WBUC
HEALTH
16
°rl^	O.b. depository
INDEX S€pi^!979
U.S. DEPOSITORY
M - SIKBSISILfflSI® sep i (•1979
Utfealth Planning
Bibliography
A Review of Planning Methods
and Criteria for Neonatal
Intensive Care Units:
An Annotated Bibliography
of the Clinical and
Health Planning Literature
Series
U﻿HEALTH PLANNING SERIES
The Bureau of Health Planning is a primary resource for
current information on a wide variety of topics related
to health planning. To facilitate the dissemination of this
information to health planners, the Bureau issues publications
in the following series:
Health Planning Methods and Technology
This series focuses on the technical and administrative
aspects of health planning. Included are such areas
as methods and approaches to the various aspects of the
health planning process, techniques for analyzing health
planning information and problems, and approaches to the
effective dissemination and utilization of technical
information.
Health Planning Information
This series presents information on the analysis of
issues and problems relating to health planning including
trend data, data analysis, and sources of data to support
health planning activities.
Health Planning Bibliography
Bibliographies on specific health planning subjects are
published in this series. Subject areas are selected
by the frequency of inquiries on specific topics and from
suggestions by Bureau staff and health planners throughout
the Nation.
"A Review of Planning Methods and Criteria for Neonatal Intensive
Care Units:	An Annotated Bibliography of the Clinical and Health
Planning Literature" is the fifteenth publication in the Health
Planning Bibliography Series. A list of all publications in the
Health Planning Bibliography Series appears on the inside of the
back cover of this publication. Documents can be ordered from
the National Technical Information Service (NTIS), 5285 Port Royal
Road, Springfield, Virginia 22161.﻿A Review of Planning Methods
and Criteria for Neonatal
Intensive Care Units:
An Annotated Bibliography
of the Clinical and
Health Planning Literature
Prepared under
Contract No. HRA 231-77-0108
by
Boston University
Center for Health Planning
Boston, Massachusetts
August 1979
HRP-0301601


L
U S DEPARTMENT OF HEALTH. EDUCATION. AND WELFARE
Public Health Service
Health Resources Administration
Bureau of Health Planning
National Health Planning Information Center .
DHEW Publication No. (HRA) 79-14038
5﻿t>344-‘-fz.!2_
PUBLIC
HEALTH
LIBRARY﻿FOREWORD
The major purpose of this annotated bibliography is to
assist State and local agencies in the planning and review of
neonatal intensive care units. It covers the important background
and more recent clinical and health planning literature as well
as the National Guidelines for Health Planning. A subject index
which identifies areas of frequent interest for planners has
been included to assist the reader in the use of this material.
Neonatal intensive care units are a widely accepted although
relatively new institutional resource for the optimal manage-
ment of high-risk newborn infants. These special care units
require highly trained personnel, specialized equipment, and support
services which are not available or appropriate for all insti-
tutions. It is, therefore, especially important to determine the
number and types of patients, appropriate setting, and geographic
location for each unit in an area to ensure appropriate utilization
and provide the most effective services both in terms of patient
care and costs.
The material for this bibliography was prepared for the
Health Resources Administration as part of the work under Contract
No. 231-77-0108, A Review of Planning Methods and Criteria for
Neonatal Intensive Care Units.
Colin C. Rorrie, Jr., Ph.D
Director
Bureau of Health Planning
iii﻿OnifDOTEOS
This report was prepared by the Boston University Center for Health
Planning. The Project Director was Charles L. Donahue, Jr. The principal
staff members responsible for writing this document were Marc Wine and
Saul Spivack. Other staff members involved in this project were Rosemary
Chiusano, Cynthia Brown and Paula Dolan.
