social benefits in kind or in cash covering sickness and invalidity accounted for 38.8 percent of the total amount of social benefits paid out by public authorities (Social Security and the state) and that Social Security benefits in kind alone amounted to 28.7 percent of total benefits paid out by these organizations. Structure of National Current Health Expenditure The breakdown of 1970 estimates is shown in Table 7. The breakdown shown in Table 7 does not, however, give a really valid picture of the relative importance of each activity. In fact, due to the various duplications mentioned above, particularly in the case of prevention, teaching and research, the above percentages correspond to the restricted definition of these terms. Using the broad definition of these terms (duplication included) the percentages change as follows’r: Source of the funds making up the National Current Health Expenditure Because of the importance of health consumption in the total figures for health expenditure, the source-of-funds structure for the former determines to *31 = F. 5.53. *This of course implies that the total for these percentages exceeds 100 percent. 176 LLI ln-paticnt Hospital Care Table 5 NATIONAL HEALTH CONSUMPTION BREAKDOWN BY FINANCING SECTORS . _ _ Health Care Services Activmes Total Transp. Inpatient Hospital Care Ambulatory Care and Home Visits Medical Goods Total of Sick Categories . people of care Public Private Total Physic. :2: Auxil. Dentists Thermal Relhtahr- Total Dmgs [5:53:35 Total Values % FianCinl 5891015 | 2 3= 1+2 4 5 6 7 8 9 10=4a9 11 12 13=1|+12 1140=+3l+3 IS 16=14+15 Mia] Seeuritx General and related regimes ......... 6,714 4,434 11,148 3,437 607 885 1,205 127 48 6,309 5,003 262 5,265 22,722 198 22,920 50.1 Re’gimes for agriculture. . . . . . . . r . . :. 1,050 573 1,623 487 63 85 174 6 1 816 821 40 861 3,300 42 3,342 7.3 Other régimes ................... 977 675 1,652 578 93 112 171 13 _l 968 892 40 932 3,552 g 3,579 7.8 | 7 Total . . . . .................... 8,741 5,682 14,423 4,502 763 1,082 1,550 146 50 8,093 6,716 342 7,058 29,574 267 29,841 65.2 State and Local Authorities Medical assistance ................. 1,527 141 1,668 210 7 7 7 7 - 217 178 — 178 2,063 7 2,063 4.5 Veteran medical assistance ........... 81 27 108 63 3 8 1 1 7 76 126 7 126 310 2 312 0.7 Army and prisons health services ....... 111 7 111 65 1 20 ‘ 7 7 86 19 7 19 216 — 216 0.5 School health services ............. 7 7 7 9 — 9 7 7 7 18 7 7 7 18 7 18 Maternity and children centers ........ 7 — — 30 7 29 7 7 7 59 7 — — 59 7 59 0.2 Health services for public administration salaried person .................. — 7 7 17 7 9 — 7 _7 26 2 2 - 26 -' 25 2 7 Total .............. 1,719 168 1,887 394 4 75 8 1 — 482 323 7 323 2,692 2 2,694 5.9 37PrivateAdministrations ........... pm pm pm pm pm pm pm Pm pm pm pm pm pm pm pm pm 7 Private Insurance: 4 7 Mutual benefit and profit insurance societies ....................... 332 256 588 428 98 55 173 29 — 783 624 37 661 2,032 — 2,032 4.4 Business Enterprises 5 7 ln-plant health services ........... 7 — 7 456 7 228 7 7 7 684 7 7 7 684 7 684 1.5 6 7 Households ................... 502 518 1,020 2,053 144 171 2,625 118 — 5,111 4,014 338 4,352 10,483 7 10,483 23.0 I+2+3+4+5+6 = 7 National Health Consumption ........... 11,294 6,624 17,918 7,833 1,009 1,611 4,356 294 (50) 15,153 11,677 717 12,394 45,465 (269) 45,734 100.0 ‘ 0.5 million p m = negligible. Table 6 NATIONAL HEALTH EXPENDITURE AND NATIONAL RESOURCES Resident Population (annual average): 50 673 103 inhabitants 1. Gross National Product (GNP) ................ 808,436 106 F. 2. National Income ......................... 619,304 106 F. 3. Households’ consumption .................. 457,512 106 F. 4. Government and other public administration consumption .......................... 100,084 106 F. 5. Social benefits paid by administrations .......... 1 19,836 106 F. of which: Social benefits paid by Social Security 104,145 106 F. 6. Assistance transfers paid by public administrations 16,592 106 F. 7. National Current Health Expenditure ............ 52,043 106 F. of which: National Health Consumption ........ 45,734 106 F. of which: Personal Health Consumption ........ 44,730 106 F. of which: Collective Health Consumption ....... 320 106 F. of which: In-plant health services ............ 684 106 F. 8. National Current Health Expenditure per capita . . . . 1,027 Francs of which: National Health Consumption per capita . 902 Francs of which: Personal Health Consumption per capita . 883 Francs 9. Sickness benefits in cash and kind .............. 46,516 106 F. of which: Sickness benefits in kind ............ 29,897 106 F. 10. Medical assistance transfers (in kind) ............ 2,375 106 F. 11 = 7/1 National Current Health Expenditure/ GNP . . 6.44 % 12 = 71/] National Health Consumption/GNP ........ 5.66 % 13 = 72/1 Personal Health Consumption/GNP ........ 5.53 % 14 = Personal Health Consumption/Gross National Market Production ................ 6.24 % 15 = 72/2 Personal Health Consumption/National Income 7.22 % 178 Table 6—Continued NATIONAL HEALTH EXPENDITURE AND NATIONAL RESOURCES Resident Population (annual average): 50 673 103 inhabitants 16 = 72/3 Personal Health Consumption/Households’ Consumption ................ 17 = 73/4 Collective Health Consumption/Public Con- sumption ................... 18 = 9/5 Sickness benefits in cash and kind/Social bene- fits paid by administration ........ 19 = 91/51 Sickness benefits in kind/Social benefits paid by Social Security ............ 20 = 10/6 Medical assistance transfer/Assistance transfers paid by public administration ...... . . . 9.78 0.32 . . . . 38.82 28.71 14.31 % % % % % Table 7 SUMMARY STRUCTURE OF NATIONAL HEALTH EXPENDITURE 103 Frs. % Persona1 health consumption ................ 44,730,022 85.95 Collective health consumption .............. 319,899 0.62 In-plant health services .................... 684,000 1.31 = National Health Consumption .............. 45,733,921 87.88 Other current health expenditures ............ 6,3 08,669 12.12 of which: Collective sanitary prevention* ...... (484,238) (0.93) Medical teaching (includ. expenses paid by households)* .............. (481,656) (0.93) Medical research* ................ (585,231) (1.12) General Administrative Costs (includ. private insurance services paid by households) ................. (4,757,544) (9.14) 52,042,590 100.00 *Strict definition, see also Part 1. 179 Table 8 NATIONAL CURRENT HEALTH EXPENDITURE BREAKDOWN BY ACTIVITIES Breakdown of values Strict Definition Exten. Definition National Health Consumption ..... 87.88 87.88 Sanitary prevention ........... 0.93 2.90 Medical teaching ............. 0.93 1.47 ’ Medical research ............. 1.12 2.00 Health general administration ..... fl fl National Current Health Expenditure 100.00 * M Frs. $2,042,590.103 a great extent the structure for the latter (Table 9). For 65.2 percent of current health expenditure the source was Social Security, for 8.2 percent the state and local authorities, for 1.4 percent businesses, for 3.9 percent insurance companies and mutual benefit societies, for 0.3 percent private organizations and for 21 percent private households. For activities other than health consumption which were mentioned previously, the breakdowns are very different, e.g.: With the exception of expenditure included under the heading “health consumption” (immunization, etcetera) the source of funds for preven- tive health measures was almost exclusively the state and local authorities (90.4 percent) with the rest supplied by Social Security (9.6 percent)* Medical teaching (in its restricted sense) was financed for 86.1 percent by the state and 11.8 percent by private households—private organiza- tions and companies played a very minor role (2.1 percent) Medical research (in its restricted sense) was covered for about 90 percent by the state, 5 percent by companies, and 5 percent by private organizations (foundations, and so on). However, if one takes account of research financed out of the price of medical goods and services (particularly pharmaceutical research paid for partly out of the price for pharmaceutical products), the sources of funds for research in the *It should be remembered, however, that, in this report a very restricted definition is given to preventive health measures. 180 I81 Table 9 STRUCTURE OF ACTIVITIES BY FINANCING SECTOR $13?an T535115? 5:13;? $53322. 1335??? Consumption defsiflrilgon) definition) definition) tion Exgzfigiture State ....................... 5.87 75.42 86.08 89.48 4.24 8.04 Local authorities ---------------- 0.02 14.99 - 0.07 - 0.16 Social Security ................. 65.25 959 — 0.06 85.26 65.23 Business Enterprises .............. 1.50 - 0.45 5.13 - 1.38 Private Insurances ............... 4.44 - - - - 3.90 Private Administrations ........... - - 1.68 5.26 2.32 0.29 Households ................... 22.92 - 11.79 - 8.18 21.00 Total ....................... 100.00 100.00 100.00 100.00 100.00 100.00 broadest sense are as follows: state 51.0 percent, companies 45.6 percent, private organizations 30.0 percent and others 0.4 percent‘ Administrative expenses were split between Social Security (85.3 percent), private organizations (2.3 percent), the state (4.2 percent) and private households (8.2 percent). Of the total Current Health Expenditure, the source for almost three- quarters (73 percent) was collective public funds (the state, local authorities, and Social Security) and» for one-quarter (27 percent) the source was private funds—with private households contributing one-fifth, while companies, private organizations and insurance companies accounted for only a very small share. SOCIAL SECURITY GENERAL AND RELATED REGIMES PRIVATE BUSIN. ENTERPRIE Figure 4. Breakdown ofcurrent national health expenditure by financing sectors, percent, 1970. 3. THE DEVELOPMENT OF HEALTH CONSUMPTION IN FRANCE FROM 1950 TO 1974 The statistical analysis of National Health, Expenditure did not begin until 1970 and therefore long-range trends from 1950 are available only for health-care consumption. For part offthese 24 years, no data at all are available 182 581 Table 10 TRENDS OF HEALTH CONSUMPTION — 1950 - I974 Average 1950 1955 1960 1965 1970 1974* annual growth Values Current prices 106 Frs ............ 2926 6526 11,934 24,169 44,462 78,285 Index ........................ 100.0 223.0 407.9 826.0 1519.2 2675.5 +14.7% Prices** Index ........................ 100.0 141.9 185.7 233.2 293.5 369.0 + 5.6% Vol me (constant prices 1962) 106 Frs. ...... 6023 9466 13226 21,331 31,183 43,770 u Index ........................ 100.0 157.2 219.6 354.2 517.7 725.1 + 8.6% Health Consumption Frs ................ 69.9F 150.3F 216.2F 495.7F 771.6F 1484.1F per capita Index ............... 100.0 215.0 373.7 709.1 1103.9 2123.2 +13.6% Preportion in G.N.P. at current prices (%) .......... 2.9% 3.8% 4.0% 4.9% 5.5% 5.8% Proportion in total households’ consumption at current prices (%) .............................. 4.5% 6.0% 6.6% 8.4% 9.7% 10.3% *Provlsional. ”Implicit price index as a result of evolution of values and volumes. on collective health consumption (particularly maternity and child care and prison health care) and thus the estimates give a figure somewhere between Personal Health Consumption and National Health Consumption. In this report the term “health consumption” or “health expenditure” will be used in order to simplify the analysis and because of the fact that Personal Health Consumption, in other words market health goods and services, accounts for practically 98 percent of the total. In addition, statistics covering the period 1950 to 1959 enable expenditure to be broken down by only three types of treatment, i.e.: Ambulatory care and home visits (including physicians). Inpatient hospital care. Medical goods. Each of these three categories corresponds to the breakdowns used above (See Procedures and Table 1 in Part 1). Lastly, the health accounts use two price indices for pharmaceutical products; i.e., one based on the Consumer Price Index (CPI), and a second which is an average price for pharmaceutical items. The first index serves to deflate the value and estimate the volume (quantity plus quality), the second serves to deflate the value and estimate the quantities only. In this report only the CPI index will be used. Trends in the Value, Volume and Price of Health Care Between 1950 and 1974, health care consumption at current prices rose from F. 2,926 million to F. 78,285 million, which represented an increase of 14.7 percent per year and an increase in expenditure per person from F. 70 to F. 1,484; i.e. 13.6 percent per year. (F. 70 = $20.10 and F. 1,484 = $307.60)*. In terms of share of GNP it rose from 2.9 percent in 1950 to 5.8 percent in 1973 and has accounted for a growing share of household expenditure (1950 = 4.5 percent, 1973 = 10.3 percent). These increases have been at a regular rate over the years as will be seen from Table 10 and Figure 5. After a very inflationary period between 1950 and 1952 (+ 16 percent per year)‘t, the trend in the implicit price index for health consumption (see Figure 6) has been fairly regular (+ 4.7 percent per year). Over the whole period from 1950 to 1974 the index increased on average by 5.6 percent per year, which is practically the same rate as the total Consumer Price Index (+ 5.4 percent). Over this 24-year period, the increase in total volume at constant prices has shown a remarkably steady growth curve (Figure 7) at an annual rate of 8.6 percent. Volume per capita increased by 7.5 percent per year. However, the steady development of total health consumption is the result of a variety of trends for the different categories of health care as is shown in Table 11 and Figures 5, 6 and 7. *1950: $1 = F. 3497,1974: $1 = F. 4.824. TBetween 1950 and 1952 the Consumer Price Index (CPI) increased by an average of 14.4 percent per year, indicating a relative increase for health costs of 1.6 percent per year. 184 581 Index ‘500 4000 3000 2000 1000 800 700 600 400 1IllllllllllllIlllIIlHlllHIlll 300 200 IHHIIHIIHHIIIII] 100 1 l l | l | | 3000.4 Inpatient Hospital Care +15.2% 2675.54 National Health Consumption +14.7% ' / Ambulatory & Home Visits +14.6% ’ \Medical Goods +14.0% 2324.6 / Total Household Consumption +10.5% I | | | | l l I | 1 | l I l 1 I l | 1950 51 52 53 54 55 56 57 58 591960 61 62 63 64 65 66 67 68 691970 71 72 73 74 1950 = 100 Figure 5. Trends in values in health consumption (current prices); 1950 = 100. 981 Index 600 — 534-9 Ambulatory and Home Visits +12% ’/ 500 _— z : /472.5 _ E l,’ / Inpatient Cost per Day +6.7% I 1’ I 400 L ,-” ,/ Implicit Price Index ._ a ,_ ,’ / of Health Consumption +55% : ’1 /’ 369.0 : ’/’ /r /35246 : ’lz’ /’ I Consumer Pricelndex +54% _ ___r , I 300 : I/ / / _ / l r — 1’ ’/ I _ ’1’ ,_—-—/ ’/ .._ I, 1/ / ,— I’ / a/ —— a — ’I’ ” / 200 — , ,/ I I — ’1 ’/ — I _. ,’ —’/ 169.8 . _ ’--_’,I .’.——./ Medical Goods +22% _ ,I’ I / ’ I ._ _. /' - ,,/ 3/ Hz \.——-— —- , -_ . ' ./ _. - . _‘_~/ I. 100 I I l L l l l I l | I l l I l I I l I | l | I I 1950 51 52 53 54 55 56 57 58 59 1960 61 62 63 64 65 66 67 68 69 1970 71 72 73 74 Figure 6. Trends in price ind’eXEEfESO = 100.’ L81 TRENDS OF THE CATEGORIES 0F CARE Table 11 Index 1950 = 100 Average 1950 1955 1960 1965 1970 1974* annual growth Inpatient hospital care (public & private) Value (current prices) .................... 100 186.1 392.3 843.3 1601.3 3000.4 +15.2% Price index ........................... 100 131.4 184.6 251.9 340.0 472.5 + 6.7% Volume (constant pnce 1962) ............. 100 141.6 212.5 334.8 469.9 635.0 + 8.0% Ambulatory care and home visits Value (current prices) .................... 100 242.8 426.6 808.6 1482.9 2610.4 +14.6% Price index ........................... 100 158.9 225.1 309.1 409.1 534.9 + 7.2% Volume (constant price 1962) .............. 100 152.8 189.6 261.6 362.1 488.0 + 6.8% Medical goods Value (current prices) .................... 100 9.0 407.1 823.1 1454.2 2324.6 +14.0% Price index ........................... 100 127. 6 139.0 143.0 158.3 169.8 + 2.2% Volume (constant price 1962) .............. 100 195.1 292.9 575.6 918.6 1369.0 +11.5% *Provisional. Whereas inpatient hospital care, ambulatory care and medical goods have increased in value by an average of between 14 and 15 percent per year,* trends with regard to price and volume have been very different since 1952. The average annual increase in cost for ambulatory care was 7.2 percent from 1950 to 1974 (1952-1974 = 6.4 percent), for inpatient hospital care the figure was 6.7 percent (1952-1974 = 6.0 percent), and for medical goods (i.e., essentially the retail price of drugs and drugs sundries) the figure was 2.2 percent (1952-l974=1.1 percent).** Note also that the average price for ambulatory care shows a fairly regular trend. The same is true for medical goods, particularly from 1952. By contrast, however, the average cost of inpatient hospital care has evolved in successive phases of from 5 to 6 years, e.g.: 1952-1958, 1958-1963, 1963-1969, 1969-1974, each phase starting with a slow rise which then accelerates up to the end of the period. The effect of this is visible in the different trends in volume for each of the three categories of health care, that is: The volume of inpatient hospital care has increased at a fairly consistent rate of about 8.0 percent per year, with the exception of 1956 (15.0 percent) and 1968 (0.5 percent). The trend in the volume of ambulatory care and home visits can be divided into three phases, viz., 1950-1958 (9.1 percent per year), 1959-1962 (0 percent: a slight decrease followed by a slight increase), 1962-1974 (7.7 percent per year). With the exception of a period of stability between 1958 and 1960, the volume of medical goods consumed increased at an average rate of between 11 and 14 percent per year (1950-1958 = 14 percent per year, 1958-1960 = 1.3 percent per year, 1960-1974 = 11.6 percent per year)‘l'. An analysis of the value of total Health Care Consumption by category of health care (Table 12) shows that their share has not varied greatly between 1950 and 1960; i.e., inpatient hospital care 37 percent, ambulatory care and home visits 33 percent and medical goods 30 percent. However, it will be noted that, since 1960: The share for inpatient hospital care has tended to increase (1960 = 37 percent, 1974 = 43 percent) The share for ambulatory care and home visits has gradually decreased (1960 = 34 percent, 1974 = 32 percent) and primarily in the case of physicians and dentists *Obviously these trends have not been completely uniform and parallel. Inpatient hospital care showed only a slow increase between 1952 and 1955 (5.8 percent per year) and the same is true for the increase in ambulatory care between 1959 and 1961 (4.9 percent per year), as Well as for pharmaceutical products between 1958 and 1960 (6.4 percent per year) and for health care as a whole in 1968. “It should be stressed that the increase in the price index for pharmaceutical products based on the Consumer Price Index is certainly underestimated but that this increase is below that for both inpatient hospital services and practitioners. It is a Laspeyres index which only partly takes into account the rapid changes in the type of drugs used and the trend toward more recently introduced (often more expensive) products. TBecause of the difficulties involved in establishing a valid price index for pharmaceutical products, the estimated volumes and trends over a long period are to be treated with caution and are certainly overestimated. 188 The share for medical goods is progressively diminishing (1960 = 29 percent, 1974 = 25 percent). In the case of the breakdown by volume, the trend analysis is less clear because of the result of the probable underestimation of the increase in the price index for pharmaceutical products, which has caused a rapid increase in volume. If the price index for pharmaceutical products is adjusted roughly on the basis of the index of average prices for the quantities involved (see under Statistical Sources, above) the trends in the breakdown of volumes shown in Table 12 would appear to require revising in the following manner: A steady increase in the volume share for pharmaceutical products from around 25 percent in 1950 to around 30 percent in 1974. An increase in the volume share for inpatient hospital care from 36 percent in 1950 to 40 percent in 1974. A decrease in the volume share for ambulatory care from 39 percent in 1950 to 30 percent in 1974. Trends in the Source-of-Funds Structure for Health Care Consumption Since the 1974 figures are provisional, the source of funds for health-care expenditures can be studied only for the period 1950 to 1973. Over the course of these 23 years, the pattern has been as follows (see also Figure 8): Social Security: 1950 = 44.4 percent, 1973 = 68.3 percent Medical Aid and Veterans’ Medical Aid: 1950 = 14.2 percent, 1973 = 4.0 percent Insurance companies and mutual benefit societies: 1950 = 2.0 percent, 1973 = 4.0 percent Private households: 1950 = 39.4 percent, 1973 = 23.7 percent. However, this development divides into two distinct phases, as follows: (3) 1950-1959: During these first 9 years the respective shares for the various “sources” did not change a great deal. At most, there was a slight decrease in the share for medical aid with a corresponding slight increase in the share for private households, due to the fact that, with the rise in income levels, less use was made of the medical-aid system, designed basically for the low income groups. (b) 1959-1973: The Social Security’s share increased while the shares for medical aid and private households decreased—this was due to several factors: 1. With effect from 1960 a new agreement between private practitioners and the Social Security resulting in a higher coverage of medical health care expenditure by the Social Security (80 percent on average) in the case of treatment provided by physicians, dentists and auxiliaries.* . The setting up in 1961 of the Social Security regime for farmers. . The setting up in 1970 of the Social Security re’girne covering self-employed non-agricultural workers (tradesmen, independent busi- nessmen, professionals, etcetera). WM *By 1973 about 95 percent of the physicians and 99 percent of the dentists and auxiliaries were registered with Social Security. 189 061 Table 12 TRENDS OF THE HEALTH CONSUMPTION STRUCTURE BY CATEGORIES OF CARE (VALUES AND VOLUMES (CONSTANT PRICES 1962) — Not available ~Percent— 1950 1955 1960 1965 1970 1974* Value Volume Value Volume Value Volume Value Volume Value Volume Value Volume Inpatient hospital care 38.0 38.5 31.7 34.8 36.5 37.4 38.8 36.5 40.1 35.0 42.6 33.8 Public ........ ~ — — - (23.2) — (25.0) —— (25.1) — (26.8) 4 Private ........ — — — — (13.3) — (13.8) — (15.0) — (15.8) — Ambulatory care and home visits ....... 32.9 41.0 35.8 39.8 34.5 35.3 32.2 30.3 32.1 28.7 32.1 27.6 of which: Physicians (19.9) - (19.5) — (19.1) - (16.1) ~ (16.3) — (15.8) — Dentists. . (10.0) — (10.5) — (11.1) — (11.2) — (9.8) — (9.5) — Medical goods ..... 29.1 20.5 32.5 25.4 29.0 27.3 29.0 33.2 27.8 36.3 25.3 38.6 Drugs ........ (26.1) -— (29.6) — 26.4 — (26.9) — (26.2) — (23.8) — Eyeglasses, ortho- pedic appliances (3.0) — (2.9) — 2.6 — (2.1) — (1.6) - (1.5) - HEALTH CON- SUMPTION ...... 100.0 100.0 100.0 :100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 *Provisional I6I Index 3000— 2000 - 1000 — 900 ~ 800 — 700 600 500 400 300 ‘200 100 1950 51 Medical Goods ............ +11.5% ./ 725.1 National Health Consumption +86% Inpatient Hospital Care +8.0% Ambulatory 8: Home Visits +63% | I | l l l l I I l l l | l l l l l l l l l l 5 52 53 54 55 56 57 58 59196061 62 63 64 65 66 67 68 691970 71 72 73 74 Figute 7. Trends in volumes of health consumption (constant prices 1962); 1950 = 100. 4. An increase in the number of active wage-earners or salaried employees in the total population which increased the number of persons insured under the compulsory schemes, particularly the basic regime and allied regimes which in 1970 covered about 70 percent of . the total population compared with 57 percent in 1960. 5. A decrease in the contribution from state and local authorities in the form of medical aid for the same reasons as applied between 1950 and 1960, together with the transfer to Social Security in 1970 of part of the medical-aid contribution in cases of mental illness. Although very minor, the share for private insurance and mutual benefit societies tended to increase between 1950 and 1969. The principal reason for this would seem to be the growth in the salaried and wage-earning population, since a good number of mutual benefit societies are set up to cater for specific professions and to supplement payments made by Social Security. Their share has, however, progressively decreased since 1970 as a result of extension of the coverage of the Social Security scheme which in turn has to a large extent replaced the mutual benefit societies which provided health insurance for the professional classes. 1 00 90 80 70 lnsu rances 60 4o 30 20 10 0 1950 51 52 53 54 55 56 57 58 59196061 62 63 64 65 66 67 68 69197071 72 73 Figure 8. Trends in health consumption by financing sectors, percent, 1950-1973. 192 4. CONCLUSIONS Following this brief summary, many points of which would merit a more detailed analysis, several conclusions can be drawn, some of them in the form of questions. The desire to have concepts and methods which are as precise as possible is mainly so as to be able to define the limits of the area under study, because activities involving the population’s health are in fact complex and often scattered over other fields of economic activity. Obviously, such a method is based on a series of arbitrary definitions which as such can be criticized and challenged. However, the fact that this method is part of the National Accounts system and based on the internationally-accepted SNA definitions, means that it has at least some coherence with the methods used in macro-economic analysis. Personal health consumption accounted for about 86 percent of the total National Current Health Expenditure. Not only is it the major element but in addition, it is also the principal source for other health-care activities insofar as the cost for treatment (per diem hospital costs, price of drugs, and so on) provides part of the funds required for prevention, medical teaching and medical research. It should be stressed, however, that although the French health system may be considered as being basically a “market” system, it is nevertheless under public supervision—the state not only controls the public hospitals, the per-diem hospital rates, the coordination of investment in both public and private sectors, but also the training of medical personnel and a large portion of medical research. Furthermore, Social Security exercises an indirect form of control on the creation and development of the mechanisms of supply and demand through the system of registration for private practitioners and private clinics and by virtue of the fact that it is involved in the fixing of prices and in financing consumption. In fact, the health-care system in Franceis a market system which to a large extent is under public “tutelage.” A detailed analysis of the statistics since 1970 highlights the prominent part in the health-care delivery system played by hospitals on the one hand and private practice on the other, as well as the pharmaceutical industry and retail trade. Since private practitioners also work in private clinics and sometimes on a parttirne basis in public hospitals, there is a large amount of overlapping between these two systems of health care. The most important feature in the development of health care consumption between 1950 and 1974 has been the increased share for inpatient hospital care both in terms of value and volume. Next in importance has been the steady decrease in the share for ambulatory care, particularly that provided by physicians and dentists, both in terms of value and volume. And one would rank as the third feature the decrease in the share in value terms for drugs and drug sundries despite an increase in their share of the volume. Such trends raise several questions to which the overall accounts cannot provide an answer, for example: To what extent do practitioners, faced with the problems of a limited amount of time and a fixed scale for visits, unload part of their responsibility 193 for treatment by prescribing more drugs and/or transferring patients to hospital services? In view of the developments in medical technology, to what extent are individual practitioners tempted to support (or replace) their diagnoses and their treatments by the use of highly sophisticated methods provided by the pharmaceutical industry, hospitals, analyses, etcetera? To what extent are physicians subjected to pressure from pharmaceutical lobbies to prescribe more drugs? The increasing share for inpatient hospital care is partly due to the increase in admiésions and in the average number of days per hospital patient; but above all it is the result of an increase in the cost of the “production factors,” e.g. the increase in the number of staff per bed, the development of complex equipment for X-ray, analysis, monitoring, renal dialysis, and so forth. But, to a degree which is difficult to estimate, this is also the consequence of the increasing coverage of cost by public bodies which is higher in the case of hospital treatment than for ambulatory care. To what extent are the public and private hospitals complementary or competitive? Insofar as they are competitive, to what extent is this competition a source of increased cost? The increase in the amount of health-care consumption financed from public'funds (principally by Social Security which currently covers about 75 percent) introduces a new dimension not only in the behavior of health-care “suppliers” but also that of health care “consumers” and thereby in the relationship between physician and patient, hospital and patient, physician and hospital as well as in the structure of expenditure, without it being always possible, however, to isoalte the precise effect that this “financing” factor has on global trends. Lastly, certain of the observations and questions raised here might give the impression that the French health system was progressively developing into a type of national health scheme, but this is a question for the politicians to answer, since the statistical data provide only one of the elements on which to base such a decision. 194 BIBLIOGRAPHY FOULON, A. (1972), Les comptes nationaux de la Sante—Méthode— “Economie et Santé” No. 2, Paris: MinistEre de la Santé Publique et de la Sécurite Sociale. FOULON, A. (1973), La consommation médicale nationale en 1970 (Volumes 1 & 2), CREDOC Report, limited issues, Paris: CREDOC. FOULON, A. (1973), Evaluation de la consommation medicale de 1966 21 1971, CREDOC Report, limited issues, Paris: CREDOC. DURIEZ, M. and FOULON, A. (1974), La depense nationale de santé en 1970, CREDOC Report, limited issues, Paris: CREDOC. DURIEZ, M. and FOULON, A. (1974), La consommation medicale nationale, Evolution 1970-1974, CREDOC Report, limited issues, Paris: CREDOC. DURIEZ, M. and FOULON, A (1975), Evolution de la consommation medicale en France de 1950- 1974, CREDOC Report, limited issues, Paris: CREDOC. N. A. C. E. -O. S. C. E (1970), Nomenclature génerale des activites economiqu’és des Communautes Européennes, Luxembourg: N A. C. E. ROSCH, G. (1957), Les dépenses médicales en France de 1950 a 1955, “Annales du CREDOC” No.1, 1957, January-March, 1957, Paris: CREDOC. ROSCH, G. (1961), “La consommation medicale, ” Revue Economique No.2, March 1961, Paris: CREDOC. S.E..C - O.S...CE (1970), Systeme Europeen de Comptes Economiques Intégres, Luxembourg: S. E C. UNITED NATIONS ORGANIZATION (1970), Systéme de Comptabilité Nationale, New York: UNO. 195 DISCUSSANT’S COMMENTS ON CHAPTERS 6 AND 7 A. PETER RUDERMAN University of Toronto, Ontario, Canada I do not wish to criticize either of these papers technically, because Ithink the authors should be congratulated for a job of applied statistics in the great French tradition. When you read these papers, you know exactly what they did and why. Every term is defined clearly and lucidly and I thought the exposition was a model of what this sort of study should be. Even though the papers were translated word-for-word from French into English, there is an element of cultural untranslatability when you talk about any country in terms of the experience of another, so I shall address myself particularly to the Americans in the audience and tell them a little bit about some of the ways in which the context differs in France and, perhaps, in Europe in general. On drugs, just tell me briefly, do your series for drugs include drugs dispensed in hospitals? Mme. Sandier: No. (Dr. Ruderman): That is what I thought. For outpatient drugs, part of the reason they gave you for difference in utilization is quite correct; obviously if you have more ambulatory care and home visiting and fewer cases treated in the hospital, you are going to have more drugs prescribed outside of the hospital and entering these series. There are two other characteristics. One, I think, is cultural, perhaps French, perhaps generally European, but they prescribe a lot of placebos. You can’t go to see a doctor in France without getting at least five or six items on the prescription! And this may indeed be part of it. Secondly, you have to understand differences in marketing and distribution in Europe. In France the word is concurrence dé loyale. In Germany they talk about unlauterer Wettbewerb and these things mean disloyal competition or impure competition. There are laws that prevent businessmen from competing in price terms in many countries. The trend has been declining in recent years in Europe, but still, in general, you have rigged prices and you have what Jane Aubert at the University of Rennes used to call a sort of pre-capitalist economy in which the small merchant wants to have a guaranteed high profit with very little risk. So you must look at the marketing and distribution system to understand prescription drug costs. The difference in styles of medical practice also deserves emphasis because the American doctor is much more test, X-ray, and lab-oriented, partly because 196 of his training and partly because of the constant threat of malpractice suits, so that he always takes the last picture and does the last test just in case some lawyer some day asks him why he didn’t. Now, in French medical teaching, it is getting a little more this way but still, classically, the student is told to rely, first of all, on his ears, which means to take a good history, and secondly, on his hands, which means to make a good examination, and X-ray and lab reports count as supplementary information— nice to have but not vital. That is a tremendous difference in philosophy. You might also take into account that much of the attention that is paid to the heart in American medical schools is paid to the liver in France. As I remember it, there were at one time some 22 diseases of the liver taught in France which were considered in North America not to exist! There are different schools of medicine, different styles of practice, and this too, has to be taken into account. What fascinates me indeed is the fact that despite the manifest differences in so many of these aspects of health care, health behavior, you do have a comparability of general trend. Now, when it comes to care seeking behavior, another area which I hope you will look into some day, it would be interesting to correlate your economic indices for health care for the two countries with the epidemiological information, and you may turn out to have more cirrhosis in France, let us say, and more neurosis in the United States (at least these words rhyme) but whatever it is, it has to be studied and included in a total analysis. Another point of interest is that we have had a progression in international comparisons at this conference. First, you in the United States looked at Canada, which is a country where people do things in a slightly different way. You then moved on to the United Kingdom where, in addition to doing things in a very different way, there are tremendous institutional differences. Now we are talking about western Europe, which, as Professor Blanpain very sensibly brought to your attention, can be viewed as a rather homogeneous entity that has its own old and tried way of doing things. You have to remember that health-care insurance started in the 1880’s in Germany. These are the countries in a sense that had made their original commitment to the insurance basis by the turn of this century, and that was a long time ago. If I were to be asked to look for a model, I would say that the US. has probably got more to gain by studying the French, the German model, the Low Countries, possibly Scandinavia, but I should imagine Denmark rather than Sweden because the Swedes, like the English,have gone very far in terms of structural change. But France is probably structurally very similar. It is a good example of a non-system that sort of works and in this connection, remember that non-systems cost money. I am grateful to F oulon for putting in the figure for general administration as 9.14 percent. That would be a little higher than the Canadian figure, and would be double the cost of administering the National Health Service in England and the difference is probably attributable to bookkeeping costs, because in France, as in Canada and unlike England, you have to audit and you have to monitor and control the physician’s bills and you need review committees and taxing committees and you have to keep book on people’s contributions and measure their eligibility, 197 which means you have another set of clerks working on that all of the time and these activities inevitably raise the cost of the system. Indeed, one of the arguments for the British approach, in countries that can face the political shock that may be involved, is that it is an awful lot cheaper just to give everything to everybody than to start picking and choosing and setting rules that then have to be interpreted and administered. I hope things have improved in France in recent years. But a few years ago, it took one doctor to check on nine others (at least that was the figure that was commonly talked about). These review doctors worked, if not full-time then parttime, in Social Security, reviewing bills and things like that, and there was for a long time a state of open war between the medical profession and the Social Security which I understand has finally more or less ended. That is a final lesson that you can learn from France, incidentally, that as Americans with a rather offensively free-enterprise-oriented medical profession, you have now the example in France of an equally offensively free-enterprise-oriented group of doctors who have now come to heel. It took a long time, but the Social Security finally are exercising some control over these unpleasantly independent types, so if they could do it in France, perhaps that is the real lesson—that you can also do it in the United States. In Canada, the doctors have always been a little more amenable to reason, and in Britain there is another philosophy altogether, but in the French case, from the point of View of cultural identity, the medical profession is very similar in its outlook and aims as individuals to the medical profession in the United States, and therein lies the best lesson of all. Thank you, very much. 198 Chapter 8 HEALTH COSTS AND EXPENDITURES IN SWEDEN — THE PROBLEMS OF A PRIVATE GOOD IN THE PUBLIC SECTOR INGEMAR STAHL Institute of Economics, Lund University, Sweden Introduction A quick glance at the trends of health costs and expenditures in Sweden during the last decade will give a picture which is probably common to most modern industrialized economies; i.e. an increasing share of total economic resources being allocated to health expenditures in the very broad sense. (1, 2 / Although the statistical information is incomplete, it is not possible to discover any similar increases in indicators of the health status of the population. Substantial increases in a rough indicator—expected life-span——took place long before heavy expansion of the health-care system in the early ’sixties and afterwards. Another indicator, the number of reported sick days, shows a long-run upward trend. Experiences of this type are common to most countries and are not particularly embarrassing to the economists, who might in any case tend to look upon marginal health-service consumption in similar fashion to the consumption of other commodities. Planners and politicians in the health field are much more prone to accept 3 “requirements approach”—based on somewhat ambiguous concepts of “need”—in the hope of getting some type of final answer to the “health needs” of the population. A good example of this concept of health planning is given in an article on Swedish health planning by the director-general of the Swedish National Board for Health and Welfare, Professor Bror Rexed: (3) “The needs of acute somatic inpatient care for the forseeable future is believed to be covered inside the existing volume of hospital beds. . . Because of the increasing number of old people in our population, the need for long-term medical care will increase over the coming years. . . A strong priority will be for the expansion of primary medical care in outpatient health centers. . . It is believed that long-term psychiatric care will diminish in the future due to intensified acute treatment. The expansion of future psychiatric care will be in daytime clinics in the general hospitals and in a strongly expanded outpatient psychiatric organization. . . . 199 . . . The idea is to set up a public—health planning system providing also a continuous feedback of information to central authorities. The proposed planning system contains three planning levels with 30-year, 15—year and 5-year planning cycles. . . . . . . Long-term goals are being defined for the Swedish health—care system. We believe that only by paying regard to such goals and by using planning, rationalization and educational instruments to direct the development will it be possible to control development of the health delivery system. Only in this way can we channel its effects towards the most pressing health needs. . . . . . . There are still a number of health needs which are inadequately met in the population. There are new ways of prevention, which have not been fully utilized. We can foresee that medical research will give us new weapons in the diagnosis and treatment of disease, and we will have to use these possibilities to make the health delivery system work even more effectively. The present level of national expenditure for health will therefore be sur- mounted. . . .” By this rather extensive quotation I hope at the same time to show essential parts of the conceptual framework (needs, plan, etcetera) that is dominating decision-making in the Swedish health-care system, and some broad outlines of future plans (switching over from hospitals to local health-care centers, and so on). ‘ In the official documents, words like consumers’ demand, price and income elasticities, relative priCe changes, productivity increases, and similar, are very rare. Still, simple regression analysis of total health consumption as a function of total domestic consumption for the years 1963-72 gives a direct “expendi- ture elasticity” of 1.68 (SD = 0.06, R2 = 0.990) calculated on the basis of 1968 prices. This means that when total domestic consumption expenditures rose by 1 percent then consumption expenditures in the health sector rose by 1.68 percent. An equally simple trend analysis with time as the only independent variable gives an annual increase of 5.70 percent in fixed prices (SD = 0.19, R2 = 0.99). Is it possible that the elaborate planning system and the complicated political decision process as a final total result only produces extremely stable statistical trends or relationships? To understand some of the more specific features of the Swedish health-care system it will be necessary to break down the statistics from the national accounts and scrutinize the underlying organizational and decision structure. A good starting point is to look at the Swedish political structure and its development. In an analysis of the health field I think the following points are most important for an understanding of the development: (A) Since 1932 the Social Democratic Party has been dominant in government, except for short spells of coalition governments, but even then the Social Democratic Party retained dominance. The main ideology during this time embraced a keynesian approach to stabilization, a policy that included a highly selective labor market policy, the building up of a large-scale social 200 security system (with compulsory pension and health-insurance schemes and' large fiscal transfers from the working part of the population to families with children and old-age people) and, finally, heavy expansion especially during the 1960’s in the “social good” sectors such as health, housing, education and, later on, day-care centers for children. Liberal economists like Keynes and Galbraith probably had more influence than Marx. On the other hand direct government involvement in industry has been very low, with private ownership completely dominating. No industrial sector has been nationalized, and except for public utilities and transport, state ownership is limited to single firms, competing with private firms, in mining, steel, timber and pulp, shipyards and pharmaceutical products. Totally, state firms employ only about 5 percent of the labor force in industry, and in the expanding manufacturing industry state ownership is completely insignificant. Public policy has thus been very much oriented towards the growth of sectors like health. Two related features should also be remembered. There has been a deliberate policy in the social sectors to move away from the price system: free or almost-free delivery to the consumers of services in health and education, and subsidized self-cost pricing in housing; these have been political goals in a drive for anti-commercialism in the sectors mentioned. This means that different kinds of rationing devices have developed to bridge the gap between limited supply and consumers’ demand at zero-cost. In the health system direct queueing, waiting lists and considerations of “need” are frequently-used allocation devices. At the same time the insurance character of pension or health-insurance schemes has moved away from an individualized relationship between premiums paid and corresponding benefits towards a taxation format. In the compulsory health-insurance system there is at present small possibility of individualizing premiums and benefits (except for the income-insurance part of the health-insurance system where premiums and benefits are both income-related). The compulsory insurance system, and the suppression of market mecha- nisms and replacing the market with other allocation and planning devices, strongly demand an effective administrative or bureaucratic structure. In a rather small and homogeneous society as the Swedish, administrative allocation procedures might be relatively successful, although there is a strong tendency to create uniform, “well-organized” systems with small possibility of differenti- ation. With a small exaggeration one could say that the organizational charts of school, university and hospital systems look as if they had been copied from a military master-plan. (B) In this organization, the immediate responsibility for the production of health services in hospitals, in outpatient care centers or by district physicians lies with the counties (or as an exception with two major cities). The counties have the power to impose a direct and proportional income tax. In most counties this tax amounts to about or slightly above 10 percent of taxable income, and the overwhelming part of the county tax is for financing direct costs for health care and investments in hospitals, and so fOrth. Direct elections to county councils create the necessary contact with the electorate, although 201 health problems are not very much of a concern in election campaigns (except for local questions regarding the localization of hospitals, health centers, etcetera). The central state involvement in the health field is limited mainly to planning and general supervision by the National Board for Health and Welfare (NBHW). All county plans and allocations of physicians between hospital departments have to be approved by the NBHW. Another state responsibility is the training and licensing of physicians at the six medical faculties at the universities. The direct financial involvement—except for part costs of training hospitals—is limited to the “evening out” of tax receipts from high-income counties to low-income'counties. The social insurance system (SIS)~covering pensions as well as health insurance—is formally a public corporation but for all practical purposes may be regarded as part of the central government. The main part of the health insurance scheme is an income-related income-insurance during illness. To a small degree, however, the SIS contributes to the costs of the county health-care system and to a somewhat larger degree to the still-remaining private health-care sector. As probably the reader has inferred from the preceding text, the private sector in health care is rather limited and decreasing in relative importance. Since a 1974 reform, private physicians are paid a fixed sum from the customer per treatment (irrespective of the treatment specifics) of 20-30 Sw.Crs. The rest of the physicians’ fees are paid by the SIS according to a standard scale. Of about 12,000 physicians in 1972 more than 8,500 were publicly employed. Possibly in the form of an adaptation to circumstances caused by a decreasing private sector, in recent years enterprises and public authorities have shown a strong tendency to integrate some health services into their firms and administrations. In 1973 dental care was integrated in the SIS. This part of the health-care system has been dominated by private dentists. According to the new rules, 50 percent of standard fees are paid by the patient up to 1,000 Sw.Crs. The rest is paid by the SIS. For expensive treatments (above 1,000 Sw.Crs.) and for preventive treatments, the SIS pays 75 percent of the fees. A similar cost-sharing plan is used in paying for prescriptions of pharmaceu- tical drugs, with the patient paying only up to 15 Sw.Crs. for each prescription and the rest being paid by the SIS. The pharmacies, which earlier had a very traditional organization that included public licensing of private ownership, Were nationalized in the 1960’s and organized as a state public utility with close connections with the NBHW. However, the state-owned industry KABI—specializing in antibiotics and globulin—operates on a purely competi- tive basis. Thus, the main tendency has been toward increasing public control of the health sector, coupled with increasing subsidization of consumer demand. At the same time there has been stress on a thoroughly-planned, organized, coherent or uniform structure, partly in an attempt to solve the delicate allocation problems which occur when private goods are distributed outside the market and by administrative mechanisms. 