The Boston University Center for Health Planning would like to thank
the Health Resources Administration Project Director, Alan L. Pinkerson, M.D.,
for his advice and support throughout the project. The Center would also
like to thank the following individuals for their review and comments on
the draft report: Leo Stern, M.D., Chairman of Pediatrics, Brown University
and Pediatrician-in-Chief at Rhode Island Hospital; Mary Terrell, Office of
State Health Planning, Massachusetts Department of Public Health; Mark
Chambers, Southeastern Wisconsin Health Systems Agency, Inc., John Suskiskis,
New Jersey Department of Health; Charles Kratt, Central Jersey Health Planning
Council, Inc.; Donald S. Shepard, Ph.D., Adjunct Research Associate, Kennedy
School of Government and Lecturer in Health Services, Harvard School of
Public Health; Joe Liberatore, Health Systems Agency of San Diego and Imperial
Counties; Andrew Fleck, M.D., M.P.H., New York State Department of Health.
iv﻿PREFACE
One of the Section 1502 priorities of Public Law 93-641 calls for the
'development by health service institutions of the capacity
to provide various levels of care (including intensive care,
acute general care and extended care) on a geographically
integrated basis."
Perinatal health services have been chosen for special attention in
the development of these regional health systems. Standards respecting the
appropriate supply, distribution, and organization of health resources have
been developed for obstetric and neonatal intensive care services. These
standards are required by Section 1501 of the National Health Planning
and Resources Development Act as a part of the National Guidelines for
Health Planning. Health Systems Plans established after December 31, 1978,
must be consistent with these resource standards.
The standards for neonatal intensive care require that the services be
planned on a regional basis for linkages with obstetric services. The
population-based need should not exceed four intensive and intermediate
level beds per 1,000 live births per year for a defined neonatal service
area. An adjustment of this rate upward is justified when the rate of high
risk pregnancies in an area is unusually high. A single Level II or Level
III neonatal intensive care unit should contain a minimum of 15 beds. An
adjustment of this standard downward may be justified for a Level II or
Level III unit when geographic remoteness makes travel time to an alternate
unit a serious hardship. These resource standards are largely based on the
report "Towards Improving the Outcome of Pregnancy." This report was
issued by the Committee on Perinatal Health, representing the American
Academy of Pediatrics, the American Academy of Family Physicians, the
American Academy of Obstetrics and Gynecology, the American Medical Associ-
ation, and the National Foundation-March of Dimes.
Health planning agencies could benefit from technical assistance materials
as they try to implement the changes called for by the Standards of the
National Guidelines for Health Planning. Health planning agencies need assist-
ance in some of the following areas:
« in understanding the clinical and health planning background
literature
o in understanding how neonatal intensive care units relate to a
perinatal regional system
o in understanding the use of and adjustments to the Standards of
the National Guidelines for Health Planning
v﻿•	in understanding the methods and techniques for planning of
neonatal intensive care units
•	in organizing the planning process for the formal designation
of hospital units in terms of level of care and service areas
•	in educating those involved in the planning process and those who
will be affected by it
This Volume, An Annotated Bibliography of the Clinical and Health Planning
Literature, should assist health planning agencies in their literature
reviews. An effort was made to abstract information that would be of assist-
ance to health planning agencies. The focus of the review was on literature
relating to the planning of neonatal intensive care units. It is not a review
of the perinatal planning literature.
The clinical articles were abstracted from books and journals. The
health planning documents were largely unpublished. Each health planning
agency in the United States was contacted and asked to submit reports or
other documents that it had developed related to the planning of neonatal
intensive care units. These documents were the basis of the health planning
reviews.
The documents are separated into two groups. The clinical literature
is first and is followed by the planning literature. The documents in
each section are organized alphabetically according to the name of the lead
author. A listing of all articles is in the bibliography. The index is
organized by subject.
vi﻿CONTENTS
Foreword iii
Acknowledgments iv
Preface v
Chapter 1. Clinical Literature 1
Chapter 2. Health Planning Literature 116
Bibliography of Clinical Literature 181
Bibliography of Health Planning Literature 188
Index 191
vii﻿
﻿CHAPTER 1
CLINICAL LITERATURE﻿


﻿Alden, Errol R., M.D., Ted Mandelkorn, M.D., David E. Woodrum, M.D., Richard
P. Wennberg, M.D., Colby R. Parks, M.D., and W. Alan Hodson, M.D.
1976 "Morbidity and Mortality of Infants Weighing less than 1,000 grains in an
Intensive Care Nursery". From the Division of Neonatal Biology, Department
of Pediatrics, School of Medicine, University of Washington, Seattle,
Washington.