202 In the following paragraphs I will concentrate on the econOmic aspects of the Swedish health-care system, as many of the institutional and planning as- pects have been recently covered in a publication from the DHEW {4). As an excuse for some of the shortcomings it must be noted that statistical data are very much “input”-based while the information of actual output from the sector is scarce. In the next paragraph we will study the growth of medical consumption. This is followed by a discussion of some of the recent investments trends in the health sector. These paragraphs are then followed and summarized in a section on financing, including a presentation of the relevant tax and insurance schemes. The succeeding section is a presentation of some of the more important input-factors and a sketchy review of the organization. The last paragraph includes a short discussion on some output measures. In a concluding remark (and Apppendix) there is an attempt to summarize an economist’s interpretation of the Swedish health-care system. Growing Health-Services Consumption In 1963 (the first year of the revised national accounts system) the total private and public consumption in the health sector amounted to 4,200 million Sw.Crs.* out of a total private and public consumption of 69,000 million Sw.Crs. (6.1 percent). In 1972 the share spent on health care had increased to 9.3 percent and in current prices health consumption amounted to 13,900 million Sw.Crs., with a total consumption amounting to 148,700 million Sw.Crs. (see Statistical Appendix Tables 1 and 2). Measured as a share of GNP at factor costs, health consumption in 1963 was 5.1 percent and in 1972, 8.1 percent. This reveals a feature well-known in most industrialized countries, and although the share is high on an international basis, the figures are probably well explained by the rather high per-capita income in Sweden. As was earlier stated the growth rate or expenditure elasticity was extremely stable. If the statistics are recalculated on the basis of fixed prices (1968 prices) there are some changes. The share of total consumption in 1963 would then be 6.9 percent and this would have increased to 8.4 percent in 1973. These changes are formally explained by the high share of fast-rising wages in the health sector and by the fact that national accounts only measure inputs and not outputs in the health sector. It is therefore still an open question whether the difference between current price and fixed price calculations reveal a slow productivity growth (increasing relative prices for health) or is just a result of the primitive measurement techniques in which calculations are based on input indices and not on non-existing output indices. Of some interest is the fact that the share of private health consumption (mainly including visits to private physicians or publicly-employed physicians in ambulatory care, and dental care and direct purchases of drugs) in relation to total private consumption was almost constant during the period concerned. *In the following we will constantly give figures in Sw.Crs. At current exchange rates 4 Sw.Crs. are equal to $1 US. A purchase-power parity concept might indicate a rate of 5 Sw.Crs. to 31 US. 203 Public health consumption (hospitals and outpatient care at hospitals) increased its share of total public consumption (from 17.9 percent to 23 percent in current prices) as well as of total consumption (from 4 to 7 percent). It is stressed that the division between private and public consumption is somewhat artificial as parts of publicly-supplied care are included in private consumption, and private consumption as here defined (“non-hospital care”) has been subsidized at an increasing rate. Still, a small amount of privately-paid fees goes into the public consumption part. The main tendency is clear: an increasing part of the steady growth in health services consumption was produced by publicly-owned hospitals. As will be discussed later, the share of the costs paid from taxes or by the health insurance scheme also increased. In the following Table the growth rates are summarized: ANNUAL GROWTH RATES 1963 - 1972 Current prices 1968 prices Total health consumption 14.2% 5.5% Total consumption 8.9% 3.3% GNP 8.5% 3.8% Source: National accounts. (5) Investments in the Health Sector As private hospitals are insignificant in the broad setting and all new hospitals are built for the public sector, investments in the public-health sector will give a sufficiently good indicator of the total investment volume. Table 3 in the Statistical Appendix presents the development of investments. Invest- ments in buildings make up the great part—roughly about 80 percent. It should also be observed that there is no stable relationship between investments and consumption. Development shows an increasing investment volume until 1970 and a decreasing one thereafter. The immediate explanation is that the 1960’s represented a period when the increase of public interest in health services was largely directed towards reinvestment in (and in many cases completely rebuilding or extending) old hospitals, and creating a limited number of new hospitals. During the 1950’s the annual investment volume in buildings in the health sector was around the 300 million Sw.Crs. level (1968 prices). But in the 1960’s the volume increase drastically to the billion Sw.Cr. level. The total capital invested in buildings in the health sector from 1950 to 1963 was 14,000 million Sw.Crs. (1968 prices). As rental values were included in the presentation of public consumption only to a very limited extent, the actual figures for health consumption—about 10,000 million Sw.Crs. in l972—is made more complete in a comparison with general private consumption ‘only when 204 an imputed rental value of 1,000-1,500 million Sw.Crs. is added. The disregard of imputed rental values thus makes up for an underestimate of about 10 percent of the value of total consumption. Still, after a correction of this type the labor-intensive character of health care is not substantially changed. Financing the Health-Care System A starting point for a description of the financing of the health-care system might be the situation of an individual. After the changes made in 1974 the mainlines of the system remain as follows: (A) In the public sector of the outpatient care—district physicians, health-care centers and outpatient departments of the hospitals—the individual himself pays a fee of 12 Sw.Crs. (to be raised to 15 Sw.Crs. during 1975). This fee is below the price for a haircut or a cinema ticket and can be regarded as a token. In some counties there are proposals for a reform with an annual maximum—100 Sw.Crs.—total individual fees to be paid by any one person; such fees are independent of the type of care. If the person prefers to visit a private physician he will pay 20 - 30 Sw.Crs. directly to the physician and, as in the public sector, this fee is independent of the type of medical services offered. Some special types of medical services are completely free of charge, including health controls of children and pregnant women, and the prescription of contraceptives. In any case the difference between 0 and 20 Sw.Crs. may in most cases be insignificant in comparison with some of the indirect costs: queueing time in a waiting room, waiting lists and so on. It is obvious that one effect of the low price is a constantly excessive demand for services, which is dealt with by some of the rationing devices mentioned. However, relatively little is known about the quantitative importance of these rationing systems, but the existence of queues exerts pressure on the politicians to expand the capacity. A specific feature of the Swedish system, partly due to the (previous) shortage of physicians, and partly due to the‘constant excessive demand, is that home visits by physicians are oflow frequency; patients go to see the physician in his clinic. (Travel costs are also subsidized by the health-insurance system.) Dental care has been integrated in the system since 1974 and, as previously mentioned, the patient pays half of the fee up to 1,000 Sw.Crs. and above this amount (and for some preventive care) 75 percent. The reform in 1974 increased demand and created a situation of excess demand and longer waiting times. Dental care for children is managed by the public dental system while care for adults still is largely the realm of private dentists. Care in a hospital is free except that the health-insurance system makes a small deduction from sick-payments (10 Sw.Crs. day) which is transferred to the hospital. The only part in the health system in which private fees might be of considerable importance is in nursing homes for old people, for which the counties use an income- and wealth- related fee scale. Pharmaceutical drugs on prescriptions are sold to the patients at a standard fee of 15 Sw.Crs. (to be raised to 20 Sw.Crs.). There is continuous discussion 205 with the argument that this solution takes away some cost-control incentives from physicians and create new incentives in the form of over-prescribing with large quantities or with multiple drugs. As can be seen from this review of the main rules, two different principles are used: in some cases, e.g. dental care for adults, we have a cost-sharing system; in most other cases the patient pays a small fixed sum regardless of the type of treatment. (B.) The health-insurance system pays to county administrations a fixed fee per consultation in outpatient care of 48 Sw.Crs. (to be raised to 60 Sw.Crs.). Private physicians are paid according to a scale: the difference between the stipulated tariff for a specific service and the 20 or 30 Sw.Crs. paid by the patient. Dentists are also paid the difference according to administrative tariffs. Pharmacies are paid from the system in a similar way. This means that, since private physicians and dentists were strongly integrated into the system in 1972 and 1974, a system of price control has since been introduced or sharpened—with tariffs partly negotiable between the professional organiza- tions and the National Board for Health and Welfare. The salary level of physicians and dentists is thus partly under the control of the authorities, and administrative control has also been extended to control the entry of professionals, especially new private dentists. It is also likely that the new payment system gives larger possibilities for a tax control of private physicians and dentists, as the non-subsidized market for health care is very small. For private physicians there is a maximum number of consultations that can be paid from the insurance system; at present this is about 6,000 per year. (C.) Since the beginning of the 1970’s all physicians employed in the public sector have a fixed salary, according to a common salary scale. A similar salary scale is applied to dentists in public service. The main intention behind the change from a salary scheme with a fixed salary per month (plus a sum per consultation) to a completely fixed salary was to improve the quality of the services and to avoid some of the “piece-work” nuisances in the medical profession. Very likely, however, this change had a supply-decreasing effect regarding the number of possible consultations, and may thus have increased the excessive demand simultaneously created by the low, fixed patient-fees. It is however difficult to separate the effects, as at the same time the number of physicians was rapidly increasing because of an earlier expansion of the education system. High marginal taxes—about 80 percent-marginal tax rate—in the income brackets applicable to the medical profession, in conjunction with a more effective tax control system, might also have increased the preferences for shorter and more regular hours. The county councils pay the remaining cost when fees have been collected from the patients and the health insurance has paid its fixed sum per consultation. However, there is no individual charging or invoicing for each consultation and although the amounts paid to the councils (12 + 48 Sw.Crs.) . in most cases are below the actual costs, there may be consultations giving a surplus to county councils. 206 To summarize, the larger part of outpatient services costs are paid by the insurance system (and the patients) including drug prescriptions, while the overwhelming part of hospital care costs is paid by the counties. (D.) This picture of the financing system with a sharing between the counties and the health insurance system is not due to any longterm rational consideration, but rather can be looked upon as a result of an historical process. As has earlier been mentioned the county tax is a proportional direct tax on taxable income. The fees structure of the health insurance system— including the sickness benefit scheme, which is discussed further on—are rather complicated. From a formal viewpoint the health insurance system is a part of the Social Insurance System, including the compulsory pension scheme. The SIS is run as public corporation (excluded from the state budget) but it can for all practical purposes be regarded as part of the state sector—as all important decisions have to be made by parliament and are enforced by the Social Insurance act. First, there are direct premiums paid by the individuals (together with the state tax, the county tax and the local municipality tax) which at present amounts to a fixed sum of 300 Sw.Crs. per insured per year, and a strictly proportional fee of about 1.6 percent of taxable income up to a level of 60,000 Sw.Crs. For self-employed persons the proportional premium is somewhat higher. Second, there is a wage tax paid by firms, which at present is about 4 percent of taxable income up to the level of individual fees. This tax or “premium” will increase to 7 percent in coming years and the income ceiling will be dispensed with. Of the total sum 10 to 31 percent is intended for health-care financing and the rest is for sickness benefits. Third, the state pays 55 percent of the health-care costs of the insurance system and an equal part of the basic benefits in the sickness benefit scheme. The state tax is either a proportional indirect tax, tax on some commodities (alcohol, tobacco) or a direct progressive income tax. The possibilities for an individual to get a complete overview is certainly very limited. Roughly, however, the premium system, viewed in total works as a fixed basic sum plus a slightly progressive tax when all the different types of premiums and taxes are added. Above a certain point there is, however, a regressive character. (B) As was mentioned above, health care financing is partly integrated with the sickness benefit scheme. The latter system operates broadly as follows: The system is compulsory for all employed. Since 1974 the benefit scheme has changed from tax-free benefits to taxable benefits. The benefits are related to income lost during sickness and the compensation level attained is 90 percent of income lost. In the system having tax-free benefits it was possible in some cases to reach a level about 100 percent—due to “kinks” in tax and other social-benefit scales. Benefits become payable the first day after the insurance office is notified by the person concerned. (It is thus possible to phone a message in the evening and get the benefits the next day.) As with the premiums, there is a ceiling of 60,000 Sw.Crs. (in constant prices). High income earners are in most cases covered for the top of their incomes by direct 207 agreements in salary contracts. For housewives, students, and similar, there is a possibility of voluntary insurance integrated in the general system. In the case of long-term sickness there is a possibility that the individual can be transferred to the pension scheme of the SIS. It should be stressed that there is no differentiation in different risk-classes, which gives the system a touch of a tax-transfer system rather than an insurance system. Another part which is basically a tax-transfer is the “parenthood-insurance” completely integrated in the health insurance system. This gives either parent the possibility of being at home with a 90 percent income-coverage for 6 months (with a proposed extension to 7 months). In some cases parents are eligible to sickness payments if a child falls sick (there is a maximum of 10 days per annum). Also, there have been recent changes in benefit schemes, but details of premiums for health insurance and few statistical data are yet available from operations in 1974. It may still be of some value to look at earlier data—the new developments are all in the direction of larger coverage by public payments and increasing emphasis on the “tax-transfer” character of the system. From the national accounts from 1972 we get the following picture: Sector/Item Million Sw.Crs. Private consumption 3,400 (a) Drugs 1,260 Households 520 Insurance 740 (b) Other products and therapeutic equipment (mainly households) 310 (0) Physicians’ services 1,820 1. Private physicians 240 Households 120 Insurance . 120 2. Public physicians outside hospitals 550 Households 100 Insurance 450 3. Private dental care 950 Households 930 Insurance 20 Source: National accounts (5) 208 The total payments from the households (excluding the fees for public hospi- tals’ outpatient care) is thus about 2,000 million Sw.Crs., making up less than 60 percent of the health care labeled as private consumption. About one-half of these payments were for dental care. The reforms since 1972 have the effect that at least one-half of dental care costs is paid through the insurance system, and probably that the insurance part of private physicians’ payments has also increased. The privately-financed part of total health consumption would in 1972 be about 15 percent (but with an almost complete concentration on dental care, private physicians and drugs and equipment consumption). The financing system will have obvious allocation effects, as it not only changes income distribution but also relative prices. According to the degree of subsidization we will find hospital care at one end of a range going over to outpatient visits in the public system, visits to private physicians, drug consumption, therapeutic equipment and dentists. The recent reforms have probably meant that dental care has moved to the middle of the scale and that the distance between private and public outpatient care has decreased. To understand the structure of the Swedish health care system and the relative distribution of excessive demand, and queues or waiting lists, these changes in the relative prices have to be observed. It should also be noted that cost—sharing between the insurance system and the counties probably has very small effect upon allocation and distribution. The tax character of both financing systems is almost equal. The only trend created by cost-sharing may be that county councils have a tendency to increase the production of services with a high degree of subsidization from the insurance system—Le, cheap services. However, it is unlikely that this tendency is of any importance when taking into account the workings of the hospitals’ budget and managerial system. Public sector financing is shown in the next Table. It should be observed that the figures presented are from budgetary and financial statistics and are not completely coherent with the national accounts data. PUBLIC HEALTH CARE (NET CF PATIENT FEES) 1972 Operating costs Investment costs Million Sw.Crs. % Million Sw.Crs. % State 2,680 21.9 230 15.3 Insurance system 1,800 14.7 - Counties 7,770 63.4 1,280 84.7 Total 12,250 100.0 1,510 100.0 Per capita, in Sw.Crs.: 1,510 185 Source: Health statistics. (6) 209 A breakdown of the figures for the insurance system gives the following result: Million Sw.Crs. (1972) Medical care and dental care (outpatient) 580 Hospital care 232 Travel expenses 123 Other expenditures for care 125 Pharmaceutical drugs 740 1,800 Source: Social Insurance statistics. (7) (Compared with the national accounts data for financial statistics include parts of the subsidies to private consumption in the public sector). The complete revenues and expenditures of the insurance system are given in the following Table: Insurance system revenues Million Sw.Crs. % Fees from individuals 2,003 34.2 —of which for medical care 569 9.7 Wage tax 2,952 50.4 —of which for medical care 987 16.8 State contribution 843 14.4 Interest, et cetera 46 1.0 5,844 100.0 Source: Social Insurance statistics. (7) Insurance system expenditures Million Sw.Crs. % Sick benefits 2,980 52.6 Transfers 60.8 Maternal benefits 466 8.2 Health care 965 17.0 (incl. travel expenses) SubSidies Pharmaceutical drugs 1,801 740 13.1 31.9 Other expenditure 96 1.8 Administration* 41 5 7.3 5,662 100.0 Source: Social Insurance statistics. ( 7) *Not included in the health sector in the national accounts. 210 The main conclusions are as follows: Almost all of the medical care in the public sector is administered by the counties. They are financed by a county tax (for main part), state subsidiesflmainly as lump-sum transfers from the state to the counties, but with some exceptions in which there are incentive effects—and contributions from the insurance system, mainly depending on the number of consultations or “bed-days” produced, but otherwise disregarding the type of service. Thus, the insurance system in this context does not operate as a private insurance company paying a part of or the whole bill for a specific treatment for an individual patient. As will be discussed later, this fact may have some influence on the system’s incentives and cost control. The private part—outpatient care and dental care—is based on a combination of private fees and insurance contributions. There is a marked difference between medical and dental care; in medical care insurance pays the residual above a fixed fee while dental care is based on proportional cost-sharing. The compulsory sick-benefit scheme is financed out of insurance premiums of tax character. There is from an economical point of view no difference between the direct premiums and the wage-tax premiums paid by the employers. Added to these there are contributions from the state, financed by the state tax. It is obvious that the complete structure of this system is not the result of a careful, once-and-for-all rational consideration. There are serious difficulties in getting a complete picture, especially as different political bodies and administrations are involved. But at present there are no serious attempts toward a more streamlined design, except that the tax-transfer character of the sick-benefit schemes has become more obvious. Organization and Factor Inputs In this section we give some basic data about the organization and factor inputs in the health-care system. A breakdown of a subset of the public hospitals’ costs for 1972 is given in the following Table: Hospital costs 1972 Million Sw.Crs. % Wages, salaries and pensions > 4,706 72.6 Drugs 215 3.3 Nursing requisites 200 3.1 Administration and external Services 718 l 1.1 Depreciations etc. 627 9.7 Other costs 13 0.2 6,479 100.0 211 Wages are obviously the main factor of production. The rules for calculating depreciations are in this case based on the hospital accounts and there is a difference between this calculation and the principles used in the national accounts. From the national accounts (1972) we get the following picture: Private sector Public sector Million SW.Crs. Million SW.Crs. Prod. value 1,248 1 1,677 Inputs ‘ -169 —2,802 Value added 1,079 8,875 Salaries 459 8,323 Surplus 615 - Depreciation - 414 Corrections 5 138 This gives the result that salaries in the health sector made up roughly 7.1 percent of the total wage bill in the Swedish economy. (Surplus in the private sector is mainly incomes for private physicians and dentists being considered as “entrepreneurs.”) The same ratio was 5.2 percent in 1967 and 4.4 percent in 1963. A similar comparison of total employment (Table 4 in Statistical Appendix) gives the result of a share increasing from 4.2 percent in 1963 to 5.1 percent in 1967 and 7.3 percent in 1972. This means that employment has gone up faster than the share of the total wage bill. An explanation is partly given by an increasing share of parttime employment in the sector. Out of 218,000 employed in a “health occupation” 1972, almost 209,000 were women. The comparison between “employed in the health sector” and “employed in health occupations” also indicates an increasing number of non-medical personnel. A very strict definition of medical personnel and its breakdown is shown in Table 5. Taking the total number of hours worked, the health sector increased its share from 4.1 percent in 1963 to 4.8 percent in 1967 and 6.9 percent in 1972. This is a somewhat slower increase than the share of total employment and the . share of the wage bill, indicating a relative decrease in working hours made up by an increase in parttime work as well as the sign-unknoWn effect of average working hours and a decrease in relative wages. As a total this latter effect is, however, rather insignificant but it may hide quite different trends. For a long time, when there was a serious excessive demand for physicians the salary level was relatively high. Studies of salaries for physicians indicated a total net benefit (gross incomes minus education costs minus incomes foregone) for medical training of the size of 1.9 million SW.Crs. in 1968, and 1.1 million SW.Crs. in 1960 (1968 prices) as compared with medium-level technical training. No other profession turned up with benefits of this size (dentists 0.6 million SW.Crs. in 1968, secondary-level teachers 0.1-0.3 million SW.Crs. in 212 1968) (8). At the same time the relative position of low-income earners in the field has improved. Since the late 1960’s there has been a leveling out of wage differences between different groups, which is to a large extent due to increases in personnel trained in the education system. The number of physicians in the Swedish system has been rather low by international standards. In 1972 Sweden had about 700 inhabitants per physician, when countries like Austria had 540 inhabitants per physician. The limited output from medical schools during the first decades after World War II was probably caused by a combination of pressure from the strong professional medical association and planning process thinking based on a need concept— one simply could not foresee the rise in consumption while using a requirements’ approach. The physician shortage partly explains the typical Swedish distribution between hospital care and outpatient care. The total number of visits to a hospital outpatient department or a district physician amounted to 14.6 million visits in 1972 (Table 7) which is about 1.8 visits per inhabitant per year. Even if the private sector is included—with no statistics available—the general picture is not altered. At the same time by international comparison the number of beds was high. The increased number of bed- days produced was mainly produced in hospitals for chronically sick persons (Table 6). A likely explanation of the specific Swedish structure, then, is that with a limitational factor of production—physicians—increased consumption took the form of care in its original sense, which required to a large extent other factors of production. Increases in the number of physicians make it possible to increase outpatient services. There are predictions that at present graduation rates the number of physicians from the faculties will increase from 15,800 in 1974 to 26,000 in 1985. This will of course change the relative prices of different types of services. One must, however, be very skeptical about the possibility of decreasing costs by increasing the number of outpatient services in a hope of relieving the hospitals. A personal opinion is that just one of the rationing factors at present—patient time in queues and patient waiting lists—will decrease and that suppressed demand at patient zero cost will get out in the open. The relatively important increase in pure care is possibly explained by the fact that the public sector (at low prices for the patients) has taken over care-production—services that in many countries still are produced at home by relatives, and so on. An indication of the changing structure of the direction of investments is given by the following comparison between the composition of the county councils’ investment budgets in 1963 and 1974: 26: 2.9.71 Mental care (including physically retarded) 10.5 19.3 Long-term care 7.6 33.6 Hospitals 81 .9 33.7 Outpatient care 13.4 100.0 100.0 213 From 1960 to 1972 the number of visits for hospital outpatient care rose from 5 million to 9.3 million visits, i.e. an increase of about 85 percent. During the Same time bed-days increased from 38 million to 43 million, i.e. an increase of 13 percent. In real prices there is, however, more than a doubling of costs during the period. However, using present data it is not possible to equate these changes with quality changes, with increased relative prices due to slow development of productivity and errors in the statistical measurement of price indices. In Table 8 some data of costs-per-day and treatment are given in current prices. To get around problems of this kind it would be necessary to use more sophiSticated output measures such as the costs for different treatments. In the Swedish system data of this type are rarely produced. The budget allocation process at a hospital is mainly carried out by staff allocated to different departments, and very little internal pricing is used, and in this financing system there is no call to produce cost data for individual patients or treatments. This fact may also have an effect on the cost-consciousness of the different departments. But experimental studies with the use of PPBS-systems and internal pricing are under way, and in the long run these may give a better picture of the cost-structure of Swedish medical care. 214 APPENDIX A STATISTICAL APPENDIX 215 LIZ Table 1 HEALTH CONSUMPTION IN SWEDEN 1963-1972, (1968 PRICES) A B C D E F G H I . 31211111153111 dgtztgsltic A/B $133115: Iiln Total pub1ic D/E 35.133.263.115; $132365: G /H the pnvate pr1vate' % the pubhc consumptlon % A + D B + E % sector consumptlon sector 1963 ........ 1,905 65,373 2.9 4,155 22,643 18.4 6,060 88,016 6.9 1964 ........ 2,017 68,328 3.0 4,389 23,236 18.9 6,406 91,564 7.0 1965 ........ 2,154 71,209 3.0 4,633 24,431 19.0 6,787 95,640 7.1 1966 ........ 2,234 72,760 3.1 4,933 25,724 19.2 7,167 98,484 7.3 1967 ........ 2,327 74,778 3.1 5,438 26,879 20.2 7,765 101,657 7.6 1968 ........ 2,483 77,847 3.2 5,751 28,504 20.2 8,234 106,351 7.7 1969 ........ 2,615 81,138 3.2 6,087 29,889 20.4 8,702 111,027 7.8 1970 ........ 2,529 82,283 3.1 6,605 32,344 20.4 9,134 114,627 8.0 1971 ........ 2,619 81,586 3.2 7,121 33,493 21.3 9,740 115,079 8.5 1972 ........ 2,577 83,435 3.1 7,243 34,239 21.2 9,820 117,674 8.4 1973 ........ 2,711 85,342 3.2 n.a. — — — — — Source: National Accounts. 812 Table 2 HEALTH CONSUMPTION IN SWEDEN 1963 — 1972, (CURRENT PRICES) A B C D E F G H I 3:335:11; dofiégtlic A/B 21:15:31; Total puinc D/E 1381212311811 3:33:32 G/H the pnvate prlvate. % the pubhc consumptlon % A + D B + E % sector consumptlon sector 1963 ........ 1,413 53,328 2.6 2,798 15,662 17.9 4,211 68,990 6.1 1964 ........ 1,561 57,539 2.7 3,212 17,354 18.5 4,773 74,893 6.4 1965 ........ 1,777 63,087 2.8 3,704 19,688 18.8 5,481 82,775 6.6 1966 ........ 1,956 68,576 2.9 4,470 22,846 19.6 6,426 91,062 7.1 1967 ........ 2,202 73,443 3.0 5,183 25,654 20.2 7,385 99,097 7.5 1968 ........ 2,483 77,847 3.2 5,751 28,504 20.2 8,234 106,351 7.7 1969 ........ 2,698 83,713 3.2 6,407 31,337 20.4 9,105 115,050 7.9 1970 ........ 2,867 89,900 3.2 7,718 36,104 21.4 10,585 126,004 8.4 1971 ........ 3,170 95,729 3.3 9,255 41,565 24.1 12,425 137,294 9.1 1972 ........ 3,402 103,097 3.3 10,481 45,556 23.0 13,883 148,653 9.3 1973 ........ 3,948 112,268 3.5 n.a. — — 7 fl — Source: National Accounts. 613 Table 3 INVESTMENTS IN THE HEALTH SECTOR IN SWEDEN 1963 - 1973, (1968 PRICES) A B C D E F 13131313511313? Tiroltlzlsgglelggg K /B inf/zgirlnzgltflil; Total invest- D /E health sector 1112382223? ths::lt:lth ment 1963 ............................ 471 18,042 2.6 603 26,488 2.3 1964 ............................ 538 19,563 2.8 689 28,564 2.4 1965 ............................ 641 20,435 3.1 805 30,128 2.7 1966 ............................ 760 20,878 3.6 936 31,230 3.0 '1967 ............................ 913 22,306 4.1 1,128 32,959 3.4 1968 ............................ 1,166 22,718 5.1 1,444 33,355 4.3 1969 ............................ 1,181 23,666 5.0 1,453 34,868 4.2 1970 ............................ 1,239 24,169 5.1 1,594 36,134 4.4 1971 ............................ 1,142 23,604 4.7 1,413 35,740 4.0 1972 ............................ 1,014 24,985 4.1 1,305 38,057 3.4 1973 ............................ 889 24,237 3.7 1,152 37,957 3.0 Source: National Accounts OZZ EMPLOYMENT IN THE HEALTH SECTOR Table 4 A B C D E F G Total employ- Health sector, A/B Private health Health occu- Physicians Nurses ment in Sweden total % sector pations active active 1960 ............... 3,615,700 132,200 3.7 17,500 — 7,100 — 1961 ............... 3,645,000 138,400 3.8 18,600 — 7,400 i 1962 ............... 3,064,800 144,800 4.0 19,800 — 7,600 20,800 1963 ............... 3,682,200 153,200 4.2 22,000 1,298,000 8,000 21,600 1964 ............... 3,736,000 163,500 4.4 21,400 133,200 8,200 22,600 1965 ............... 3,761,800 169,900 4.5 21,400 142,600 8,500 24,000 1966 ............... 3,767,100 176,100 4.7 20,600 148,600 8,800 24,600 1967 ............... 3,726,900 188,400 5.1 20,200 153,100 9,200 26,200 1968 ............... 3,766,400 208,200 5.5 26,200 177,500 9,800 28,700 1969 ............... 3,837,300 224,800 5.9 26,300 187,300 10,400 31,800 1970 ............... 3,912,500 242,200 6.2 26,800 203,900 10,600 33,900 1971 ............... 3,906,000 266,400 6.8 26,200 211,200 11,200 36,600 1972 ...... ‘ ......... 3,906,200 285,400 7.3 28,700 227,700 11,900 42,300 1973 ............... 3,921,800 292,100 7.5 29,100 237,800 — — Sources: National Accounts, Labor Force Statistics, Health Statistics. Table 5 ACTIVE MEDICAL PERSONNEL AT END OF YEAR 1960 1965 1970 1972 Physicians ..................... 7,130 8,520 10,560 11,920 Nurses ....................... 21,410 23,190 32,700 }42,890 Nurse-midwives and midwives ........ 1,700 1,760 2,160 Physiotherapists ................. 1,790 2,310 2,600 3,510 Occupational therapists at somatic hospitals .................... 210 420 660 710 Social workers at somatic hospitals . . . . 230 360 720 870 Auxiliary nursing personnel at somatic hospitals ..................... 32,860 35,790 53,960 61,450 Nursing personnel at mental hospitals, etc .................. 11,000 12,870 14,950 16,780 Nursing personnel at homes for mentally retarded ............... 3,870 5,430 200 — Administrative staff at hospitals ...... 1,870 2,730 3,820 4,300 Domestic staff at hospitals .......... 15,200 16,520 18,490 21,600 Dentists ...................... 5,090 6,080 6,720 6,990 Dental technicians ............... 1,850 2,300 2,600 2,700 Dental chairside assistants .......... 5,090 6,800 7,500 8,000 Pharmacists .................... 770 790 810 750 Bachelors of pharmacy and pupils ..... 1,480 2,370 2,700 2,800 Technical personnel in pharmacies ..... 4,960 5,800 6,500 6,400 Total ........................ 116,510134,040 167,650 191,670 Source: Health Statistics. 221 ZZZ Table 6 INPATIENT HEALTH CARE, 1972 A B C D E F G H Somatic Of which Care Of Care Of Number of Bed-da 5 care chronic sick mental mentally . Beds total Admissions y diseases retarded hospltals 1,000 Bed—days Bed—days B dd B dd 1000 1000 e ‘ ays e ' ays ’ ’ 1,000 1,000 1960 ............... 928 116,700 1,001,200 37,700 19,300 6,100 12,400 4,700 1965 ............... 902 125,700 1,132,200 38,200 - — — — 1970 ............... 891 133,600 1,341,100 41,160 24,000 10,700 12,300 4,600 1972 ............... 875 136,200 1,383,400 42,500 25,700 12,400 12,200 4,600 of which private ....... 253 7,700 Source: Health Statistics Table 7 OUTPATIENT HEALTH CARE, PUBLIC SECTOR, 1972 Number of patients Number of visits 1,000 1,000 Hospitals ................ 3,570 8,820 (1960) ............ (2,000) (5,075) District physicians total number 1,400 .......... 5,780 District nurses total number 1,900 .......... 4,828 Maternal and child care ....... 3,500 Source: Health Statistics Table 8 RUNNING COSTS IN ACTUAL PRICES (SWEDISH CROWNS) Cost/day Cost/patient 19601 1970 1972 19601 1970 1972 Teaching hospitals ....... 92 409 517 1,344 4,837 6,218 Other general hospitals . . . . (72) 272 310 972 3,615 4,710 Mental hospitals ......... 25 l 13 150 — — — Long-term hospitals ...... 29 1 17 152 —— — ~ Source: Health Statistics 1 Excluding pensions and depreciations. 223 APPENDIX B A Theoretical Note on the Determination of Supply of a Private Good in a Public Sector For the sake of simplicity we assume that there is a homogeneous commodity called health care that can change continuously. To avoid problems of uncertainty, liquidity effects and information, we also assume that health care takes the form of an optimal insurance of the type discussed by Arrow (1962) (9). In a perfect market, each individual could be described by a demand curve or a marginal willingness to pay curve (no distinction is made between compensated and uncompensated demand curves). The health-care system is assumed to work with constant marginal costs without any fixed costs. For our individual the optimum could then be described in the traditional way (Figure l) as Q° MARGINAL WI LLINGNESS TO PAY (MWP) A MARGINAL COST (MC) MC ll 0 : QUANTITY (0) 0° Figure 1. Individual demand in a market solution. In a system of Swedish character the total size of the health-care system is decided upon in a political process with a majority decision rule, and the financing is carried through by a proportional income tax. In an equally perfect “political” system (without log-rolling and compensations in bargaining situations) we could in a simple way determine an optimal voting rule for the consumer. This simplified picture of a voting system is not as bad as it first may look because of the single-purpose character of Swedish county councils. In the political process, the problem is to determine the average health consumption, 224 C for each individual, assuming that the health-care system will treat everyone equally. _ Assume that the benefits of a consumption Q for the izth individual is Bi _ 63 (Q) with the marginal willingness to pay given by? If the individual income is Yi and the average income is Y then the tax rate t is given by t .Y= c . 6 and the individual county tax Ti = t . Yi = %Yi- The individual marginal tax Yi with regard to health care is then Y .C =aiC where ai is the individual’s income as a fraction of average income. Optimal voting behavior is then determined by a maximization of Bi (5) with regard to the costs C6 . ai which aBi gives T = aiC ll MWP MC MARGINAL TAX PAYMENT, MTP ___ l..__,___ o ‘l’. 0 Q Q Figure 2. The voters’ problem or the county council’s problem. 225 9ZZ Low-l ncome Earner High-Income Earner Figure 3. The rationing problem. For an individual with income coefficient a' the optimal amount is Q' while it will be Q" if the income coefficient is a". Ceteris paribus, low-income earners will vote for a larger size of the health-care system than high-income earners. The final result will be determined by the marginal voter in a majority decision rule system and the strict equality (1) will hold only for the marginal voter(s) and one-half of the electorate will have inequality, stating that they would prefer the ruling tax system to have an increase in health expenditures while the other half would prefer fewer health expenditures. A model of this type could be used and developed to study allocation and income redistribution effects under varying assumptions regarding income and price-elasticities or inequality in income distribution. However, this part is only half the story. Under normal conditions the voting system will give a supply quantity which is below aggregated individual demands at zero-cost. Thus, supply must be rationed by queues, waiting lists or considerations according to need. A possible interpretation of this rationing system is that everyone is treated as the “average” patient, with little regard given to potential willingness to pay at the actual income-distribution. A hypothetical demand curve would be constructed showing the individual demand at highly equalized income distribution (Figure. 3). According to such a rule, the marginal willingness to pay at the hypothetical equalized income level would be the same for two patients [(a) the low income earner and (b) the high income earner in Figure 3] . A rationing behavior of this type, which in any case might be a good description of the intentions of the Swedish health-care system (disregarding the “old boys” network, informa- tion, and bargaining power as ingredients in a rationing system giving favors to high income earners) creates some interesting problems. The low-income earner might feel that he is treated too well; if he had the opportunity to get a lump-sum transfer instead of a subsidy in kind he would prefer to reallocate his increased budget to less health consumption and more consumption of other commodities. The opposite will hold true for the high-income earner who will be dissatisfied with what he thinks to be a low standard at his income level (“too many beds in the ward” etcetera). Still, he would not prefer to buy the health-care services himself, because the zero-cost offer of public-health services is a better offer than would be to pay both the public tax and the fees for private services. 227 REFERENCES l. OECD (1970), “Public Expenditure Trends,” Economic Outlook, Occa- sional Studies, July. 2. KLARMAN, HE. (1974), “Economic Determinants of Health Care Expend- iture: Comparisons over Time and among Nations.” In Health Cost Explosion: Which Way Now?, Stockholm: Institut Henry Dunant (in press). 3. REXED, B. (1974), “Sweden’s Innovative Plan,” Health Cost Explosion: Which Way Nowl, Stockholm: Institut Henry Dunant (in press). 4. NAVARRO, V. (1974), National and Regional Health Planning in Sweden, Fogarty International Center, National Institutes of Health, Washington: U.S. Government Printing Office. 5. NATIONAL ACCOUNTS (1974),Nationalrdkenskaper, 89. 6. HEALTH STATISTICS (1972), Hdlso- och sjukvilz’rd. 7. SOCIAL INSURANCE STATISTICS (1974), Allman fo'rsa‘kring. 8. STAHL, I (1974), U 74-en samha'llsekonomisk analys av utbildningen (Economic Analysis of Higher Education), Stockholm: SNS (in Swedish). 9. ARROW, K. J. (1963), “Uncertainty and the Welfare Economics of Medical Care,” American Economic Review, Vol. 58, PP 941-973. BIBLIOGRAPHY ANDERSEN, R., SMEDBY, B. and ANDERSON, 0.W. (1970), Medical Care Use in Sweden and the United States—A Comparative Analysis of Systems and Behavior. Center for Health Administration Studies, Research Series 27. ANDERSON, O.W. (1972), Health Care: Can There be Equity? The United States, Sweden and England. London & New York: John Wiley. HALSO- OCH SJ UK VXRD INFUR 80-TALE T - FORSLA G TILL E TT PRIN - CIPPROGRAM, (1973), (Health Care for the 1980’s—a program) Social- styrelsen (National Board for Health and Welfare, Sweden) (in Swedish). LINDGREN, SA. (1973), “Sweden”, In Health Services Prospects, An International Survey, London and Oxford: Nuffield Provincial Hospitals Trust. SWEDISH MEDICAL CARE IN THE 1980’s, WAYS AND MEANS (1974), Landstings—forbundet (Federation of Swedish County Councils) (in English). THORBURN, T. (1967), “The Organization and Financing of Medical Services in Sweden”, In Health Services Financing, London: British Medical Association. 228 DISCUSSANT’S COMMENTS O.W. ANDERSON Center for Health Administration Studies, University of Chicago, Chicago, Illinois, USA. It is very nice to follow Professor Stahl because of the great amount of time I have spent in Sweden each year since about 1960. I don’t know whether he was complaining or not, but he sounded a little querulous because Belgium, the Netherlands, France and other parts of the world and Europe have been “neglected” by American researchers. I think the main reason I have “neglected” them, although I am trying to make up for it, is that I set up a research design of three countries ranging from what I call the loosely-struc- tured, multi-nucleated decision-making systems, i.e., the United States, to the highly-structured one of Great Britain and, impressionistically, the Swedish one was more or less in between. The latter seems to be moving in part, at least, towards the British model. Stahl’s is a very straightforward and factual paper, bringing up-to-date trend data in Sweden in a no-nonsense manner, which seems to be a sort of Swedish style, with very little interpretation. The Swedes seem to dislike flying off on interpretation, so maybe it takes an American of Scandinavian descent, like me, too conditioned in part by the American culture to fly off, although hanging on as well as possible to the data and not wandering too far from it. So, I will try to expand on the base that Stahl established. Furthermore, I wish to speak about some of the things that have taken place in this conference since yesterday and try to make sense of these diversities and seeming anomalies. The discussions on the health services have revealed that we are still at a stage of working out of a data base, at data universe from which we can push more deeply into determining what relationships are. Understandably, we are at a fairly descriptive stage at this point, with a lot of speculation, and there may be some danger of philosophizing beyond the data base that we have. I think we should dare to be agnostic until we do have some good control studies and well-designed types of research project. Further, within the Swedish structure,l am very pleased for a change to see an economist—a Swedish economist—examining the Swedish health services. He did give us some idea as to why there seems to have been a relative lack of interest in the Swedish health service on the part of Swedish economists. And as he said, there has been tremendous growth in the Swedish health services~as is true of all systems—but this growth has been particularly, Ithink, character- istic relative to the Swedish health system. Now, my own hypothesis for the tremendous growth in the Swedish health services is the decentralized nature of the financing and, particularly, the fact 229 that the counties and their politicians (the county councillors) have, in effect, only one political issue with the exception of the municipalities. That is to have the best hospital system in Sweden. I have been told in Sweden that the counties competed with each other, because they had nothing else to do, as it were, and were given taxing power more or less commensurate with their responsibilities. The reason for the relative scarcity, relative to other countries, of doctors, is that the state owns all the medical schools. It is reasonable to assume that in the past there was agreement (I won’t call it collusion) between the Swedish Academy of Medicine and parliament to limit doctors. Here there was a central decision-making point, and each position in the medical school had to be, in effect, approved by budget by the parliament. However, this has changed, considerably. The Swedish doctor population is now expanding, and interestingly enough, expanding in the face of Sweden having the lowest of this or that in the health indices in the world. But the country still feels it has to have more doctors. One then wonders whether the health service is related to health indices, or not, as output measures. What the Swedish experience also shows is that it may make no difference whether you have a largely governmentally financed and administered system, or a largely private one—as to how much money is put into the services. It seems to me. that it is more a matter of an implicit social policy to have a tight system or a generally-funded system and the latter is certainly true of the Swedish experience. When I was in Sweden, I felt that I was experiencing a generous concept in relationship to the tremendous economic growth taking place in Sweden, a generous concept of a health-services system. So it may be quite irrelevant whether it is a governmental system or a largely private system. Professor Stahl asks, after showing the trends, “Is it possible that the elaborate planning system and the complicated political-decision process, as a final total result only produces extremely stable, statistical trends or relationships?” He didn’t answer the question, at least not directly. However,I would say, flatly, yes. In my historical and cross-sectional observations of a number of systems, I am struck by the very powerful, underlying forces of technological development, rising expectations, increasing affluence, and so on. Thus, regardless of who owns the system, or how it is funded, there are the same basic fundamental trends. And even though there are interventions attempted in these trends, we still come out with quite similar patterns. It then remains to be seen, as the various countries move more and more into regulation and planning, what will be happening to these trends in the next 10 years. From the evidence, then, I personally have little faith in directed change as such, and I think that directed change has only marginally affected the basic forces already in operation. In this conference we have talked a lot about the public and private sectors as two interrelated or sometimes separate concepts. NowI would like to see 230 some more work done on private-public relationships and what they mean in the evolution and the development of health systems. I have a very quick breakdown of various types of private-public relationships. For example, one might be called “using the private sector as a safety valve for a highly-structured system,” which I think is true of the British example. Four or five percent of the population belong to BUPA or the private, non-profit health insurance scheme. It siphons off some discontent with the larger system. This is what I call the pressure-cooker concept. Another is to use this as a supplement to the public sector. Still another is to be competitive with the public sector, setting up, say two more-or-less equal systems. In President Lyndon Johnson’s days and at the beginning of the Nixon Administration, there was the term “partnership of the private and public sectors.” In any case it seems that all systems and including, 1 think, the United States, the trend seems to be going toward the public sector. The private sector is a sort of annoying residual that we have to live with and which eventually will be abolished or very much weakened. As we again look at the trends, I don’t think that we know what we will experience. We have not yet experienced a crunch—a real crunch between resource allocation and the provision of services. I think that this is what we will be experiencing in the next 10 years. The trend, I would say, seems to be toward the United Kingdom as an ideal, or the USSR, and in that context I wish to quote myself in a remark I made at last year’s Geneva conference. “The deeply implicit, if not clearly explicit desire, is for the policymakers, the planners and the administrators of a particular country to assume that there is a rational model of a health-service delivery system which can transcend the historical and contemporary matrix of a particular country and which provides a fit between population needs, financial resources, personnel resources, and professional judgment.” It seems to be assumed that there was something akin to an industrial model which works, in the sense that one can take a Fiat factory in Italy and rebuild it in Leningrad (as was indeed done), and teach Russian workmen how to go through all of the highly specified and mechanical motions and, in due course, Fiats roll off this Russian assembly line, although with a different name. Policymakers and the bureaucracy endeavor to get a rational health services system, which by all appearances works, because the performance indicators are so crude, that a system is seen not to underserve, or for that matter, overserve, until there are gross incidences of underserving—maybe measured by politically and publicly intolerable queues—and overserving may be measured by, say, 50 percent bed-occupancy through the system on an average, or by doctors being occupied only one half of the time, even though they may spend an hour with each patient. Tolerance levels of system reforms are therefore far lower in Canada, the United States, and Sweden, than I think, in Great Britain; and, given the performance indicators that have systematic reference points such as an automobile manufacturing plant, my own generalization has been to get an equilibrium concept of a “workable” health-services system. At any given time, a health-services system is in a temporary, however defined, equilibrium of personal professional judgments and public perception 231 funding, as related to other perceived needs of the facilities—the British found this in their highly-structured system (which is evolved from a relatively undemanding public); the Swedes found this by balancing local and central administration and funding; and the Canadians are still in an extremely tenuous balance between provincial and federal cost-sharing. All systems seem to be moving toward more structure and boundaries and more centralization of control—presumably to even out inequities—but really to enable a government to predict and to control costs, therefore anticipated more salaried personnel, more planning of facilities and personnel—hence lesser freedom for individual units in relation to the whole, and a great reduction in optional action outside of the system. The strain is toward eliminating optional action in order to facilitate ease of prediction and control. There could be eventual countervailing forces (public and professional), and rebellion against increasingly rigid systems, (unless perhaps the systems are as generous in their provision of services as seems to be the Russian case). I think a governmental system can be also loosely structured. Finally, I believe an aspect of this is the suggestion brought out in Geneva. I haven’t heard it here yet, but it deals with trying to bring the doctor closer to the manager, to have the doctor think of scarce resources in relation to a single patient. If this trend goes far enough, it will or can result in unconscious collusion between the manager and the doctor against the patient. The solution is the impotent ombudsman, rather than individual doctor advocates, for each patient in a structured adversary relationship against the system. Now, I think there are two performance indicators, (of which health indicators is not one, except in very specialized instances) both of which are measurable, and which stem from the social and political values of equalized access and actively-shared costs. Through small sample data and periodic surveys of the public, we can determine to what extent each health service system is “fair.” Are the uses of services evening out by geographic area, family income level, and morbidity characteristics? How responsive is this system in this regard? Great Britain, incidentally, has not done this type of evaluation, at least not officially. The National Health Service is interested largely, you might say, in operational analysis, efficacy measures and so on. Sweden has engaged in this type of evaluation. And so has the United States. 