GENERAL ABSTRACT:
A five-year experience with 161 infants with birth weights less than 1,000
grams was evaluated to ascertain predictive factors for mortality, cause of
morbidity, and status of survivors. There was an overall mortality of 87%
which was worse if the initial hemocrit was less than 40%, the blood pressure
was less than 40 cm H20, the Apgar score was low, or if hyaline membrane
disease was present or assisted ventilation was required. The survival rate
was improved by two-to threefold if the Apgar score was greater than six and
if the infant required no resuscitation. The survival rate was two to three
times normal if rupture of the amniotic membranes preceded labor and if the
infant was able to maintain his core temperature. Respiratory difficulties
accounted for the highest incidence of morbidity:	apnea 84%, hyaline membrane
disease 66%. Only two infants of the twenty surviving had definite abnormal
developmental quotients. Intensive care does not appear to have affected
survival but may be responsible for an improved outlook for survivors.
ANNOTATIONS:
Hyaline membrane disease, apnea, respiratory failure, and infection are
the most common causes of morbidity in this weight group. Sixty-six percent
of all infants had hyaline membrane disease and only 5% survived. The occur-
rence of apnea had no predictive value for survival but was the most common
problem (84% of all infants). There was an association between the time of
onset of apnea and mortality. Apnea occurred in the first day of life for
nonsurvivors but in the third day of life for those who lived. There were no
survivors among those infants who required positive pressure ventilation
because of hyaline membrane disease. However, infants ventilated because of
late occurring apnea (greater than 24 hours) did occasionally survive. Twenty-
three percent of the survivors received positive pressure ventilation because
of apnea. These infants were ventilated at a much later age (234 hours) than
those who died (38 hours).
American College of Obstetricians and Gynecologists
1978 Health Care for Mothers and Infants in Rural and Isolated Areas.
Chicago, IL:	American College of Obstetricians and Gynecologists.
GENERAL ABSTRACT:
This committee was to organize expanded work groups and produce a document
which could help health professionals, State agencies, legislative bodies, and
1﻿consumers improve the quality and availability of health care services in
rural and isolated areas. The work was to be based on the premise that
mothers and infants in rural and isolated areas should have the same opportu-
nities for care as those in more centralized and regionalized urban areas.
Areas studied by the committee were Appalachia and rural south, the Rocky
Mountains, the Great Plains, and Western Indians. The Western Indian topic
was limited to Indians in the northern great plains and the upper southwest
part of the United States. The problems of the upper northeast, the Pacific
northwest, and other isolated areas were felt to be similar enough to the
four designated study areas to allow easy adaptability. Hence, this paper
summarizes the activities of this group and presents the model designed to
link the rural and isolated population of the United States with the peri-
natal health system.
ANNOTATIONS:
These annotations are meant to be a guide to the specific contents of
the paper. These are only general statements concerning planning standards
and more specific area-facility recommendations are within the body of the
report.
General Principles
In designing the perinatal care system for rural and isolated areas, the
following were identified as guiding principles:
•	Perinatal health care systems should be designed around the culture
and resource strengths of the area. The area should be assessed
to determine cultural background of the people and health care
resources available before any system can be formulated.
•	Once the strengths have been outlined, the system should build
upon them and eliminate severe deficiencies. It will interface
with the population's present system of values.
•	Rural perinatal health care systems should be built around a model
that can be expected to provide total health care.
•	Some agency or organization should provide manpower, materials
and methods to insure a proper foundation for a successful
functioning system.
•	Providers within the health care system should be given responsi-
bility in relationship to the training they have received.
•	The financing of the system should be shared by the population
at large. A larger proportion of costs should be shared by
the total population as the care increases in complexity.
2﻿Data System
The multiple data sources mandate that support for the information system
be widespread. The administrators of the perinatal program and the evaluation
team ultimately are responsible for the design and implementation of an
effective information system. The system should be capable of telling the
range of providers, higher level administrators, legislative bodies, and
other funding sources, and the public whether the program is attaining its
intended objectives. Examples of Indicators:	sociodemographic characteris-
tics; reduction in percent of women with unfavorable course during pregnancy;
reduction in neonatal morbidity; compliance with objectives; long-term reduc-
tion in perinatal mortality and morbidity. Vital statistics registration and
reporting system include fetal and neonatal death records and birth records.