232 Chapter 9 HEALTH COSTS AND EXPENDITURES IN BELGIUM, DENMARK AND THE NETHERLANDS ANDRE PRIMS Professor of Hospital Administration and Medical Care Organization, The University of Leuven, Belgium Introduction First of all, I thank Dr. Leavitt, Director of the Fogarty International Center, for inviting me to participate in this International Conference on Health Costs and Expenditures. It is a great honor and a real pleasure to deliver a paper on Health Costs and Expenditures in Belgium, Denmark and the Netherlands. Second, in order to better understand the answers I will give to the questions listed in the conference outline, I will indicate by way of introduction the general characteristics of the consumers, the providers, the institutions, the regulating agencies and the financing of medical care in the three countries of the European Economic Community. I. THE CONSUMERS (A) Belgium Belgium has a population of about 10 million people. In 1972 the birthrate was 13.8 per 1,000 and the death rate 12 per 1,000. The rate of natural growth of the population was 0.18 percent. Every Belgian citizen is compulsory insured against sickness and disability. The health insurance scheme covers all illness risks for employees, the elderly, widows, orphans and the handicapped. Independent workers (self-employed) and some other categories of the population are covered only for the so-called high risks. (B) Denmark Denmark has about 5 million inhabitants. In 1972 the birth-rate was 15.1 per 1,000 and the death-rate 10.1 per 1,000. The rate of natural growth of the population was 0.50 percent. The Danish health-insurance system is compulsory for the whole population but it is divided into two groups according to annual income. The first group, which represents 80 percent of the population, has a right to free medical assistance; the second group (20 percent of the population) has to partially pay for medical services provided; however, hospital care is free to everyone. 233 (C) The Netherlands The Netherlands has about 13 million inhabitants. The birth-rate is 16.1 per 1,000 and the mortality-rate 8.5 per 1,000. The rate of natural growth of the population is 0.8 percent. Every citizen is compulsory insured for long-term care, but for general care the health-insurance system is compulsory only for employees with an income below a certain amount. This category of people represents approximately 52 percent of the population. For the rest of the population (11 percent) there exists a free system of health insurance, and there is also a special insurance scheme for the elderly (7 percent). About 30 percent of the population is not insured. 2. THE PROVIDERS (A) Belgium Physicians in Belgium are practically all free and independent entrepreneurs. Only a small number of specialists have employee status in hospitals or other medical institutions. Physicians may join the health insurance scheme on the basis of an agreement, and this means that their fees will be to a certain extent limited. The patient is completely free to choose his doctor (general practitioner or specialist); there is no compulsory referral system. With few exceptions, the system is that the physician is paid on a fee-for-service basis. In Belgium there are 15,500 physicians (1 for every 630 inhabitants). (B) Denmark Most Danish physicians are appointed as employees in hospitals and other medical institutions, and the general practitioner is regarded as a kind of public servant. One group of the insured population has to retain general practitioners for at least 1 year. For the second group there exists a free choice of doctors (G.P.’s or specialists), and the referral system is compulsory. For hospital care patients have to go to the hospitals of the counties in which they live. Doctors are salaried but can receive supplementary income through treating patients of the second insured group. In Denmark there are 7,000 physicians (1 per 700 inhabitants). (C) The Netherlands Every compulsorily-insured person has to have his own general practitioner who is remunerated partly by means of a fixed sum based upon the number of insured people on his list, and partly by fees for certain services. The referral system is compulsory. In Netherlands hospitals some specialists are salaried employees, but most are independent and remunerated on a fee-for-service basis. In the Netherlands there are 18,000 physicians (1 for each 760 inhabitants). 234 3. THE INSTITUTIONS (A) Belgium Hospitals (compulsory planning) Number of beds: In general hospitals: 53,000 (5 beds per 1,000 inhabitants). In geriatric hospitals: 3,000. In psychiatric hospitals: 25,000. Public and voluntary hospitals: General hospitals: public 41 percent, voluntary 59 percent. Geriatric hospitals: public 60 percent, voluntary 40 percent. Psychiatric hospitals: public 20 percent, voluntary 80 percent. Other medical institutions Polyclinics. Homes for the elderly. Health centers for preventive medicine. General practitioners Fifty percent of general practitioners are in solo practice. (B) Denmark Hospitals {compulsory planning) Number of beds: In general hospitals: 30,116 (6.12 beds per 1,000 inhabitants). In nursing institutions: 1,300. In psychiatric institutions: 12,500. There are only 11 voluntary hospitals with 1900 beds; all other hospitals belong to the counties, except 3 which are owned by the state. Practically all psychiatric hospitals are state-owned. Other medical institutions Polyclinics. Homes for the elderly. General practitioners Most work in health centers. 235 (C) The Netherlands Hospitals ( compulsory planning) Number of beds: In general hospitals: 71,500. In nursing homes: 27,000. In psychiatric hospitals: 27,000. Public and voluntary hospitals: 78 percent of the beds are in voluntary hospitals and 22 percent in public hospitals (profit-making hospitals are not permitted in the Netherlands). Other medical institutions Polyclinics. Homes for the elderly. Health centers for preventive medicine. General practitioners Only 4,500 out of 17,381 physicians are mostly in solo practice. 4. THE REGULATING AGENCIES (A) Belgium There is a Ministry of Public Health and Family the role of which is mainly concerned with regulating, coordinating and financing. The National Institute for Health Insurance has more importance in policy-making than the Ministry, because this Institute’s primary concern is with financing health care. The medical profession and the sickness-funds are very active pressure groups, and private initiative is still playing a fundamental role in the health field, although the field has its base in public financing. (B) Denmark In Denmark there is no Ministry of Health. Health matters are the responsibility of the Minister for the Interior. The health-care system is completely regionalized; counties are responsible for the provision of health services for their populations. In Denmark the medical profession is a very strong pressure group; private initiative is practically non-existent. (C) The Netherlands In The Netherlands there is a Ministry of Public Health and Environment with a role similar to that of Belgium’s: regulating, coordinating and financing. 236 However, unlike in Belgium, the health insurance system depends upon this Ministry and not upon the Ministry for Social Security. Participation is very well organized; private initiative is very important. 5. FINANCING (A) Belgium Health care in Belgium is financed by: (1) The state—preventive medicine; subsidies for building medical institutions; 25 percent of the hospital bill, and so forth. (2) The health insurance system which is financed by the contri- butions of the insured and by state grants. (3) The patients (only hospital care as such is free in Belgium, for treatment and drugs a certain part of the bill must be paid by the patient, and, if the doctor has no agreement with the health-insurance scheme he is completely free to charge the patient). (B) Denmark In Denmark health care is financed by: (1) The state. (2) The counties. (3) The patients if they belong to the second group of the insured people. (C) The Netherlands In The Netherlands health care is financed by: (1) The state to a small degree (preventive medicine; subsidies for building medical institutions, etc.). (2) The Health Insurance (completely for long-term care, partially for normal care). Health insurance for long-term care is financed by the state and by contributions; health insurance for normal care is financed only by the contributions of employers and employees. (3) The patients (the non-insured; the supplements of the insured). 6. COST OF RECEIVING HEALTH CARE SERVICES (A) Cost of Receiving Health-Care Services Physician fees Belgium: Of the fees paid physicians who have an agreement with the health insurance scheme, a small part is paid by the patient. Physicians who have no link with the health insurance scheme are free to charge the patient, who is reimbursed by the sickness fund at the fixed rate. Denmark: For the first group of insured people medical services are completely free of charge. For the second group of insured people the physician is free to charge the patient, who is reimbursed at the fixed rate. 237 The Netherlands: For compulsorily-insured people medical services are completely free of charge. In the case of non-compulsory-insured people the physician is free to charge the patient, who can obtain some reimbursement if he is inde- pendently insured. Hospital charges Belgium: Hospital treatment is free of charge; 25 percent of the bill is paid by the state and 75 percent by the health-insurance scheme. A supplement has to be paid if the patient is hospitalized for non- medical reasons in a room with only two beds or one bed. Denmark: Hospitals are free of charge for everyone; the bill is paid by the county. The Netherlands: For compulsorily-insured, hospital care is free of charge if the patient is hospitalized in a ward. Other persons have to pay their own hospital bill. Prescription drug charges Belgium: Drugs accredited by the health insurance system are paid for at cost; partial reimbursement is made in hospitals. Denmark: In Denmark a distinction is made between three kind of drugs—the necessary ones (75 percent paid by the county), less-important drugs (50 percent paid by the county); other drugs must be paid for by the patient. The Netherlands: All drugs are free of charge. (B) Relationships Between Quality of Health-Care and Health Costs In none of the three countries is there a means of or criteria for determining relationships between the quality of health care and health costs. 238 (C) Major Factors Influencing Changes in Costs These are the same in the three countries. Examples: (1) The extension of the compulsory health-insurance system; (2) the growing health-consciousness of the population. 7. COST'OF PROVIDING HEALTH-CARE SERVICES (A) Hospital Capital Costs and Hospital Current Costs Belgium: Capital costs I no figures available. Current costs: 1964— 100 percent. 1974—2268 percent. Den mark: Capital costs : Increase: 1966-67—1967-68: 1968-69—1969-70: 1970-71—1971-72: Current costs: Increase: 1966-67—1969-70: 1969-70—1970-71: 1970-71—1971-72: The Netherlands: Capital cost : Million guilders 196$ 1967: 1969: 1970: 197k 1972: 149.5 210.2 323.9 341.9 415.2 345.3 Current cost : Increase: 1970: 1971: 1972: 1973: (B) Relationship Between Changes in Technology and Hospital Costs 20.4 percent. 18.9 percent. 21.6 percent. 18.5 percent. 29 percent. 38 percent. 31 percent. 15 percent. 19 percent. 17 percent. Everyone is aware of the fact that changes in technology have influenced hospital costs, but these cannot be quantified. 239 8. HEALTH-CARE COMPONENTS In all three countries hospital care contributes most to the total increases in health costs and also the most rapidly. Example: In Denmark between 1966 and 1971 there was a 15 percent increase attributable to general practitioners. 9. EXPENDITURES 0N HEALTH-CARE SERVICES (A) Expenditures No data are available regarding private expenditures. Public expenditures: Belgium: Below are shown details of the evolution of expenses related to the health insurance system: Year Total expenses in Index Rate of growth million francs 66 = 100% % 1958 5,421.9 27 1964 12,144 60.8 1965 17,428.8 87.3 43.5 1966 19,970.7 _ 100 14.5 1967 20,4975 102.6 5.5 1968 23,469.8 117.5 14.5 1969 27,7203 138.8 18.7 1970 32,154.9 161.0 16.0 1971 35,621.6 178.4 10.8 1972 41,564.2 208.1 16.7 1973 49,3036 246.9 18.6 1974 57,800.4 289.4 17.2 1975 71,122 356.1 23.05 240 Denmark: During the year 1971/72 running costs were as follows: Million Kroner Rate of growth 1966-71 Medical doctors in practice 730 15 Dentists in practice 250 16 Medicine 340 1 7 Preventive services 1 00 l 2 School dental health 1 10 12 Home visiting nurses l 10 15 General hospitals 3,600 16 Psychiatric hospitals 5 20 l 3 Specialized somatic hospitals 180 12 5,940 Average 15 a 16 The Netherlands: The evolution of expenditures between 1953 and 1972 in million guilders (absolute figures): 1953: 753.9‘ 1958: 1,329.7 1963: 2,176.5 1968: 5,001.7 1970: 6,971.0 1972: 9,969.4 (B) Major Factors Influencing Expenditures In the three countries the major factors are: (1) Number and the qualification of the personnel. (2) Salaries and the working conditions. (3) Advances in technology. (4) Increases in the claims of patients. 10. RELATIONSHIP BETWEEN THE REIMBURSEMENT MECHANISM AND COSTS AND EXPENDITURES There most certainly are relationships between the reimbursement mech- anism and cost and expenditures, but there are no objective criteria with which to measure these relationships. 11. CHANGES IN HEALTH-CARE POLICY It cannot be said that in any of the three countries there were very important changes in health-care policies in the last decade. 241 12. CURRENT TRENDS lN HEALTH COSTS AND EXPENDITURES (A) Governments and consumers are concerned about increases in health costs, and some measures are taken or are under study. The most important are: (1) More emphasis will be put on: (a) preventive medicine. (b) Domiciliary care. (c) Ambulatory care. (2) Compulsory planning of health services with the aim of : (a) Diminishing the number of hospital beds. (b) Coordination. (0) Integration. ((1) Regionalization. (3) Participation in the decision-making process. (4) Partial payments by the patients. (B) Cost-Control Mechanism In all three countries governments are trying to implement some sort of cost-control mechanism, thus far without success. There are no initiatives from either the medical profession or the consumer. 242 DISCUSSANT’S COMMENTS JAN BLANPAIN The University of Leuven, Belgium My comments on the paper: “Health costs and expenditures in Belgium, Denmark and the Netherlands” will focus on studies related to health-care costs undertaken in a number of these countries in view of adding an extra dimension to the general discussion. My comments in this respect are based on a four-country survey in Europe which together with Dr. Delesie I conducted during the past 2 years and finished recently. The survey was made at the request of the Nuffield Provincial Hospitals Trust in Belgium, the Netherlands, France and West Germany. It intended to present an overview and analysis of health services research efforts undertaken in these four European countries. Our working definition of health services research (HSR) defined it as “organized and rigorous inquiry into aspects of the effectiveness of medical care.” Effectiveness of medical care is consistent with its availability, its quality and its efficiency. Aspects considered relevant for the effectiveness of medical care in this broad meaning are: health beliefs and health practices of consumers of health care; the planning, production, deployment, organization, financing, management and evaluation of health care services. Defined as such, HSR includes cost studies. Concern over the spiraling and runaway development of health-care costs is voiced everywhere in terms of crisis from all possible platforms—in parliament and in campaigning by political leaders, in conferences and editorials by spokesmen of the professions and special interest groups, in newspapers by consumer representatives, and so forth. However, a major paradox present in each of the countries surveyed, is the discrepancy between the efforts at voicing concern and calling for investigation and the actual efforts devoted to health-costs studies. This discrepancy between problem-stating efforts and problem-defining and solving efforts is due in the first place to the rather limited resources mobilized for HSR. By and large the research efforts in the health-care field are predominantly taking place in the biomedical sphere with emphasis on normal and pathological structures and processes in human beings and reliance on disciplines like biochemistry, biophysics, physiology, microbiology, cell- biology, immunology, etcetera. Even the identity of HSR is still questionable and so far has not emerged as a separate entity in the taxonomies that the countries in the survey have developed to describe their national research efforts. HSR is hidden under headings like Public Health, Management, Sociology, and Systems Analysis. 243 Table HEALTH AND HSR EXPENDITURE—1970 The Netherlands Belgium France West Germany Percentage GNP to health .......... 6.4% 6.5% 6.3% 6.2% Health research in million $ ........ 58.75% 44.65 216.20 326.65 HSR in thousand $. . . 3811.7 740.25 5146.5 4700 HSR as percentage of health research ..... 6.5% 1.7% 2.4% 1.4% HSR as percentage of _ health expenditure . . 0.11% 0.03% 0.05% 0.03% The Table indicates the limited amounts, both in relative and absolute terms, of the resources spent on HSR. It must be noted that some of the figures are estimates and were arrived at by combining data from scattered sources (which in itself documents the underdeveloped state of the health costs data base). Within this limited effort—and the question arises whether this effort is not subminimal and insufficient to have any serious social impact— cost-oriented studies on the average represent less than one-third of the studies undertaken, if the main disciplinary approach is used to classify HSR projects. In the four countries surveyed, only in France is relatively more importance given to economy-oriented projects than to organization-oriented or behavioral-sciences-oriented projects. The reasons for this being out of phase of HSR efforts in general, and health-cost studies in particular, vis-‘a-vis rising expenditures for health and for health-related research are multiple and interdependent in such a way that what seems to be the cause can also be considered as an effect and vice versa. The traditional research model whereby research efforts are situated within educational settings, primarily universities, presents several characteristics which are to a large extent responsible for health-cost studies lagging behind. A substantial part of the research money flowing through universities comes from Ministries of Education and is not earmarked in terms of purpose or orientation. It tends to flow toward traditional strongholds within the universities and is used to a substantial degree for the career development of educators and researchers in given scientific disciplines along lines of either personal preferences or topics traditional for the setting. HSR, as a newcomer in this game, is at the very end of the receiving line. This particular situation has blocked within departments of economics the development of health economics as a discipline or as career, leading toward 244 the paradoxical situation that in given cases health-cost related research could not be undertaken because of a lack of experienced or even interested economists. University-based HSR settings on the average only count a few individuals and present a rapid turnover of junior staff using the research episode as a once-and-for-all effort in the career build-up. Health-costs studies undertaken in these university-basedHSR settings are usually short-term studies limited in resources and scope. Examples of such studies are doctoral studies at the Institute for Health Care at the Tilburg University in the Netherlands where four graduating econometricians de- veloped, with the data of a local hospital, an input-output model by which inputs like salaries and materials are costed to output like operations and diagnoses, by the intermediate use of cost centers. Valid data were only available to relate inputs to cost centers. A delphi method was applied to relate cost-centers to outputs. Another example is doctoral work in progress at the Department of Hospital Administration and Medical Care Organization at Leuven University, where the influence of physician payment systems on the utilization of medical care is being investigated. The main objective of this study is to explore to what extent the difference in payment systems of general practitioners between Belgium (with a fee-for-service system) and the Netherlands (with a capitation system) is responsible for differences in medical-care utilization and medical- care expenditures in the two countries. The central hypothesis is that the per capita payment system of G.P.’s in the Netherlands favors referrals to specialists in hospitals and hence more expensive care versus the fee-for-service system in Belgium favoring extramural care by G.P.’s. So far, analysis of the data seems to confirm the hypothesis. Other favorite cost-study topics in university-based HSR settings in the survey countries are hospital cost studies, cost studies of given categories of morbidity such as cardiovascular diseases or mental health care, tentative analysis of alternative modes of delivery of care, determinants of health-care demand, and projections of future health-care expenditures. This picture is changing, however. In the first place special governmental agencies were created to guide the nation’s research efforts, to allocate resources accordingly and eventually to decide on the purpose and orientation of all government-financed research. Belgium, France and West Germany have definitely chosen this direction. In the Netherlands—where for some time already an intricate system of interlocking advisory committees has confronted governmental and non-governmental individuals, the last ones mainly being university professors—a certain degree, within universities, of alertness and response through research to health services problems, has taken place earlier in comparison with the other countries and has delayed the impact of the central governmental research policy agency. A second development has been the emergence of special purpose HSR settings either directly organized or financed by government or by voluntary agencies such as hospital associations. The Ministries of Health in the respective countries have recently increased their influence in this respect. Other Ministries, as in Belgium (the Ministry of Social Affairs) and in France (the 245 Ministry of Economy and Finances) have stepped in to develop special purpose mission-oriented HSR. This combined development of special governmental agencies for research policy and the emergence of special purpose HSR settings is changing the characteristics of the traditional HSR project. Resources, both in terms of manpower and finances, are substantially larger per project and permit intensive as well as long term investigation. From being retrospective, descriptive and analytical in approach the HSR projects become more prospective and applications-oriented towards the solution of given problems. From being micro-oriented as they deal with sub-elements of particular facilities of the health-care system the HSR projects become geared towards the macro-level. In terms of cost studies this is particularly the case in France and in the Netherlands. In France, cost-studies undertaken by the section of Medical Economy of the Center for Research and Documentation of Consumption (CREDOC) some of which are reported in this volume by A. Foulon and S. Sandier, are examples of this important shift in HSR. For the Netherlands I would like to point out the major on-going project directed by Prof. Van Praag of the Economic Institute of the University of Leiden, in which at the request of the Ministry of Health and Environmental Hygiene a macro-econometric model is being developed to help long-term planning of the Dutch health-care system. The supply, demand and price levels of all health-care activities are investigated in four levels: (1) general practitioners; (2) specialists and out- patient services; (3) short-term inpatient care; and (4) extended inpatient care and postclinical care. Demand is subdivided in autonomous demand (originated by the patient himself) and derived demand (generated by health-care providers) for the four levels. The supply of health care is also broken down for the four levels and includes long-term projections for the number of general practitioners, the number of specialists, the number of hospital beds by type and level, etcetera. To exemplify the level of detail which is sought in the model: the number of future general practitioners is included as a function of the 18 to 25 year age bracket of the population and the expected future average income for general practitioners. This research effort, which will take many years, has already led to interesting interim reports permitting the Dutch government to base its hospital bed/population ratios for the future on a more solid base. Given the strong tradition (through J. Tinbergen) of econometrics in the Netherlands this particular project is from an international and disciplinary viewpoint worth following. A second major ongoing study in the Netherlands is also worth looking at. It is an analysis of hospital costs directed by Professor Groot of the University of Maastricht at the request of the Dutch National Hospital Institute and the Ministry of Health and Environmental Hygiene. Based upon historic cost data of all Dutch hospitals it intends to develop an explanatory model of hospital costs taking into account future and probable developments. A number of leading health economists have been advisors for this project. 246 In conclusion I stress the importance of increased efforts in the systematic exchange of information on ongoing or projected health-cost studies in particular and HSR in general. Unless a more systematic vehicle of information is created we will witness a growing series of ad hoc surveys like the one which is currently being undertaken by ILO and being considered by other supra- or international agencies. A number of these costly surveys collect redundant information and miss the potential for ongoing exchange. A specific international journal seems the answer to meet the growing needs for relevant information for both the policymakers and the scientific community regarding health-care costs. 247 Chapter 10 HEALTH COSTS AND EXPENDITURES IN THE FEDERAL REPUBLIC OF GERMANY AND THE UNITED STATES UWE E. REINHARDT Woodrow Wilson School of Public and International Affairs, Princeton University, New Jersey ABSTRACT Presented is a synopsis of the health-care system in the Federal Republic of Germany. Both the health-care delivery system and the health insurance system are surveyed and, where practicable, compared to their United States counterparts. The paper is~based on work during the early phases of an ongoing research project on European health systems “A Cross-National Study of Health Systems,” sponsored by the Ford Foundation (Grant No. 730-0094) and conducted at the Woodrow Wilson School of Public and International Affairs at Princeton University. Future research in this project will be devoted to more penetrating analyses of certain subsectors of the health-care system. The author has benefited from a discussion with Dr. Heinz Allekotte of the University of Cologne and, particularly, from discussions with Frau Regiemngs- direktor Schneider of the Bundesministerium fiir Jugend, Familie und Gesundheit (Federal Ministry for Youth, Family and Health) of the Federal Republic of Germany. Frau Direktor Schneider provided the author with many of the source materials on which this paper is based. Valuable research assistance during the earlier phase of the research was provided by Mr. Karl- Heinz Wacket, a West-German exchange student visiting the United States. 1. INTRODUCTION From the viewpoint of policymakers in the United States and Canada the West-German health system should be of considerable interest. First, unlike the British National Health Service, the West-German health-care delivery system resembles in important respects those found on the North American continent. Second, f virtually the entire West-German population is now covered by comprehensive health insurance without cost-sharing on the part of patients at the time health services are consumelehe system under which this coverage is extended is generally acknowledged to be the oldest such system in the world, dating back to 1886 and having grown to its present form in stages/Finally, the West-German health system incorporates some features occasionally proposed as remedies for shortcomings in the North American health systems. 249 For these reasons alone the evolution of the West-German system and its current modus operandi furnish a potential data source for policymakers now anxiously speculating on the future impact of national health insurance in the United States. Oddly enough, one finds virtually no mention of the West-German health system (or of the somewhat similar French system) in the current debate over national health insurance in this country. Still more disturbing is the fact that years of generously-funded health services research in the United States has virtually overlooked the data sources represented by these European health systems. This paper presents an overview of the health system in the Federal Republic of Germany (hereafter referred to also by the acronym FRG or by West Germany). Given the orientation of this conference, the presentation is primarily descriptive and necessarily synoptic. An attempt is made, where feasible, to contrast the West-German data with similar data for the United States. However, it is noted at the outset that both the synopsis of the West-German system and the international comparison are constrained by the data available at this early stage of the research effort. The discussion is divided into three major parts. Section 2 below contains a sketch of the West-German health-care delivery system. That section is followed by a description of the health-insurance system in the FRG. Finally, Section 4 presents the cost and expenditure record of the WesbGerman health system and contrasts that record with comparable American statistics. 2. ORGANIZATION OF THE HEALTH-CARE DELIVERY SYSTEM IN THE FRG West Germany currently has a population slightly in excess of 61 million (the 1971 figure being 60.68 million). As is shown in Table 1 below, relative to the US. population, them/est German population has a higher average age and has proportionately fewer males. These differences notwithstanding, the age-sex composition of the two ‘nations” populations can be said to be fairly similar. According to statistics reported by the Ministry of labor and Social Security (Ministerium fu'r Arbeit und Soziale Sicherung) of the FRG, the nation’s population is served by approximately 500,000 healthworkers, representing a ratio of between 850 and 900 personnel per 100,000 population (see Table 2 below).* This labor force is complemented by some 7,000 public, private non-profit, and private for-profit hospitals with close to 700,000 beds, or 1,120 beds per 100,000 population. This health-care delivery system represents, just like its counterpart in the United States, a complex mosaic of ownership interests and functions. In the following subsections, the major components of this system will be examined more closely. *According to US. Bureau of Labor statistics, the United States health sector employs a total of 4.8 million health workers, or 2,305 per 100,000 population (see Table 2). Although it is quite possible that the Us. health delivery system is relatively more labor-intensive than the West-German one, so large a difference suggests that the statistics may not be fully compatible. 250 Table l AGE AND SEX DISTRIBUTION OF THE POPULATION— U.S. AND FRG, 1970 Difference in U.S. FRG Pegéntage omts U.S. - FRG 1. Total Population ........ 207 million 61 million ~ 2. Age Distribution: Less than 15 ......... 28% 23% + 5.0% 15-44 .............. 41% 41% 0.0% 45-64 .............. 21% 23% - 2.0% 65 and over .......... 10% 13% ~ 3.0% 3. Sex Distribution: Male .............. 49.8% 47.6% + 2.2% Female ............. 50.2% 52.4% - 2.2% Sources: Statistisches Jahrbuch fiir die Bundersrepublik Deu tschland, 1972, Section C of the “Internationale Ubersicht,” Table 1, p. 24 and Table 4, p. 30. Table 2 EMPLOYMENT IN THE HEALTH-SERVICES SECTOR, U.S. and FRG1970 U.S. FRG Total Reported Employment .............. 4,771,500a 521,466 Employment per 100,000 Population Physicians ......................... 150 173 Dentists .......................... 45 51 Pharmacists ........................ 60 97c Nurses ........................... 28213 327 Medical and Dental Technicians .......... l 19 38 All other .......................... 1,662 176 Total per 100,000 Population .............. 2,305 859 aIncludes 109,000 phaxmacists in drug stores. Registered nurses only. cIncludes employees of pharmacists. Sources: U.S. Data: United States StatisticalAbstract, 1974, Table 116, p.76. FRG Data: Bundesminister fiir aIbeit und Sozialordnung, Ubersicht fiber die Soziale Sicherung, p.152. 251 The Organization of the Physician Sector West Germany currently has somewhere between 110,000 to 120,000 active physicians (including interns) or close to 180 physicians per 100,000 population. In 1970, that ratio stood at 173 (see Table 4) and exceeded the comparable U.S. statistic by about 15 percentuAbout 10 percent of these physicians were engaged in research and administration. The remaining 90 percent were distributed among types and loci of activity as follows: Table 3 DISTRIBUTION OF WEST-GERMAN PHYSICIANS BY TYPE AND LOCUS OF ACTIVITY Private Hospital Practice Staff TOTAL General Practitioners .......... 5 5% 59% 51% (Praktische Arzte) Specialists ................. 45% 41% 39% (Fach'arzte) Total .................... 100% 100% 90% A more detailed statistical profile of West-German physicians is contained in Tables4 to 6 below. It will be noted that, relative to the United States, a substantially higher proportion of physicians in the FRG are general practitioners, and fewer of them are surgeons. It is also seen in Table 6 that the West-German physician population is unevenly distributed over the FRG’s Linder (states), although the degree of unevenness is far below that found in the US. The typical American or Canadian physician treats his patients in his own practice and in the hospital(s) with which (s)he is affiliated. Hospital privileges in this North American pattern are enjoyed by only about 7 percent of the West- German physician population—by the so-called Belegsarzte. Other West- German Kphysicians work full- time either in their own private practice or in the hospital Physicians in private practice (Niedergelassene Arzte) typically treat their patients on a fee- for- -service basis, with the bulk of the fees (about 85 percent) coming from third-party payers. Physicians in hospitals (Krankenhau- sa'rzte), on the other hand, are salaried, and only the chiefs of staff enjoy the privilege of treating private patients on the hospital’s premises, for a fee. As a result of this rigidly enforced dichotomy between the ambulatory and the hospital physician in the FRG, the former typically loses contact with patients during the latters’ hospital stay. A corollary is thatJValthough West-German patients have the right to choose their own physician for ambulatory health services,* freedom of choice does not extend to treatment *In fact, an insured patient can normally change physicians only every three months, at least under the Social Health Insurance system. 252 Table 4 NUMBER AND DISTRIBUTION OF MEDICAL DOCTORS, U.S. AND FRG, 1970 U.S. FRG 1. Active Physicians (M.D.’s):a Number .......................... 311,203 105,976 Number per l00,000 population .......... 150 173 2. Distribution by Locus of Activity:b Office-based Practice ................. 61.8% 47.9% Hospital-based Practice ................ 27.7% 42.4% Other, including Research and Administration . 10.5% 9.7% 3. Distribution by Specialtyzc General Practice ..................... 19.3% 58.2% Specialists: ........................ 80.7% 41.8% Internal Medicine .................. 13.2% 11.0% Pediatrics ....................... 5.7% 3.4% Obstetrics-Gynecology .............. 6.3% 4.4% General Surgery .................. 9.5% 4.5% All Other ....................... 46.0% 18.5% a For the U.S., this figure includes interns and residents, but excludes 19,621 inactive M.D.’s and 3,204 M.D.’s with unknown address. Doctors of Osteopathy are also excluded. A population of 207 million is assumed. For the FRG, the figure includes 99,654 “berufstatige” (active) physicians and 6,3 22 “Medizinclassistenten” (interns). A population of6l million is assumed. b Includes physicians who spend a substantial proportion of their time in hospital- based practice. c Excludes interns (Medizinclassistenten). For the U.S., the base is defined as 311,203 M.D.’s minus 11,449 interns (= 299,754). For the FRG, the base is 105,976 M.D.’s minus 6,322 “Medizinclassistenten” (= 99,654). Sources: U.S. data—American Medical Association, The Profile of Medical Practice 1971, Table 2. F RG data —Statistisches Jahrbuch fiir die Bundesrepublik Deutschland 1972, Table 11, p.66. in the hospitalijlhis arrangement can obviously bias the primary physician’s decisions on hospitalization, although the direction of the bias is not clear a priori, as will be brought out later. Little is actually known about the organization of physician practices in the FRG. Surveys of the sort regularly undertaken in the U.S.—for example, those undertaken by the trade journal Medical Economics or by the American Medical Association—do not appear to be customary in West Germany. According to recent estimates, there were only about 505 group practices in *The exception would be private patients of chiefs of staff or of Belegsa'rzte. 253 the FRG as of 1971,l’with an unknown number of physician-members. This figure, however, does represent a 65 percent increase (17 percent per year) over the reported number in 1969. One may thus be observing the onset of a rather rapid drift towards group practices. (In this connection, see Gesellschaft fur Sozialen Fortschritt e. V. (1974), p. 75.) As to the organization of individual practices, it is known that the staffing per practice has increased over time, as may be inferred from Table 7 below. Although the time series on equipment shown in this table undoubtedly contains a price-change component, it can be assumed that physician equipment and floorspace was added in complementary fashion to increases in staffing. From production-function estimates reflecting the experience of American physicians (see, for example, Reinhardt (1972) and Kimbell and Lorant (1973)) it can be inferred that the pattern exhibited in Table/7 must have increased the average productivity of West-German physicians. Indeed, since the bulk of the practices reflected in Table 7 are solo practices,it can be concluded that the typical private practitioner in the FRG is likely to carry a larger patient-visit load than his American colleague who, on average, employs onlgl/jlabout 2.5 to 3 persons (including the physician) per practicej, ‘ e political organization of physicians in the FRG invoh/es a mixture of private and public associations at the state and federal levels. At a minimum, every physician practicing in a particular state (Land) must, by law, belong to that state’s Arztekammer (State Medical Association). In contrast to the private State Medical Associations in the US, those in the FRG are publicly chartered bodies to whom the state entrusts the regulation and supervision of the physician’s professional conduct. This regulatory power extends to the individual physician’s decision concerning the organization of his practice. For example, the establishment of group practices or even the employment of a physician assistant (Praxisassistenten) requires authorization by the Arzte- kammer (in this connection see Schicke (1971), p. 143). Furthermore, the state expects this public body to assume responsibility for the delivery of health services where needed. The problem of geographic maldistribution of medical manpower is thus one for which the State Medical Association has responsibility—an arrangement clearly without parallel in the United States. In addition to compulsory membership in his/her state’s medical associa— tion, a physician who has been accepted as a participating member of the Social Health Insurance System? must join his state’s so-called Kassen'arztliche Vereinigung (Association of Sickness Fund Physicians, hereafter referred to simply as K.V.) This association is the representative body which negotiates *In this connection, see also Schicke (1971), pp. 192-3. Schicke imputes to West Ger- man physicians roughly twice as many patient-visits per year as to American physicians. In all probability, however, that is an overestimate, because the US data include only the physicians’ office visits and exclude the substantial amount of time which American phy- sicians spend in the hospital. It will be recalled that private practitioners in the FRG spend virtually all of their working hours in their office practice. Even so, Schicke’s general con- clusion seems to be supported by the available data. TAs will be brought out later, to receive reimbursement from the health insurance system, a physician must have applied for and received admittance to the system. Since 1960, admission to the insurance system has been and still is the right of any qualified West-German physican. 254 professional and economic matters with a parallel association of the state’s health insurance funds (described in detail later). In addition, the K.V. exercises certain self-regulatory functions. For example, the health insurance funds look to it for assurance that insured patients receive all necessary, covered physician-services when and where needed and delegate to the K.V. the task of curbing excess prescription/utilization of physician services. The publicly-chartered State Medical Associations form a voluntary associa- tion at the federal level, the so-called Bundesa'rztekammer (Federal Medical Association). Represented in this national body are not only the State Medical Table 5 PROFILE OF PHYSICIAN POPULATION, FRG, 1970 Percent of Total Number of Active M.D.’sa Number (%) 1. Total Number of Active M.D.’s, excluding Interns ........................ 99,682 — 100% 2. Niedergelassene Arzte (M.D.’s in private, office-based practice): Praktische Arzte (General Practitioners) . . 27,999 55% 28% Facharzte (Specialists) ............. 22,740 45% 23% Total ....................... 50,739 100% *51% 3. Belegséirzte (M.D.’s in office-based practice with hospital privilege; included also in (2) above) ......................... 6,885 — 7% 4. Krankenhaus'arzte (M.D.’s in full-time hospital-based practice): Praktische Arzte (General Practiioners) . . 22,938 59% 23% Facharzte (Specialists) ............. 15,717 M) 16% Total ...................... 38,655 100% *39% 5. Arzte in Sonstiger Stellung (M.D.’s in other (salaried) positions, including research and ‘ administration) ................... 10,268 — * 10% 6. Other Categories: Female Physicians: in private practice .............. 7,850 42% 8% in hospitals ................... 8,980 47% 9% in other employment ............ 2,105 11% 2% Total .................... 18,935 100% ' 19% Foreign Physicians .................. 4,793 — 5% aPercentages preceded by an asterisk add to 100%. Source: Thelen (1974), Appendix, Table 6. 255 Table 6 GEOGRAPHIC DISTRIBUTION OF PHYSICIANS, U.S. AND FRG U.S. FRG 1. Range of Physician-Population Ratios, by State (Land): Highest .......... . . . . 232 (New York)a 247 (Hamburg)b Lowest .......... . . . . 89 (Mississippi) 134 (Rhein. Pfalz) Range ....... 143 113 2. Average Over Stages, and Standard Deviation Average .............. 145 168 Standard Deviation ...... 65 33 Coefficient of Variation. . . . .44 .19 aActually, the District of Columbia had the highest ratio (525). It has been excluded becguse many physicians in the D.C. are not in patient care. Although the ratio of West Berlin (277) is the highest, that city has been excluded because of its unusual status. Sources: U.S. Data: U.S. Department of Health, Education, and Welfare, The Supply of Health Manpower, DHEW Publication (HRA) 75-38, December 1974, Table 10, pp. 28-9. FRG Data: Statistisches Jahrbuch fL'ir die Bundesrepublik Deutschland 1972, Table 11, p. 66. Table 7 STAFFING AND EQUIPMENT IN MEDICAL PRACTICES, FRG, 1963-71 /'-~- 1963 1967 1971 1. Average Number of Personnel/Practicea . . . . 3.7 4.1 4.6 Index, 1963:100 ....... . .......... 100 111 124 Annual rate of change, 1963-71 ........ — 2.7% — 2_ Average Value of Floorspace and Equipment/Practice (DM)b ......... . . . 13.8 16.4 19.7 Index, 1963:100. . . . .......... . . . . 100 119 143 3 Includes physicians. b Current dollars. Source: Gesellschaft fic'r Sozialen Fortschrift (1974), Table 21, p. 74. 256 Associations, but some 25 or so other physician associations, each formed around some focal professional and/or economic interest, and with voluntary membership. This federal association serves in the main two distinct functions. First, it represents the professional, scientific and economic interests of the medical profession as a whole vis-a-vis the government, the media, and the public at large. Second, it attempts to function as a clearing house for ultra-professional affairs and, in particular, as a mediator in intra-professional conflicts (usually over economic issues). An analysis of these intra-professional conflicts makes fascinating reading, but that subject matter clearly lies outside the main focus of this paper (in this connection, however, see Thelen (1974)). Owing perhaps to the success of FRG physicians in establishing a system of parallel associations—some concentrating mainly on the scientific and medical aspects of the profession and others concentrating unabashedly on the profession’s economic interest—and probably also because of the relatively high average productivity of West-German practitioners, the medical profession in that country has done remarkably well on the economic plane. Some pertinent data on physician incomes, and some illuminating contrasts with comparable U.S. data, are presented in Tables 8 to 10 below. It is seen that at the market exchange rate of DM 2.50 per US$, FRG physicians in private practice have attained absolute parity with their colleagues in the U.S. (see Table 9). Since market exchange rates in this context can be misleading, however, a more illuminating contrast is presented in Table 10, where physician incomes are expressed relative to the average total compensation (including employer contributions to social security and other fringe benefits) of employed persons in the industrial sector. As noted in the Table, there may be some upward bias in the difference between the two nations’ relative physician incomes. Even so, one is persuaded that private medical practitioners in the FRG enjoy a marginally higher position in the nation’s income distribution than do their American colleagues. The preceding income data represent only those West-German physicians who are in private medical practice—the closest analogue to the typical American physician. As noted earlier, hospital physicians in the FRG are paid salaries according to established tariffs. Because a variety of supplementary income sources are available to hospital staff (among them, overtime), it is difficult to infer total earned income from the established tariffs. From all available accounts, however, the typical hospital physician (other than a chief of staff) receives an annual income far below that attainable in private practice (see Schicke (1971)). Quite probably, as a result of this differential, there has in recent years been an exodus from institutions into private practice, and some influx of foreign physicians as replacements in hospital positions. At the same time, hospital salaries appear to have risen somewhat in response to these market signals. Not surprisingly, physicians already in private practice have recently proposed that a lengthy, mandatory term of employment on hospital staffs be imposed on newly-trained physicians as a condition for the eventual privilege of establishing a private practice. It may be assumed that the proposed measure was not dictated mainly by a concern for the quality of the novice’s medical training. 257 SSZ Table 8 GROWTH IN AVERAGE GROSS- AND NET-INCOME PER PRACTICE, FRG, 1967-1971 Percentage Percentage Specialty Gross Increase Net Increase 1971PNet Incogne Income during Income during as (er cent 0 1967-71 1967~7l 1971 ross Income DM % DM % % All Specialties ..................... 178,538 47.7 115,586 44.2 64.7 General Practice .................. 157,453 49.6 105,179 44.8 66.9 Internal Medicine ................. 227,418 53.7 139,248 51.5 61.2 OBG Specialists ................. 176,361 39.3 112,659 30.8 63.9 Pediatrics ...................... 144,971 47.4 91,680 43.9 63.2 Surgeons ....................... 207,380 10.0 120,384 2.5 58.1 Source: Adapted from Gesellschaft fd'r Sozialen Fortschrift (1974), Tables 3 and 5. Table 9 NET INCOME PER PHYSICIAN, [1.8. AND FRG, 1971 U.S. F.R.G. US. $ Indexa DM Indexa All Specialties .............. 45,278 100 113,543 100 General Practice .......... 39,823 88 102,958 91 Internal Medicine .......... 42,869 95 136,322 120 Pediatrics ............... 38,503 85 90,704 80 DEC Specialists ........... 51,062 113 112,659 99 Surgery ................ 54,045 119 120,384 106 Radiology ............... N.A.b 13113 177,582 156 a . , . , . . . Specialty s 1ncome X 100 7 income for all spec1alt1es. For reasons not stated, the income of radiologists has, since 1969, not been made public in the AMA summaries. The index is based on 1968 data. Sources: U.S.—American Medical Association, The Profile of Medical Practice 1972; FRG~Gesellschaft fiir Sozialen Fortschrift e. V. Derwandel der Stellung des Arztes im Einkommensverfilge (1974), p. 22. The Organization of the Hospital Sector Tables 3, 5, 11, and 12 (of which the last two appear below) present information on the organization of the West-German hospital sector. From Table 11 there appears a striking similarity between the ownership distribution of FRG hospitals and that of US. hospitals. In contrast to the United States, however, where private ownership of hospital beds has remained a constant 3.8 percent over the period 1950-73, private ownership in West Germany has increased from 6.2 percent in the mid 1950’s to 8.8 percent in 1970. These private hospitals are either small clinics with inpatient facilities, owned and operated by physicians, or so-called Kur-Krankenh‘auser (inpatient facilities, extended-care facilities or rest-homes typically located in Kurorten (resort areas) and without a close counterpart in the United States (see Schicke (1971) p. 97). On a per-capita basis, the FRG has 44 percent more hospital beds than does the United States, and the FRG occupancy rate of these beds exceeds the United States ratio by 13 percent (see Tables 11 and 13, 1970 data). Depending upon one’s faith in Roemer’s Law, this relatively high bed endowment in the FRG can be viewed as either a cause or a result of the relatively high number of patient-days per capita in West Germany. Since the average rate of hospital admissions in that country is actually below the comparable U.S. statistic, the explanation for the high patient-day rate lies entirely in the unusually high average length of hospital stay per episode of illness. As will be seen in Table 13, that statistic, averaged over all types of 259 hospitals, exceeds the U.S. counterpart by 67 percent. For short-term, acute hospitals, that percentage is as high as 130(!). In his analysis of the West-German hospital sector, Schicke (1971) points out that relative to the United States and the United Kingdom, West Germany has fewer nursing-home places per aged citizen and also fewer beds in long-tenn, chronic hospitals. As a result, he argues, some of the functions performed by these specialized institutions in the U.S. and the UK. are perforce performed by short-term acute facilities, a factor contributing to the relatively long stays in acute, short-term facilities in the FRGj vaiously that is a sensible proposition and one that could readily be tes ed with data on case-mixed treated. But one also suspects that the average length of stay for given cases tends to be much longer in the FRG than on the North American continenty‘ For example, the average length of stay in FRG hospitals specializing in obstetrics is 11 days, suggesting that a delivery in the F RG may Table 10 PHYSICIAN INCOME RELATIVE TO AVERAGE COMPENSATION PER EMPLOYEE IN INDUSTRY, U.S. AND FRG, 1959-71 1959 1963 1967 1971 1. United States: Median Income/ d Physician ........ $22,100 $28,380 $34,730 $42,700 Average Compensation/ d Employed Persona . . 4,970 6,030 6,880 9,030 Ratio a + b° ...... 4.45 4.71 5.05 4.72 2. Federal Republic of Germany Net Income/Practiceb DM35,290 DM49,760 DM80,150 DM115,580 Average Compensation/ Employed Persona . . 6,460 8,940 11,700 17,720 Rati0a+bc ...... 5.46 5.57 6.85 6.52 :Includes employer contributions to social security. CNet Income/Physician is only slightly below Income/Practice. The FRG-U.S. differential in this ratio is overstated somewhat because of the factor mentioned in note (b) and the fact that U.S. physician incomes are medians and not aver- ages. Given the somewhat skewed distribution of physician incomes, the median is apt to be somewhat below the mean of the distribution. d1964 data. Sources: U.S. Data: Physician Incomes, in United States Statistical Abstract 1974, Table Table 112; average compensation of employed persons, U.S. De- partment of Commerce, Survey of Current Business, various years, Tables 6.1 and 6.4. FRG Data: Gesellschaft ft'ir Sozialen Fortschrift e. V. (1974), Table 10, p. 32. 260 Table 11 HOSPITALS AND HOSPITAL BEDS, U.S. AND FRG, 1970 U.S. FRG 1. Number of Hospitals, All Types .......... 7,123 3,587 2. Number of Beds, All Hospitals ........... 1,616,000 683,254 Publicly Owneda .. .1. . '. . . . . 1 ....... 58.0% 54.6% Non-public, Non-profitb ............ 38.4% 36.6% Privatec ........................ 3. 6% 8. 8% Acute, Short-term ................. 52.5% 67.0%d Other ......................... 47.5% 33.0% 3. Average Number of Beds per Hospital ...... 227 177 4. Number of Beds per 1,000 Population ...... 7.8 1 1.2 aFor the U.S., this includes Federal, State and Local Government hospitals; for the FRG, it includes the category of “Oftentliche Krankenh'auser.” bFor the U.S., this includes the category of “Nongovernmental Nonprofit” hospitals; for the FRG it includes the “F reien Gemeinniitzigen Krankenha'user." 6For the U.S., this includes “Private, For-Profit” hospitals; for the FRG, it includes the “Privaten Krankenh‘auser.” d1969. Source: United States StatisticalAbstract 1974, Table 118, p. 77 and Table 122, p. 79; Statistisches Jahrbuch fiir der Bundesrepublik Deutschland I 9 72, Table 13, p. 67; Bundesminister filr Jugend, Familie und Gesundheit, Gesundheitsbericht, Stuttgart: Verlag Kohlhammer, 1971; pp. 34-35. require about two to three times as many hospital days as it does in the United States. It could be argued that these differentials simply reflect cultural differences. Further on in this essay, however, it will be suggested that economic incentives are apt to contribute to the observed differentials. Average hospital stays such as those in the FRG would wreak a financial catastrophe in the United States, other things being equal. They are fiscally tolerable in West Germany, because patient days there are relatively inexpen- sive. Table 19 indicates that even as late as 1971, the health insurance funds under the FRG’s Social Health Insurance system paid hospitals an average of only DM 63.60 per patient day. Although this “price” may not include a a ital-recovery component, it is much below the comparable U.S. figure. fish—b 12 suggests that a good part of the cost differential is attributable to the relatively low staffing per bed and per patient in FRG hospitals Quite possibly, West-German hospitals are also less capital-intensive than are .S. hospitalsu: although this observation is conjecture at this stage of the author’s investi- gation. To what extent the relatively higher resource-intensity of an American patient-day facilitates a shortening of hospital stays is a matter of conjecture as well. The technical and economic nature of such a trade-off, if it is feasible at all, is an area inviting further research. 261 Table 12 STAFFING OF HOSPITALS, US. AND FRG, 1965 AND 1970 U.S. FRG 1. Full-time Hospital-based Medical Personnel, including Interns and Residents, 1970” Number ....................... 66,103 44.977b Number per 100 beds .............. 4.1 6.6 2. Full-time Equivalent, Non-medical Personnel per 100 Patients: 1965: All Hospitals .................. 139 68 Short-term Acute Hospitals ........ 246 NA. 1970: All Hospitals ................. 23l N.A. Short-term Acute Hospitals ........ 310 NA. aNot directly comparable because the U.S. figure excludes the hospital-based activity of office-based physicians. bIncludes 6.322 Medizinelassistenten (Interns). Sources: Unitr'dStatos SmtisticalAbstracr 1974. Table 120, p. 78; R. K. Schieke (1971), Table 19, p. 126. At the political level, the hospital sector in the FRG lacks the organized, regional and national superstructure which is found in the nation’s physician sector or, for that matter, in the United States hospital sector. One suspects, therefore, that the hospital sector in the PRC lacks the political strength to represent its economic interests before political bodies as effectively as can the medical profession. Perhaps as a result, the PRC hospital sector appears to face perennial fiscal problems that have only recently (in 1972) received the attention of the federal government.* These and other financial aspects of the West-German hospital sector, however, had best be examined in connection with that nation’s health insurance system. 3. THE HEALTH-INSURANCE SYSTEM IN THE FRG As noted in the introduction, 99.1 percent of the West-German population now has comprehensive health insurance coverage. Of the insured, roughly 8.5 percent is insured by private insurance companies who set premiums on the basis of standard actuarial principles. The remainder (90 percent) obtains insurance coverage under what is known as Soziale Krankenversicherung *In I972, the legislature passed the so—ealled Krankenhausfinanzierungsesetz (Hospital Finance Act) providing for a number of fiscal reforms to be discussed in Section 3. 262 Table 13 HOSPITAL USE, US. AND FRG, 1970 U.s.a F.R.G.b 1. Number of Admissions per 1,000 Population: All Hospitals ..................... 156 145 Short-term, Acute ................. 146 129 2. Number of Patient Days per 1,000 Population: All Hospitals ..................... 2,331 3,620 Short-term, Acute ................. 1,173 2,375 3. Average Length of Stay: All Hospitals ..................... 14.9 24.9 Short-term, Acute ................. 8.0 18.4° 4. Occupancy Rated .................. 78% 88% aTaken or calculated from United States Statistical Abstract, 1974, Tables 123 and 124, p. 93. bTaken or calculated from Statistisches Jahrbuch fiir die Bundesrepublik Deutsch land 1972, Table 15, p. 68 for a population of 61 million. 0Interestingly enough, the average length of stay in FRG hospitals specializing in gynecology and obstetrics is as high as 11 days. dFor the U.S., this includes only General and Special Hospitals, excluding Psychiatric and Tuberculosis Hospitals. For the FRG, all hospitals are included. (Social Health Insurance) or also as Gesetzliche Krankenversicherung (loosely translated, “Legally Mandated Health Insurance”) which carries the acronym G.K.V. and will be so identified hereafter in this paper. It may be noted at the outset that a literal translation of the word Krankenversicherung is the term “sickness insurance.” This term is actually more descriptive of the kind of insurance it represents. For the most part, the insurance covers the financial losses occasioned by illness and, in the FRG, ensures patients access to health services in case of illness. Attempts to finance the consumption of preventive care through the insurance mechanism are relatively novel, even in the FRG. Since 1970, however, all health insurance funds in the G.K.V. must cover a series of preventive services for children up to age 4 years (a total of seven examinations during that age interval) and an annual examination for the detection of cancer available to women over the age of 30 and men over the age of 45. Preliminary data indicate that the West-German public has been slow to avail itself of these preventive measures.* *Thus, in 1972 only 54 percent of the theoretically available examinations for children were actually demanded by the insured. Similarly, only 11 percent of the men and 27 percent of the women eligible for cancer-screening tests actually availed themselves of these services. There is some evidence, however, that these percentages are increasing over time. (See Bundesminister fur Arbeit und Sozialordnung (1975), pp. 163-4). 263 Social Health Insurance (G.K.V.) in the F RG The Social Health Insurance (G.K.V.) in the FRG is the oldest branch of that nation’s highly progressive social security system. The legal foundation for the G.K.V. dates back to 1883. In its early stages, the system covered only blue collar workers and thus excluded large segments of the population. Among the benefits then provided by the G.K.V. were a modest range of necessary health services and, more importantly, sick benefits in the form of costs intended to compensate partially for loss of wages as a result of illness. 6 evolution of this insurance program during the ensuing three-quarters of a century has proceeded in what strikes one as predictable directions. First, the system has embraced a steadily increasing proportion of the population. As will be seen from Table 14, one would not violate reality unduly by treating the G.K.V. as the health insurance system in the F .R.G. Second, the G.K.V. has covered an ever-widening range of health-care related goods and services. The benefit package offered by the G.K.V. now includes items that might be contemplated for insurance coverage on the North American continent only in Canada, but would probably not enter the minds of even the more progressive American policymakers. In this subsection, the more prominent features of this health insurance system are examined in detail. Table 14 HEALTH INSURANCE COVERAGE, FEDERAL REPUBLIC OF GERMANY, 1971 Social . Health Private TOTAL Insurance InHeraltlcie System S“ an (G K V )a Companies 1. Number of Health Insurance Funds in the G.K.V. ; or Private Health Insurance Companies .......... 1,863 1,801 63 2. Number of Insured (millions) ...... 60.1 54.9 5.2 3. Percent of total Population (60.68 million) Insured .............. 99.1% 90.5% 8.6% 4. Percent of Population Voluntarily Insured ................... 22.3% 14.0% 100% 5. Percent of Population in “Ersatzkassen” .............. 26.6% — — a“Gestatzliche Krankenversicherung” alias “Soziale Krankenversicherung” (Social Health-Insurance). Source: Bundesminister ffir Arbeit und Sozialordnung, 1975, Ubersicht fiber die Soziale Sicherung, pp. 151-3. 264 The Insurance Carriers in the G.K.V. Unlike the British or Canadian social health insurance systems, the G.K.V. is in effect a mosaic of some 1,600 (as of December 1973) health-insurance fu’nds (Krankenkassen). These funds may be organized by locality (the so-called Allgemeinen Ortskrankenkassen), by the trade or craft of their membership, by employer (for large industrial enterprises) or by still other criteria. Within the legal framework established for the G.K.V. as a whole, each health-insurance fund functions as a legally and fiscally autonomous entity, albeit on a non-profit basis. The law prescribes lower limits to the funds’ benefit packages, prohibits favorable risk selection on the part of the funds, prescribes minimum and maximum fees or charges for health services, and provides for other forms of public regulation and supervision of the funds. Within these constraints each fund is expected to maintain solvency with recourse only to membership contributions. Public subsidies of the sort common in other nations are the exception rather than the rule. The typical health insurance fund is directed through an assembly of representatives recruited, in equal strength, from the fund’s members and from the latters’ employers. The notable exception to this pattern is furnished by the so-called Ersatzkassen (Substitute Funds) which are directed solely by representatives of the membership. These “substitute funds” enroll predomi- nantly white collar workers. In general, they offer their members relatively more comprehensive benefit packages and pay providers relatively higher remuneration. They also levy relatively higher contributions on their members. The health insurance funds in the FRG are organized at both state and national levels. The health insurance funds within the state form a statewide association that negotiates on behalf of the funds with the state’s Kassena'rzt- lichen Vereinigung—the association of all the state’s physicians (or dentists) participating in the G.K.V. The main item negotiated at this level is a lump sum transfer of funds to the physicians’ (dentists’) association, which in turn distributes the fund to its members on either a capitation or, now more commonly, a fee-for-service basis. At the federal level the associations of health insurance funds provide countervailing power to the federal associations of health-care providers and to federal lawmakers. Membership (Enrolhnent) in the G.K.V. Depending upon a member’s occupation and/or economic status, (s)he is either a compulsory or a voluntary member of a G.K.V. health insurance fund. As is shown in Table 14, about 86 percent of the current membership is enrolled on a compulsory basis. Included in this group are: All blue-collar workers White-collar workers whose income is below a certain level (currently DM 22,500 and, in general, a level equal to 75 percent of the maximum income limit for contribution to social security (Rentenversicherung) itself a dynamic cutoff). With few exceptions, all retired persons. All unemployed persons. Virtually all farmers. Sundry other groups of modest economic status. 265 Persons who do not have to join the G.K.V. on a compulsory basis may nevertheless do so; indeed, a number of distinct groups among such persons have a legal right to gain admittance to the G.K.V. on their own initiative. Finally, sundry groups who must, in principle, join the G.K.V. on a compulsory basis may apply for an exemption which, if granted, frees them from membership in the G.K.V. As Table 14 suggests, any such flight from the- G.K.V. can at best have been a trickle. As noted, the typical health insurance fund is fully financed through premiums (contributions) by members. The source of these premiums varies according to the economic position of members. For actively employed persons, contributions are calculated as a percentage of gross earnings up to the income limit (DM 22,500 in 1973) at which insurance ceases to be compulsory. On average, the contribution by such members amounted to about 9 percent in 1973, although there were significant variations across funds, with some Ersatzkassen levying as much as 12 percent and the Mariners Health Insurance Funds (See krankenkassen) as little as 6.2 percent. In general one«half of an employed person’s contribution is paid by the employee. The other half is borne by the employer (who will, of course, shift significant portions of that contribution backwards). For employees with abnormally low incomes, the employer makes the entire contribution. Contributions for retired persons are made by the pension funds of which these persons are members. These contributions are calculated as a function of the pension received by the insured. (During 1974, the average annual contribution per retired member was about DM 1,150.) The retired themselves are currently not required to absorb any part of this contribution; during the period 1957 to 1970, they were required to contribute 2 percent of their annual pension. As can be seen in Table 16, the payments health-insurance funds make on behalf of the aged exceed the contributions received on their behalf by a substantial margin (30 percent in 1970); this is one problem which is currently under review by West-German legislators. In the meantime, the G.K.V. follows the principle of distributing the health insurance of retired persons to those funds to which these persons belonged during their working lives. This procedure goes some way towards eliminating adverse risk selection, but it still exposes particular funds to rather capricious burdens over time. Finally, the contribution of sundry other groups—e.g., the unemployed, persons on social welfare, members of the armed forces—are assumed by the public sector, notably by the federal sector. The public sector also contributes to a fund a stipulated sum (currently DM 400) for every case in which the fund paid maternity benefits. On balance, however, direct public contributions are a minor source of income for the G.K.V. system; it is essentially a self-fmanced program. It should be noted that {Member’s contribution to a health-insurance fund extends all insured benefits (excluding cash payments for loss of wages due to illness) to all members of the insured ’s family) Since the insured’s contribution to a fund is a function solely of his or her gross incomemor a fixed sum for retired and other members—it follows that the health-insurance funds observe actuarial principles only at the aggregate level. Within the membership of a particular fund there is cross-subsidization not only from well to sick persons, 266 and from the working population to the aged, but also from small to large families. This circumstance furnishes an incentive for breadwinners in small families, or single persons, to seek health insurance at actuarially fairer premiums if their income is sufficiently high to place them outside the compulsory G.K.V. membership. The private-insurance sector writes policies on precisely this basis. The Benefit-Package under the G.K.V. Health insurance funds operating within the G.K.V. framework offer their members an impressive catalog of benefits.* Among these are: (a) Preventive health services (annual checkups and special diagnostic tests) for children and other particular age groups. (b) All medically necessary health care in case of illness, including: Services rendered by physicians and dentists, including routine dental care. All services rendered by hospitals, including services rendered by hospital physicians and, since January 1974, unlimited hospital stays. All drugs and sundry health appliances prescribed by physicians or dentists (including all or part of the cost of dentures, prostheses, eyeglasses, and so on). The services of health personnel in patients’ homes (rarely, because such instances normally involve outright hospitalization). Stays in a rest home or extended-care facility if prescribed by a physician. Rehabilitative services. (0) Cash payments to cover loss of income as a result of illness, with payments beginning after salary or wage payments by employers cease, and amounting to 80 percent of gross income. ((1) In addition to all medical services in connection with a delivery, cash payments for working mothers to compensate for income loss before and after the delivery (payments cover 6 weeks prior and 8 weeks after delivery). (e) Cash contributions to the cost of funerals in case of death. (f) Cash payments to compensate for income loss when the insured member must stay home from work to care for a sick child. (g) The cost of substitute personnel to participate in household chores or the care of children if parents are unable to do so because of illness. (h) The cost of personnel employed by farmers as a substitute for the labor of an insured farmer who is temporarily ill. In earlier stages of the G.K.V. cash payments to compensate for income loss (item (c) above) constituted the main expenditure of the funds. In 1925, for example, the ratio of expenditures for this purpose to expenditures on bona fide health services was 1:1. By 1973, that ratio had fallen to 1210.3. This *Virtually all service benefits are available to insured patients without coinsurance or deductibles at point of consumption. 267 development reflects only in part,the ever-increasing relative cost of health services. For the most part, it reflects the fact that, by law, employers in the FRG must now pay employees absent because of illness their wages for at least the first 6 weeks of absence. Since few episodes of illness occasion a 6-week leave, employers have absorbed the bulk of the income-compensating cash payments to sick employees. The legal framework of the G.K.V. prescribes minimum benefits that must be made available by all G.K.V. health—insurance funds. With a few exceptions, the preceding catalog is included in that minimum package. Individual funds may, however, offer additional benefits at their discretion, although also within limits prescribed by law. For example, an insurance fund may offer its members full coverage of the cost of dentures or major health appliances. The funds may also offer hospital accommodation superior to the so-called third-class accommodation (rooms with more than four persons) provided for in the minimum package. The Ersatzkassen (Substitute Funds) typically do offer such additional benefits, hence their generally higher premiums. To the extent that certain desired benefits—cg. first-class accommodation in hospitals or the choice of a particular physician within the hospital—are not made available by G.K.V. funds, members of the G.K.V. may purchase supple- mentary insurance from private insurance carriers. For that reason, the number of insurance policies issued by private insurance carriers suggests that roughly 10 percent of the West-German population have some type of private health insurance coverage, when in fact only about 8.5 percent have only private health insurance (see Table 14). The Relationship between the G.K.V. and the Providers of Health Services Because of the comprehensive benefit package offered by G.K.V. funds, that system constitutes the main source of revenue for the bulk of health-care providers in the FRG and for the producers of health products as well. In principle, the health insurance funds disburse the funds collected from members (minus an average of about 5 percent in administrative costs, see Table 16) in the following order of magnitude:* (1) Per diem fees for hospitalized patients. (2) Payments for drugs and sundry appliances. (3) Payments for physician services and dental services. The daily reimbursable charges for inpatient care are negotiated between the funds and hospitals according to guidelines prescribed by law. Prior to 1972 these daily charges left the hospitals with perennial deficits that had to be made up from other sources, often private contributions. Since 1972, however, the daily charges have been defined to cover only operating expenses (Benuzterkosten or “user-costs”); capital costs are recovered from public sources: one-third from the federal government and the balance from the state government. Since 1974 there has prevailed one uniform daily charge for all hospital patients regardless of the patient’s insurance statussl' *Thelen (1974), p. 103. ,_ . TSee Bundesminister fur Arbeit und Sozialordnung, Ubersicht fiber die Soziale Sicherung (1975), pp. 167-8. From this reference, one infers that this uniform daily charge applies to all hospitals, although the matter is not discussed explicitly. 268 The prices of pharmaceuticals are determined primarily by the pharma- ceutical industry and its retailers, the pharmacists. In principle, the attending physician has discretion over the drugs he prescribes for a given treatment, but he must be prepared to defend his choices as cost—effective. Indeed, the health-insurance funds may require him to do so. A similar condition applies to dentists. As far as the patient is concerned, all medically necessary drugs are free at the time of their acquisition, aside from a trivial fee per prescription (currently a maximum of DM 2.50). Presumably it is thought that the physician’s mandate to prescribe cost-effective drugs will induce the appro- priate amount of competition among pharmaceutical companies. How success- ful this attempt to stimulate the market can actually be remains a question inviting further analysis. Of the total of approximately 100,000 professionally active physicians in the FRG, roughly 50 percent are in private medical practice (see Table 5). As is shown in Table 15, more than 90 percent of these participate in the G.K.V. system, although about 15 percent of these physicians’ gross income is derived from the treatment of private patients (without G.K.V. coverage). Only about 7.9 percent of private practitioners devote their practice exclusively to private patients.* It will be noted from Table 15 that these proportions have shifted over time in favor of the G.K.V. To some extent this development may reflect the fact that, prior to 1960, admission to the G.K.V. system was a prerogative bestowed by the health insurance funds, who used it inter alia to constrain the physician-population ratio within their domain to what was deemed a maximum prudent limit.1‘ In 1960, the West-German Supreme Court ruled that Table 15 PRIVATE MEDICAL PRACTITIONERS IN THE G.K.V., FRG, 1960-1971 Number of . Percent Treating only Year M.D.’s in Private Petrceplt [Signs/ted Private (Non-G.K.V.) Practice 0 e ' ' ' Patients 1960 ...... 49,225 81.9 18.1 1965 ...... 50,215 90.2 9.8 1971 ...... 50,717 92.1 7.9 Source: Gesellschaft fiir Sozialen Fortschrift, e. V. (1974), Table 2, p. 20. *Whether the typical G.K.V. physician gives his private patients superior treatment is a matter of some debate. In many instances, physicians do reserve particular blocks of hours for private patients and do them the courtesy of shorter queues in the waiting room. Whether such courtesies carry over to the application of medical science is a matter of conjecture. Tlt would seem, then, that as long ago as 1900 astute German thinkers anticipated the notorious Canadian Dr. Evans who many years later established that the demand for physician services is essentially physician-induced. In this connection, see virtually any of Robert Evans’ papers, including, most likely, that presented at this conference. 269 arrangement unconstitutional. Since then all professionally qualified (appro- bierte) physicians and dentists can gain admittance to the G.K.V. upon application. G.K.V. physicians or dentists are not paid directly by the health insurance funds for services rendered. Instead, the funds annually negotiate, ex ante, lump-sum transfers to the state associations representing these professionals (the Kassenfirztlichen Vereinigungen and the Kassenzahn'arztlichen Vereini- gungen). By law, the latter are free to disburse these funds among their members on either a capitation or a fee-for-service basis. Capitation reimburse- ment was predominant some decades ago; fee-for-service reimbursement currently predominates. Whatever method is used, the underlying document on which payment is made is the so-called Krankenschein, a voucher surrendered by the patient to the physician with whom primary contact is made in case of illness. Each insured patient has four such vouchers for physician services and another four for dental care, per year. Implicit in this arrangement is, of course, that once a patient has contacted a primary physician and surrendered the quarter’s voucher to that physician, she or he is expected to receive all further ambulatory treatment in the quarter from that physician, unless the physician endorses the voucher over to another physician by way of referral, or has the patient hospitalized.* To be paid for his or her services, the physician submits all vouchers received during a quarter to his Kassena'rztliche Vereinigung (State Association of Sickness Fund Physicians) for payment. This submission is made at the end of each quarter. In a sense, then, the state association of G.K.V. physicians functions as what is known in the United States as a “medical foundation.” There are, in effect, eleven such foundations in the PRC, one for each Land (state). Since the association is thus forced to preside over a massive annual intra-professional, zero-sum game, it is to be expected that it assumes certain policing powers over individual physician members. Thus it is not uncommon that physicians with a much above average number of vouchers (or services) will find their fees per voucher or service reduced relative to physicians with lower reported “output” rates. The notion appears to be that over-achievers tend to trade quality for the sake of quantity, which may be deemed unethical for at least one of two reasonsxl' Furthermore, since the medical profession in the FRG does assume responsibility for an adequate geographic distribution of medical services, physicians in relatively over-doctored urban areas may find their fees reduced relative to those paid for identical services by physicians in under-doctored rural areas. This arrangement is interesting from the American perspective. It can fairly be said that the American medical profession would find revolution- ary the notion that the profession as a whole can be held responsible for an adequate distribution of medical services among the American people. The idea of using fee differentials (under national health insurance) as a policy instrument in this context is deemed more revolutionary still, apparently even *Interestingly enough, between 1969 and 1973, insured G.K.V. members were entitled to a refund for unused vouchers. This arrangement has since been abolished. The change in policy might be used to subject the Evans’ thesis to an empirical test. TThe second reason is that such conduct may work to the disadvantage of the patient. 270 by policymakers deeply concerned over the alleged maldistribution of medical ~ services in the United Statesnt The Flow of Funds in the G.K.V. System Figure 1 below presents a schematic summary of the flows of funds and of goods and services under the Social Health Insurance System of the FRG. The actual magnitude of the monetary flows in 1970 can be seen in Table 16. Added perspectives on these monetary flows are provided in Tables 17 to 21. As already noted, practically 100 percent of the G.K.V.’s resources are obtained from membership contributions. These funds are distributed in almost equal proportions to physicians, hospitals, the producers of pharma- ceuticals and health appliances, and others (including dentists). In connection with dental services, incidentally, it will be noted that the virtually total coverage of dental care in the FRG does not appear to constitute an enormous drain on the national economy. The figure of DM 1.6 billion in Table 16 (even if one raised it to DM 2 billion to account for dental appliances and dental services not paid for by the G.K.V.) constituted about 0.29 percent of the FRG’s gross national product in 1970. If these statistics are any guide at all, one may surmise that even the United States and Canada could survive complete coverage of dental services. The ratio of payments to contributions for the G.K.V. system as a whole is 0.954. The remainder of about 5 percent represents administrative costs. This percentage is strikingly similar to the experience reported by the provincial medical-care insurance plans in Canada. In 1972, for example, the administra- tive costs of the Saskatchewan Medical Care Insurance Commission amounted to 4 percent of total receipts (Annual Report 1972, p. 7). In the same year, the Régie L’Assurance-Maladie du Quebec (Quebec Health Insurance Board) spent 5.6 percent of its revenues on administration (Quatrieme Rapport Annuel 1972/73, p. 10). Unfortunately, the West-German figure excludes the adminis- trative costs incurred by the Kassenairztlichen Vereinigung (state associations of G.K.V. physicians) in the distribution of funds to the individual physicians. This function includes the examination of the physicians’ claims and the monitoring of their treatment pattern; the associated costs are likely to be non-trivial. The Canadian figures include these costs and are therefore not directly comparable to the West-German data.* By way of contrast, Table 17 presents data on the flow of funds through the private health-insurance system in the FRG. In this system the retention factor (funds retained for administration and profit) is understandably higher. As is noted in a footnote to the Table, virtually all of the 6,000 or so policies written by the private sector are sold on an individual basis, nourishing in the process a commensurate army of marketing personnel. TFOI a proposal to use fees in precisely this manner under a US. national health insurance system, see Reinhardt (1975). *The author is indebted for this point to Professor Erik Holst. 271 ZLZ Table 16 SOURCES AND USES OF FUNDS OF THE COMPULSORY HEALTH INSURANCE SYSTEM (G.K.V.) OF THE FEDERAL REPUBLIC OF GERMANY, 1970 Retired Persons and ' Other Members and All Mem bers and Families Families Families Billions DM % Billions DM % Billions DM % 1. Sources of Funds: Contributions by and for Members .......... 5.0 19% 20.0 77% 25.0 100% 2. Uses of Funds: Payments to Physicians .................. g 1.6 29% 4.0 } 71% 5.6 23.5% Payments to Dentists ................... 1.6 1.6 6.7% Drugs, etc. (incl. Dentures) ................ 2.3 40% 3.4 60% 5.7 23.9% Payments to Hospitals .................. 2.1 35% 3.9 65% 6.0 25.1% Maternity Benefits .................... — 1.1 100% 1.1 4.6% Payments at Death (Sterbegeld) ............ .3 75% .1 25% .4 1.6% Cash Payments in Connection with Sick Leave (Krankengeld) ....................... 7 2.3 100% 2.3 9.6% Administrative Costs .................... — — — 1.2 5.0% Total .............................. 6.3 26% 16.4 69% 23.9 100% 3. Contribution and Payment per Member: Contribution (Premium) ................. DM 627 DM 881 DM 816 Payment (All Types) ................... DM 821 DM 763 DM 779 Payment/Contribution .................. 1.309 0.866 0.954 Source: Statistisches Jahrbuch fiir die Bundesrepublik Deutschland 1972, Table l-C, p. 382. Table 17 SOURCES AND USES OF PRIVATE HEALTH INSURANCE COMPANIES, FRG, 1971a 13:33:: Payments Payments/ Billions DM BllllonsDM Premiums Health Services ............. 2.64 1.93 .731 Cash Payments to Cover Lost Income (Krankentagegeld) ...... .31 .19 .629 Supplementary Benefits to Cover Coinsurance (Selbst—standiger Teil) 1.50 .76 .506 Total .................... 4.45 2.88 a The data represent a total of 54 companies and policies for between 5.5 and 6 million insured persons (including family members). More than 90 percent of these policies were individual; only about 10 percent represent group coverage. Source: Statistisches Jahrbuch derBundesrepublik Deutschland 1972, Table 5, p. 374. 4. EXPENDITURES ON HEALTH SERVICES IN THE FRG Tables 18 to 21 present a statistical overview of health-care expenditures in the FRG. Where possible and useful, these data are accompanied by reasonably comparable US. data. It warrants mention at the outset that such comparisons are fraught with pitfalls and are at best suggestive. The statistical material available to the author at this time does not contain a time series of national health expenditures in the FRG. The absence of such a series from the country’s statistical yearbook (Statistisches Jahrbuch fur die Bundesrepublic Deutschland) suggests that such aggregates are not regularly compiled as is done with great care, for example, in the United States. One of the more recent attempts to construct such a cost picture for the PRC is contained in the Gesundheitsbericht (Health Report) of the Bundesminister fl'fr Jugend, Gesundheit und Familie (1971). Table 18 has been adapted from that report. According to Table 18, the FRG devoted about 6.1 percent of its gross national product to the provision of health care in 1968. The comparable statistic for the United States is 6.6 percent. Both nations devoted about 80 percent of that total to medical treatments, although the breakdown within that category shows some differences. For example, the data in Table 18 strongly suggest that the United States devotes a relatively greater proportion of its resources to the hospital sector than does the FRG. Indeed, Table 18 is likely to understate this difference, because the PRC expenditures of DM 10.6 billion on hospitals includes 273 VLZ NATIONAL HEALTH EXPENDITURES, FRG AND U.S., 1968 Table 18 Sources (in Billions of DM) Comparable U.S. Object of Expenditure Sector G.K.V.a Dggigfeby Otherc .T-OTAL Percent ' ' Datf’fercent of Householdsb (Billions DM) of GNP Billions $ GNP . Preventive Care and Public Health Services .......... 2.9 1.3 0.2 0.1 4.5 .83 6.7 .77 . Medical Treatments: a. Physicians and Dentists. . . . 0.3 6.4 2.3 — 9.0 1.67 14.7 1.70 b. Hospitals ............. 3.7 4.3 0.7 1.9 10.6 1.96 23.2 2.70 0. Drugs, etc. (incl. Dentures) . 0.5 3.8 1.9 — 6.2 1.15 7.9 .91 Total ................ 4.5 14.5 4.9 1.9 25.8 4.78 45.8 5.31 . Education and Research: a. Education ............ 2.1 — 0.2 0.1 2.4 0.44 2.3 .26 b. Research ............. 0.1 — — — 0.1 — 1.8 .21 c. Total ................ 2.2 — 0.2 0.1 2.5 0.44 4.1 0.47 . Grand Totale ........... 9.6 15.8 5.3 2.1 32.8 6.07 56.6 6.55 (30%) (48%)” (16%)b (6%)c (100%) SLZ Table 18—(Continued) 8May include funds obtained from the public sector to cover health insurance of retired persons and recipients of social welfare. blncludes cash indemriities paid by private insurance companies to private households. Such indemnifies are estimated to have amounted to DM 2 billion in 1968. cEmployers and other organizations (including religious organizations). dThe GNP for the FRG in 1968 was DM 540 billion. The U.S. figure stood at $864 billion. eExcludes cash payments for sick leave and early retirement. In' the FRG these amounted to DM 15.7 billion in 1968, of which DM 4 billion was paid by the G.K.V., DM 3 billion by employers, DM 0.1 by private insurance companies, and DM 8.6 billion by the public sector. fIt is difficult to attain complete comparability with published data. The following distribution was adopted: Line 1: Other professional services; government public health activities; expenses for prepayment and administration; other health services. Line 2a: Physicians’ services (including those rendered in the hospital which are excluded for the FRG); dentists’ services. Line 2b: Hospitals (excluding payments for services by affiliated physicians. In the FRG virtually all physician services are included in Line 2b); nursing home care. Line 2c: Drugs and sundries; eyeglasses and appliances. Line 3a: Medical facilities construction. Line 3b: Research. Sources: For FRG data: Der Bundesminister fiir Jugend, Familie und Gesundheit, Gesundheitsbericht (1971), p. 158. For U.S. data: United States StatisticalAbstract 1972, Table 91, p. 66. 9LZ Physicians in Private, Ambulatory Care Distribution of Lump Sum Receipt by Capitation or State Associations of G. K.V. Physicians Pharmaceutical Industry Including Retailers (Pharmacists) '-"I Hospitals, Including Hospital-Based Physicians l l I I I I I I l I l l I I J |_—_..__.._______—_____—_...__—__—-__— Practice Feefor-Service and of Dentists Reimbursement l I I : Annual Lump Payment for Drugs and I Sum Transfer Medical Supplies at | Administered Prices I l I I I Payment at I G.K.V. Sickness Funds Predetermined I Per Diem Rates I I I l . «o» I I I I Other (Federal : Pension Funds Essigeadmnasmbers Government, Local I Governments, Etc.) : A I L .. __________________ J I L- _ _____________ l.____ _______ _..____...______....____ LEGEN D Flow of Funds ————— -— Flow of Goods and Services Figure 1. Flow of funds and of goods and services under the G.K.V. system. SECULAR CHANGE IN HOSPITAL LENGTH OF STAY AND “COSTS” Table 19 PER DAY,(3) FEDERAL REPUBLIC OF GERMANY, 1950-1971 Y Episodes of Length of Stay G.K.V. Expenditure ear Hospitalization per Episode per Patient-Day 1950 ........ 2.9 million 23.3 days DM 6.50 1960 ........ 4.1 million 22.2 days DM 17.20 1971 ........ 5.8 million 20.7 days DM 6360 Annual Growth 1950-60 ...... 3.5% -.5% 9.7% 1960-71 ...... 3.2% -.6% 11.9% Sources: Bundesminister fiir J ugend, Familie und Gesundheit, Gesundheitsbericht (1971), pp. 167-8, and Statistisches Jahrbuch fu'r die Bundersrepublz'k Deutschland, various years. Table 20 SECULAR GROWTH OF EXPENDITURES BY HEALTH INSURANCE FUNDS, FEDERAL REPUBLIC OF GERMANY, 1950-73 (1950 = 100) Expenditures on: Year EXILZtgiltures Physicians Drugs and Institutional1 Sickday and Dent1sts Sundry Care Benefits Billions DM 2.278 .568 .438 .438 .469 1950 index 100 100 100 100 100 1960 417 412 358 310 548 1970 1,105 1,261 1,153 1,371 491 1973 1,877 1,901 2,253 2,581 840 Annual Growth: 1960-70 9.7% 11.2% 11.7% 14.9% -1.1% 1970-73 13.2% 10.3% 16.7% 15.8% 13.4% Source: Bundesminister fiir Arbeit und Sozialordnung, Ubersicht uber die Sozz‘ale Sicherzmg (1975), pp. 175. 277 Table 21 SECULAR GROWTH IN EXPENDITURES PER INSURED MEMBER BY HEALTH INSURANCE FUNDS, FRG, 1951-72 Object of Expenditure: Year . . . . Drugs and Hospital Physrcrans Dentists Sundry Care m4 _D_M w m 1951 ........... 26.19 7.56 15.76 24.84 1960 ........... 71.26 20.24 34.96 N/A 1968 ........... 149.80 61.71 89.54 132.12 1972 ........... 231.20 86.98 133.11 251.20 Annual Rate of Growth jg fl E E 1951-60 ......... 11.2 10.9 8.9 9 8 1960—68 ......... 9.2 13.9 11.8 ‘ 1968-72 ......... 10.8 8.6 9.9 16.1 Source: Bundesminister fur Jugend, Familie und Gesundheit, Gesundheitsbericht (1971), pp. 167-8, and Statistisches Jahrbuch fiir die Bundesrepublz'k Deutschland, various years. virtually all inpatient physician services. The bulk of such expenditures is excluded from hospital expenditures for the United States and is included instead in the share accruing to physicians and dentists. This observed difference in hospital expenditures is all the more remarkable in view of the relatively much higher rate of patient-days per capita in the FRG. As is noted in Table 13, the per-capita utilization of patient-days in the FRG exceeds that in the United States by 55 percent counting all hospitals, and by 102 percent in acute, short-term hospitals. All of this differential was seen to be attributable to a difference in the average length of hospital stays. It has already been observed in an earlier section that the typical patient-day in the FRG appears to be far less resource-intensive than it is in the United States. For example, the division of total reported expenditures on hospitals in the FRG for 1968 by total reported patient-days for that year yields an average cost per patient-day of DM 48. (In this connection, see also Table 19, column 3, on G.K.V. expenditures per patient-day.) This figure, it will be recalled, includes the cost of inpatient physician services. Although a fully comparable statistic for the United States is not available, it can be observed that a similar calculation for that country yields an average cost (excluding the cost of inpatient physician-services performed by private physicians) of roughly $50 per patient-day. As was noted in Table 12, a good part of this cost differential seems to be attributable to differences in underlying staffing patterns. It would appear from the Table that the average number of 278 non-medical personnel per patient in the United States exceeds the comparable average for the FRG by about 104 percent. Whether commensurate differences exist also in physical plant and equipment is a moot question at this time of writing. Similarly, available statistics do not contain information on the relative remuneration of allied health-workers in the two countries and on the occupational mix of non-medical hospital personnel. A part of the observed cost difference may reside also in these factors. Tables 20 and 21 present secular movements in expenditures for health care by the G.K.V. funds. Table 20 presents total expenditures; in Table 21, these statistics are converted to a per member (not to be confused with “per capita”) basis.* As is shown at the bottom of Table 21, expenditures per insured member have increased in all categories during the past two decades, at average annual compound rates of somewhere between 8 and 11 percent. Secular movements in the price of health services account for a good part—perhaps as much as half—of these increases in expenditures (see Table 22 below). But it is clear that there have also been significant increases in real consumption of physician services and drugs. What impact a nation’s health system has on the population’s overall health status will always remain a complex question, if only because health services as such constitute but one input into the health production process. It would therefore be reckless to evaluate the West-German health system against data on morbidity and mortality. Even so, Tables 23 and 24 below present some illuminating information of particular targets not currently being attained in the FRG. Table 23, for example, exhibits information on death rates by leading cause for the US. and the FRG. It will be seen that, on the surface at least, death rates for illnesses subject to medical control tend to be remarkably higher in the FRG than they are in the US. Of particular interest in this connection are infant mortality and maternal death rates. As is seen in Table 24, on these indices the FRG truly lags behind the US. experience, especially if one bases that judgment on a comparison of the U.S.-White rates with the West-German rates. After all, the US. death rates are significantly raised by what Americans prefer to refer to as the “heterogeneity” of the American people. Such “heterogeneity” is not reflected in the West-German statistics. Precisely why the United States, with its much-maligned, alleged “non-system” of health care exhibits a superior record on these scores is an interesting question, but one that goes beyond the compass of this paper. It also, of course, lies beyond this author’s professional competence. Table 25 presents statistics on the morbidity reported for employed persons covered by the G.K.V. system. Two factors stand out from the Table. First, there has been a pronounced secular increase in the reported number of illness episodes resulting in work-loss (line 2). Some growth has occurred also in the number of workdays lost due to illness per insured employee, although this growth has been dampened by some decline in the average days of work-loss per illness episode (lines 3 and 5). Line 6 presents the so-called Durchschnitt- lichen Krankenstand (average percentage of insured employees absent at any point in time due to reported illness). According to that statistic, roughly 5 *By “member” is meant only the primary insured. A member may account for several covered persons because members’ families are automatically insured under the G.K.V. 279 08Z Table 22 SECULAR PRICE-INDEX MOVEMENTS IN THE U.S. AND THE FRG, 1950-1973 (1967 = 100) Consumer Medical Care Index of Preisindex Preisindex Index for Year Price Price Physician fiir die ffir Physician Index Index Fees Lebenshaltunga Gesundheitspflegeb Feesc (2A) (3A) (1B) (2B) (3B) 1950 ........... 