The specific role of the primary perinatal care provider is to:
•	Supervise/coordinate with health advocates who are linked with
the patients for whom responsibility exists.
•	Provide service needs-
•	In cooperation with the health advocate, determine if educa-
tional needs are met.
•	Participate in evaluation.
Support needed by the primary perinatal care system is:
•	Education in working with health advocates.
•	Automated methods of screening, monitoring, communication,
and decision making to provide the best perinatal informa-
tion to the patient.
•	A mobile unit or neighborhood clinic which is easily accessible
to the patient.
Insurance
Insurance coverage for perinatal care has often been inadequate, especially
in meeting the costs of transport and care of high-risk patients. Insurance
companies should be required to offer comprehensive coverage for the perinatal
care consumer.
Coordination and Consolidation Facilities
Protocols to merge or close small inefficient hospital obstetrical units
may be inconsistent with rural needs, particularly in view of great distances
and difficult weather. Quality professional care and convenient, attractive
facilities at low costs are important. Centralization of delivery services
in sparsely populated areas may prove counter-productive.
3﻿There is a lack of coordination and sometimes duplication of services
in existing facilities. Satellite facilities should be established and
linked to existing facilities to help improve the quality and accessibility
of care. Consolidation of laboratory facilities across regional and State
lines for mass screening and technically difficult procedures may provide
better service.
Families are often uprooted from their homes and support systems. Need
of these persons include emotional support, financing, child care assistance,
temporary housing, and tertiary level health care. Housing facilities are
needed near the delivery center and/or tertiary care center.
Laboratory Support Capability
At Level I hospitals, a portable x-ray capability within the nursery is
considered ideal, but not mandatory. Semi-microlaboratory techniques should
be available for blood glucose, hematocrit, electrolytes, calcium and bilirubin.
Ultrasound capability is not essential at the Level I hospital, but must be
available at Level II and III centers.
Personnel
Recommendations in regard to Level I unit personnel include:
•	To make consultation readily available and to maintain coverage
of the obstetrical services, there should be at least two ob-
stetricians and up to five nurse midwives or physicians with
obstetrical training and experience.
•	To maintain anesthesia coverage, there should be a minimum of
two anesthesiologists or nurse anesthetists.
•	There should be the capability of performing delivery by Cesarean
section within at least 30 to 60 minutes when the need occurs.
•	Every Level I unit should have a registered nurse on duty at all
times whose primary responsibility is the organization and super-
vision of nursing services in the labor/delivery area, the new-
born nursery and postpartum units. The nurse should possess the
knowledge and clinical competency in the nursing care of mother,
fetus, and infant during labor and delivery and in the postpartum
and neonatal periods.
•	Facilities with less than 500 deliveries a year should have a
staff of at least one professional nurse per shift in the ob-
stetrical newborn unit. Where there are over 500 deliveries
per year, there is a need for at least two professional nurses
per shift for attendance during delivery and the immediate
recovery period.
Minimal standards of proficiency for health care personnel and providers
should be maintained. This must be a priority matter, superceding any
political issues. Accreditation and registration standards, plus continuing
education standards, must be applied, specifically, Joint Commission on
Accreditation of Hospitals' standards.
4﻿American College of Obstetricians and Gynecologists
1973 Memorandum to the Committee on Health Care Delivery from Ervin E. Nichols, M.D.,
FACOG, Associate Director; Maternity and Neonatal Services and Facilities
Guidelines for Levels of Complexities of Care. Chicago, IL:
American College of Obstetricians and Gynecologists.
GENERAL ABSTRACT:
The purpose of this document is to offer broad categories, or levels of
complexity of care and guidelines as to the types of patients to be cared for
and the services and facilities required at each level. The levels of care
are Level I, II, and III for maternity and newborn care.
ANNOTATIONS:
Facility Requirements for Neonatal Intensive Care:
•	Facilities for immediate resuscitation for the newborn to include
short-term assisted ventilation, cardiac massage and infusion of
alkali. This should ideally occur in the delivery room. Minimum
requirements are about 40 sq. ft. of space in the delivery room.
Resuscitation equipment should include laryngoscopes, endotracheal
tubes, resuscitation table with radiant heating and infusion equip-
ment.
•	Short-term assisted respiratory capability should be available to
enable stabilization of the infant and preparation for transfer
when indicated.