72.1 53.7 55.2 68.8 N/A N/A 1955 ........... 80.2 64.8 65.4 75.7 N/A N/A 1960 ........... 88.7 79.1 77.0 82.8 N/A N/A 1963 ........... 80.6 1965 ........... 94.5 89.5 88.3 94.9 90.9 N/A 1970 ........... 116.3 120.6 121.4 108.2 114.0 108.8 1971 ........... 121.3 128.4 129.8 113.8 121.4 121.9 1972 ........ ‘ . . . 125.3 132.5 133.8 120.3 129.3 129.0 1973 ........... 133.3 137.5 138.2 N/A N/A N/A Average Annual Growth Rates 1950-60 ......... 2.1% 3.9% 3.3% 1.9% — — 1960-70 ......... 2.7% 4.2% 4.6% 2.7% — ’ I8Z Table 22—Continued SECULAR PRICE-INDEX MOVEMENTS IN THE U.S. AND THE FRG, 1950-1973 (1967 = 100) Consumer Medical Care Index of Preisindex Preisindex Index for Year Price Price Physician fur die flir Physician Index Index Fees Lebenshaltunga Gesundheitspflegeb Feesc (2A) (3A) (113) (23) (33) 1965-70 ......... 4.1% 5.6% 6.4% 2.6% 4.5% 4.3% 1970-72 ......... 3.7% 4.7% 4.9% 5.3% 6.3% 8.5% 1972-73 ......... 6.0% 3.8% 3.3% ~ _ — aPrice Index for Maintenance of Living Standards. bPrice for Health Care. cThis is an index developed from the fee schedules (E-Adgo) of the Ersatzkassen in the G.K.V. Sources: US. data: (1.5. Statistical Abstract 1974; Table 98, p. 68; Table 665, p. 411. PRO data: Calculated from data in the S tatistisches Jahrbuch fir die Bundesrepu blik Deu tschland I 974, Section XXI-D, Tables 3 and 4, pp. 4634, and Gesellschaft fiir Sozialen Fortschn'tt (1 974), Table 18, p. 66. Table 23 DEATH RATES, PER 100,000 POPULATION, BY LEADING CAUSE, U.S. AND FRG, 1970 12%]; Cause of Death U.S. FRG B9 Whooping Cough 0.0 0.1 B14 Measles 0.1 0.1 B21 Diabetes Mellitus 18.5 32.1 B23 Anemia 1.7 2.3 B27 Hypertension 11.0 20.6 B31 Influenza 1.9 11.9c B32 Pneumonia 28.5 32.4 B33 Bronchitis, Emphysema, Asthma 14.8 38.8 B35 Appendicitis 0.7 2.8 B40 Miscarriage 3.8 6.8 B41 Other Complications during Pregnancy, or Delivery” 21.0 45.0 aClassification Code of the 1968 ICD List B. bPer 100,000 live births. 0For 1971, the reported figure is 2.9. Sources: Statistisches Jahrbuch fii'r die Bundesrepublt'k Deutschland, 1972 (for data on the FRG) and 1973 (for data on the U.S.), Section E of “Internationale Ubersicht,” Table 1, p. 34. percent of the employed West-German labor force absents itself from work at any point in time because of illness. This absentee rate is considerably higher than that in the United States. In 1970, the United States reported an average of 5.4 work-loss days (disability days) per employed person aged 17 and over. As is seen from Table 26 (line 3), a fairly comparable figure for West Germany was 17.1 disability days per employed person—or roughly 216 percent higher than the U.S. figure. There is reason to believe that this difference in reported disability days reflects more than underlying morbidity. There are likely to be cultural differences in the definition of what constitutes a legitimate reason to be absent from work, and financial incentives may contribute to the observed differential in disability days as well. As was noted above, the West-German social security system and related labor legislation insulate West-German 282 Table 24 INFANT MORTALITY AND MATERNAL DEATH RATES, U.S. AND FRG, 1960 - 1970 U.S. FRG Total White Non—white Total 1. Infant Mortality:a 1960 .............. 22.9 43.2 26.0 33.8 1965 .............. 21.5 40.3 24.7 23.9 1970 .............. 17.8 30.9 20.0 23.4 2. Maternal Deathz‘D 1960 .............. 26.0 97.9 37.1 97.7c 1965 .............. 21.0 83.7 31.6 NA. 1970 .............. 14.4 55.9 21.5 51.8 aDeaths of infants under 1 year of age, exclusive of fetal deaths, per 1,000 live births. bPer 100,000 live births. cFigure for 1961/62, quoted from R. K. Schicke (1971), p. 183. Sources: U.S.—United States StatisticalAbstract, 1974, Table 84, p. 60. FRG —Statistz'sch es Jahrbuch filr die Bundesrepublik Deutschland, various years. Data on Infant Mortality —Table 1 of Section D of “Internationale Ubersichten," p. 31*. Data on Maternal Death Rates—T able 7, “Miittersterbefalle nach Alter- sgruppen und Todesursache,” Section III, “Gesundheitswesen,” p. 62 of the 1972 edition. employees from any negative financial consequences of disability days. In the ' United States, such protection is less complete.* *In terms of its macro-economic impact, the high number of disability days in West Germany is probably more than offset by the country’s relatively low unemployment rate and the low number of days lost due to strikes. The following data are illuminating: U.S. FRG 1. Unemployment, percent of potential labor 5.3% 0.7% force (1970) 2. Days lost due to work stoppages, per 1000 941 4 employees (1970) 3. Disability days per employed person (1970) 5.4 17.1 Sources: U.S. Department of Labor, Bureau of Labor Statistics, Handbook of Labor Statistics, 1974, Table 39, p. 103; Table 59, p. 143; Table 153, p. 368; Table 164, p. 412; and Table 171, p. 418. U.S. Bureau of the Census, Statistical Abstract of the United States, 1973, Table 120, p. 80. Thelen (1974), Appendix Tables 11 and 15. 283 Table 25 DAYS LOST FROM WORK BY MEMBERS OF THE COMPULSORY HEALTH INSURANCE SYSTEM (G.K.V.a) OF THE FEDERAL REPUBLIC, 1950-1970 1950 1960 1970 1, Members (excluding Retired Persons and Family Members of Employed Members), in Millions .......................... 15.7 21.6 22.6 Male .......................... 10.4 13.7 14.8 Female ........................ 5.3 7.9 7.8 2. Episodes of Illness Resulting in Work-Loss, per 100 Members (as Defined in Line 1): Male .......................... 45.0 74.6 89.0 Female ........................ 47.0 64.6 86.5 3. Number of Days of Work-Loss per Member: Male .......................... 11.2 17.4 17.1 Female ........................ 10.9 15.6 17.1 4. Number of Hospital Days per Member Male .......................... 1.8 1.8 1.7 Female ................ ' ........ 2.4 2.0 2.0 5. Number of Days of Work-Loss per Episode of Illness: Male .......................... 23.8 23.4 19.2 Female ........................ 26.6 24.5 19.8 6. Average Percentage of Insured Members reported to be incapable of work because of illness, at any point in timeb Male .......................... 4.0% 5.9% 5.4% Female ........................ 4.1% 6.0% 5.2% aGesetzliche Krankenversicherung (G.K.V.) Durchschmittlicher Krankenstand. Source: Calculated from Thelen (1974), Appendix Tables 11 and 15. 5. CONCLUDING REMARKS It has been the objective of this paper to provide a concise description of the health-care delivery system and of the health-insurance system of the Federal Republic of Germany. In the course of that description it was brought , out that the West-German health system appears to be amply endowed with 284 medical manpower and inpatient facilities—at least by international stand- ards—although the PRC appears to employ relatively fewer allied health personnel than are employed in the United States. Furthermore, it was noted that the Social Health Insurance System (the G.K.V.) in West Germany is a remarkable one, both in terms of the proportion of the population it embraces and in terms of the generous benefits offered to enrollees at effectively zero marginal-cost when health services are consumed. Finally, it was noted that although the average West German uses roughly 50 percent more hospital days than does the average American, and also consumes substantially more ambulatory physician visits,* the proportion of gross national product devoted to health care in general appears to be roughly the same in the FRG and in the United States. The reason for this apparent paradox seems to reflect differences in the resource intensity of a “hospital day” and a “physician visit.” It was noted in Section 4 that the typical patient-day in West Germany appears to be only about half as resource-intensive as an American patient-day. In connection with physician visits, it has been noted by Schicke (1971) that West-German physicians appear to spend considerably less time per patient visit (perhaps only half as much) than dorAmerican physicians (p. 193). Although it was noted in Section 3 that West-German physicians appear to have pushed the delegation of tasks to support staffs further than have their American colleagues, and may thus well be more productive, it may also be the case that West-German and American “physician visits” are not qualitatively com- parable. If one can judge by the expenditure data of the G.K.V., the cost of health services in the PRC has been rapidly escalating during the first part of this decade (Tables 20 and 21). Expenditures per insured member appear to have increased at an average annual compound rate in excess of 10 percent between 1970 and 1973, and the trend appears to have continued unabated since that time. According to one recent estimate reported in the press, for example, the average contribution to the G.K.V. fund will have to increase by more than 40 percent over the next 3 years, to a total of about 13 percent of an employee’s gross earnings. There appears to be resistance to such levies on the part of the West-German work force (The German Tribune, Nov. 14, 1974, p. 12). A number of economic incentives built into the West-German health system are apt to make that system more expenditure-prone than it probably needs to be purely on the basis of medical need. First, the absence of any cost-sharing on the part of consumers effectively eliminates any financial stake consumers might have in an economic use of the nation’s health resources. Although a reform proposal in 1960 included a provision for cost-sharing of some type, the measure was never passed into law, and it is not likely to be in the near future. Indeed, the recently-released report on the health system by the Minister of Labor and Social Security (Bundesminister fi’ir Arbeit und Soziale Sicherung, 1975) evinces little concern over the cost issue, and instead maps out plans for an even more generous *According to Schicke (1 71, p. 185) West Germans consume an average of about 10.2 patient-visits (arztliche Konsultationen und Besuche) per year. This statistic may be compared with a per capita office visit rate of 4.5 for Americans. 285 benefit package. As noted earlier, even the erstwhile system of refunds for unused health services vouchers has recently been abolished, presumably on a philosophy that is opposed to cost-sharing as a matter of principle. West-German providers of health services also face few if any effective incentives to control utilization on their own volition. As noted, the individual physician loses control over his patients as soon as the latter are hospitalized. In light of this circumstance it is perhaps not surprising that virtually the only utilization statistic on which the F RG scores lower than the US. is the number of hospital admissions per capita (see Table 13, line 1). Once patients are hospitalized, the decision to release them rests with the house staff, and the latter has no obvious incentive to hasten the patient’s departure. Indeed, the fiscal arrangement between the G.K.V. funds and the hospital sector gives the latter every incentive to prolong patient stays, for at per diem charges that are fixed ex ante any additional patient-day contributes to the recovery of overhead. Although a salaried chief of medicine may be thought to be impervious to the hospital administrator’s economic calculus, it is unreasonable to pretend that the house staff would not in some way identify its own position with the hospital’s fiscal position. Under a health-insurance system in which private practitioners are paid on a fee-for-service basis and patients bear no out-of-pocket costs at point of consumption, the individual physician clearly has an economic incentive to over-service his or her patients (unless, of course, the practice faces strong excess demand). The incentive will be the stronger, the lower the absolute level of the fee schedule is. It was noted in Section 3 that some control over excess prescription of services is exerted by the fact that the typical West-German practitioner practices under a form of statewide medical foundation. That this control mechanism is not unduly harsh can be inferred from a comparison of the relatively high West-German physician incomes with the relatively low fees for particular services.* Until recently, the education of medical personnel in West Germany had been subject to the so-called Numerus Clausus under which the absolute number of medical-school places was legally constrained to a given limit. Some writers have identified the West-German medical profession as the source of the underlying legislation (Thelen (1974)). It is hypothesized that physicians seek limits on their number to preserve their high incomes. During the last few years, the Numerus Clausus restrictions have been relaxed, and thought has been given to increasing the West-German physician supply to levels as high as 300 physicians per 100,000 population. The notion now appears to be that such increases will induce competition among physicians and thus trigger downward pressure on physician incomes and fees. The danger of such a policy is all too obvious to merit extended comment. First, there is always the possibility that any excess supply of physicians will, indeed, lead to physician-induced increases in the demand for health services. One need not deny the existence of a significant price-elasticity of demand for *According to the relatively higher fee schedule observed by the Ersatzkassen, an office visit (Beratung) in 1972 carried a fee of DM 5, an examination (Enigehende Untersuchung) DM 9.20, and a home visit during the day (Besuch am Tag) 'DM 12.40 (see Gesellschaft fiir Sozialen fortschritt (1974), p. 66). 286 physician services to reach that conclusion. Furthermore, under a national health insurance system of the West-German variety, a politically well- organized medical profession may well be successful in maintaining the income of its members (through commensurate increases in negotiated transfers, i.e., fees) almost irrespective of the overall number of its members. How West Germany will fare under the proposed manpower policy—should it in fact be implemented—would be highly instructive for American policymakers. Equally instructive would be a closer examination of the social contract between West-German society and its medical profession. As was noted in Section 3, West-German society has granted its medical profession a lucrative monopoly over the provision of primary care and of specialist services. In return for this privilege, however, society expects the medical profession to assume responsibility for making adequate health services available when and where needed. This arrangement is formalized in the contract between the health insurance funds and the Kassenaerztlichen Vereinigungen (state asso- ciations of physicians practicing under the G.K.V. system), and in the public charter given the Landesaerztekammern (state medical associations). The level of physician remuneration under the G.K.V. also is agreed upon quite formally through negotiations between the health-insurance fund and organized medicine. One of the noteworthy features of these negotiations is that they are not an open-ended commitment on the part of the public as is the case in Canada (where only fee schedules are negotiated) or in the United States (where physicians are commonly paid on the basis of “customary” local fee schedules). Instead, under the West-German system an aggregate lump sum transfer is negotiated ex ante, and for the ensuing year that lump represents society’s financial commitment to physicians. Furthermore, there is thus an agreed-upon bottom line figure of which all parties are aware. On division of this lump sum by the total number of practicing physicians, average annual physician income can be made highly and accurately visible in the media—and therein surely lies some virtue. By contrast, the social contract between physicians and society in the United States is far less explicit and more one-sided. Like their West-German counterparts, physicians in the United States have been granted a virtual monopoly to function as entry points into the health-care system, and like their West-German colleagues, American physicians are paid well for their efforts. In return for these privileges, however, organized American medicine has not been asked, and has not offered, to assume responsibility for the provision of adequate health services when and where needed. Whatever services Americans do receive from the physician sector, they receive them by the physicians’ good graces. It would probably not occur to an American community without a physician to call upon the state’s medical society for relief or, if such a call were made, that medical society wOuld undoubtedly feel no compulsion to solve the community’s problem. Why American society has been at once so generous and so timid in contracting with its physicians is a fascinating question. Sooner or later the United States will introduce a national health insurance system, following, perhaps, the example recently set by Canada. Whatever the particular form of that system will be, it can be taken for granted that 287 physicians will continue to occupy their highly favorable position in the nation’s income distribution picture. A national health insurance system, however, is likely to make more visible the flow of funds from consumers to the physician sector, and it is likely to make the contract between the public and that sector more explicit. Those who administer the system may well feel compelled by this visibility to seek from the medical profession a more formal commitment to respond to society’s need, as that need is defined by society. In thinking about this possibility, American policymakers might gain insights from the West-German experience. On paper, at least, that country appears to have struck a more balanced contract between society at large and the medical profession. The question is how well that social contract performs in practice. 288 BIBLIOGRAPHY AMERICAN MEDICAL ASSOCIATION (1971, 1972, 1973), Reference Data on Profile of Medical Practice. Chicago: American Medical Association. BUNDESMINISTER FfiR ARBEIT UND SOZIALORDNUNG (1970, 1975), Ubersicht uber die Sozz'ale Sicherung. Bonn: Der Bundesminister fiir Arbeit und Sozialordnung; (June 1970 and June 1975). BUNDESMINISTER FUR JUGEND, FAMILIE UND GESUNDHEIT (1971), Gesundheitsberich t. Stuttgart: Verlag W. Kohlhammer. BUNDESMINISTER FI'J'R JUGEND, FAMILIE UND GESUNDHEIT (1973), Problemanalysen and Reformschwerpunkte fur das Gesundheitswesen der Bundesrepublik Deutschland. Stuttgart: Verlag W. Kohlhammer. EVANS, ROBERT G. (1973), Price Formation in the Market for Physicians’ Services in Canada, 1957-1969. Study prepared for the Prices and Incomes Commission, Canada, 197 2. Ottawa, Canada: Information Canada. GESELLSCHAFT FUR SOZIALEN FORTSCHRITT e.V. (1974),Der Wandel der Stellung des Arztes im Einkommensgefuge. Berlin: Duncker & Humbolt. KIMBELL, L. J. and LORANT, J.H. (1973), “Physician Productivity and Returns to Scale.” Paper presented before the American Economic Association Meetings, New York, December 29, 1973 (Mimeo). LICHTNER, SIGRID, and PFLANZ, MANFRED, (1971), “Appendectomy in the Federal Republic of Germany: Epidemiology and Medical Care Patterns.” Medical Care, 9 (July-August 1971): 31 1—330. MONSMA, GEORGE N., JR. (1970), “Marginal Revenue and the Demand for Physicians’ Services,” in Herbert E. Klarman (ed.) with the assistance of Helen H. Jaszi, Empirical Studies in Health Economics, Proceedings of the Second Conference on the Economics of Health. Baltimore: Johns Hopkins Press. REINHARDT, UWE E. (1972), “A Production Function for Physicians’ Services.” Review of Economics and Statistics 54 (February 1972): 55-56. REINHARDT, UWE E. (1975), “Alternative Methods of Reimbursing Non- Institutional Providers of Health Services.” In Controls of Health Care: Papers of the Conference on Regulation in the Health Industry. Washington, D.C.: Institute of Medicine, National Academy of Sciences, pp. 139-174. SCHICKE, ROMUALD K. (1971), Am and Gesundheitsversorgung im Gesellschaftlichen Sicherungssystem: Bundesrepublik Deutschland-England- USA. Freiburg im Breingau: Verlag Rombach. STATISTISCHES BUNDESAMT DER BUNDESREPUBLIK DEUTSCHLAND (1972), Statistisches Jahrbuch fur die Bundesrepublz’k Deutschland 1972. Wiesbaden: Verlag W. Kohlhammer. 289 STOCKHAUSEN, JOSEF (1973), Der Arztliche Beruf in der Bundesrepublik Deutschland 1973. Koln: Deutscher Arzte-Verlag. THELEN, WILLI (1974), Numerus Clausus und Arzteschaft: Zur Kritik der Arztlichen Interessenpolitik. Giessen: Verlag Andreas Achenback. U.S. BUREAU OF THE CENSUS (various years), Statistical Abstract of the United States. Washington, DC: U.S. Government Printing Office. U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE (1974), The Supply ofHealth Manpower. Washington, DC: U.S. Superintendent of Documents, December 1974. 290 DISCUSSANT’S COMMENTS ROMUALD K. SCHICKE Hanover Medical School, West Germany Risks and Responsibilities in Social Welfare It may be considered a truism that the degree of personal responsibility for health and economic wellbeing is defined by each society in the light of its unique historical, social, and economic forces molding their contemporary security system and adjusting to changing needs. The locus of responsibility to assume different risks, in turn, leads to developing new, and modifying existing insurance schemes—private (individual or collective) on one hand, and collective-social on the other—the latter scheme being offered either as a matter or right or privilege comprising the fabric of a more—or-less comprehensive security scheme according to the society’s prevail- ing value system. The component elements—comprising private insurance, social insurance, and social welfare schemes as a matter of right, and public assistance as a privilege—indicate different constellations in a number of societies. It seems that one could differentiate between more dynamic systems where the allocation of responsibilities is in the process of evaluation where, for example, claims on the health services as a social right are not as yet confirmed by the de facto distribution of responsibilities on one hand (for instance, in the United States) and countries where such questions need not be posed or have already been resolved for some time (in the FRG, for example) on the other. In the Federal Republic of Germany (FRG) the social insurance system is the prevailing one, and priority is given to the collective social insurance scheme as a matter of earned right; whereas the “welfare state” type of protection applies solely to some underprivileged and disabled population groups for which the state has assumed responsibility; and the public assistance system is allotted a marginal, tertiary role to fill the gaps of the other schemes. Parenthetically, approximately 2.5 percent of the total population were “on welfare” in 1971. When compared internationally, the public assistance concept was relatively early superseded by the health insurance system introduced on a large scale by the end of the nineteenth century. By comparison, the predominantly public assistance system of Britain (embodying certain punitive features of the “work-house” concept rooted in Elizabethan Poor Law) was gradually but hesitatively transformed into a Welfare State scheme. It is noteworthy that within the British health security scheme insurance is assigned a tertiary role. In the United States where historically the public-assistance-type scheme was conceptually adopted from Britain, it exists rather residually, since the 291 locus of the responsibility for economic wellbeing and (nota bene) for health have become parts of a societal credo emphasizing self-reliance, independence, and opportunity for everyone. The dynamic growth of the private insurance system, as compared with more moderate growth in the public sphere, tends to lend further strength to private insurance mechanisms including those in the realm of health. Parenthetically, even for aged persons eligible for Medicare, as much as 33 percent of the funds for their health care came from private sources in 1970. ‘ Expenditure on social security schemes including health programs reflect responsibilities shared between the person and society, the burden’s allocation and its gravitation. Also, the mode and financing of insurance programs and the source of funds evidence the role of sharing arrangements, responsibilities, and deduced rights in the society in question. The Financing of the Health Scheme In the realm of sickness insurance in the FRG, the responsibility rests, as already pointed out, within the collectivity and solidarity of traditional occupation-related risk groups with basically self-financed (employer- employee) programs reflected in 1,627 separate sick funds (as of December 1973). While in the FRG approximately 95 percent of the insurance schemes are financed by the private sector, in the United States the role of government financing increased considerably from a share of 13.6 percent in 1929 to 38.9 percent in 1973. In the FRG it seems that expansion of the self-financed program (which covered at its inception in 1883 approximately 15 percent of the population) to include 90.5 percent of the population under statutory insurance (8.6 percent are privately insured and 0.9 percent are without insurance) in 1971, has been mainly due to the “dynamically” conceived financial mechanism directed toward gradually absorbing higher income groups and thus spreading the risk. Needless to say, such a dynamic feature also provides for some redistribution of resources. But this does not seem to apply to the United States, where government-sponsored programs (Medicaid, Medicare) deal either with a population at risk with limited resources, high utilizers of health services, or both categories which rely mainly on governmental health services, and, not rarely, the latter are competing for resources with other governmental programs alien to health. Such also may apply to the United Kingdom and Canada to a varying degree, or to all systems financed mainly by general revenues rather than by employer-employee contributions. The growth of the statutory health system in the FRG has been accompanied by the addition of benefits and new programs. More recently (since 1970) there has been a reallocation of responsibility, and by the same token a reallocation of funds from the social health insurance scheme to the employer—who has been assuming payment of sick-leave benefits to the extent of full average wages, also for blue-collar workers, for a duration up to 6 weeks. The parts of the sick funds thus have been reassigned to expand benefits in the 292 area of preventive medicine, for example, cancer-detection screening programs for women 30 years and older and for men 45 and over. Preventive services have also been initiated for children up to the age of 4, and the increasing costs of such care have also been met by this means. The expansion of health insurance has not been confined solely to expanded benefits. The latest action has been assumption of the bulk of cost for dental prosthetics; and since 1972 there have been included in the statutory scheme further population segments such as certain groups of farmers. Presently under consideration are laws that would extend the statutory health insurance scheme to students. Regrettably, the expansion of health services has also witnessed a considerable increase in statutory health insurance expenditures, which have risen from DM 15.8 billion in 1965 to DM 42.7 billion in 1973. N.B., similarly increased expenditures (1974) have been made on the consumption of alcoholic beverages (DM 27 billion) and tobacco (DM 15.4 billion), for a total of DM 42.4 billion. The overall costs for health services increased during 1965-1972 by 113.9 percent or at the average rate of approximately 11 percent annually and such a trend may continue. While in 1965 approximately 6.8 percent of the GNP was spent on health services, this amounted to 8.3 percent in 1972 (8.1 percent in 1971). It is anticipated that limits to “growth” or the ability to finance may soon be reached. Reportedly, slightly lower expenditures for health services were made in the US. (7.6 percent of GNP) and Canada (7.1 percent of GNP) in 1971. In this connection a word of caution may be in order. While such data may be generally comparable, differences in accounting, benefit structure, and morbidity-indicated possible differences in the utilization of health services in Table 1 COSTS OF HEALTH SERVICES IN THE FRG, 1965-1972 1965 1970 1972 1965-72 (in Millions) DM DM DM Increase % Prevention ............ 1,793 2,805 3,827 134.4 Treatment ............ 20,489 35,400 46,354 126.2 Occupational disease and accidents ........... 4,662 7,172 9,160 96.5 Invalidity ............ 4,565 6,877 8,051 76.4 Total ............ 31,509 52,254 67,392 113.9 GNP (in Millions) DM ...... 460,900 682,800 813,500 Health Services in % of GNP. 6.84 7.64 8.28 293 Table 2 COST OF SICKNESS AND EXPENDITURES FOR HEALTH BY CATEGORIES IN THE FRG IN 1968 Category Cost in Billions of DM % A. Education & Research .......... 2.5 5.15 B. Curative Sector .............. 25.8 53.20 Hospitals ................ (10.6) (21.86) Physicians & Dentists ......... (9.0) (18.56) Drugs & Medications ......... (6.2) (12.78) C. Consequences of Illness ........ 15.7 32.37 Wages & Salaries, Financial aid, etcetera ........ (9.1) (18.76) Premature Invalidity due to Illness or Accident ................ (6.6) (13.61) D. Prevention and allied programs . . . . 4.5 9.28 Total ................ 48.5 100.00 extent and kind may, when compared internationally, provide some sources for inherent distortions. The financing of the tremendouslyrising expenditures in the health sector, aggravated by stagflation, has been insured by the annually-rising ceiling of income liable to contribution (DM 2,100 monthly or DM 25,200 yearly in 1975 compared with DM 22,500 annually in 1974). The present level of contributions approaches 5 percent of liable wages paid by each—the employer and the employee. It therefore becomes evident that the risk of ill health in the FRG is a truly collective one, where contributions are stratified according to the ability to pay rather than flat rates, family rates, experience rating, or other modes used elsewhere. Also, the copayment feature does not constitute a significant element and principally applies to medications purchased at the drugstore (at 20 percent of cost or maximally DM 2.50 per prescription), in the ambulatory sector, and for« amenities such as private accommodations in hospitals, not indicated by the medical condition, in the stationary one. Besides, since January 1974 hospital care has been offered on an unlimited basis (previously limited to a maximum of 1.5 years). A law providing flat-rate rebates for not using ambulatory physicians’ services, introduced in 1969, was repealed in 1973 as not being expedient. One could add that it seemed rather conceptually contradictory if the aim is to provide “free” preventive services on one hand, and to discourage the 294 utilization of ambulatory physicians’ services on the other. Nevertheless, this attests to the fact that even old-established systems occasionally are subject to some experimentation. By comparison with the United States it seems that issues related to cost “explosion” have been more dormant in the opinion of the FRG public. The reasons for this may be manifold but it seems that the level of cost-conscious- ness in the realm of health is rather low among users. In the FRG patients have been fully isolated from participation in the financing mechanism, which predominantly is left to bilateral collective arrangements between the sick funds and the physicians’ collective representative body at the Land (state) level in the ambulatory care sector. Similarly, the setting of hospital per-diem rates and payments is of concern to hospital authorities on one hand, and the Land’s health office and the insurance funds on the other. By contrast, remuneration aspects in the US. have basically been preserved within the sphere of patient-physician contact and in the social health- insurance scheme. Additionally, cost awareness is promoted among the users, not seldom through insurance carriers. The remuneration mode of the FRG social insurance system seems to encourage quantification of various services rendered on a rather detailed fee-for-service basis. This enhances providers’ incomes in the free practice sector, where both general practitioners and specialists operate. The volume of reimbursement is guided by an itemized list of services (approximately 1700 items in 1967) which may be subject to modification for general conformity by the Insurance Physicians’ Association. More significant deviations from the “average” may be subject to financial regression measures. The relationship between general practitioners and specialists is not free from certain com- petitive aspects, partly because most diagnostic work is carried out on an extra-institutional basis, extensively by specialists who derive a considerable portion of their incomes from technical diagnostic procedures rather than from consultation fees. The prevailing organizational separation of the institutional sector from the ambulatory one—the former dominated by governmental, state and local institutions—reinforces the structural and functional segregation of the providers, and, at times, contributes to the duplication of costly diagnostic servrces. Out of the total statutory health-insurance expenditures in 1973 the main spendings in the ambulatory sector were comprised of 21.2 percent for physicians’ services, 11.2 percent for dentists’ services and prosthetics; and drugs and medications dispensed amounted to 20.5 percent. Owing to higher consumption and relatively higher per-item costs, it seems that medications are more expensive in the FRG than in Britain or some other countries. In part this is due to predominantly quasi-administered prices and lower levels of competition, especially in the distribution sector. Indirectly, the consumption of medications is affected by higher physician-patient consultation rates (7.6 per inhabitant, as compared with Smden’s 2.5 and the US. rate of 4). In general, it may be noted that the ratio of monetary to service benefits within the statutory health scheme has markedly declined from 120.6 in 1885 to 1:103 in 1973. Nevertheless, in 1973 10.8 percent of expenditures were 295 devoted to the continued payment of salaries and wages during sickness or disability. The largest individual cost item constituted the hospital and allied stationary services, which absorbed a 28.3 percent share of the expenses incurred in 1973. In this connection it may be added that the majority of hospital physicians and other personnel are remunerated by salaries guided by official tariff arrangements, similar to those of civil servants and coupled with limited career opportunities due to a relatively smaller number of higher level medical staff positions offered. These organizational aspects of medical staff composition may have some bearing upon the decision process of patients’ care evidenced in a longer length of stay than elsewhere (short-term hospital stays in 1970 were: FRG 18.3 days, France 17.8, Denmark 13.1, Sweden 12.7, and USA 9.0 days). While in 1960 there were 42 subordinate physicians to 10 chiefs in the hospital, the ratio changed to 63 subordinates per 10 chief physicians in 1975. It may also be noteworthy that in 1975 there is one hospital physician per 12.6 hospital beds. Slightly more than one-half of all physicians (56,327) are engaged in hospitals, while 52,913 physicians are practicing in the ambulatory sector— rendering a ratio of one physician per 1,170 population or 85.5 physicians per 100,000 inhabitants. Among the factors contributiong to sizeable cost increases is the growth in the hospital sector in terms of bed/population ratios. Unlike in Britain (10.0 beds/1000 population in 1962 and 9.0 in 1972) or the US. (9.2 beds/1000 population in 1960 and 7.9 in 1970), where the trend has been toward reduction in the number of beds, the supply of hospital beds in the PRC increased from 10.5 in 1960 to 11.4 in 1972. The relatively greater share of costly short-term beds (7.7 in 1972) may be of further special concern. The growth trend in the hospital sector has, however, not been slowed by notorious underfinancing in years past. In 1972 and 1973 legislation was enacted providing for covering capital cost and depreciation expenses of 15 percent of total costs to be equally shared by federal, state (Land) and local government funds, while 85 percent, mainly operating costs, are being covered by the insurances’ per diem payments. The co-fmancing of depreciation and capital costs is predicated upon instituting modern cost-accounting procedures which will be mandatory by 1978, and contingent upon a hospital-needs plan prepared at the Land level. N.B., formalized criteria and standards for the economic operation of hospitals are still lacking. Hospital Rates Establishing Boards, as already mentioned, are organized at the Land level. Hospital costs soared between 30 and 50 percent ranging between DM 130 and DM 170 (1974) in a number of general city and county hospitals throughout the PRC between 1973 and 1974. To be vigilant regarding economies there seems to be a latent need to install utilization and similar committees in hospitals—next to providing adequate academic education facilities for top hospital-management personnel as a step towards the modern hospital management methods evidenced in the United States. 296 The separations and gaps between hospital and ambulatory sectors need to be narrowed, if not closed, to achieve greater economies of resources, and to appreciably enhance organizational cooperation. In conclusion it may suffice to point out that despite the common cost problems that haunt the health-care scene (but which are not confined to it), because of its unique circumstances each society has to employ its own measures toward the containment of costs, but in doing so it should be substantially guided by the experience and concepts of other countries, especially those lending themselves to generalization and abstraction inter- nationally. Appendix A STATUTORY HEALTH INSURANCE EXPENDITURES IN THE FRG IN 1973 Category DM % Physicians 8,644,415,419 21.2 Dentists 2,614,999,658 6.5 Other professionals 8,163 ,480 — Drugs, medications and appliances 8,350,691,878 20.5 Dental prosthetics 1,901,383,349 4.7 Hospital and other stationary care 11,476,872,436 28.3 Wage and salary benefits 4,365,096,196 10.8 Control physicians and dentists 160,532,334 0.3 Convalescent care 52,854,117 0.1 Screening and prevention 727,698,834 1.8 Family assistance 326,224,761 0.7 Maternity services 564,426,723 1.3 Maternity cash benefits 720,884,443 1.8 Maternity screening and prevention 72,462,607 0.1 Housekeeping services 11,351,635 — Death benefits 796,201,251 1.9 Total 40,794,259,121 100.0 297 BIBLIOGRAPHY COOPER, B. S. and McGEE, M. F. (1971), “Medical-care outlays for three age groups: Young, intermediate and aged,” Social Security Bulletin, 34:12-13, May. COOPER, M. H. and COOPER, A. J. (1972), International Price Comparison, A Study of the Prices of Pharmaceuticals in the UK and Eight Other Countries in 1970, London: National Economic Development Office. EICHHORN, S. (1973), “German Federal Republic,” In I. Douglad-Wilson and G. McLachlan, (eds.), Health Services Prospects—An International Survey, pp. 81-84 London: The Lancet and The Nuffield Provincial Hospitals Trust. GEHB, K. (1975), Die 'arztliche Versorgung in der Bundesrepublik Deutsch- land, Ergebnisse der Arztestatistik zum 1. Januar 1975, Deutsches Arzte- blatt, 18:1263-1268, May. GESETZ zur wirtschaftlichen Sicherung der Krankenh'auser und zur Regelung der Krankenhauspflegesatze—KHG. June 29, 1972 (BGBl I S. 1009) GOVERNMENT OF CANADA (1973), National Health Expenditures in Canada 1960-1971 (1973j—with comparative data for the United States, Health Economics and Statistics Division, Health Programs Branch, pub- lished by authority of Ministry of National Health and Welfare, Ottawa, October. LEHMING, R. (1975), “Kosteninflation im krankenhaus,” Deutsches Arzte- blatt, 41221-222, January 23. RICE, D. P. and COOPER, B. S. (1972), National Expenditures, 1929-1971, Social Security Bulletin, 35: 7, January. SCHEWE, D. et a1. (1975), Hbersicht fiber die soziale Sicherheit, ed. Der Bundesminister fur Arbeit und Sozialordnung, 9th ed., Bonn: January. SCHICKE, R. K. (1971), Arzt und Gesundheitsversorgung im gesellschaft- lichen Sicherungssystem, BRD—England-USA, Rombach: Freiburg i.B. SCHICKE, R. K. (1972). “Die regionalisierte Gesundheitsversorgung Schwe- dens,” Der Krankenhausarz t, 45(7):402-410. SCHICKE, R. K. (1972), The Physician in Three Societies’Social Security and Health Schemes—A Comparative Conceptual Framework, Paper presented at the Third International Conference on Social Science and Medicine, Elsinore, Denmark, August 14-18, 1972. SCHICKE, R. K. (1973), “The pharmaceutical market and the prescription drugs in the Federal Republic of Germany: Cross-national comparisons,” International Journal ofHealth Services, 3(2):223—236. SCHICKE, R. K. (1973), “Der organisatorische Wandel der Gesundheitsdienste in England seit Beginn dieses Jahrhunderts,” Das 5ffentliche Gesundheit- swesen, 35(9):500-508. 298 SZAMEITAT, K. and WUCHTER, G. (1970), “Was kostet die Gesundheit?,” Baden-Wfirttemberg in Wort und Zahl, Statistische Monatshefte, 18(5):]26-131. VERORDNUNG zur Regelung der Krankenhauspflegeséitze—BPflV—vom 25. April 1973 (BGBlI 8.333-356) WORTHINGTON, N. L. (1975), National Health Expenditures, Calendar years 1929-73, Office of Research and Statistics, US. Dept. of Health, Education, and Welfare, Research and Statistics Note Nr. 1, pp. 1-13, February 19. 299 Chapter 11 HEALTH COSTS AND EXPENDITURES IN ROMANIA CRISTIAN-ADRIAN HAVRILIUC Institute of Hygiene and Public Health, Bucharest, Romania 1. MEDICAL CARE ORGANIZATIONAL SYSTEM Medical care, provided through the government’s medical institutions, is a constitutional right in Romania. The whole medical-care system, which is based on unitary organizing and functional principles, provides the population with: Access to all forms of medical care. Unity of effort toward solving territorial health problems (prophylac- tic, curative and recuperative). Constant medical care for patients during all stages of the evolution of disease by the continuous cooperation of physicians in the same or in different medical units. Efficient use of specialized staff and the technical devices of investigation and treatment. Constant development and permanent improvement of medical care in keeping with contemporary medicine. All activities relating to health care are led, guided and controlled by the Ministry of Health. The Ministry carries out state policy regarding public health, and guarantees observance of the laws of the Romanian Council of Ministers in this area. In this function the Ministry cooperates with other central bodies of the state administration, and with leading Romanian cooperative associations and other social organizations (Figure 1). The Ministry of Health is the only technical body invested with the right to establish public health standards. Responsible for implementing national policy and adjusting its health-care details to meet the specific needs of the various territories, are the Popular Councils of each district and their specialized bodies, the District Health Directions.* The network of health units is structured on three levels. (1) The basic level of ambulatory medical assistance provided by general practice is represented by territorial dispensaries and enterprise factory dispensaries. In rural areas the dispensary territory is represented from the *District = administrative structure with a population of about 500,000 inhabitants. Bucharest, the capital of the country, is itself considered a district. 301 administrative point of view by the village In urban areas the delineation is conventional according to accessibility and addressibility criteria. In big factor- ies in which the production process may have health-risk factors, there are enterprise dispensaries. And, depending on local conditions, school dispensaries are also organized. COUNCIL OF MINISTERS MINISTRIES AND MINISTRY OTHER CENTRAL BODIES OF HEALTH EXECUTIVE BOARD OF THE DISTRICT POPULAR CO N I LOCAL U C L DISTRICT TECHNICAL BODIES L HEALTH DIRECTION \\\\\\\\\\\\\\\\‘ WI/I/l/I/l ' \\$\\\\\‘k\‘\\\\\\\\\\\\\\\\\\‘ COMMUNITY EXECUTIVE BOARD (URBAN) \\\\\\\\\\\\\\\\\\\ / g TOWN HOSPITAL g f t I COMMUNITY , EXECUTIVE BOARD V////////////////////////////» (RURAL) MEDICAL UNITS m Administrative Hierarchy _ Technical Hierarchv Figure 1. Health care management bodies (Source: Ministry of Health of Romania). 302 (2) The second level of specialized medical care is represented by polyclinics (outpatient clinics) and hospitals. The polyclinic is organized for a territory that will include several dispensaries, and the hospital will serve a territory that includes one or more polyclinics. Each district has one or more structural modules of this type (Figure 2). The third level is represented by superspecialized medical care, carried out by hospital and specialized institutes such as the Center for Heart Diseases, LOCAL HOSPITAL Bed Departments Polyclinic Diagnosis and Treatment Departments Creches l Orphanages Administration Department Urban Medical Dispensaries Factory, Medical Dispensaries School Medical Dispensaries R ural M edical Dispensaries all Z221 :22: pm Examination Examination Rooms Examination Examination Rooms Rooms FAP = Red-Cross First-Aid Point Figure 2. Structure of a local hospital (Source: Ministry of Health of Romania). 303 Oncological Institute, Neurosurgery Hospital, Geriatrics Hospital, and similar. These units are directly responsible to the Ministry of Health. The hospital in the district capital as a rule is the biggest one, it coordinates the activities of all hospitals and thus of all medical units in the district territory. The leading body of this district hospital is represented by the District Health Direction, which at the same time is the specialized body of the District Population Council. In fulfilling its leadership function the District Health Direction makes use of technical units, as follow: Antiepidemic center; Emergency service; District blood bank; Forensic medicine laboratory; Health education laboratory; laboratory for analysis of health and health evaluation. Other district units falling with the purview of the District Health Direction include sanitaria, tuberculosis preventoria, créches and orphanages. The District Health Direction has a double subordination (Figure 1), for it is administratively dependent on the Executive Board of the District Popular Council, and technically dependent on the Ministry of Health. This double subordination can be found at all levels. Health units are administratively dependent on the Executive Board of the local Popular Council, and technically upon the chief of the hospital or polyclinic. Medical unit structure is determined by territorial needs, and the standards and organizational framework established by the Ministry of Health. Thus, at the dispensary level we have one general practitioner for 2,000 inhabitants in the rural areas, and one for 3,000 inhabitants in the urban areas. In villages with at least 5,000 inhabitants, stomatology consultation rooms are also organized at the dispensary level. When a dispensary has at its disposal more general practitioners, they provide medical care for both adults and children. The same situation holds good for the nurses (one for 900 inhabitants), they provide adult and child care, and are also active as midwives and hygienists. Optionally, depending on necessities in the rural dispensaries, some other departments are organized, including a laboratory, pharmaceutical dispensary, bed department for adults, bed department for children, maternity home, and dietetical office. Polyclinics are organized with at least six specialities and laboratories, and in these units, there is one specialized physician per 4,500 inhabitants. Hospitals are organized in departments that comprise between 50 and 150 beds and are designed for a certain speciality. There is one specialist physician per 15 to 25 beds. The number and types of specialities in a department are established in response to need. For a given territory there should at least be departments of internal medicine, surgery, pediatrics and obstetrics-gynecol- ogy. In order to provide the population with high-quality services differentiated according to medical needs, and toward the provision of good equipment and 304 the efficient use of materials and staff, hospitals are organized hierarchically. This hierarchy is achieved first within a district, and then in adjacent districts by the establishment of some specialities. For example, each district has one orthopedic department, and there is one department of nutritional diseases for every four or five districts. The hierarchy is made possible by unitary technical management that places the District Health Direction at the district level and the Ministry of Health above. The same technical assistance unit ensures the dynamics of the system, providing a patient with sequential care as f0110Ws: (a) In the territorial dispensary or enterprise dispensary; (b) In the territorial polyclinic if a specialized examination is required by the physician, and then to (c) the territorial hospital if hospitalization is necessary. After the hospitalization period, the patient is followed-up at the polyclinic or in the dispensary, depending on the case. However, in practice, when required by the medical situation, the circuit is modified and the patient is sent straight to the unit which can provide the best medical care for his disease. Thus, the most serious cases are sent, at the very beginning, to the most appropriate units while less serious ones do not uselessly burden the beds in specialized hospitals. Professionally the system allows for and facilitates the cooperation of specialists at various levels. Besides cooperation in solving problem cases, general practitioners at the basic level are periodically trained in polyclinics and hospitals to acquire or improve some techniques, and physicians from polyclinics take turns in hospital activity, and vice versa; the staff structure in polyclinics and hospitals is unique. Romania presently has about 33,000 working physicians (Figures 3-8). Administrative dependence on the local Popular Council provides the population with broad participation in solving their health problems and in health-care management. The Popular Council is responsible for budgeting the health units, and this is in keeping with the national budgeting system which is completely descendent from the nation budget through to district and, finally, territorial budgets (towns or villages). At each level the budget is distributed on the basis of activity, and is designed on the grounds of technical-economic indexes (in the subject case these are such as indexes of use, number of beds, etcetera) as proposed by the health units and District Health Directions. The proposals are analysed and approved by the Treasury and by the Ministry of Health, and the budgets are balanced according to incomes and expenditures. The Ministry of Health also controls the following medical units: Medical institutes and centers. Senior staff training center. Senior nurses training center. Sanitary high school; Sanitary colleges. Medical information center. Central Medical library. Medical Publishing house. Hospitals, sanataria, preventoria. Pharmaceutical supply units. Medical Computing and statistical center. 305 90$ MEDICAL DISPENSARY Pediatrics Adults Stoma— tology BED DEPARTMENTS >. C .t‘ :3 g es: 5: L: SE 0‘5 0; <30 u.< LLO E3: OTHER DEPARTMENTS Dietetical Dispensary Office Laboratory Pharmaceu— tical Examination Rooms in Villages Examination Rooms in Factories Red-Cross First-Aid Points Red-Cross F irst-Aid Points Figure 3. Organizational structure of the rural medical dispensary (Source: Ministry of Health of Romania). Medical education is controlled by the Ministry of Teaching and Education and by the Ministry of Health. i Health care activity is planned. At the level of the Ministry of Health, the plan is expressed by “long-term programs” for various problems and by annual plans of “technical-organizational measures.” Decisions are based upon studies carried out in the units of the Ministry and scientifically coordinated by the Romanian Academy of Medical Sciences. 27613 7664 1938 1950 1960 1970 1974 Figure 4. Number of physicians (excluding dentists) (From: Statistical Summary of Socialist Republic of R omam’a). 307 762 1938 1950 1960 1970 1974 Figure 5.1nhabitants per physician (excluding dentists) (From: Statistical Summary of Socialist Republic of Romania). 308 For 100,000 Inhabitants (I) 50 100 150 200 2?0 i l l U.S.S. R. W/ / WM 231 Czechoslovakia [7/ d 200 Hungary W A 191 Bulgaria V/ A 183 Austria [7/ A 182 Italy [7 i/ / / M179 E:;:;T\gepublic of V / A170 Belgium [7/ A 155 Greece V M 155 German Democr. Rep. W A 152 Denmark 145 Norway mm 141 Switzerland 138 Spain 134 France mm 130 Sweden V////////////////A 130 . Romania 129 United Kingdom 122 Netherlands 122 Yugoslavia 95 Year 1974 No Physicians Finland 95 in Romania = 27,613 Portugal 84 (excluding dentists) Albania /A 59 (1967) i \ \ Ms §\\ \\ «a Figure 6. Physicians’ density (excluding dentists) in some European countries, 1969 (Source: Ministry of Health of Romania). 309 117,426 1 1,360 1938 1950 1960 1970 1974 Figme 7. Nursing staff (From: Statistical Summary of Socialist Republic of Romania). 310 1950 = 100% 53% 6.1% 5.8% \\\‘ \\ 3.4% \\ \\ \ \\ \\ \\\\\\\\' 1.1% / \\\ \ \\ \\ \\ i / Population Physicians Medical Hospital Hospitalization Hospitalizations Procedures Beds Days \ Figure 8. Average annual growth rhythm, 1951-1972 (From: Statistical Summary of Socialist Republic of Romania). 2. HEALTH COSTS AND EXPENDITURES IN ROMANIA Our epoch is characterized by a marked rise in the demand for medical care. Society manages to meet a great deal of this demand, but it is difficult to maintain the availability-demand balance in all the organizing and budgeting variants of the health-care systems. Two series of costs are apparent as a result of the fact that budgeting is in the charge of the state or of some groups: (1) The costs of providing health-care services. (2) The costs of receiving health-care services. 311 There is at present a difference between these series of costs, and this stems from there being higher demand for health-care services than there are possibilities of increasing health-care resources. A circle is thus closed. In practice this circle is determined by the permanent granting of ever-increasing requests for health care, by cumulation and marked differences between the two series of costs, by the development of science and technology, by the rise of the cultural and health educational level, the living standard of the population, the index of prices, and by birth-rate increases and increased expectation of life. In these circumstances “cost” becomes a valuable tool in managing health care. (1) The Costs of Receiving Health-Care Services In Romania medical care is practically free of charge; the implied expenditures being covered from the state budget. The whole population benefits from free medical care, drugs, and sanitary materials for: Prophylactic activities. Cases of medico-surgical emergency. Cases of transmissible diseases. Some chronic diseases categorized by the Ministry of Health. Outpatient medical care for general practice and specialities, as well as hospitalization, drugs and materials necessary during hospitalization, are free of charge for: Public employees. Members of handicraft cooperatives. Members of agricultural collective farms. The retired. Family members of those in the above-mentioned categories. Children up to the age of 16 years. Pupils and students. Pregnant women and women lately confined. Mothers accompanying children up to the age of 2 during their hospitali- zation. Expenditures for drugs necessary in outpatient treatment are covered by the patient, but the following categories benefit from tax-free drugs: Children up to the age of 16. Pupils and students. Pregnant women and women lately confined. Mothers of three or more children. Retired persons with less than 715 lei (and their family members). Under this system, practically the whole population of the country benefits from different forms of tax-free services. For the persons not excepted, prices are modest. For instance, 7 lei is the cost of a general practitioner appointment; 15 lei for a specialized examination; from 10 to 20 lei for a hospitalization day. The incomes thus derived go into the state budget and not into the budget of the medical unit. These prices were established in 1958, and were at that time less than cost, judging from the producer’s point of view. 312 In addition to the medical units within the territorial network, there also are polyclinics which provide examination, investigation and treatment for fees. Services in these units are rendered exclusively at the request of the population. Such units are to be found in university centers and in large towns. They are conceived as an additional way to meet health-care demand and to permit citizens free access to their favorite highly-skilled physicians. These units are staffed by personnel from the network health units, operating in their spare time. These outpatient clinics are organized on the self-budgeting principle, and their movable and immovable goods are provided by the Popular Councils (as stipulated by law). Ten percent of the income of these polyclinics goes to the local budget. Also, prices are established by law: 60 lei for an appointment with a professor, 40-50 lei for a specialist physician and 30 lei for a dentist, and so on. Cases requiring urgent hospitalization are not charged. Drugs for outpatient treatment for those who do not benefit from the free system, are bought from state pharmacists. Drug prices are governed by the general price policy; but in the last 20 years there were no changes in drug prices. Every year the Ministry of Health establishes the classified list of drugs which can be used in treatment, and about 90 percent of these drugs are Romanian products. The bulk of drugs at the disposal of the population grows yearly. In 1974 the population was offered three times more drugs than in 1960 and 1.9 times more than in 1965. Of these drugs the share of them granted free of charge increased from 42.7 percent in 1960 to 57.8 percent in 1974. Comparisons may be made because of price stability and because both hospitals and the population get drugs at the selling price (Figure 9). This increase of the share of free drugs may be explained by the development of hospital and outpatient units; the extension of categories bene- fiting under the free system; and a general tendency of growth of drug use in Total 57.8% / Tax Free (Covered by the State Budget) ’4 42.2% . . (Covered by the Patlent) 57.3% 42.7% 1960 1965 1970 1974 Figure 9. Dynamics of drug consumption, 1960-1974 (From: Aspects of Health Economics). 313 l I 1950 1960 1970 1975 Figure 10. Health expenditures from state budget in millions Lei (From: Aspects of Health Economics). therapeutics. Studies estimate that in 1980 about 45 percent more drugs will be used as compared to 1975, and that 62 percent of these will be tax free (Figure 10). Romanian prices law interdicts the establishment of higher prices for new products which differ from the existing ones only slightly and in which the difference does not increase their use value for consumers. The same provisions apply to fees. For those products and services of importance to the national 314 Total 61.9% 611% Tax Free 59 7% 60.5% .— ’ (Covered by the State Budget) 59.0% - , 583% —-"?' 38 5% 38.1% f 41 0% 403% 395% ' (Covered by the Patient) k 41 7% 1975 1976 1977 1978 1979 1980 Figure 11. Forecast of drug consumption (From: Annual Statistics of Socialist Republic of Romania). economy and public use, unique prices are settled on the whole country—irre- spective of the producer and the individual costs. This explains why prices for health services and drugs remain firm and universal throughout the country. However, the development and issuance of some new products of higher quality did lead to alterations in costs for groups of drug products (Table 1). (2) The Costs of Providing Health-Care Services In Romania the state provides medical care through state units and the budgeting of these units is achieved through a “medical-care budget.” This budget has been increased every passing year, so that in 1975 it represents 9,965.8 million lei (Figure 10). At the same time, health care includes some other activities financed by the state, which are part of some other budget section in the Romanian financial system. So that, in 1975, the provisions are (Figure 10): State social insurances* ................... l8,079.0 million lei State allowances for children“? ............... 7,810.0 million lei Physical education and sports .................. 264.4 million lei Social welfare .......................... 1,631.8 million lei. It is obvious that some other activities such as communal services, pollution control, credit grants for private dwellings and so forth, contribute to the promotion of health; these activities are mentioned in other parts of the state budget. ‘ The health budget for 1975 represents 19.3 percent of all social and cultural expenditures, and 4.1 percent of the state budget (Table 2). The 51,665.8 *From the state budget for “State social insurance” the following categories are paid: old-age pensions, incapacity pensions, as well as medical support for temporary incapacity for work. To obtain these rights the employees are not obliged to subscribe. TA money support granted according to the number of children and the income of the parents. 315 Figure 12. Health expenditures per inhabitant, in Lei (From: Professor Dan Enichescu: Pilot Study of Public Health Problems in Romania). 1950 1960 1970 1975 million lei representing social and cultural activities cover 21.3 percent of the state budget (Table 3). However, the analysis of health expenditures during 1950-1975 shows a consistent level, as part of the overall total budget, of the health share—even though the 1975 amount indicates an increase of more than 15 times as compared to the budget of 1950, and health expenditures per inhabitant rose during the same interval 12 times—from 40 lei in 1950 to about 475 lei in 1975. During the period (1950-1975), taking as the basis the year 1950 (100 per- cent), the following increases have been registered: Percent Population increase ............................... 127.7 Employed population increase ....................... 120 Increase in the volume of investments .................. 1,800 National income increase ........................... 828 Gross industrial output increase ...................... 1,600 Gross agricultural output increase ..................... 278 Increase in retail sales through socialist trade .............. 905 Increase in volume of foreign trade .................... 1,300 316 191,910 179,402 ’ ’0 168,815 I P P 169,616 / 155,354 I 142,245 133,850 sf Medical Care Beds <—————I I 104,422 Hospital Beds 33,763 I I l 1938 1950 1960 1970 1972 1974 Figure 13. Medical-care beds and hospital beds (From: Annual Statistics of Socialist Republic of Romania). It is estimated that the dynamics of health expenditures will register an average annual rhythm of growth of about 8 percent. Cost of investments In the Romanian budgeting system, investments are achieved from specially designed budgets. Thus, health-care investments are not included in the health budget. In 1973, expenditures for health-care investments, social care and culture reached 846 million lei (at the prices of the year 1963 (Table 4)). Of these, 774 were from state funds, and the balance from voluntary contributions of the 317 Table 1 AVERAGE COSTS OF SOME PHARMACEUTICAL PRODUCTS (Lei) Year 1966 Year 1972 Pharmacodynamic group 3:11:11; Average :(‘i’s'tfilfg Average products COSt products COSt 1 2 3 4 5 Antibiotics ............ 19 24.00 31 50.00 Chemotherapy ......... 7 7.65 1 1 62.60 Androgenous hormones 2 18.50 4 40.50 Progesteronal .......... 2 10.50 4 41.00 Antiepileptics .......... 2 4.00 5 16. 80 Neuroleptics .......... 13 7.60 18 15.60 Cardiotonics ........... 4 3.50 12 8.00 Source: Enachescu, Trends in Health Expenditure, 1974. Table 2 STATE BUDGET STRUCTURE FOR SOCIAL AND CULTURAL ACTIVITIES (minions Lei) 1955 1960 1965 1970 1975 1 2 3 4 5 6 6,539.9 13,245.7 20,925.0 34,019.0 51,665.8 TOTAL BUDGET . . . . (100.00) (100.00) (100.00) (100.00) (100.00) 2,523.5 4,025.0 7,270.5 10,156.9 13,916.8 Education and culture . (38.6) (30.4) (34.7) (29.9) (26.9) 1,798.5 2,954.8 4,794.6 6,930.2 9,965.8 PUBLIC HEALTH . . . . (27.5) (22.3) (22.9) (20.4) (19.3) 270.7 989.6 883.7 1,249.9 1,631.8 Social welfare ....... (4.1) (7.5) (4.3) (3.7) (3.1) Physical culture and 59.2 10.0 66.0 186.5 262.4 sports ........... (0.9) (0.1) (0,3) (0_5) (0.5) State allowances for 74.2 1,726.4 2,715.2 42.32 7,810.0 children ---------- (1.2) (13.0) (13.0) (12.4) (15.1) 1,813.8 3,539.9 5,195.011,263.0 18,0790 State soc1a11nsurance . . (27.7) (26.7) (24.8) (33.1) (35.1) Source: Annual Statistics of Socialist Republic ofRomania, 1970-74. 318 Table 3 STATE BUDGET STRUCTURE (million Lei) 1955 1960 1965 1970 1975 1 2 3 4 5 6 TOTAL 42,9157 55,422.5 93,057.3 130,900.2 242,822.2 EXPENDITURE. - . (100.00) (100.00) (100.00) (100.00) (100.00) Budgeting of national 26,460.4 35,658.6 63,494.6 81,019.0 155,902.6 economy ------- (61.7) (64.3) (68.2) (61.9) (64.2) Budgeting of social— 6,539.9 13,2457 20,9250 34,0190 51,6658 cultural activities . . (15,2) (239) (22.5) (25,0) (21.3) 4,226.7 3,392.1 4,734.9 7,066.8 9,723.0 State defense ..... (9.8) (6.1) (5.1) (5.4) (4.0) 1,391.1 1,734.8 2,285.2 2,457.9 2,791.3 State administration . (3.2) (3.1) (2.5) (1.9) (1.2) Source: Annual Statistics ofSocialist Republic ofRomam'a, 1970-1974. Table 4 CAPITAL COSTS (millions Lei, 1963 prices) 1965 1970 1966-1970 1971 1972 1973 l 2 3 4 5 6 7 Public health, social care and physical culture ........ 519 1,143 4,100 777 906 846 Science and ' scientific services . 190 438 1,978 578 539 584 Education, culture, and art ........ 834 1,591 6,571 1,517 1,827 1,940 Communal economy and dwelling . . . . 4,508 5,316 25,615 5,229 6,230 6,700 —Of which dwelling accounts for ..... 3,151 2,994 16,642 2,578 2,963 3,741 Source: Annual Statistics of Socialist Republic of Romania, 1970-74. 319 population, from the funds of mass and cooperative organization, and from the funds of agricultural collective farms. These investment funds were used for building (452 million lei), and for equipment and instruments (242 million lei). In the past 4 years (1971-1974), 25 hospitals with about 17,500 beds and 24 outpatient clinics have been built. In 1974 there were 191,910 beds for medical care (that means 9.1 beds for 1,000 inhabitants), and 50,000 beds for prophylactic care (2.3 beds per 1,000 inhabitants). In the period 1976-1980 another 45 hospitals with 23,400 beds and 41 outpatient clinics are to be built, and this number may be increased by the end of the 5-year plan. The investment policy aimed, in the beginning, at building outpatient units and then hospitals. At present in each district capital there is a hospital at which—because it has the equipment and a highly-skilled staff—the entire medical care of the district is coordinated. Some irregularities that existed in the medical units distribution network have now been eliminated. In the future, newly-built hospitals will complete or replace old ones. It has been estimated that it is more expensive to renew old hospital buildings and to adjust them to modern requirements than it is to build new hospitals. Older hospital buildings will be used as social-care units. Cost studies (made starting from 1964) have shown variations (according to the importance of the hospital) both in expenditures for a physical bed and for an effectively-occupied bed. Based on these studies the optimal number of beds for a district hospital is considered to be 700 (Table 5). These hospitals, of modular structure, permit the dimensioning of departments according to need. At the same time, hospitals may now be combined with outpatient clinics, and this allows better utilization of investigation services, of the specialist teams, and other facilities. To decrease the cost of investments typified projects have been used in order to obtain hospitals of 240, 450, 700 and more than 1,200 beds. Current cost Although in Romania the costs of medical-care services are not recovered from beneficiaries, it is current practice to analyze the costs in order to aid in decision-making and to ensure optimal use of the available funds. First, it is important to compare expenditures in medical outpatient practice with expenditures in hospitals. It is generally accepted that of the total health funds 50-65 percent are used in hospital care, and this is in relative contradiction to the reduced volume of services, but is justified by the specific activity and the nature of expenditures. Thus, each year only 4 million persons are treated in bed-units, being hospitalized for a total of 44 million days, while outpatient care amounts to 150 million appointments and treatments, for an allotment of 30 percent. In the cost-structure analysis it will be noticed that a large share is represented by staff remuneration (about 58 percent); drugs (about 25 percent); and some 12.4 percent of hospitalization costs represents the provision of food. Other cost elements represent housing expenditures, services and repairs. The investments do not appear in this structure, but are in another budget. 320 Table 5 STUDY CARRIED OUT ON 111 HOSPITALS IN 1964 . Effectivel Hospital beds Phys‘cal bed occupied beym (Computing units) (Computing units) 1 2 3 30 - 6O ................. 1.4 1.9 61 - 100 ................. 1.6 1.9 101 - 200 ................. 2.2 2.5 201 - 300 ................. 1.8 2.2 301 - 400 ................. 1.7 1.9 401 - 500 ................. 2.0 2.2 501 - 750 ................. 1.6 1.9 Over 750 ............ _.> 2.0 2.2 TOTAL .......... 1.9 2.2 Source: Ministry of Health of Romania. Hospital Urban Ambulatory Units Rural Ambulatory Units tb. Dispensaries, Sanataria, Preventoria Emergency Stations (ambulances and planes) Other Figure 14. Cost structure by unit types, percentage (From: Aspects of Health Economics). Regarding the analysis of medical-care cost structure, the following items are of importance: Payment (remuneration) for medical staff work is made according to the law, and is the same for all medical units in the country. Remuneration is differentiated according to the quantity and quality of the work carried out,- taking into account the following: Skill level and title; 321 complexity and difficulty of the speciality; specific work circumstances; years of service results; function; and etcetera.’ These differentiating criteria in the payment of medical staff have been in experimental use since 1970. At that time a newly-introduced system awarded an increase of remuneration averaging 18.4 percent. A new increase of about 20 percent in medical staff pay will be made during this year. About 5 percent of the physicians also have university duties, and these are paid from the educational budget. At national level the relationship between the net maximal and minimal remuneration is about 6:1; in the medical field it is only 35:1. The food allotment quantum is also established by law, being differentiated in accordance with a department speciality profile starting from the part food plays in patients’ treatment. In the years 1970 and 1973 two increases, each of 4 percent, were made in the allotment quantum. In the matter of drug costs (in addition to a trend towards increased consumption) studies conducted in hospitals (university clinics, district and town hospitals), and comparatively by medical specialties, pointed out great variations between one unit and another in all the indexes studied (Table 6)—cost per bed; cost per patient; cost per hospitalization day. Drawing a parallel between costs and the efficiency index (average hospitalization period, the use index, and so forth) their correlation has not been emphasized, although theoretically we should have expected that hospitals with a high-use index should also have increased “bed drugs” costs. Similarly, hospitals with a long average hospitalization period should have high HOSHTAL AMBULATORY Wages Food .0 oo.’.9 w c? O. 0.:0 ‘ O .9 O O 0.0 o o o. ’9 "m" 9 £03.33. .9 co 9" Drugs 9 o 020:9} 0.9 9099909 9 3 .030 I « O O «‘9‘. 0.0% o. ~c§ 0 K0. Administrative Expenditu res Services, Repairs Other Figure 15. Current cost structure for hospital and ambulatory units, percentage (From: Aspects of Health Economics). 322 EZE Table 6 DRUG COST INDEXES BY HOSPITAL AND DEPARTMENT TYPES, 1973 (in conventional currency units—Lei) Internal Surgery Obstetrics Pediatrics Contagious Diseases per per per per per per per per per per per per per per per . bed patient day bed patient day bed patient day bed patient day bed patient day . _ max. 134.50 7.11 0.47 303.75 15.14 0.93 100.00 2.34 0.33 129.10 6.02 0.40 101.19 4.15 0.33 University CllnlCS min. 28.86 1.65 0.10 28.33 1.05 0.10 40.98 1.18 0.16 18.68 2.01 0.21 53.85 1.49 0.17 District capital max. 70.65 2.26 0.24 87.88 2.83 0.34 49.09 1.20 0.15 94.25 4.92 0.33 57.95 2.29 0.20 cities ------- min. 41.40 1.85 0.17 56.63 2.59 0.19 34.52 0.95 0.14 41.75 2.30 0.15 46.55 1.92 0.17 T max. 157.30 4.92 0.53 207.32 4.11 0.43 88.72 2.26 0.35 104.97 4.23 0.39 114.95 2.96 0.23 own ........ min. 23.25 0.90 0.08 33.15 0.99 0.12 16.42 0.44 0.07 17.34 0.88 0.06 23.61 0.93 0.07 R 1 max. 98.42 2.58 0.29 103.29 3.91 0.36 82.59 1.58 0.24 94.17 3.86 0.38 61.55 2.67 0.20 ura ........ min. 17.30 0.81 0.07 23.20 0.73 0.09 14.53 0.52 0.06 12.33 0.36 0.05 32.60 1.17 0.13 Source: Ministry of Health of Romania. expenditures in their “drugs/ patient” index, and hospitals affiliated with university clinics and district hospitals—since they deal with more important cases—should show higher costs in their “drug/hospitalization day” index. The above-mentioned studies permitted some detailed analysis in the units that indicated significant deviations from average value. Aspects of medical occupational activity, organization and management were analysed. The Medical Computing and Statistical Center of the Ministry of Health has annually studied these costs, since 1968. The Institute of Hygiene and Public Health (as well as the District Health Direction) is interested and concerned with these studies. The noticed differences are considered to be the effect of: (a) Discrepancies between the skill and the complexity of medical teams; (b) lack of interest in problems of health economy still shown by some physicians; (c) discrepancies between requests on the part of patients (in the objective context of ignoring the real needs) and the lack of some quantitative indexes of the health status; and (d) unequal development of investigating services. These differences provide information on opportunities of improving the relationships in the hospital capacity-equipment-competence area of service. Going deeper into the question of costs per cases (evaluated per hospitaliza- tion day and average hospitalization period, in current pathology) some maximum values are pointed out for cases of: burns; craniocerebral trauma; children’s bronchopneumonia; operated gastroduodenal ulcer; abnormal births requiring surgical therapy; rheumatic fever; and cholecystolithiasis. High costs are due to the use of increased quantities of antibiotics, cortico- therapy, laborious anesthesia, and so on. In comparing the costs of some cases of chronic bronchitis, ischemic heart diseases, and chronic kidney infections treated with new and old drugs, a net cost rise was evidenced in connection with introduction of new therapy (Table 7). Although in the cases analyzed efficiency indexes were not improved, the results were better from the point of view of: (a) The clinical situation at the end of hospitalization; (b) patient’s satisfaction; (c) more rapid integration into normal life; and (d) the reducing of relapses. The organization of several programs for priority health problems will facilitate implementation of the necessary information system for cost-benefit analysis. (For example, in Romania before 1948 about 200,000 new cases of malaria were registered yearly. The malaria-eradicating program cost 253 million lei but it brought tenfold benefits.) 3. Changes in Health-Care Policy Against a background of firm organizational principles, Romania’s health- care system was subjected to change as part of the system’s permanent adjustment to new social-economic circumstances. In examining the investment strategy we can see that during the early period the building of polyclinics was a priority. These provided the material basis necessary for specialized outpatient medical care, and hospitals were built later. 324 SZE Table 7 COMPARATIVE COSTS DEPENDING ON A NEW THERAPY (Lei) Classical Drugs Newly-Introduced Drugs 8 s g» s s 8 g» s 3 '5‘ '5 3’0 1.. 3:? 0.3. .5 '5 g", H '3 n- S S = “>’ S a S S s “>’ 2 ga 0% 63% ua ”a 83% 0% 3:3 353% 8a gag zaaa 3E5 §§5 Eééfi 3E5 §§s gags o < “ o ” a < "‘ u “ Chronic bronchitis ...................... 17.10 19.9 340.10 30.70 22.3 684.45 Ischemic heart disease .................... 2.60 18.6 48.15 19.10 28.1 535.50 Chronic kidney infection .................. 10.05 24.7 248.90 12.90 29.0 372.50 Source: En‘échescu, Trends in Health Expenditure, 1974 A further and recent stage was characterized by the building of hospitals combined with polyclinics, in district capitals, in order to obtain more efficient use of medical staffs and materials. Simultaneously, in university centers there were built specialized hospitals (for phonoaudiology, neurosurgery and so on). For the next period, investments plans will aim at the building of hospitals with 240-450 beds in smaller towns, to replace existing units. The hierarchy of hospital units appeared as the natural result of efforts to increase the efficiency of medical services and of the principle of providing the best type of specialized medical care where needed. The network of medical units was adjusted to requirements by diversifica- tion of the types of units. At the same time, health centers were instituted by the unification of two or more dispensaries; the polyclinic got closer to the hospital as it began to use the same investigation services and staff scheme. The organization of polyclinics for consultations paid for by patients did not mean an abdication from the principle of free and comprehensive medical care, but indicated a supplementary way of meeting some of the need of the population. . An important change in the organization and work of health units was their administrative subordination to the local Popular Councils, thus involving the population directly in the solving of health problems. In an initial phase, District Health Directions were administrative depart- ments within the District Popular Councils. Their present status with leading responsibility for the district hospital allows for the better handling of medical-care problems. Thus the District Health Direction has become a strong technical body with special management tasks. The change of the remuneration system in 1970 completed the occupational promotion system as it involved physicians in permanent training. Among the recent changes is the inclusion of members of collective farms in the category of those receiving free outpatient care and free hospital care and treatment. Another change was the imposition of conditions on the selling of some drugs (only by medical prescription) to control self-medication. Meanwhile, a pharmacovigilance system was set up, with all health units being involved. The foundation in 1969 of the Academy of Medical Sciences as a scientific forum contributing to the development of medical sciences and to the promotion of health care, allowed for the carrying-out of valuable studies for the organization of medical care, for substantiation of Ministry of Health decisions, and so forth. The completed forecast studies represent the basis for longterm programs and for planned annual revision of technical-organizational methods. Particularly important has been the integration of teaching with research, and medical care activity in the units within university centers is also most beneficial. All these changes had direct implications on cost—implications reflected in improvement of the efficiency index of the units. 326 LZS * Felt HEALTH NEEDS Lb- Unfelt Expressed Requirements Satisi‘ied Unsatisfied Medical Consumption Potential Unexpressed Requirements Known Satisfied Consumption -_------—--—---J Medical Unsatisfied Consumption Potential Unknown Consumption Figure 16. Health needs and requirements (From: Enichescu: Trends in Health Expenditures, 1974). 1 Cardiovascular Diseases 2 Respiratory Diseases 3 Cancer and Other Tumors 4 Digestive Diseases 5 Endocrine and Nutritional Diseases 6 L Neurological Diseases 7 Mental impairments j 8 Rheumatism 9 < Other Diseases and Symptoms __l Figure 17. Diseases proposed to become the object of prevalence screening (From: Pro- fessor Dan Enichescu: Pilot Study of Public Health Problems in Romania). 4. Present Trends In Health Costs Morbidity represents one of the main factors determining the volume and the type of medical activities, and thus directly influences the cost of health. If we can estimate the volume of future allotments for health depending on planned development of the economy, then we must as much as possible know the medical needs of the population in order to establish some priorities in meeting the requirements. Although it is clear that there cannot be equality between registered consumption and medical needs, (Figure 16), in Romania special activities were organized in an effort to determine the health status. Thus between 1959 and 1961 a medical social inquiry was carried out on a sample of 5 percent of the population. The inquiry was repeated in 1964, and on the basis of the results 328 of these inquiries a longitudinal health status study was organized relating to 200,000 inhabitants. The Public Health Chair in the Faculty of Medicine at Bucharest organized during the same period similar studies on two samples of 9,000 persons each. At the request of the Ministry of Health, the Institute of Hygiene and Public Health and the Chair of Public Health have now designed and tested a methodology for multiphase screening using standardized examinations. The testing of this methodology was carried out on a group of 11,000 persons, with future plans to study about 100,000 adult persons, within a representative sample. The main chronic diseases were the goal of these examinations (Figure 17). Valuable results (Figures 18 and 19) were thus obtained as regards the prevalence by age, sex, occupation, civil status, and etcetera. These data and the data in the statistical report system allowed for the carrying out of some estimates of the consumption of specialized, general practice, stomatology services, hospitalizations, number of hospitalization days and laboratory tests—through 1977 (Figures 20-27). .,. Total n u. 'U C c 8 Iv g m m a Male u, u w a a a c ._ é o B Female § E. _ -— u) ‘2 E ‘3 w :5 g . a U ‘7; 5 ° a 2 111.17 S G) E E — .2 ,. {U E § 3 a) seal g .2 n: o E 3 a 8 to g .3 .9 D 1g '5 a .2 ‘5’ 0 E5 .: .9 8 a 53.69 E 2‘ E o 43.1: . E g ,. T» In .93 3' n: .2 a w k m w :01 a 9 3 9 D W. .3, 26.01 a”; to, Z O ;.:« O "a '0 v .0 . 3r 3 '5: Cancer and Other Tumors 9’0 . . v 20%. 62323253232326“ .xzzzzzzzszaz N a .3 u ': en E S" a :1 a a P a .N .. O ,6: p .9. Figure 18. Screening results, urban. 329 600 —- 500 400 300 Diseases per 100 Persons 200 100 011111111111! 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 69 overfi Age Figure 19. Prevalence by age, diseases per 100 persons (From: Professor Dan Enachescu: Pilot Study of Public Health Problems in Romania). The information obtained will be used to establish priority directions and to substantiate health programs, and these will represent the basis for the plans of unit development and equipment, and staff training. Within these estimates it is believed that the investment in medical services will increase considerably, particularly in the area of equipment. Because of the use of new pharmaceutical products the cost of treatment will also increase. The use of special technologies in the treatment of some diseases will call for the organization of superspecialized hospitals at interdistrict or national level. Experience in this direction shows that even if the cost in these hospitals is higher, it is justified by better results. But the organization of these units is a problem to be discussed in relation to established priorities. On the occasion of the testing of the new screening methodology, the costs of some investigation variants were calculated. These data are now being used 330 as variables in an attempt to formulate a mathematical pattern of the health-care system. Interdisciplinary teams are studying the organization and cost problem—-by means of various methods of operations research and systems analysis. In order to involve physicians in the solving of health-economy problems, various forms of postgraduate training are being organized. Similarly, medical students are also being made aware of health-economy needs and notions. Of particular interest in the area of health planning was the organization in Romania, in 1971, of an international course on planning, in cooperation with the World Health Organization. A second course to be held in September 1975 is in process of preparation at this time. 240 Forecast for 1977 231%(169,240) ’2’ 220 — SE =8.09% /’ y =a+ bxt / y =94.7618+ 7.8071xt / — t =1 . . . . 18 ’ Hypothetical Straight Line 180— 32 _ Empirical Curve 140— _ / 100— fillllllllIlllllll? 1960 1965 1970 1975 1977 Figure 20. Medical examinations and treatments, total (From: Enichescu: fiends in Health Expenditures, 1974). 331 260 Forecast for 1977 241% (123,783) =8.61% / a+bxt / 94.7012+8.1718xt / 1.. . . . 18 / ”<‘< > fi—Hypothetlcal Straight Lune 160 — Empirical Curve 140 — 120 — / 100 — filllllllllllllllll? 1960 1965 1970 1975 1977 Figure 23. Medical examinations and treatments, total by specialist physicians. 334 42° ’ Y = 20.71% 431% (310,076) / y = a + b x t y =80J615+183917xt ,/ 380-— t-1....18 // 3w— // 300 — 260-— ” '_ Hypothetical / <—Emplr1ca| Curve Straight Line 7 220 180 140 10 / 1960 Forecast for 1977 Illllllllllllll 1965 1970 1975 Figure 24. Stomatological examinations and treatments. . 7 1977 335 220 200 — if = 4.31% Forecast for 1977 y = a+bxt 168% y = 92.0939+4.2621xt 523 061 t = 1....18 ' 180 - _ / 160 // % Hypothetical 140 Straight Line 120 IIllIIJIIlIIJIIli I 1960 1965 1970 1975 1977 Figure 25. Hospitalization days. 336 % 260 Forecast for 1977 242% (585,300) ’ 24° — Y = 6.01 , y = ax bt . I y = 89.69x14051t 220_ t=1....18 / / l 200— / / / 180— / 160— 140— Hypothetical —./ 120 _ Curve Empirical Curve 100 —/ ®L111I1L11|111141 1960 1965 1970 1975 1977 Figure 26. Hospitalizations. 337 800 700 ~— 600 — 500 — % 400 — 300 — 200 — 10 0/ 07. 1 960 Forecast for 1977 761% (120,975) 1 / 35:25.3696 I y=axbt / y =85.89X1.128t / t =1 ...18 / / / / / / / Empirical Curve Illllllllllllllll? 1965 1970 1975 1977 Figure 27. Laboratory tests (From: Enichescu: Trends in Health Expenditures, 1974). 338 Chapter 12 COMPARISON OF HEALTH-CARE COSTS AND EXPENDITURES FOR WESTERN EUROPEAN AND NORTH AMERICAN COUNTRIES: REPORT ON PAPERS AND DISCUSSIONS AT CONFERENCE HERBERT E. KLARMAN New York University I am somewhat embarrassed that I do not have a new book to advertise. If you come to believe by the end of this paper that maybe Ihave just started one, I shall not resist the conclusion. . I tried hard not to write the paper before the conference, and am glad to report success. In actuality I prepared this report last night, with all the papers in hand and the discussions out of the way. If a paper seems to get undue attention, it is likely to be one that reached me early. That is only fair. This summary of the papers and discussions has three purposes. One is to try to convey what was said at the conference. The second one is perhaps not nearly so important: it is to augment the record, wherever I believe it to be useful. The third one is to take advantage of my position on the program to challenge emerging consensus, when it strikes me as perhaps premature. Throughout I shall try hard to distinguish between report of what others wrote or said here and my own comments and observations. This paper consists of five sections. Three are taken without apology from Professor Zubkoff’s suggestions yesterday afternoon. He wanted to know what has happened to health-care expenditures and t0 the proportion of the Gross National Product (GNP) that health care absorbs. Second, he was interested in the factors involved. One can think of this question in terms of accounting for the increases and, more difficult, also in terms of explaining them. Third, he wanted to know about the control efforts that have been tried. These three sections requested by Zubkoff are surrounded by an introduc- tory section that depicts the background for the conference, as I see it. The paper is concluded by a section on implications and looking ahead, with the usual suggestions for further research. 1. REASONS FOR CONCERN In the United States we have witnessed an unusually high rate of increase in health-care expenditures, particularly after 1965 when Medicare and Medicaid were enacted. This was accompanied by a continuing rise in the percentage of health-care expenditures to the GNP. 339 A second factor in our concern over health-care expenditures is the increased expenditure of funds from the federal budget. Here I would add that perhaps there is a feeling of responsibility that the increased federal funds may have exerted some influence on what has happened to health-care unit costs and prices. Third, whatever other people may be interested in, it has always been the concern of economists, certainly economists in the health field, to ask, “Are we producing whatever services are being rendered as economically or at as low a cost as possible?” Economists always inquire into such matters as economies of scale, possible and actual substitutions among inputs, and so forth. A fourth factor, I believe, is that more recently, particularly under the influence of Professor Fuchs (1, 2) and certainly under the influence of Professor A. L. Cochrane from Britain,(3/ we have begun to ask the further question, Are we getting value for our money? If we are spending a lot more money, what do we have to show for it? Those of us who were at Geneva in the fall of 1974 may recall that that conference dealt with the health-care cost explosion and how to contain it. The concern was over the explosion in health-care costs. Since then something has happened which seems to relieve the element of crisis. More and more of us have become aware of the fact that at least in the United States the percentage of health-care expenditures to the GNP has been stable for 4 years, at 7.7 percent. (4) In my paper for the Geneva conference I found only 3 countries among the 10 for which I had data that had managed to achieve some degree of stability for so long a period. (5, pp. 10-11) One is the well-known case of the United Kingdom. The second country should not surprise one, Israel, with so large a burden of defense spending. The third country is Japan. What happened in Japan was a high rate of inflation in the economy as a whole. Medical-care expenditures were rising at about the same high rate, yielding a constant ratio to the GNP. Without claiming to understand Japan’s experience, I have gone out on a limb and adopted it as a leading indicator for the period of high inflation in the United States. This decision was not so brave as it may appear. A few years ago Sam Shapiro and I attended a meeting on infant mortality, at which he claimed to be the only expert to have predicted the decline in infant mortality in the United States beginning in 1966, after a decade or so of flatness. Naturally he was asked what made him so good a prophet. He answered that he always looks at the Swedish data; in his experience anything that happens there will happen here after a lag of severalyears. You will recall that Professor Evans employs Saskatchewan for the same purpose in Canada. I do not pretend to understand this interval of stability in this country. I gather from Mme. Sandier (Chapter 6) that there has been a decline in the rate of increase in health-care expenditures in France in 1974. Pointing in the other direction, Professor Evans reports that in Ontario expenditures started to take off again in December 1973. 340 2. TRENDS IN EXPENDITURES What can we say about trends in expenditures, as reported at this conference? In the paper by Messrs. Sandier and Rosch (Chapter 6), which compares France and the United States, it is reported that the rise in expenditures in France was at a higher rate than in the United States, so that the gap in per capita personal health-care expenditures between the two countries narrowed. In West Germany in the 1960’s, Professor Reinhardt reports (Chapter 10), health-care expenditures were rising at the rate of 9.7 percent a year; in the more recent period, 1970-73, the annual rate of increase was 13.2 percent. In Sweden, Professor Stahl reports (Chapter 8), the annual rate of increase in current prices is 14.2 percent. If allowance is made for inflation in the economy as a whole, the annual rate of increase is 5.5 percent. It is fair to say that, as ProfeSSOr, and soon to be Dean, Ruderrnan expressed it during the discussion, the trend toward rising health-care expenditures is universal. What has happened to health care expenditures as a percentage of GNP? Comparable figures for all countries for a single recent year are lacking. My notes show the most recent figure for Canada, as of 1971, is 7.1 percent (Evans); for Sweden, the most recent figure, as of 1972-73, is 8.1 percent (Stahl); for West Germany, as of 1968, 6.1 percent (Reinhardt); for the United States, as of 1973 or 1974, 7.7 percent (Rice); for the Netherlands, as of 1972, 7.3 percent (Maxwell); for France, as of 1973, 6.3 percent (Maxwell); and for the United Kingdom, as of 1973, 5.3 percent (Maxwell). Mr. Maxwell has drawn this generalization. He states that in the decade of the 1950’s the percentage of GNP devoted to health care rose by 1.0 percentage points; in the decade of the 1960’s, by 1.5 percentage points; and in the 1970’s he expects it to rise by 2 percentage points. In light of the stability in the percentage of healthcare expenditures in the United States in the period 1971-74, which I cannot really explain, I have little confidence in any projection for the remainder of this decade. Subsequently, I shall discuss several factors that affect total expenditures, some pointing upward and some downward. Mr. Maxwell observes two common, principal features of the health-care industry. One is its labor intensive nature, with salaries and wages at 70 percent of total expenditures. The second feature is that hospital expenditures usually exceed one-half of total expenditures. My impression from the papers we have heard here is that the second generalization may not hold today; perhaps it applied in the past. As for the first generalization, all of us assert it;however, evidence is scanty and sketchy. (I, p. 333) 3. FACTORS INVOLVED What are the factors involved in the increase in health-care expenditures?, Professor Zubkoff asked. To begin with, it is important to distinguish, as Professor Fuchs did in his work on physicians (6) and Professor Martin 341 Feldstein did in his work on hospitalsl7), between an accounting framework in which the data are aligned and classified as an explanatory model of behavior. It is fair to say, without derogation (given my own work in this area), that the papers at this conference present much more of an accounting framework than efforts at explanation. This is not surprising, since, as Professor Anderson has remarked, we are predominantly still in a descriptive phase. We have heard some attempts to explain. These take the form of hypotheses at this time, and are discussed below. Accounting Framework Bearing in mind this distinction between an accounting framework and attempts at explanation, neither should be applied directly to expenditure figures for the health—care sector as a whole. In fact, a considerable amount of detail has been accumulated at this conference on total expenditures distributed by object, such as hospitals, physicians, drugs, and so forth. In particular, data have been presented in systematic fashion on hospital expenditures and on physician expenditures, which permit a calculation of sources of increase. An apt example is the accounting for the increase in hospital expenditures by Mme. Sandier. What does she report? The increase in population has been a small factor. The aging of the population is a negligible factor. Per capita utilization is a more important factor in France than in the United States, but still less important than the change in the prices of inputs. The last is the most important factor. There is still another factor which is almost as important, according to these calculations, namely, the increase in the quantity of resource inputs, which have to do in part with the changing nature of the hospital product. Here I take the liberty of introducing my own comments on these findings. Obviously, I am not familiar with the data underlying the calculations in France. In the United States, I know that the prices of inputs factor is measured by the average annual wage of employees. Given the fact that over the years the quality of inputs in hospitals has tended to decline (the composition of employees has shifted toward the less skilled), the change in average wages for all employees understates the increase in wages obtained by any individual category of employees. If this is true in France, as I believe it to be true here from Martin Feldstein’s data and in Canada from Professor Evans’ observations at this conference, we may be witnessing a wider phenomenon, namely a strong capability on the part of health-care providers to make hay financially. In addition, there is the technical point that the change in the quantity of inputs is overstated, if the change in wages is understated. If these comments are apt, perhaps the extent of changes in technology have not been so great during this period as is widely reported. (Let me note, too, that the increase in wages may be understated—or overstated—for still another reason. If parttime employees now work fewer—or more—hours than formerly, the basis for calculating full-time equivalents should be changed.) Proceeding from the labor component to the non-labor components, it strikes me as inappropriate to take all non-labor expenses and divide them by a 342 price index. Too many heterogeneous items are involved, as shown by the following examples from the United States experience. There is good reason to believe that in the middle 1960’s, appreciable amounts of additional depreciation drifted into the non-labor component, not because anything had changed in the use of capital but because more hospitals began to account for it. We have seen a shift toward more disposables in hospitals; this signifies a decline in labor expenses and a simultaneous increase in the value of purchased products. Similarly, a shift has taken place toward the purchase of laundry service. I suggest that all of these items should be examined separately before they can be aggregated in a meaningful way. Explanatory Hypotheses As for the efforts made here at explanation, they are specific and therefore disparate. My notes show that Professor Cooper’s explanation of the relatively stable ratio of health-care expenditures to the GNP in the United Kingdom is two-fold. One is that services have been rationed by the availability of staff and facilities. The second is that as the single source of financing, the state has exercised its monopsony power to hold down the incomes of health-care providers. Professor Ruderman observed that it is also important to recognize that the United Kingdom has become a poor country, especially in comparison with Canada. It struck me when Professor Ruderman spoke—and I believe that the point was made later in the discussion—that perhaps what is important is not whether a country is rich or poor but rather what has happened to its rate of economic growth during the interval in question. With a high rate of economic growth, it is feasible to increase expenditures for health care without taking anything away from other sectors of the economy. In Sweden, Professor Stahl believes that a monopsonistic situation may exist. Even so, expenditures mounted because the supply of physicians was restricted. The latter permits the exercise of monopoly power. When monopoly and monopsony confront one another, the outcome is of course highly uncertain, depending on relative power and on the preferences of the two parties. Professor Anderson attributes the rise in expenditures in Sweden to the fact that the county councils have but this single mission. In consequence, the councils were actually competing with one another to determine who would have the best program or who would have the most costly program. He adds that in the expanding Swedish economy it was possible to implement a generous concept of what health care ought to be. Dr. Prirns states that the major explanation of the rise in health-care expenditures is the relatively large size of the personal-service component. To me the labor-intensive nature of health care has always implied that the health-care sector is bound to lag in productivity gains in a growing and increasingly productive economy. However, Dr. Prirns also had in mind the shift within the hospital from a staff with large numbers of Sisters of Catholic orders to a staff of lay employees who expect to earn a living. As Professor Fuchs observed, such a shift does not entail an increase in real resource cost. 343 These explanations appear plausible to me, despite their variety. They deserve to be treated as interesting hypotheses to be tested, which can keep many of us busy for a long time. In the paper on health-care expenditures in the United States, Mrs. Rice discusses the several explanations that have been offered for the rise in hospital expenditures in this country. In preparation, we are told, she drew heavily on an excellent review article by Karen Davis, (8) and I subscribe to that laudatory judgment. Ultimately, Dr. Davis reduces the alternative explanations to demand-pull, technology, and third-party reimbursement. Although Professor Fuch has not worked directly on the hospital problem, his work on physicians puts strong emphasis on the notion that it was medical technology that explains the behavior of expenditures. Here it suffices to note that Professor Fuchs did not go into the post-Medicare period. I know from past conversations with him that he has taken an agnostic position on the factors that were operative after 1965. In his work on hospital expenditures Martin Feldstein incorporates the technological explanation into the demand-pull model. The demand-pull explanation holds basically that a major effect of health insurance is a dual-price system, with the patient facing close to a zero price at the time of illness. Professor Feldstein then observes that in hospitals we have witnessed the proliferation of technology, because nobody cared how much it cost when it looked so cheap. Professor Evans has told us here, and elsewhere (9), that he has seen no evidence for the operation of the demand-pull theory in Canada. Only a small increase in utilization was associated with insurance. When hospital costs rose, the dominant factor was an increase in wages. Some of the conference participants will recall my predilection for the cost reimbursement explanation. From the outset it struck me as plausible, because that is the way hospital administrators talked and acted beginning in the summer of 1966. One of the implications of cost reimbursement on a strictly retroactive basis is a likely drop in the rate of occupancy. In turn, this calls for a modification of Roemer’s Law, of which I am also a strong proponent. In the United States we have witnessed a decline in the rate of occupancy in the 1970’s. Notwithstanding, it is only fair to note that the econometricians who have searched for the cost-reimbursement factor have not been able to find it. The only explanation I can offer is that all hospitals moved to cost reimbursement on a wholesale scale at the same time, thereby precluding the variation required for multivariate statistical analysis. There are those who would ask, as a practical matter, Why is it important to distinguish among the alternative explanations? It seems to me important, because if cost reimbursement is a significant factor, it can be just as readily be abandoned or modified as adopted. It has the virtue of reversibility. Moreover, if reimbursement is important today, it will be even more important under any program of national health insurance. It is practicable to take action on reimbursement without doing anything else. There is no need to make changes in the delivery system, nor is it necessary to agree on the kinds of medical care organization sought. 