•	Intravenous therapy with infusion pumps.
•	Assisted ventilation on short-term basis with bag and mask.
•	Treatment of infections, except those concerning and affecting the
central nervous system.
•	Exchange transfusion team.
•	Twenty-four hour clinical laboratory facilities to include micro-
blood gases, bilirubin and blood sugars.
•	Twenty-four hour radiology services for the newborn.
•	Ultraviolet light therapy for hyperbilirubinemia.
5﻿American College of Obstetricians and Gynecologists
1971 National Needs in Obstetrics and Gynecology. Chicago, IL:
American College of Obstetricians and Gynecologists.
GENERAL ABSTRACT:
This paper summarizes the major national problems in obstetrics and
gynecology. Also, it suggests some viable approaches to solutions. The
statements herein may be useful as a general guide for introducing and
identifying problem areas for planning.
ANNOTATIONS:
The ACOG has adopted these recommendations for OB/GYN care in hospitals:
•	The need for hospital obstetric services should be identified by
geographic areas and related to population density.
•	Hospital obstetric services should be consolidated. In larger
communities (100,000 population or more) more than 1500 deliveries
should occur yearly in each unit. In smaller communities (30,000
population or more) more than 500 deliveries should occur annually
in each unit. In rural areas, obstetric services should be coordi-
nated and consolidated as much as possible.
•	Provision should be made for the transfer of patients with major
complications from the smaller to the larger units where facili-
ties for special care are available.
•	Regional units for high-risk OB patients and newborns should be
developed and evaluated.
The following recommendations are offered for financing expensive OB/GYN
services:
•	OB/GYN services should be provided for medically indigent women by
insurance payments made on their behalf and realistically calculated.
•	Differential reimbursement should be established by third-party
carriers for normal and high-risk OB/GYN patients. Peer review
should be part of the continuing evaluation of such reimbursement.
•	Reimbursement should be made on a realistic basis according to the
relative value of the services provided.
•	Catastrophic insurance for OB/GYN, such as high-risk patients,
should be developed.
•	Reimbursement by third-party carriers for comprehensive OB/GYN care
should include remuneration of services performed by allied health
personnel for normal patients and for education, with well-defined
standards and under medical supervision.
6﻿•	Reimbursement by third-party carriers should be encouraged on a
continuing basis for health maintenance under the team approach,
both in the private and public sectors, depending upon the
personnel necessary and utilized.
•	Traditional methods of payment for services by fees should be
maintained but study and evaluation of new systems should be
continued.
Berger, Gary S., M.D., Dennis B. Gillings, Ph.D., and Earl Siegel, M.D.
1976 "The Evaluation of Regionalized Perinatal Health Care Programs".
American Journal of Obstetrics and Gynecology 125: 924-932.
GENERAL ABSTRACT:
A survey, conducted in the summer of 1974, to determine the extent of
regionalization in perinatal services in the United States revealed 28
states to have programs in operation. About 16 other States had plans to
regionalize services. Evaluation was a common concern of these programs.
A model for the evaluation of Regionalized Perinatal Care Programs is
presented in this article. The North Carolina program is used as an example.
Evaluation was mandated as an integral part of the program in North Carolina,
and this model has been developed in response to that mandate.
ANNOTATIONS:
The report of the Perinatal Mortality Committee of the Province of Quebec
is of particular importance for evaluation studies because of the adequacy of
controls in the analysis. Among infants born in 1967 through 1969 in hospi-
tals registering more than 1,000 deliveries annually, the perinatal mortality
rate per 1,000 was lowest in hospitals with intramural neonatal intensive care
units (14.7), followed in ascending order by hospitals without NICUs, but
which referred infants to the regional centers (16.9), and hospitals which
provided neither intramural nor referral neonatal intensive care (19.1).
The same pattern prevailed for both fetal/neonatal mortality rates, with the
latter controlled for birth weight of the infant (1,000 to 2,500 grams and
more than 2,500 grams). This is the most persuasive report to date from a
methodologic veiwpoint documenting the association between Regional Perinatal
Care and declining perinatal mortality rates.
Predicted Effect of RPCP on Perinatal Deaths in North Carolina
Predictions of the effect of RPCPs on perinatal deaths in North Carolina
indicated that 30-50 percent of perinatal deaths are preventable in theory.