344 .. . My impression is that Professor Evans would question such an approach given his emphasis on ownership. I surmise that Professor Fein might go along with the approach, only because he deems it impractical to postpone enactment of national health insurance until reform of the delivery system is achieved. Let me turn now to the relatively small size of the hospital sector both in France and West Germany, neither of which has been explained at this conference. The smallness of hospital expenditures in France compared with those in the United States is further accentuated when a simple adjustment is made. The French data include physicians working in hospitals. In the United States data physicians who work on salary are included, but physicians who treat private patients in the hospital and charge separate fees are not. Professor Ronald Andersen was kind enough to get some unpublished figures for me from his 1970 nationwide survey. That survey shows that approximately 42 percent of expenditures for physician services in this country are devoted to inpatient care. For fiscal year 1974 that amounts to a figure of the order of $8 billion, yielding an increase of approximately 8 percentage points in the hospital expenditure figure for the United States. If this item were added to hospital expenditures, then the difference between France and the United States in the relative role of institutional expenditures—hospitals and nursing homes—widens from 11 percentage points to 19 (Chapter 6, Table 5). Conceivably the difference might be still larger if allowance were also made for payments to private duty nurses, which are not included in hospital expenditures in the United States; however, a firm basis for making such an adjustment is lacking. I trust that this exercise in adjustments is consistent with Dr. Altman’s admonition to this conference not to be too precise with data. I believe that the discussion has benefited from his advice, while at the same time reaping the advantage of its reaching the participants so late that our papers did indeed contain a good many precise details. More than offsetting the relatively smaller hospital expenditures in France and West Germany are the much larger expenditures for drugs in the two countries, compared with the United States. Mme. Sandier (Chapter 6) thought that 80 percent reimbursement for drugs under health insurance may play a part. Professor Ruderman pointed to the prevalence of physician home visits in France and to the large number of ambulatory visits by physicians—10 per person per year—in Germany. I was astonished by these data and believe that attempts at explanation have scarcely begun. We have heard here discussion of the reasons for the increasing role of public financing of health-care expenditures. Professor Fuchs speculated broadly on the bases of modern egalitarianism, including the capacity of those who are worse off to make life miserable for those who are better off. With respect to health care there may be a vestige of the time when a good deal of illness was communicable. More important, perhaps, is the fact that equality in medical care does not alter any fundamental class or power relationships in society. Other attempts at explanation were more specific. Thus, health care is regarded as unique in certain respects. A major part of health care is now 345 purchased through third parties. Once that has taken place, what difference does it make whether the plan is private or public? At some point there are reasons for making health insurance compulsory. Then the public sector becomes a serious alternative. Another aspect of the uniqueness of health care, according to Professor Weisbrod, is the difficulty of changing the production process without also changing the product. You may recall Professor Evans’ emphasis of the importance of public ownership. Professor Stahl invoked his majority voting model. Given certain facts about the income distribution, income elasticities, and the availability of a management system, he would predict the increased role of the public sector. Professor Weisbrod’s attempt at explanation involves the distinction between goods for which the demand is homogeneous and those for which the demand is heterogeneous. The former are a good candidate for the political process. In this context, let me offer a small addendum to the United States data. In this country 40 percent of the total health-care expenditures fall under the public heading. Actually, that is an overstatement of 2 percentage points, owing to the treatment of workmen’s compensation and the premiums paid by enrollees under the voluntary part of Medicare. However, there are offsets that are even larger. 1 refer to the values of the tax deductions under the individual and corporation income. taxes. The combined figure is $5 billion, or approximately 5 percentage points. I am not assessing now which single figure is most valid for official publication; it may be 40 percent. But if one wants to inquire into what will happen to sources of payment under alternative health-insurance programs, then such features as “tax expenditures” and private premiums under publicly-mandated programs must be taken into account. 4. EFFORTS AT CONTROL This section attempts to respond to Professor Zubkoff’s third question concerning the control efforts that have been made. It is agreed that the marketplace is losing ground. As Professor Altman observed, it is still necessary to make decisions concerning the allocation of resources to and within health care. The question is, How? Efforts have been made to introduce control measures on the demand side and even more so on the supply side. On the demand side, my list includes only two items: co-payment and rationing by time. On the supply side, my notes show a long list of control measures. These will be presented in their postulated order of increasing rigor and controversy. Co-payment or User Charges We have heard here a long discussion of co-payment or cost sharing. Dr. Prims strongly favored this feature, maintaining that in the long run a medical-care system cannot be maintained entirely free of charge. By contrast, Professor Evans stated that for the most part the Canadian public has not 346 displayed a great interest in the size of premiums or taxes and prefers to receive services without charge. Professor Cooper believes that consumers behave under a tax illusion. It seems to me that there exists less evidence on co-payment than the discussion warrants. Rather, a good deal of value judgment is involved. It is my hope that an explication of the issues will yield a measure of clarification. Prior to this conference I have usually cited three purposes of co-pay- ment. (10) The discussion here has added a fourth. The first reason for deductibles or coinsurance is to deter utilization. The second is to hold down the size of the premium. The third reason is to keep out of the public budget a large item of expenditure that can remain in the private sector in order to preserve the public budget for expenditures that the private sector will not make. The fourth reason, due to Professor Evans, is that providers do not mind if fee limits are not imposed on them. The third reason is largely associated with Charles Schultze of the Brookings Institution. The proposition is that we ought to try to save the public budget for public goods. I am not persuaded that it is necessarily a bad idea to exert pressure on health-care spending, just like on any other item which is competitive with various objects of expenditure. I doubt that anybody here would advocate the fourth reason for co-payment, namely, to raise providers’ incomes. As for holding down the amount of health-insurance premiums, it seems to me that under compulsory insurance, it becomes possible to take full cognizance of whatever health-care expenditures are in fact incurred. Only under voluntary health insurance do the various plans have to be competitive in the marketplace. Voluntary health insurance may have good reasons for imposing co-payment features. I do not see that the same reasons apply to compulsory insurance. One is left with the first reason. The question is whether, and the extent to which, co-payment deters use. If it does, who and what are deterred, and do we approve? Apparently there is some effect on utilization; certainly the study by Anne Scitovsky shows that. (11) However, there is reason to believe that the effect may erode over time, as happened in Saskatchewan. The reduction in utilization does not hit everybody evenly. Indeed, Professor Cooper stated that there is no reason to believe that it hits only trival uses. Professor Somers observed that in the effort to relieve the poor of the burden of co-payment, we are led to incur increased administrative costs. Professor Cooper argued that co-payment may even result in an increased use, because patients feel entitled on the basis of paying a concurrent fee. Elsewhere I have argued that perhaps we are not so serious about deterrence by co-payment, since we do not prohibit the purchase of supplementary insurance to pay for the coinsurance. (10) In a private conversation at this conference Dr. Newhouse suggested that to avoid the purchase of such supplementary insurance it is not necessary to enact a legal prohibition; it would suffice to withdraw the deduction under the tax laws (see above). My rebuttal was—and is—that the aged beneficiaries under Medicare were not getting appreciable tax deduction when they purchased supplementary insurance. Nor is the tax 347 savings uppermost in the minds of union leaders who bargain for first dollar benefits. It is time to stop. It strikes me that pending the findings of further studies the factual base is not conclusive and that a good many judgments on this issue are political in the best sense. They reflect the protagonists’ posture toward the role of the public sector in general and toward the distinguishing characteristics. of health services in particular. Those who believe in the efficacy of prices see no obvious alternative to them. Rationing by Time The other device on the demand side, rationing by time, aims to make it inconvenient for people to use services. How else can one deal with perpetual excess demands at zero price, Professor Cooper asks. I should add that to the extent that demand for a service is subject to chance variation, marginal cost for off-peak utilization is low, a low or zero charge is then appropriate, and waiting time need not be long. The point was made by Professor Holst and Mr. Maxwell that if waiting time—a queue—is used as a rationing device, it is necessary to make sure that not too many people leave the system to seek care on the outside. Recourse to outside use on a large scale can have a deleterious effect on the health-care system as a whole. Offer Preventive Services as Benefits In our discussion, as I think is true elsewhere, increasing recognition is being given to the importance of the supply side for explaining the increase in health-care expenditures and for controlling them. As Professor Evans put it, the supply side is crucial. Those of us who follow the National Health Service in the United Kingdom know that the basis of the Beveridge Plan was to provide many more preventive services. The expectation was that, after an interval, the need for curative services would fall and yield a savings in health-care expenditures. It has not turned out that way; as Professor Culyer pointed out, the expectations that were attached to preventive services were exaggerated. He noted that the costs of persuading people to behave in a manner conducive to good health is high. There is also the question of which personal health services are truly preventive. Particularly in England the effectiveness of screening has been questioned. (12) I should add the importance of paying attention to the false positives in screening and to the costs associated with diagnosing them. It seems to me that with the exception perhaps of most vaccination programs and a few screening programs, the surest way to save money on health care is not to spend it. Substitute Ambulatory Services for Inpatient Services Similar in aim is a policy of increasing the volume of ambulatory services, thereby effecting a reduction in the volume of inpatient services for which they 348 would substitute. According to the evidence presented'here, it does not quite turn out that way. Indeed, according to Professor Evans, in Canada the increase in hospital use that followed the enactment of medical-care insurance exceeded that which followed enactment of hospital insurance. This affirms the findings of many studies reported in the literature. (13/ My conclusion is: an increase in ambulatory services serves to increase ambulatory services; the volume of inpatient services may remain unchanged or it may even rise. It is fair to note, however, that the unit costs of the two types of services are quite disparate. Health Planning . Professor Stahl’s comment on the Swedish experience is skeptical: with all the planning activity going on in Sweden, the expenditures elasticity is still 1.68. Why bother? Apart from the evidence introduced here by Professor Stahl, I do wonder whether health planning in Sweden is not more appearance than reality. When I visited the Scandinavian countries with a group from the Health Services Research Study Section in 1966, my impression was that in the provinces they were not paying much attention to planning guidance from headquarters. The extent of planning, as reported in Stockholm, was vastly exaggerated. However, Professor Stahl has also argued here that with planning and the implementation of plans both done at the local level, expenditures and taxes are tied together. The result is a lower level of expenditures than otherwise. This reminds me of Anthony Downs’ theory of why the budget in a democracy tends to be on the low side. (14) However, let me offer an argument from the other side. When Henry Aaron studied health-care expenditures in a large number of countries, he found that only one factor really explained the size of the increase, namely the presence of an earmarked social security fund. (15) With a single purpose fund, it was far easier to get appropriations and increase expenditures. Regionalization Regionalization, a special application of health planning, was first men- tioned by Dr. Altman as a means of controlling the supply of expensive facilities. For economical operation they must be utilized at or near full capacity. The proliferation of such facilities results in low rates of utilization and high unit cost. It was reported that regionalization is successful in Sweden. The problem is travel distances for some patients. Dr. Prirns stated that regionalization has been achieved in Denmark to a large degree. The degree of achievement in other countries cannot'be reported, since the discussion was not complete. In the United States regionalization has been advocated for many years. In my opinion it has been retarded by the failure to recognize the adverse consequences imposed on specialists who practice both in the community and in hospitals when some of them are denied access to specialized facilities. 349 Curtailing Supply of Hospital Beds The prolonged discussion on the supply of hospital beds astonished me, for I had come to believe that the desirability of curtailing the number of beds was now the conventional wisdom. At this conference the issue was obviously controversial, on two separate grounds. One, asked Professor Yett, “Why do you want to curtail the number of hospital beds? Is it not good policy to respond to a population’s demand?” Two, according to Mme. Sandier and others, “What happens if the number of beds declines and average length of patient stay is reduced? Is not the result an increase in cost?” I should like to deal with each question in turn. The first question pertains to Roemer’s Law. According to a friend, Professor Milton Roemer attributes the designation to me, although he does not dissociate himself from the substance. At any rate, let me recall some history for you. Not long after Roemer published his two key articles on the subject about 1960, (1 6, 17) Dr. Gerald Rosenthal challenged his conclusions, arguing that hospital beds were built in response to demand. (18) Other economists agreed. Practicing hospital consultants, like Ray Brown, insisted, however, that the way it worked was that hospitals were built and then were filled. After a period of observation and review of the evidence pro and con,I concluded that the practical consultants were closer to reality, because they reflected accurately developments in New York City. There was no formal reconciliation, one way or the other, until the British studies reported on wide geographic variation in hospital use in a country without hospital construction for a generation and longer. (19, 20/ In effect, they were in a position to report on a natural experiment. The proposition was that a difference in hospital use among regions of two or three times can exist, yet the rates of occupancy will be about the same. What helped make Roemer’s Law an acceptable proposition in the United States, in my judgment, was that many of the people who might have opposed it were also ardent proponents of prepaid group practice. It was around 1960 that the low hospital use associated with prepaid group practice began to be reported. (21) Later allusions to the hospital as a dangerous place to be in were probably an afterthought. They can also be seen as a reversion to Lister’s “hospitalitis.” In my mind, at least, there is no doubt that limitation and curtailment of the supply of short-term hospital beds is a policy with a firm basis in empirical knowledge. With hospital expenditures as the largest component of health-care expenditures, the possible savings accruing from a reduction in the number of beds are substantial. Whether such a policy is likely to be carried out successfully is a different question. I used to think so, but do not feel as confident today. One factor is the present climate of job scarcity. Another is the persistence of the standard of need. The oldest of the certificate of need laws in this country is in New York State, going back to 1964. My former colleagues at the Hospital and Health Planning Council of Southern New York tell me that in the exercise of their review function, the predisposition is to approve. The may spend much time getting and analyzing data, they may delay, but ultimately they approve. They realize they are likely to wind up in court if they disapprove an application. Since they still believe 350 that they are trying to meet the quantitative standards of bed need, rather than acting to curtail the total bed supply, they feel there is little chance of prevailing in a court suit. A third factor is lack of will, as indicated by this incident. A year and a half ago I was testifying before a Congressional Committee, to whom I made a pitch for curtailing the supply of beds. One Congressman said, “You know, we cannot do this. Why don’t the local planning agencies do it?” I answered, “I suppose they don’t do it for the same reasons that Congress won’t.” The second question deals with length of stay and its effect on cost. There is no question that if average length of stay is reduced, per diem cost will rise. However, the cost per episode of illness must decline and so will total hospital ' expenditures. Although the cost of admission procedures and the cost of discharge procedures are spread over a shorter time and ordinarily there is still only one operation per episode, the hotel services and some of the nursing services are avoided when the stay is shortened. Either Ihave misunderstood the argument or something subtle escapes me. Whether reducing hospital stay by moving the patient into an extended-care facility will yield savings is another matter. The outcome depends on the respective numbers of days in the two types of facility and on the marginal cost of care in each. (I know of no such studies.) It may be, too, that extended care in a facility attached to a hospital costs more than in a separate one. Obviously, this discussion abstracts from the lower cost of time that is associated with a shorter stay. Prospective Reimbursement of Institutions Dr. Altman alluded briefly to still another device for curtailing expendi- tures, prospective reimbursement, “whatever that is.” I had thought that prospective reimbursement was something specific, but apparently it is not. Since the record of the conference discussions is bare and I did not have the opportunity to inquire into the matter with Dr. Altman, Ihave decided not to pursue it further in this summary beyond the mention of “experiments” that are currently under way under the sponsorship of the Social Security Administration. Physician Fee Schedules The discussion displayed a consensus that physicians are usually able to beat a fee schedule. They can act to improve collections. They can trade upward and render more costly services (Professor Evans). I should add, too, the increasing fractionation of services and fees, as displayed in the United States after 1965. They can refuse to accept the assignment of fees, as under Medicare, when the option is given to them. In this context it is appropriate to raise a question that Professor Fuchs has often asked, namely, “How important is it to control physician income, given the fact that they exert so much control over other, more costly items of expenditure?” The implied answer, which was negative, seemed reasonable. 351 However, during this conference I have begun to wonder whether this is still the right answer, given the large increases in the supply of physicians ahnost everywhere. It may make sense to reconsider. Allocating Fixed Budgets to Providers Another method of control, allocating fixed budgets to providers, was advocated by Mr. Maxwell. It is somewhat similar to the HMO concept; the idea is to give providers as a group a sum of money and have them distribute it somehow among its members. That is what the West-German physicians do. According to Professor Reinhardt (Chapter 10), it reflects the power of the medical association to enforce its authority. The association performs enough of the routine day-to-day work on behalf of the financing plans to be allowed to do so. The West-German system resembles the medical-care foundation form of the HMO. It makes outlays actuarially calculable. For Professor Fein the distinction is important between the allocation of finite resources to providers and having the individual physician weigh resource considerations when he cares for the individual patient. He favors the former and opposes the latter. Management of Supply Side A noteworthy departure from post World War II thinking was made by Professor Evans, when he argued in favor of public management of the supply side. This point is old enough to be new, because in effect it marks a reversion to the doctrine that public financing is linked with public production. You will recall that when the British established the National Health Service, they still thought in an old-fashioned way. If tax funds were to be spent, there would be government production. Both the Canadians and the Americans, in putting a good deal of public money into the health-care system, opted for the separation of the two functions. For the first time in a generation, it seems to me, I hear someone say, let us reconsider the basic proposition. This is a very important issue. We did not see its stark clarity when Professor Evans first put it to us. I am putting it to you now. Professor Evans’ argument is that health care is a merit good, and that its supply must be ensured. The optimum organization of services depends on an adequate flow of information from providers. In his judgment private providers will not supply the requisite information. In support of his proposition Professor Evans drew on the case of smallpox vaccination in one of the Canadian provinces. Because the provincial authorities exercised control over the delivery system, they were able to move promptly to vaccinate the large majority of children. This argument does not strike me as persuasive, however, for we have been successful in vaccinating most children against smallpox in this country, with its prevailing form of private medical practice. But, then, I owe you the counter-rebuttal, namely, that in this country when the children came for smallpox vaccinations, they received other pediatric services as well; no physician visits were lost. 352 Control Over Number of Physicians The idea of possibly limiting the number of physicians runs counter to prevailing trends and policies almost everywhere. A vast expansion is under way in Sweden (Professor Stahl). In West Germany thought is being given to an expansion of 80 percent (Professor Reinhardt). As Professor Evans observed, the increase in physician supply through a vast expansion of the capacity of medical schools and through immigration makes one wonder about the role of physician extenders. In the United States both physicians and assistants are increasing in number and all past projections have been exceeded. Notwithstanding, the current literature for the United States is replete with citations from the 1950’s and the early 1960’s of conditions of shortage. Control over Location of Health Manpower A noteworthy report to the conference was that in West Germany the geographic distribution of physicians is such as to assure patients of access to physicians caring for ambulatory patients. Mr. Maxwell thought that success in maintaining such a distribution depends on the cooperation of the medical profession. Action to achieve redistribution in favor of areas that are often underserved may be more necessary under national health insurance than otherwise. Professor Fraser has found that with the enlarged flow of health-insurance funds, Canadian physicians are increasingly in a position to cater to their non-pecuniary preferences for locating in metropolitan areas. Although the situation in the United Kingdom was not discussed, it seems appropriate to mention here that one of the major accomplishments generally credited to the National Health Service is greater geographic equality in the distribution of health-care resources. There, considerable control is lodged in the government. ‘ 5. IMPLICATIONS FOR FUTURE This section incorporates a larger proportion of personal reflections than do the others. It comprises four parts: (1) The outlook for health-care expenditures. (2) Need, demand, and want as the criteria for resource allocation. (3) Lessons from comparisons among countries. (4) A research agenda. Outlook for Health-Care Expenditures Without exception the participants at the conference projected continuing increases in health-care expenditures, basing themselves on past experience. At least by implication~and explicitly in the case of Mr. Maxwell—an increase in the percentage of the GNP spent on health care was also projected. 353 Mr. Maxwell listed the factors pointing toward an increase. One is the advance of science and technology. This factor was further elaborated by Mrs. Rice: new specialized facilities; higher utilization of old services; etcetera. A second factor was the rising income of health-care providers. Initially that reflected a catching-up with earnings in the economy at large. More recently we are witnessing a display of power on the part of providers vis-‘a-vis local and provincial authorities. A third factor is rising public expectations and higher standards of service associated with affluence. I should add a fourth factor: the prospect of attending to the caring function more fully in the future. More doubtfully I offer a fifth factor: a high rate of unemployment in the economy. Under such circumstances the health-care sector is likely to be more stable than the economy at large; therefore the percentage of the GNP spent on health care would rise. In addition, the health-care sector may be seen as an outlet for additional jobs of a public service nature. I should like to avail myself of this opportunity to posit the possibility of a slowdown in health-care expenditures, at least in the United States. This was not discussed here, but it did come to mind during this conference. Again I see five factors at play. One is a high rate of inflation in the economy; I invoke the case of Japan. Second is the possibility of a slowdown in the rate of productivity gains in the economy at large, owing to a wider concern with the environment and conservation of energy. If such a slowdown occurred, the relative lag in productivity gains on the part of the health-care sector would be offset. Third is the declining labor intensity of the hospital, as shown by the United States data. Fourth is the shift away from retrospective cost reimbursement of institutional providers, which is just emerging. Fifth is the heightened visibility of health-care expenditures and their competition with other priorities; this is close to the case of the United Kingdom. This is not an exercise in forecasting. I suggest only that both sets of factors, those pointing upward as well as downward, should be kept in View. Need, Demand, and Want as Criteria Whatever the direction of and slope of the increases in expenditures, the economist continues to be interested in the optimum use of scarce resources. This is especially true in an economy characterized by full employment. As noted early in this summary, the economist’s ultimate concern is for getting value for our money. This can be expressed in several ways. Mr. Maxwell sees the goal of the health-care system as meeting the dual criteria of need and of effectiveness. An economist would restate this equivalently as having marginal benefit exceed marginal cost. The essential point in common is that there is supposed to be a link between the input of resources and the outcome of programs; and that outcomes have values attached to them. Here the economist and the health-care administrator part ways. The economist says, let us have regard simultaneously for both costs and benefits, at the margin. The health-care administrator is more apt to seek the maximum benefit in relation to the satisfaction of a medical need. 354 Nevertheless, there may exist a measure of agreement that is not so obvious. Both economists and health-care administrators, I believe, would discard individual ability to pay as the basis for allocating health-care resources. Both would certainly agree that resources made available by whatever criterion should not go unused and be wasted. Disagreement enters, because economists believe in the possibility of substitution among the factors of production and in catering to consumer preferences. The latter need not be based on personal knowledge but may reflect social, cultural, and psychological predispositions. Furthermore, even if individual ability to pay is precluded as a criterion, the total community’s or nation’s ability to pay remains as a factor. Often the benefits of programs are subject to diminishing marginal utility and cost to diminishing marginal productivity. If so, priorities cannot be absolute; they are likely to differ between rich and poor geographic areas. The possibility of substitution among inputs denies an absolute priority to any single input. These considerations by economists are additional to, and apart from, the question whether professionals really know how to determine need in objective, that is, replicable, ways. I refer here to the classic work on need by Lee and Jones, (22) which I have discussed elsewhere. (23 ) Here a health-care administrator, with his patience exhausted, could say with truth: I do not know of any systematic full-fledged cost-benefit analysis in the health field, with all of the alternatives posed. Cost-effectiveness analyses, yes; cost-benefit analyses, no. Surely, cost-effectiveness analysis does not assist in determining priorities for the allocation of resources. (24 ) One could proceed further, as Professor Fuchs himself does. (2) Although aggregate mortality statistics do not show the marginal effectiveness of health care, disaggregated data for particular conditions sometimes do. The treatment of patients with long-term kidney disease is a good example. In addition, medical care contributes to the improved health status of many persons who use it, apart from any effects on the death rate. Perhaps changes’in life style are more important, but are they attainable and at what cost? As Professor Weisbrod put the essential question, “Does medical care contribute a worthwhile benefit?” I should go even further and question the criterion of effectiveness in certain respects. I agree with McKeown that when individuals are invited into the health-care system, as in asymptomatic screening, the issue of effectiveness is central.(12) Perhaps effectiveness is the crucial criterion for any newly proposed diagnostic or treatment measure. Certainly, when alternative modal- ities are equally available, we should apply the one with the lowest cost per unit of output. The question is, What to do in the absence of an efficacious remedy? Is it not of some benefit to a sick person to get diagnosis, consultation, attention, and consolation? Does this context afford a clue perhaps to the large expansion of the physician supply taking place in Sweden, which possesses some of the best indicators of health status? It is also useful to recall that Professor Cochrane views the caring function as important and subject to the criterion of equity, not effectiveness. His last chapter is an eloquent plea for society to try to do the decent thing, when 355 knowledge is lacking. Mr. Maxwell’s example from the field of mental retardation is a salutary reminder that long-term patients have been neglected by the professions for a long time. Finally, as suggested by Dr. Alice Rivlin, there can be more proximate measures of the success of programs than effect on health status, namely, their effects on the redistribution of services. (25) Here, it seems to me, we can report success for the Medicare and Medicaid programs, although I should caution against overestimating it. (The technical point here is that as current income is depressed by illness, the utilization of services by income class appears to be more equal than it is.) Several measures of fairness in distribution exist: equality in access, in utilization of services, or in health status. Professor Weisbrod offered the additional criterion of a person’s realizing his or her own demand function. Here, Professor Fuchs would ask, “If health care is not effective, why be fair in distributing it, more so than in distributing income?” One can return to an earlier point about the other purposes of health care. Or one can settle for the broader generalization: economists can perhaps spell out the implications of the several criteria but society will somehow make the judgment. Of course, the preceding discussion is not conclusive. It is offered here in an attempt to clarify the implications of alternative criteria for health-care expenditures, to point up the areas of agreement, and to narrow the issues of disagreement. Lessons from Comparisons It was stressed repeatedly that France and West Germany have more pluralism in their health-care systems than Canada or the United Kingdom (Professors Blanpain and Ruderman). Accordingly, they may offer many more useful lessons to the United States. Let me, at this point, summarize the cautions attached to the application of the lessons drawn from one country to another. It is obvious that converting the money of one country into another’s poses difficulty. Brian Abel-Smith has discussed this very well. (2) Dr. Prims reported on his own experience in securing comparable data counts and uniform definitions of terms. Professor Evans told us that the extent to which physicians understate incomes to tax collectors can vary over time and probably varies among nations. Dr. Prims pointed to necessary cautions in interpreting presumably standardized data. Demands and needs differ (Mme. Sandier). Equivalently, the health-care system is part of a larger social system (Dr. Prims). Professor Holst emphasized the importance of history. One does not start from scratch; history provides a base-line. It points up the importance of supply factors on utilization and expenditures. One finds it difficult to overcome one’s inheritance. An historical approach creates an attitude favoring incremental budgetary adjustments. To Professor Culyer the last point makes particular sense in the presence of uncertainty. 356 Although pluralism has advantages, it creates the danger of a dual health-care system. Mr. Maxwell insisted that a second, private system must not be allowed to threaten the primary, public one. Professor Culyer would certainly allow the second system to provide non-essential services. He thought, too, that the second system might serve to set standards for caring. Finally, Professor Holst would distinguish between the transitional effects of a program or policy and its long-term effects. In his judgment it is almost always necessary to pay a price initially in order to secure control over the market. Subsequently, arrangements can be modified. A Research Agenda The specific research items I have distilled from the discussion follow no particular order. Professor Cooper suggested that countries may have different objectives for their health-care systems. Too little is known about that and about the implications of alternative objectives. , Another suggestion arises from the recent papers on screening in The Lancet. I refer to the distinction between efficacy (under specified conditions) and effectiveness (in the actual world). (27) Perhaps the gap between them varies among nations. It is important to focus on discrete programs. For example, how many cataract operations have been performed under Medicare? Professor Anderson proposed research on the relationship between the private and public sectors. He offered several theories or hypotheses. One views the private sector as a safety valve. Another sees it as a regular supplement. A third sees it as a competitor and yardsitck. In this connection, I recall Professor Culyer’s suggestion that the private sector do all the things that are not effective. There were several suggestions that more research be done on accounting. I favor that; it is important to engage some knowledgeable accountants in this activity or, even better, to teach accounting to economists. Almost from the beginning of my career in the health field I have discerned a tendency to misuse accounting, to neglect it for management purposes and to have allocated costs erroneously serve as a basis for price setting. Still another research project is an up-dating of older studies of the cost of defined bundles of health-care services in several countries. In the past the results varied greatly, depending on whose prices were applied. (28) Has there been a narrowing in the internal structures of prices? In general, a need was seen for disaggregation. Let us study populations in high risk groups; in geographic areas; by level of education. How will research be conducted? Obviously, as Professor Anderson remarked, through carefully designed studies. Let us make comparisons among nations. Even more, let us make comparisons among regions and provinces within a nation, as in France and Canada. Let us take advantage of natural experiments when they occur. Occasionally it may be feasible to conduct a designed experiment, though field work is costly. 357 But there is no need to stand still, awaiting all the results of research. I conclude on an optimistic note. In a conversation with Mrs. Worthington she commented on one of her projects,“It did not turn out well as an explanatory model, which was the original goal. But it works beautifully for prediction.” 358 10. 11. 12. REFERENCES FUCHS, VICTOR R. (1966), “The Contribution of Health Services to the American Economy,” Milbank Memorial Fund Quarterly, Vol. 44, No. 4 (October, Part 2), pp 65- 101. FUCHS, VICTOR R. (1974), Who Shall Live? Health, Economics, and Social Choice. New York: Basic Books. .COCHRANE, A. L. (1972), Effectiveness and Efficiency. London: Nuffield Provincial Hospitals Trust. . WORTHINGTON, NANCY L. (1975), “National Health Expenditures, 1929— 74, ”SocialSecurity Bulletin, Vol 38, No. 2(February), pp. 3- 20. . KLARMAN, HERBERT E. (1975), “The Economic Determinants of Health Care Expenditures,” In David Alan Ehrlich, ed., The Health Care Explosion. Which Way Now? Bern Hans Huber. pp. 7- 17. . FUCHS, VICTOR R. and MARCIA J. KRAMER (1973),Expenditures for Physicians’ Services in the United States, 1948-1968. Washington, D.C.: US. Government Printing Office. . FELDSTEIN, MARTIN S. (1971), The Rising Cost of Hospital Care. Washington, D.C.: Information Resources Press. . DAVIS, KAREN (1972), “Rising Hospital Costs: Possible Causes and Cures,” Bulletin New York Academy of Medicine, Vol. 48, No. 11 (December), pp. 1354-71. . EVANS, ROBERT G. (— —), “Beyond the Medical Marketplace: Ex- penditure, Utilization, and Pricing of Insured Health Care in Canada,” In Richard Rosett, ed., The Role of Health Insurance in the Health Services Sector New York, National Bureau of Economic Research, (forthcoming in 1976). KLARMAN, HERBERT E. (1974), “What Kind of Health Insurance Should the United States Choose?” In Joseph C. Morreale, ed., The US. Medical Care Industry: The Economist’s Point of View. Ann Arbor, Michigan: Graduate School of Business Administration, University of Michigan, pp. 93-108. SCITOVSKY, ANNE A and NELDA M. SNYDER (1972), “Effect of Coinsurance on Use of Physician Services,” Social Security Bulletin Vol. 35,No 6(June), pp 3- 19. MCKEOWN, THOMAS, Chairman, Working Group (1968), Screening in Medical Care, London: Oxford University Press for Nuffield Provincial Hospitals Trust. 359 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 360 KLARMAN, HERBERT E. (1971), “Analysis of the HMO Proposal—Its Assumptions, Implications and Prospects,” In Thirteenth Annual Sympo- sium on Hospital Affairs, Health Maintenance Organizations: A Recon- figuration of the Health Services System, Chicago: Center for Health Administration Studies, University of Chicago, pp. 24-38. DOWNS, ANTHONY (1965), “Why the Government Budget is Too Small in a Democracy,” Reprinted in Edmund S. Phelps, ed., Private Wants and Public Needs, revised edition, New York: Norton, pp. 76-95. AARON, HENRY (1967), “Social Security: International Comparisons,” In Otto Eckstein, ed., Studies in the Economics of Income Maintenance, Washington, DC: Brookings Institution, pp. 13-48. ROEMER, MILTON I. (1961), “Bed Supply and Hospital Utilization: A Natural Experiment,”Hospitals, Vol 35, No. 21 (November), pp. 36-42. SHAIN, MAX and MILTON I. ROEMER (1959), “Hospital Costs Relate to the Supply of Beds,” Modern Hospital, Vol. 92, No. 4 (April), pp. 71-73, 168. ROSENTHAL, GERALD D. (1964), Hospital Utilization in the United States, Chicago: American Hospital Association. AIRTH, D. and D. J. NEWELL (1962), The Demand for Hospital Beds. Newcastle-upon-Tyne, England: University of Durham, King’s College. FELDSTEIN, MARTIN S. (1967), Economic Analysis for Health Service Efficiency, Amsterdam: North Holland Publishing Co. KLARMAN, HERBERT E. (1963), “Effect of Prepaid Group Practice on Hospital Use,” Public Health Reports, Vol. 78, No. 11 (November), pp. 955-65. LEE, ROGER I. and LEWIS WEBSTER JONES (1933), The Funda- men tals of Good Medical Care, Chicago: University of Chicago Press. KLARMAN, HERBERT E. (1965), The Economics ofHealth. New York: Columbia University Press. KLARMAN, HERBERT E. (1974), “Application of Cost-Benefit Analysis to the Health Services and the Special Case of Technology,”International Journal ofHealth Services, Vol. 4, No. 2 (Summer), pp. 325-5 2. RIVLIN, ALICE M. (1971), Systematic Thinking for Social Action, Washington, DC: Brookings Institution. ABEL-SMITH, BRIAN (1963), Paying for Health Services. Geneva: World Health Organization. 27. SACKETT, DAVID L. (1974), “Screening for Disease: Cardiovascular; Diseases,” The Lancet, (November 16), pp. 1189-1191. 28. GILBERT, MILTON and IRVING B. KRAVIS (1954), An International Comparison of National Products and the Purchasing Power of Currencies, Paris: Organization for European Economic Cooperation. 361 INDEX A Aaron. Henry. 349 Abel-Smith. Brian. xvii. 356 Absence of workers through illness. 279. 283. 284 Access to medical care Canada. in. 59 national health insurance. and. 75 United States. in the. 57 Accreditation of hospitals. govern- ments controlling. 16 Activities. medical care. in French- American comparative study. 119 Acute versus Chronic disorders as causes of death. 10] Administration. general health. in France. 164 Administration of health care in Sweden. 20] Administrative staff. Swedish. 22] Admission. note on term. 141 Admissions hospital. comparisons of. 144 hospital. West-German and American. 286 After-care. French and American, 137 Age and sex and levels of health-care expenditures. 126 Aged and poor in US. medical needs of. 37. 38 Aging and deterioration. 14 Air pollution and deaths in New York City. 42 Allekotte. Heinz. 249 Allocation of health comment on. 8 Altman. Stuart. 2. 345. 346. 349. 351 Amalgam restorations not requiring dentist. 72 Ambulatory care comparisons of changes in. 156 expenditures in France. 129 expenditures in United States. 125 French. 167. 175 trends in. 187. 188. 190 insurance in Canada. 67 substitution of. 348-349 American medical economy changes in the. 47 problems of the. 26 I'CSOUI‘CCS. numbers of American physicians’ laboratory-test orientation. 196-197 Americans' failure to invest in health insurance. 105 Analysis. operational. of health systems. 232 Anderson. 0. W.. 229-232, 342, 343. 357 Annual health expenditures in ten countries. illustrated. 4 Anticoagulant therapy and ischemic heart disease. 43 Anxiety and medical costs in Canada and the US. 74 Appalachian States. doctor-population ratio in the. 15 Architects. income of Canadian. 83 Armed forces health expenditures. French. 166 Arrow justification for health insur- ance. 73 Associations. medical. Canada. 71 Associations. physicians’. German. 254. 256 Austria. physician-population ratio in. 213 Auxiliaries. medical. and health-care costs. 72 Auxiliary personnel use in France and the United States. 154 national strength in West B Balancing cost sharing in health systems, 231-232 Bed capacity expansion and hospital inflation. 30 Bed-days increase in. Swedish, 214 West German and United States. 260 Beds acute. misuse of in U.K.. 17 beds-population ratios. French and American. 138 changes in ratio to population. 13 curtailing supply of. 350-351 363 Beds (c0111.) numbers of. French and American, 137. 140 personnel per. in France and the United States. 139 psychiatric hospitals. in, 140 ratios of, 6 reductions in. 6 Romanian hospitals'. 317, 321 utilization in Canadian hospitals, table of. 64 West Germany and the United States. in. 261 Belgium expenditures in, 240 extramural care by general prac- titioners in. 245 fee-for-service in, 245 health-care administrative responsi- bility in. 236 health-care provision costs, 239 health costs and expenditures in, 233-242 health insurance in. 233 hospital charges in, 238 physicians in, 234 statistics of. 233 Behavior. personal. and disease. 42 utilization of health-care services, 68 Beveridge Plan. Britain‘s, 348 Biological revolution, impact of. 48 Biomedical sphere, most research being conducted in the, 243 Blanpain. Jan, 2, 243-247. 356 Britain concern, governmental, over health- care costs in, 20 . geographic equality in health care in, 353 GNP and health expenditures in, 14, 343 health-care funding in, 16 health costs and expenditures in, 93-108 misuse of acute beds in, 17 public assistance concept in, 291 staff-patient ratio in mental institu- tions in, 15 British approach to health care. agree- ment for the. 198 British National Health Service administration costs comparisons, 197 appraisal of curing aspect of, over- due, 107 beds, decline in, 96 364 British National Health Service (cont) charges to patients in. 105, 106. 107 comment on caring in. 110 complaint and the. 93 cost-effectiveness and the. 113 cost of health services in the. 97 cost of inpatient care in the, 99 coverage of. universal. 103 curing rather than caring in. 98 demand and need in, 109 dental charges in. 105 drug prices in the, 104 establishment of the, 352 executive councils” manpower changes in. 100 expenditure and external forces in. 94 expenditure patterns and the. 110-111 finances of. 103-107 gross cost of, and national income, 96 health centers' growth in the. 103 health education. and, 102 health services as proportion of total cost of, 97 hospital buildings and the. 102 hospital services manpower changes in, 100 increases in personnel in. 96 incremental planning in, 93-94 inpatients doubled in, 97 lack of end-result data in. 111 manpower changes in, 100 monopsony power of state, and the, 109 morbidity, and, 107 need of services in, insatiability of, 98 “needology” and the. 109 objective of, 93 ophthalmic charges in. 105 planning insufficiencies in the, 94 preventive medicine. and. 101, 111, 348 priorities of, 94. 98-102, 111, 113 problems in, awareness of. 108 provision of services, growing gap in the, 102-103 revenue expenditure of hospitals in, 98 running cost of, 12 sole purchaser of medical talent, is the, 104 sources of finances of. 104 specialties, current expenditures by, 112 supply in, 94-98 British National Health Service (c0nt.) tolerance thresholds of young, and the. 103 treatment conditions in, 102 unmet need in, 102 virtues of. protection of. 108 waiting lists in. 101 waiting times for hospital admis— sions in, 101 British public expenditure, analysis of. 95 Brookings Institution, 347 Bucharest Medical Faculty, 329 Budgets, allocation of fixed. 352 Building investments in Sweden's health system, 204, 219 Business enterprises in French health- care system, 177, 181 C Canada access to health care in, 59 ambulatory medical care insurance in, 67 costs of health care buried in budgets in, 59 deductibles in health-care systems in, 72 demands for expensive services in. 84—85 dental care in, 72, 73 dental care costs in. 59 dental coverage in, 73 description of health-care system of, 76 distribution of health-care resources in, 84 drug coverage in, 73 drug purchases in, 59 estimate of health-care costs in, 59 evolution of health services in, 70 excessive demand not significant in, 74 expenditures, health-care, table of, 60 federal-provincial cost-sharing in, 61 fee schedule for physicians in, 61 growth of health-care expenditures in. 65 health-care industry management in, 62 health-care system of, 59-63 health expenditures in, 59-79 hospital building program in, 64 Canada ( can I. ) hospital costs and demand for services in. 67-68 hospital patient—day cost in, 18 hospital utilization and insurance in. 67 inadequate federal statistics in, 69 insured status and health-care in. 66 1971 first year of complete national health insurance in. 63 payment for medical services in. 71 perspective on system in. 63-70 physician surplus in, 65 physicians” earnings publication in, 77 physicians' fees in, 68 physicians—population ratio in, 65 physicians' salaries in, 69, 70 population data source, 76 private health-care expenditures in, 59, 60 promotion of dental care in, 73 provincialization of hospitals recom- mended in, 71 public management of health-care in. 74 salary and wage gains of hospital people in, 69 success of health—care programs in, 63 tax system component, health-care payment is a. 76 tenuous balance between federal provincial cost-sharing in, 232 utilization variables and postulated determinants in. 87. 88-90 West German system similiarity, and, 271 Cancer-detection German, 293 Capital expenditures and hospital in- flation. 30 Carcinoma of cervix. deaths from, 43 Care and cure and British expenditure patterns, 111 Care categories of French National, Health Consumption, 171, 173, 187 Caring and curing in British National Health Service, 98, 110 Carriers, insurance, West German, 265 Certificate of Need laws, 44, 45 Changes in Canadian health-care ex- penditures, tabulated, 65 Charges to patients in British National Health Service, 105, 106, 107 programs, West 365 Charges. user, 346-348 Checks on services in French system, 198 Childbirth and hospital care. 33 Children's allowances. Romanian, 315 Children's dental programs, 78. 79 Chronic diseases therapy in Romania, 324 Chronic diseases treatment costs in Romania. 325 Chronic disorders causing death. 101 Clerical costs of health services. 197~l98 Client-oriented program budget of British Department of Health and Social Security. 111 Cloning, 48 Cochrane. A. L.. 340, 355 Coinsurance. 44 Cologne University. 249 Community hospitals. expanding role of, 30. 31 Comparisons of health-care systems. difficulties of. 115-116 Competition. business, European, 196 Complaint and the British National Health Service, 93 Compulsory insurance. West German. 266 Compulsory sick-benefit Sweden‘s. 211 Consumer Price Index. French. 167, 184 Consumer Price Index, US. medical component of, xv Consumers and health-care decisions. 113 Consumers. Belgian. 233 Consumption French National Health. 166. 167, 183. 190. 191, 192 analysis of. by establishments, 172 breakdown by enterprise. 174-175 health, Sweden's, 203-204. 217, 218 West Germany. in, 279 Contract, social. West German, 287 Control measures against medical expenditures inflation. 43-46 Control of health resources. 346-353 Controls, administrative, and medical- care costs. 47 Controls, wage and price, and health- care. 32 Convalescent homes. French. 137 Cooper. Michael H., 93-108 Cooper, Michiko M., xiv Coronary care units. costs of, 33 scheme, 366 Cost analyses, Romanian, 320 Cost-awareness in United States users'. 295 Cost-control mechanism. 242 Cost-effectiveness and the British National Health Service. 113 Cost increases in American medicine, 24 Cost increases, health. and national income, 3 Cost index comparisons, services‘. 147 Cost of British National Health Serv- ices. services proportions in, 97 Cost-sharing in health-care systems, 231-232 Cost-sharing in West Germany. 285 Cost-sharing system. Swedish, 202 Cost studies. French. 246 Costs See also Health Costs and quality of care relationships. 238 changes in, factors causing, 239 comparisons for Western European and North American countries. 339-358 health care and use of medical auxiliaries, 72 health-care in Canada. estimate of. 59. 60 health care to patient, in Belgium, Denmark and The Netherlands, 237-238 hospital. causes of increases in. 19 technological advances. and. 32-34 of British National Health Service, 12 of inpatients in British National Health Service. 99 per day. hospital. 141 French and American, 145 running, hospital. 6 Counties, health insurance system cost sharing in Sweden. 207 Counties. Swedish, administering medical care, 211 Countries covered by papers in this Proceedings, xvi Coverage, population, in Canada’s service, 63 Cultural and social activities budget, Romanian, 318 Culyer, A. 1.. 109-113, 348, 356, 357 Curing and caring in British National Health Service, 98 hospital D Davis, Karen, 344 Denmark expenditures in, 241 health-care administrative responsi- bility in, 236 health-care provision costs, 239 health costs and expenditures in, 233-242 health insurance in, 233 hospital charges in, 238 physicians costs in, 241 physicians in, 234 preventive services costs in, 241 regionalization in, 349 Death causes United States, in, 42, 279, 281, 282 West Germany, in, 279, 281 Death rates comparisons, 282 infant, 132 maternal, 283 United States’ and West German, 279, 281 Decision-making process in Canadian system, 63 Deductibles in Canadian system, 72 Defense spending in Britain, 94 “Defensive” medicine and malprac- tice suits, 45 Demand as a criterion, 354 for high—cost health-care services, 84-85 for services in Sweden, 205 “Demand is physician—induced”, 269 Dentacare, Canadian, 86 Dental assistants, Swedish, 221 Dental care amalgam restorations not requiring dentist, 72 Canada, in, 59 children’s, in Saskatchewan, 72 costs, Swedish, 208 insurance in Canada, 73 payment for in Canada, 73 promotion of, 73 Swedish, 205 Dental coverage in Canada, 73 Dental insurance programs, note on Canadian, 79 Dental programs for Canadian, 78, 79 Dental services, British, cost propor- tion in National Health Service, 97 children, Dental visits by poor and nonpoor, 39 Dentists British increase in, 96 dentists-population ratios, French and American, 121 expenditures on, American, 26 expenditures on, French, 129, 133 numbers of Swedish, 221 registered with French Social Secur- ity, 189 services, 156 West German system, in the, 270 Deterioration and aging, 14 Diabetics, insulin versus diet therapy, 43 Discussions at conference, 339-358 Disease control in US, 41 Diseases affected by nonmedical fac- tors, 42 Distribution of medical services in West Germany, medical profession responsible for, 270 District Health Directions in Romania, 326 Doctors See Physicians Drugs charges in Sweden, 205 costs in Belgium, Denmark and The Netherlands, 238 costs in Romania, 324 costs in Sweden, 208 coverage in Canada, 73 dispensing and usage, French, 196 expenditures in France, 129, 133 expenditures in the US, 26 payments for in Sweden, 202 prices in British National Health Service, 104, 107 purchases in Canada, 59 Dummy variable in Canadian hospital care statistics, 81 E Economic growth, Swedish, 230 Economic incentives to change in health care, 71 Economic Stabilization Program, US, 26, 45 Economic waste in British National Health Service, 107 Economics and moral philosophy, 58 Economics of Health and Medical Care, The, xv 367 Economist and the British National Health Service, 109 Economists, liberal, and Sweden, 201 Economy, comparison of those of France and the United States, 119 Economy, medical federal aid and the US, 35 physician’s importance in, 34 Economy, the American medical, 23-53 Education, health, and the British National Health Service, 102 Education, medical and research costs, West German, 274-275, 294 federal subsidization of, 39-40 in France, 163-164 expenditures on, 179, 180 Swedish, 213 West German, 286 Education Ministries search, 244 Empirical Studies in Health Econom- ics, xv Employment figures, health services', 11 Employment figures in hospitalization sector, 137 Employment in health care, ratio of, 5 funding re- Employment in health sector in Sweden. 220 Endogenous trends in health care in Canada, 66 Engineers, income of Canadian, 83 England and Wales, costs of health services in, 10, 12, 21 Entrepreneurial spirit of American medical economy, 23 Environmental conditions and death rate variations, 43 Equipment and staff of American and French hospitals, 140 Equipment, hospital, and changes in services, 33 Europe business competition in, 196 placebos use in, 196 western, an homogeneous entity, 197 European System of Integrated Ac- counts, 161 Evans, Robert G., 59-79, 269, 340, 341, 342, 344, 349, 351, 352, 353, 356 Examinations and treatments in Romania, 331, 332, 333, 334 368 Exogenous effects in health care in Canada, 66 expectation of life, 131 Expenditures accounting framework, and the, 342 and price data in French-American systems comparisons, 116 annual health care, increases in, 5 Belgium, in, 240 British, 93-108 by specialties, 112 public, 94 Canadian, 59—79 increase in, 62, 66 physician, 68 table of, 60 three main classes of, 60 capital, control of, 6 control of chronic inflation of, 43-46 Denmark, in, 241 educational, 123 explanatory hypotheses, and the, 343 external forces affecting, 94 factors involved in, 341 French current, 176-182 national health, 168, 169-182 personal health, 128-129 resources and, 178-179 summary structure of, 179 surveys, 166 health-care, age and sex, and, 126 GNP, and the, 3, 4, 7, 8, 9, 10 met from general taxation, 16 National Accounts System, and the, 161 national income, and, 3 private, 38 self-limiting, 70 several countries, in, 3, 4 hospital care, on, 141 institutional, 345 medical chronic inflation of, 26 Medicare and Medicaid, and, 26 Netherlands, The, in, 241 outlook for, 353 patterns, British, and care and cure content, 111 patterns and the British National Health Service, 110-111 patterns of, 4 personal health, in United States, 124-125 physicians” services, on, 155 Swedish, 199-228 Expenditures (cont.) trends in, 155, 157, 341 trends in France and United States, 143—148 trends in health care, 3-4 trends, Romanian, 325 United States, in, 24 West Germany, in, 273-284, 285, 297 Expenses, professional, in France and the United States, 154 Experiments, health care, 357 “Explosion” of costs, 295 Eyeglasses expenditures France, in, 129 United States, in, 125 F “Fairness" of health-care systems, 232 Farmers, French, and Social Security, 189 Federal-provincial cost sharing in Canadian health-care system, 61 Federal, state and local health ex- penditures, illustrated, 36 Federal subsidies, US, 26 Federal subsidization of medical care, 35-39 Federal subsidization of medical edu- cation, 39-40 Fee-for-service in US, 23 Fee-for—service versus salary debate, Canadian, 86 Fees, physicians’ Canadian, 61 French and United States, 149 Swedish, 202 Fein, Rashi, xiv, xviii, 55—58 Feldstein, Martin, 342, 344 Fiat car factory analogy, 231 Finances of British National Health Service, 103-107 Financing of health care a complex task, 23 Belgium, Denmark and The Nether— lands, in, 237 France, in, 177 root problem of, 16 Sweden, in, 205-211, 229 Financing of medical schools, 40, 41 Fogarty, Congressman John E., v Fogarty International Center, the, v activities of, vi Foreign—trained physicians in US, 41 Foulon, Alain, 161-195, 246 France . ambulatory care expenditures in, 129 ' care categories trends in, 187, 188 central administrative power in, 117 Consumer Price Index in, 167, 184 consumption of health care in, trends in, 191, 192 cost studies in, 246 costs of physicians’ services in, 155 current health expenditure in, 168, 169-182, 176-182 development of health consumption in, 1950-1974, 182 direct charges to patients in, 16 drugs dispensing in, 196 economic agents in health-care ac— counting in, 164 expenditures trends in, 143-148 financing health activities in, 181, 182 funds sources in, 126, 174-175, 176 GNP and health-care expenditures in, 135 health care categorized by estab- lishments in, 172 health-care consumption in, 168-182 health-care services in, list of, 162 health-care system of, 115-198 compared with American, 115- 160 health consumption and resources in, 169 health consumption in, analyzed by supplier, 169-170 health expenditures in, by activi- ties, 180 health products in, listed by per- centage, 170 . health services in, listed by percent- age, 170 hospital care increase in, 193—194 hospital services in, data on, 138- 139, 144-145 hospitalization costs in, 141 medical education expenditure in, 179, 180, 181 medical teaching and research in, 163-164, 197 mortality statistics in, 42 National Accounts System, and the, 161-195 See also System National Accounts 369 France (cont.) National Accounts System, satellite, of, 161-195 National Health Consumption ac- counts system in, 166 National Health Consumption breakdown by financing sectors, 177 National health expenditure in, 168-182, 178-179 accounts system in, 166 personal health expenditures in, 128-129 pharmaceutical lobbies in, 194 physicians’ philosophy in, 198 physicians-population ratio in, 148 physicians’ services in, increase in variables in, 151 placebos in, 196 preventive medicine funding in, 180 sectors of National Accounts System in, 164 Social Security in, 164 survey in, 243 trends in health consumption in, 183, 184, 185, 186 Fraser, Roderick D., 81-91, 353 French-American medical procedures comparisons, 196-197 French-American systems compari- sons, problems of definition in, 121 Fuchs, Victor, 2, 340, 343, 344, 345, 351, 355, 356 Funding capabilities, American, 122 Funding of health care in France, activities and the, 181 Funds flow in West-German insurance system, 276 Funds source and distribution of per- sonal health-care expenditures, 136 Funds, sources of, in France and United States, 126 Funds sources trends in French health care, 189 Funeral benefits, West—German, 267 French and G General practitioners See also Physicians Belgium, Denmark and The Nether- lands, in, 235, 236 fees of, 149 increase in British, 96 number of future, 246 370 General practitioners (cont) numbers of, French and American, 150 per capita payment system in The Netherlands, 245 services now taken over by specia- lists, 34 Genetic surgery, 48 GNP and health in Health Services Re- search survey, 244 Britain’s health-care costs and, 14 health expenditures and, 343 Canada’s, health services and, 293 escalation of expenditures on health care, and the, 14 French, 119 health-care consumption in, 183 health care and the, 354 health care and the U.S., 23 health-care expenditures percen— tage of, 135 health—care resources, and, 56, 57 health costs and the, xv, 339, 340 health expenditure, and, compari- sons, 3, 4 increases in proportion spent on health services, illustrated, 7, 8, 9, 10 Sweden’s, health care and, 203 U.S., and health-care costs in, 24 U.S., health services expenditures and the, 293 United States-France comparison, 119 West German health services costs and the, 293 health services relationship to, 273 ‘ ‘ Geographical comparisons, France and the United States, 117, 118, 148 Geriatrics, British conditions of treat- ment of, 102 Germany, Federal Republic of, See West Germany Goods, medical, French trends in, 187, 188, 189 Government Social Survey, British, 103 Governments and the control of health-care, 16 Gross National Product See GNP Group practice, Canadian publication on, 78 Growth in Canadian personal income, 67 Growth of expenditures in West- German system, 277, 278 H Hastings Report, The, 76 Health Accounts, French, 161 Health administration, French, 164 Health and health education, 102 Health center care, prepaid, 44' Health establishments and French National Health Consumption, 172 Health expenditures See Expenditures, health, 4 Health care as a commodity, 224 assessment of effectiveness of, 20 awareness of limitations of, 20 concept of need of, 20 definition of need for, 20 delivery service, rational model of, 231 expenditures See also Expenditures change in, comparisons of, 156 distribution by type of expendi- ture in France and the United States, 134 growth in United Kingdom, 94 population increases, and, 13 potential growth in England and Wales, 21 trends, 3-4 facilities’ provision and utilization, French and American, 122-123 field boundaries in France, 165 formulation of demand for, 20 free in Canada, 59 governments and the total bill for, industry management in Canada, 62 insurance, started in Germany in 1880, 197 paradox of, 14 payments, Canadian, 76 potintial and hospital equipment, 1 0 programs, Canadian, success of, 63 questions in Canada, 59 resources, Canadian, distribution of, 84 sector of accounts system, French, 161-164 Health care (cont.) services costs of providing, 239 French, patient transport in, 162 regionalization of, xvii satisfactory quality judgment of, 14 structures, French and American, 122-123 systems determination of size of Swedish, 224 in the United States and France, 119-133 international, study of, 84 organization in France and the United States, 137 use of resources of, 20 Health consumption and national re- sources in France, 169 Health costs causes of increases in, xv concern over rising, xv definition differences, and, xvii economic benefits, and, xvii GNP, and the, xv Sweden, in, 199-228 United Kingdom, in the, 93-108 Health economics conferences, two major, xv _ Health insurance, West German, 262- 273 Health Interview Survey, the, 37 Health maintenance organizations, 44- 45, 47, 352 Health, medical care’s impact on, 41- 43 Health sector and medical-care sector, illustrated, 121 Health services British, sources of funding of, 104 changes in, and cost increases, 31 costs in England and Wales, 10 costs in West Germany, 268, 293 distribution of, 8 comment on, 270-271 doubts regarding effectiveness of some, 9 employment figures in, 11 expenditures and the GNP, 7, 8, 9, 10 French, 167 , divisions of, 162 inequitable availability of, 8 overprovision of, 9 poor and nonpoor, utilization by, 39 public policy alternatives in financ- ing, 46-47 371 Health Services Research classification of projects in, 244 information required on, 247 special purpose settings for, 245, 246 survey, defined, 243 university-based, 245 Health status infant mortality a measure of, 84 and medical care changes, 41 of French and American popula- tions, 127 Heart attacks, unexpected, 42 Heart disease and economic stress, 43 Hemophiliacs’ treatment in Britain, 102 Heredity and health, 42 High-risk populations, 357 Hill-Burton program, U.S., 26 Historical health insurance system in West-Germany, 264 Historical perspective on Canadian system, 63-70 Holst, Erik, 2, 271, 348, 356, 357 Home visits by physicians in Sweden, 205 Hospital beds, numbers of, in France and the United States, 137, 140 Hospital beds study, Romanian, 321 Hospital building, governments, con- trolling, 16 Hospital building in Sweden, 204 Hospital care costs increases, western European, 240 changes in, comparisons of, 146- 147 French, increase in, 193-194 illustrated, 142 payment for, 141, 143 trends in France, 187, 188 Hospital costs ' causes of increases in, 19 components of, 31-32 increases and demand for services, Canadian, 67 Swedish, 211 technology, and, 32-34 Hospital discharge rates, and the poor and nonpoor, 39 Hospital facilities increase in United States, xv Hospital inflation theories, 30 Hospital Insurance and Diagnostic Services Act, Canadian, 63 Hospital labor costs dominating in Britain’s, 75 372 Hospital patient-day cost in Canada, 18 Hospital productivity indicators, 110- 111 Hospital sector growth trend, 296 Hospital services annual rate of change in, 146-147 British, National Health Service cost, in, 97 changes in, and costs increases, 31 comparisons between France and the United States, 138-139, 144- 145 French, 167 health-care expenditures, and 5, 6 West German physicians, of, 257 Hospital stays, average, in West Germany, 261 Hospital use in West Germany and the United States, 263 Hospital utilization and insurance in Canada, 67 Hospital utilization increases in United States, 37 Hospital, waiting list for, British, 97 Hospitalization days of, average, 143 expenditures increase in United States, 123 France and United States, in, 137 health-care expenditures, and, 137 physicians’ services in France and the United States, and, 133-157 plans, Canadian, comment on, 81 Romania, in, 336, 337 Hospitals Belgium, Denmark and The Nether- lands, in, 235, 236 British, inpatients costs in, 99 building in Romania, 320 building program, Canadian, 64 charges in Denmark, Belgium and The Netherlands, 238 costs of running Swedish, 223 ' I Immigrant populations and the health services, 8 Income, average, of groups, table of, 83 Income comparisons, physicians’, 154 Income data, West German physi- cians’, 257, 258, 259 Income, national, and health expendi- tures, 3 professional Income tax, counties levying in Swe- den, 201 Incomes and prices levels, French and American, 141 Incomes, physicians comment on Canadian, 81 West German, 286, 287 Industrial model analogy for health- care system, 231 Inefficiency, hospital, 30 Infant mortality, 279, 283 French and American, 127 prime measure of health status, a, 84 publication on, 81 rate of, 132 socioeconomic class, and, 42 values for 25 countries, table of, 82, 85 Inflation chronic, in medical expenditures, 26 control of, 43-46 health care, and, 354 improvements in medical technol- ogy, and, 35 in costs of health care, 14 in hospital costs, theories of, 30 Inpatient health care in Sweden, 222 Inpatient hospital care, France, trends in, 187, 188 Inpatient costs in British National Health Service, 99 Inpatients doubled in Britain, 97 Institutions, medical, in Belgium, Denmark and The Netherlands, 235, 236 Insurance benefit package, 267-268 compulsory, Sweden‘s, 201 coverage and demand for hospital care, 32 coverage, West German, 264, 265 enrollment requirements in West German system, 265-267 Insurance Physicians’ Association, West German, 295 private, and health-care expendi- tures, 136 system of Sweden, 210 system of West Germany, 262-273 voluntary, 347 Insured status and health care in Canada, 66 International Labor Organization, sur- vey by the, 247 International systems of health care, study on, 84 West German, Internists’ fees, 149 Inverse Care Law, the, 8 Investments Romanian, 317-320 Sweden’s health sector, in, 219, 204-205 Ionizing radiation, French measures g against, 163 J Joseph, Sir Keith, comment of, 101 K Klarman, Herbert E., xvi, xvii 339— 358 L Labor cost-push inflation, medical, 30 Labor intensity of health services, 4 Labor movement, US, and social health insurance, 46 Laboratory tests in Romania, 338 Lancet, The, 357 Lawyers, income of Canadian, 83 Leavitt, Milo D., Jr., Vi, 1—2, 233 Lee, Maw Lin, xiv, xiii Leiden University, The Netherlands, research project at, 246 Length of hospital stay, in West Germany, change in, 277 Leprosy, French measures against, 162 Leuven University, Belgium, 245 Life expectancy, 127, 131, 199 Life—style changes and prevention of mortality, 42 Liver, the, French emphasis on, 197 Lobbies, pharmaceutical, French, 194 Localization of health-care responsibil- ities in Sweden, 201, 202 Long-term illnesses, shift toward, in the United States, 30 Long—term sickness in Sweden, 208 M Maastricht University, The Nether- lands, research at, 246 373 Machines and labor and hospitals costs, 32 Malpractice suits and “defensive” medicine, 45 Management of health care systems, 74 Manpower adequate, West German, 284-285 British, 96 changes in British National Health Service, 100 controlling location of, 353 in French-American systems com- parison, 121 manpower-population ratios, French and American, 121 medical, costs and, 14 use of, 8 Mariners Health Insurance, German, 266 Market conditions and physician fees, 68, 69 Market expenditure, France, 166 Market, medical, in the U.S., 23 Mass-radiography program, British, 113 Maternal death rates, West German and American, 283 Mathematical pattern of health-care system in Romania, 331 Maxwell, Robert, xvi, 3-21, 109, 341, 348, 352, 353, 354, 356, 357 Mechanism of health financing, changes in the, xv Medicaid, 23, 26, 30, 31, 35, 37, 292, 339, 356 Medical activities and French- American systems comparison, 119 Medical acts data, French and Ameri- can, 115 Medical associations’ Canada, 71 Medi-Cal, 44 Medical Care Act, Canadian, 63 Medical care delivery of, in U.S., 48 evolution of, 130, 133 expenditures increases in U.S., 23, 26-29, 30 federal subsidization of, 35-39 impact on health, 41-43 per capita spending in U.S., xv question regarding efficacy of, in U.S., 47 social imperative in U.S., and, 55 utilization review, 44 West estimates in strength in 374 Medical consumption, French, by cat- egories of care, 173 Medical education, 39-40 French, 163-164 Medical examinations See Examinations Medical graduates annually, 41 Medical insurance programs uncon- trolled, in Canada, 69 Medical profession composition com- parisons, 149 Medical professions’ growth trends, 150-151 Medical resources, allocation of, 56 Medical risk variations and health services financing, 104 Medical schools’ funding, 40 Medical schools state-owned in Swe- den, 230 Medical services and general practitioners consulta- tions, 101 distribution, Canadian publication on, 78 distribution, comment on, 270-271 free of charge in Sweden, 205 Medical staff changes in British Na- tional Health Service, 100 Medical staff remuneration in R0- mania, 321 Medical technology improvements and hospital costs, 32 Medical visit numbers, American comparison, 154 Medicare, 23, 26, 30, 31, 35, 37, 44, 45, 46, 116, 292, 339, 356, 357 Canadian, 69, 89 tax deductions, and, 347 Medicine, American, rising costs in, entering U.S. French- Medicine and the social sciences, 1 Medicine‘s areas of effectiveness, 41 Mental nurses in Sweden, 221 Mentally-handicapped, conditions of treatment of, British, 102 Monopolistic medical profession in West Germany, 287 Moral philosophy and economics, 58 Morbidity British National Health Service, and the, 107 Romanian, 328 Mortality, infant See Infant mortality Mortality statistics, French, 42 and American, 127 Mortality variations, 41 N National Accounts education expenditures, and, 123 international practice for, 115 National Accounts System described, 161 European System of Integrated Ac- counts, 161 Health Accounts, and the, 161 macro—economic analysis, and, 193 satellite French, 161-195 sectors of, in France, 164 statistical methods and sources of, 161-168 National Board for Health and Wel- fare, Swedish, 199, 206 National Health Consumption, French, analysis of, 169-170 National health expenditures annual increases in American, 26-29 France, in, 168-182 United States, in the, 24, 25 National health insurance and access to medical care, 75 debate in United States, 48 system for the United States, 287- 288 National health program costs, Cana- dian, 62 National health program for United States, 46 National Health Service, British See British National Health Service National income and national health expenditures, 3 National income, British, and cost of National Health Service, 96 Nationalization of all health-care re- sources in United Kingdom, 93 Need and the British National Health Service, 109 concept of, 20 criterion, as a, 354 insatiability of, British, 98 need laws, US, 350 Romania, in, 327 true nature of, 93 Netherlands, The concern over health-care costs in, 20 expenditures in, 241 health-care administrative respon- sibility in, 236-237 health-care provision costs, 239 The Netherlands (cont.) health costs and expenditures in, 233-242 health insurance in, 234 hospital charges in, 238 physicians in, 234 research project in, 246 survey in, 243 Tilburg University research in, 245 New medical procedures, comment on, 32 “Nominal” user charges for health services, 105 / Nonlabor inputs and hospital costs, 34 Nuftield Provincial Hospitals Trust, survey funded by, 243 Numerus Clausus and German medi- cal education, 286 Nurses, Swedish, 221 Nurses-population ratios, 12 French and American, 121 Nursing homes, United States’, 23, 137, 138 charges, Swedish, 205 0 Obstetrics, length of hospital stay in United States and West Germany, 260-261 Ophthalmic services cost, British, 97 Outpatient care, Swedish, 214, 223 Outpatient centers expansion in Swe- den, 199 Outpatients increase in Britain, 97 Over-prescribing of drugs in Sweden, 206 Overprovision of health services, 9 Over-utilization of medical-care facili- ties, 45 P Pan-American Health Organization, the, v Paramedical personnel and physi- cians’ services, 34 Paramedics in France and the United States, comparisons of use of, 154 Participants in conference, list of, ix-xiv Patient-day cost in Canadian hospitals, 18 375 Patient-day expenses increases, fac- tors in, 31 Patient-day hospitalization rates in Canada, 67 Patient-days, West American, 278, 285 Patient days-population ratio United States, 263 West German, 263 Patient fees in Sweden’s system, 209 Patient-visits by physicians in West Germany and the United States, 254 Patient-ward staff ratio in British men- tal institutions, 15 Patterns of health expenditures, 4 Payment method, West German, 270 Pediatricians’ fees, 149 Personal health—care expenditures French-American comparisons, 133, 135 in the United States, tables of, 26, 124-125 Personal income and health care costs in Canada, 66-67 Personal health consumption, French, 193 Personnel British increase in, 96 costs and health expenditures, 13 health, West German and Ameri- can, 285 medical, hospital, in West Germany and the United States, 262 non-medical, per patient, German- American comparison, 278-279 per bed in France and the United States, 139 Pharmacare, Canadian, 73, 86 Pharmaceutical companies, Sweden’s, 202 Pharmaceutical products price index in France, 188 Pharmacists-population ratios, French and American, 121 Pharmacists, Swedish, 221 Philosophy, medical, French- American differences in, 197 Physical geography comparisons, France and the United States, 117 Physicians See also general practitioners access to medical care, and, 75 associations of, West German, 254, 255, 256, 265 German and 376 Physicians (cont. ) Belgium, Denmark and The Nether- lands, in, 234 buying power of, 151 can choose working location, 148 controlling number of, 353 costs in Denmark, 241 distribution of, geographic, 256 earnings, Canadian, 77 expenditures on West German, 297 fee schedules, and, 68, 202, 237, 351 foreign-trained, in United States, 41 French and US. data on, 345 hospitals, in, 296 income of, 67, 68, 69, 83, 154, 257, 260 increase in numbers of, 67, 96, 230 “inducing demand for services,” 269 lacking in population perspective, 56 pay of, West-German, 257 philosophy of, in differing countries, 198 physicians-population ratios Appalachian States, in, 15 Belgium, Denmark and The Netherlands, in, 234 Canada, in, 65 French and American, 121, 148 in 13 countries, 11 Swedish, 213 practice characteristics, 150 profile of, West German, 255 registered with French Social Secu- rity, 189 remuneration, West-German, 287 Romania, in, 322 services of, 148-157, 152, 153 costs in Britain, 95 expenditures for, 34 expenditures in United States, 124, 126 expenditures on, French, 129, 133 hospitalization in France and the United States, and, 133-157 paramedical assistance, and, 34 rate of change in, comparisons, 156 Swedish, 208 utilization comparisons, 155 shortages of, 40 specialization of, 34 specialty distribution of, 41 strike talk in Canada, 70 Physicians (cont. ) surplus in Canada, 65, 72 Swedish, 19, 206, 212, 213, 221 United States, status in, 287-288 visits of, American, 37 length of time of, 154 West-German, 269 expenditures on, 278 West—German insurance system, in the, 269 workshop, hospital as a, 33 . Physiotherapists, numbers of Swedish, 221 Placebos use in France, 196 Planning British, 94 health, 349 Swedish, 349 Political organization comparisons, France and the United States, 117, 119 Political structure, Sweden’s, 200-201 Pollution, air, and deaths in New York City, 42 Polyclinics, Romanian, 326 Poor and aged in United States, medical-care expenditures for, 37, 38 Poor and nonpoor, medical services utilization of the, 39 Poor Law, Elizabethan, 291 Poorer communities and the health services, 8 - Population changes in British National Health Service staffing, 100 Population comparisons in France and the United States, 117 Population-doctors ratio in 13 coun- tries, 11 Population growth in Belgium, Den- mark and The Netherlands, 233-234 Population-health care personnel com- parisons, 5 Population-hospital beds ratio, 13 Population increase, Romanian, 316 Population increases and health ex- penditures, 13 ‘ Population-nurses ratio, 12 Population perspective and physicians, 56 Population-physician ratio Austria, in, 213 Sweden, in, 213 Populations, French and American, data on, 126-127 Postulated determinants and utiliza- tion variables, 87, 88-90 Practice characteristics, comparisons, physicians’, 150 Practice statistics of West-German physicians, 255 Prepaid health center care, 44 Prescription charges in Belgium, Den- mark and The Netherlands, 238 Prescription drug markets in Canada, 73 Prevalence screening in Romania, 328 Preventive measures French, 162 West-German, 263, 293 Preventive medicine British National Health Service, and the, 101 , Cinderella status of British, 111 expenditures on, French, 180 Preventive services costs in Denmark, 241 Price Formation in Market for Physi- cians' Services, 77 Price-index movements, secular, American and West German, 280- 281 Prims, Andre 233-242, 343, 349, 356 Princeton University, study conducted at, 249 Private and public sectors‘ partner- ship in United States, 231 Private expenditures for health care, charted, 38 Private financing of health services in Sweden, 209 Private good in a public sector, theo- retical note on, 224-227 Private health-care expenditures in Canada, 59, 60 Private health insurance in West- Germany, 271, 273 Private health services consumption in Sweden, 203 Private insurance in French health care, 177, 181 Private market in American health care, 23 Private outlays in American medical market, 23 Private ownership hospitals in West Germany, 259 Private practice French National Health Consump- tion, and 172 physicians in, West German, 269 Private sector as a safety valve, 231 Procedures, new medical, comment on, 32 377 Proceedings of earlier health eco- nomics conferences, xv . Producer choice and consumer choice, control of, 113 Production resources index, 143 Productivity indicators, hospital, 110- 111 Professional Standards Review Or- ganization, U.S., 47 Professionalism among physicians, 33 Professionals, income of Canadian, Providers of health services in West- German system, 268-271 Provinces’ role in Canadian health care, 61, 62 Provincialization of hospitals recom- mended in Canada, 71 Psychiatric care, long-term, in Swe- den, 199 , Psychiatric hospitals’ bed ratio, 140 Public assistance concept, 291 Public expenditure in Britain, analysis of, 95 Public funds in French health-care system, 182, 194 Public Health, Romanian, 329 Public-private medical-care program proposals, 47 Q Quality of care and costs relationships, 238 Questions asked of conference papers‘ authors, xvi Quinn, Joseph R., vi, xiv, xviii R Rational health services system, a, comments on, 231 Rationing problem relevant to health care in Sweden, 226 Rationing resources by time, 348 Ratios used in health-care systems comparisons, 116 Reforms in health-care systems, 231 Regionalization of facilities, 349 Regulatory mechanisms against medical-care costs inflation, 43 Reimbursement, prospective, 351 Reinhardt, Uwe, xiv, xviii, 249-290, 341, 352 Remuneration of medical staff in Romania, 321-322 Renal dialysis in Britain, 102 Research agenda for, 357-358 funds and the universities, 244 government-financed, 245 378 Researchers, American, and the west- ern European nations, 229 Resort area hospitals, West-German, 259 Resource intense techniques in hos- pital, 32 Resources, health allocation mechanism missing in United States, 55 allocation of, 8 control of, 346-353 irrationalities in provision of, 9 rigidity of use of, 9 Resources, national, French, and health consumption, 169 Responses, national, to increases in health-care costs, 16 Retired persons’ health insurance, West-German, 266 Review mechanisms in control of infla- tion, 43 Revenue expenditure of British Na- tional Health Service Hospitals, 98 Rexed, Bror, 199 Rice, Dorothy P., 23-53, 55, 341 Rivlin, Alice, 356 Romania Academy of Medical Sciences in, 326 budget structure in, 319 capital costs in, 319 chronic diseases costs in, 325 chronic diseases therapy in, 324 drug cost indexes in, 323 examinations and treatments in, 331, 332, 333, 334 health expenditures per inhabitant in, 316 health needs in, 327 hospital beds in, 317 hospital building in, 320 hospitalization days in, 336 increases in national factors in, 316 investments cost in, 317-320 laboratory tests in, 338 Ministry of Health of, 324 morbidity in, 328 pharmaceutical costs in, 318 polyclinics in, 324 prevalence screening in, 328 remuneration system change in, 326 screening in, 329 social welfare in, 315 staff payments in, 321-322 treatment costs in, 324 Roemer, Milton, 350 Roemer’s Law, 344, 350 Rosch, Georges, 115-160, 161-195, 341 Rosenthal, Gerald, 350 Royal Commission recommended Canadian national health insurance, 64 Ruderman, A. Peter, xiv, xviii, 109, 196-198, 343, 345, 356 S Salaries, health sector, in Sweden, 212 Salaries in Canadian hospitals, 78 Salary and wage gains of hospital em- ployees in Canada, 69 Salary increases in health service costs, 13-14 Salary problems, Canadian physi- cians’, 86 Salary versus fee-for-service debate, Canadian, 86 Salkever, David, xiv, xviii Sandier, Simone, 115-160, 246, 340, 341, 342, 345, 356 Saskatchewan Medical Care Insurance Commission, administrative costs of, 271 Saskatchewan, school dental care scheme in, 72 Schicke, Romuald K., 291-299 Schools, medical, funding of, 40 Schools, medical, Swedish state owns all, 230 Schultze, Charles, 347 Schweitzer, Stuart, xiv, xviii Science, medical, its influence on costs of care, 14 Scientific base of American medicine, 24 Scitovsky, Anne, 347 Screening in Britain, 348 Screening, multiphasic, in Romania, 329 Secular movements in West-German health care, 277, 278, 299 Self-government in Canadian health services, 61 Services, health See Health services Services, physicians’, 148-157 technology, and, 34-35 Shapiro, S., 340 Short-term and long-term hospitaliza- tion, U.S., 137 Sickness benefit system, Sweden’s, 207 Sickness, costs of, in West Germany 294 children’s Sickness insurance, West—German, 263 Skin cancer, medical care and, 41 Social benefit scales in Sweden, 207 Social benefits in French health-care system, 173 Social care investment in Romania, 317 Social Democratic Party, Sweden’s, 200 Social Health Insurance System of West Germany, 254, 263, 264 Social insurance, Romanian, 315 Social Insurance System, Swedish, 207 Social Policy, 57 Social sciences and medicine’s rela- tionships, 1 Social Security, French, 166, 177, 181, 189, 198 as a funds source, 174-175, 176 control of market mechanisms and the, 193 extension of coverage of, 192 hospital care costs, and, 141 physicians and dentists’ registration in, 189 Social Security funds and health—care expenditures, 136 Social Security, West-German, 264 Social welfare risks and responsibili- ties, 291-292 Social welfare, Romanian, 315 Social workers, Swedish, 221 Socialized medical-care system, ob- jectives of Britain’s, 109-110 Somers, Herman, 2, 347 Sources of private insurance com- panies, West-German, 273 Specialization of physicians, 34 Specialized techniques used in United States, 158 Specialties, British, expenditures by, 112 Spending and priorities patterns in British National Health Service, 98 Staff, Catholic order Sisters in, 343 Staff composition, medical, 296 Staffing of health-care facilities, 5 Staffing of hospitals, American and West-German, 262 Stahl, Ingemar, 199-228, 229, 341, 343, 346, 349, 353 State and counties’ cost sharing in Sweden, 207 State, local and federal health ex- penditures, illustrated, 36 379 State and local authorities in French health-care system, 177, 181 States” autonomy in United States, 117 Statistical perspective on Canadian system, 63-70 Statistics, mortality, 127 Statistics on French and American health-care systems, 115 Stays in hospital, comparisons, 296 Stomatological examinations in R0- mania, 335 Stress and economic recession, 43 Students, medical, increases in, 40 Substitute Funds, West-German, 265, 268 Superspecialized hospitals, Romanian, Supply and demand in American med- ical system, 24 Supply factors affecting health-care costs, 158 Supply side management in medical care, 74 Supply side of medical care, possibility of changes in, 86 Surgeons, income of Canadian, 83 Survey of four European countries, 243 Surveys, household, data from, 115 Surveys, medical expenditure, French, 166 Sweden bed-days increase in, 214 compulsory insurance in, 201 compulsory sick-benefit scheme, 211 consultation fees in, 206 cost-sharing in, 202 costs of health-care services in, 223 counties’ responsible in, 201 county taxes in, 207 decision-making in health care in, 200 dental services in, 205 dentists in, 221 drugs costs in, 205 drugs payments in, 202 drugs prescribing in, 206 economic aspects of health-care system of, 203 economist examining the health services of, 229 employment in health sector in, 220 excessive demand for services in, 205 financing health care in, 205-211 growth in health services of, 229 health-care funding in, 16 380 Sweden (cont.) health-care system size determina- tion in, 224 health consumption in, 217, 218 health costs and expenditures in, 199-228 health planning in, 349 home visits by physicians in, 205 hospital building in, 204 hospital care free in, 205 hospital costs in, 211 hospitals’ running costs in, 223 households’ payments in, 208, 209 inadequate meeting of some needs in, 200 inpatient care in, 222 insurance premiums in, 211 insurance system figures in, 210 investment budgets in, counties, 213 investments in health sector of, 204— 205, 219 liberal economists, and, 201 long-term goals of health care in, 200 medical care administered by coun- ties in, 211 medical care statistics for, 215—227 medical education in, 213 medical personnel statistics of, 221 monopsonistic situation in, 343 national accounts of, 212 nursing home charges in, 205 old people increasing in, 199 organization of health-care system, 211 organizational structure of, 200 outpatient care in, 214, 223 outpatient centers expansion in, patients fees in, 209 pharmaceutical companies in, 202 pharmacists in, 221 physicians in, numbers of, 213, 355 physicians increasing in, 230 physicians’ salaries in, 206 political structure of, 200-201 projected numbers of doctors in, 19 psychiatric care in, long-term, 199 rationing problem relevant to health care in, 226 salaries in health sector in, 212 sickness benefit system of, 207 sickness, long-term, in, 208 Social Insurance System of, 202 taxes in, 201, 207 travel costs for health care in, 205 voluntary health insurance in, 208 System reform tolerance levels, 231 T Tax deductions and Medicare, 347 “Tax illusion" and the British National Health Service, 103, 105 Tax—financed health system, Britain’s, 103 Taxation, general, and health-care ex- penditures, 16 Taxes in Sweden, 201, 207 Teaching, French medical, 163-164, 197 ' Technical factors and health-care cost increases, 158 Technology changes in and hospital western European, 239 hospital costs, and, 32-34 hospital inflation and, 30 improved, inflation and, 35 medical, and changes in society, 48 medical, impact of, 33 physicians’ services, and, 34-35 Theories of inflation of hospital costs, 30 Tilburg University, The Netherlands, 245 Titmuss’ anecdote re treatment in Britain, 57 Tolerance thresholds of young in Britain, 103 Trades unions, 348 Training and education, medical, 40 Transport of patients in French health- care system, 162 Treatment ambulatory 32 costs, versus inpatient, costs in Romania, 324 in British National Health Service, anecdote of Titmuss, 57 Trends expenditures, in, 341 Romanian, 325 French health care, in, 183, 184, 185, 186, 190, 191 health-care systems of France and the United States, in, 123 health costs and expenditures in Belgium, Denmark and The Netherlands, in, 242 health costs, in, Romanian, 328 Trivial ailments and doctor visits, 105 Tuberculosis visions, 14 Tuberculosis British, 113 and health-care pro- prevention program, U Underutilization of hospital facilities in US, 17 United Kingdom See Britain, British United Mine Workers’ provision of medical care, 44 United States access to medical care in, 57 Canadian lessons for, 74-75 costs of physicians’ services in, 155 death in causes of, 42, 282 dentists’ services expenditures in, disease control in, 41 drug expenditures in the, 26 Economic Stabilization Program in, 26 effect of increased medical care in, 41 expenditures trends in, 143-148 funds sources in, 126 GNP and health-care expenditures in, 135 GNP and health costs in, 26, 339, 340 GNP in 1973, 119 health-care system compared with French, 115-160 health-care study model for the, 197 health costs and expenditures in, 249-290 hospitalization costs in, 141 hospitalization expenditure increase in, 123 hospitals and beds in, 261 hospitals in, compared with West- German, 259-260 hospital services in, 138-139, 144- 145 immunization in, 41 income of physicians in, 259 infant mortality in, 42, 279, 283 medical-care policy in, 57 medical economy, changes in, 47 “medical foundation” in, 270 medical market in the, 23 national health expenditures in, 24, 25, 274 381 United States (cont) national health insurance, and, 287 need laws in, 350 personal health-care expenditures in, 26, 124-125, 133 physicians and medical services in, comment on, 287 physicians’ distribution in, 256 physicians‘ income in, 259 physicians-population ratio in, 148 physicians’ services in, increase in variables in, 15 policy makers of, and West-German example, 288 price-index movements in, secular, 280-281 public policy and health—care financ- ing in, 46-47 regarded as “medical jungle”, 74 remuneration in, 295 rising costs of health care in, 24 scientific base of medical system in, 24 specialized techniques used in, 158 trend toward public sector in, 231 US. Social Security Act, 23 Universities and research funds, 244 University duties of Romanian physi- cians, 322 University personnel factors, 55 Untreatable cases and modern tech- nology, 33 User charges, 346-348 Utilization of health-care services, behavior in, 68 Utilization of hospitals, comment on Canadian, 81, 83 Utilization of physicians’ services in France and the United States, 155 Utilization variables’ correlation coefficients, 87, 88-90 and prestige V Vaccination in Canada, 352 Van Praag, Professor, 246 Veterans’ Medical Aid, French, 189 Voluntary health insurance, Swedish, 208 Voluntary insurance, 347 Voting behavior and the Swedish health-care system, 225, 226 382 W Wage and price controls and health care, 32 Wage increases and hospital costs, 31 Wage increases of Canadian hospital employees, 69 Wage rates and inflation in Canada, 67-68 Wages of health service personnel, 13, 14 Waiting time rationing and, 348 Sweden, in, 213 Ward staff-patient ratio in United Kingdom mental institutions, 15 Weisbrod, Burton, xiv, xviii, 346, 356 Welfare services, sources of finances of, British, 104 West Germany alcoholic beverages expenditures in, 293 children’s medicine in, 263 contributions to insurance system in, 271 copayment feature in, 294 cost sharing in, 285 costs of health services in, 293 coverage by health insurance in, 264 death in, causes of, 282 distribution of physicians in, 256 education and research expendi- tures in, 274 education, medical, in, 286 expansion of services in, 293 expenditures for health in, 294 expenditures growth in, 277, 278, 285 financial aspects of hospitals in, 262 financial parity with American phy- sicians, and, 257 financing health scheme in, 292 funds in private health insurance in, 271, 273 funds of compulsory insurance in, 272 goods and funds flow under insur- ance system of, 276 growth in hospital sector of, 296 health-care system of, 249 health costs and expenditures in, 249-290 health insurance expenditures in, 297 health-insurance system in, 262- 273 West Germany (cont.) health services expenditures in, 273-284 health services providers in, 268 Hospital Finance Act of, 262 hospital, length of stay in, change in, 277 hospital physicians’ pay in, 257 hospital sector organization in, 259-262 hospitals and beds in, 261 hospitals staffing in, 262 infant mortality in, 279, 283 insurance benefit package in, 267- 268 insurance enrollment requirements in, 265-267 nst of insurance benefits in, 267 manpower in, adequate, 284—285 Mariners Health Insurance Funds in, 266 monopoly of medical profession in, 287 national health expenditures in, 274 physicians in, 255 physicians’ incomes in, 260, 286, 287 preventive care in, 263 preventive measures in, 274, 293 price-index movements in, secular, 280-281 private health insurance in, 271, 273 private practice physicians in, 269 profession responsible for medical services provision in, 270 retired persons insurance in, 266 social insurance prevailing in, 291 Substitute Funds in, 265, 268 West-Germany (cont. ) survey in, 243 system’s similarity to Canadian, 271 tobacco expenditures in, 293 treatment expenditures in, 274 Western European nations “ne- glected” by American researchers, 229 Wilson, Douglas, 23-53 Work hours and health expenditures, 3, 6 Work-house concept, British, 291 Work-loss days, disability, United States’ and West-German, 282, 283, 284 “Work to rule” by British consultants, Workers’ absence through illness in West Germany, 279, 283, 284 Workshop, hospital as a physician’s, 33 World Health Organization, the, v X X-ray examinations increase, 143, 148, Y Yerby, Alonzo, 2 Z Zubkoff, Michael, 339, 340, 346 383 cnaamabaaa