However, analysis of North Carolina's vital statistics suggests that these
predictions of reduction in mortality rate may take several years to achieve,
and targets for the early stages of the program should be less ambitious.
7﻿The lowest rate within each group (see article for data) may be taken
as a realistic target for the early stages of the program. When these
lowest rates are applied to the total population at risk, an estimated 4%
reduction in the perinatal mortality rate is obtained.
Evaluation Plan
The evaluation plan should be flexible. It should measure achievement
of objectives as they may change from year to year, which is likely to happen
as this new program accumulates experience. The evaluation should be
designed not only to determine whether the program did or did not reduce
perinatal morbidity and deaths, but also to study the process by which
regionalized perinatal care is supposed to lead to the desired improve-
ments in perinatal health.
The basic concept of the proposed evaluation design is that the program
will move in small steps towards achieving its desired long-range outcomes
(goals). The type of work (activity) and the amount of that work to be
performed by a certain date (target) constitutes an objective and the totality
of objectives at any one time determines the step. The longer-term, outcome-
oriented goals, which remain fairly constant over, say a five-year period,
are set at the beginning of the program. The shorter-term, process-oriented
objectives are determined more frequently (e.g., yearly) and work toward the
long-term goals. Thus, the objectives defined for any stage in program
development are based on the progress achieved in preceding steps.
Berger, Gary S., M.D., J. Richard Udry, Ph.D., and Charles H. Hendricks, M.D.
1975 "Regionalized Perinatal Care:	An Estimate of Its Potential Effect on Racial
Differences in Perinatal Mortality in North Carolina". North Carolina
Medical Journal 36:	476-479.
GENERAL ABSTRACT:
In North Carolina, mortality rates for non-whites exceed those for whites
at the perinatal period. Effective July 1, 1974, a program of regionalized
perinatal care was undertaken in North Carolina with the goal of reducing
morbidity and mortality through provision of adequate prenatal care and
obstetrical delivery of high-risk patients in regional centers equipped to
handle complications, and through development of regional centers for
intensive care of the neonate. Because of the increased relative risk of
perinatal mortality among non-whites, one result of regional perinatal care
might be to narrow the gap in perinatal mortality between the white and non-
white population in North Carolina.
8﻿ANNOTATIONS:
In 1970-1972, the crude perinatal mortality rate in North Carolina was
25.7 for whites and 41.0 for non-whites. The excess mortality rate for non-
whites is consistent with the pattern that has been documented in all vital
statistics reporting on this.
When controlled for birth weight and length of gestation, perinatal
mortality rates for non-whites were not significantly higher than for whites,
and there were no significant differences in outcome by adequacy of number
of prenatal visits or by type of hospital in which delivery occurred (with
or without an NICU).
These observations suggest that progress in eliminating the gap between
the races in perinatal mortality in North Carolina will require control of
the excess rate of prematurity among the non-white population, rather than
simply providing more intensive prenatal or neonatal care services.
Blake, Anthea, Ann Stewart and Diane Turcan
1977 "Perinatal Intensive Care". Journal of Psychosomatic Research 21:	261-272
GENERAL ABSTRACT:
By 1966, intensive care had been introduced at University College Hospital,
London. From that time, a follow-up study of the survivors has been in progress
and includes repeated physical examinations, psychological and behavioral
testing. Parents, also, have been interviewed about their attitudes and per-
ceptions of their children.
ANNOTATIONS:
The results show that 17 (18%) of 205 infants weighing less than 1500 grams
at birth and now aged 18 months have major handicaps. Among these 205 children,
no difference in major handicap rate was found in infants born at University
College Hospital and those referred from elsewhere; between those whose birth
weight was appropriate for gestational age and those small for gestational
age; or between those who received total parental nutrition and" those who did
not. No difference in outcome was found between infants weighing 500-1000
grams and 1001-1500 grams.
Also, there has been follow-up of other groups of high-risk infants whose
survival rates have increased since the introduction of intensive care. No
major handicap has been detected in 123 infants weighing 1501-2500 grams
examined at 18 months of age. Three (5%) of 64 infants weighing greater than
1000 grams who were mechanically ventilated for hyaline membrane disease be-
tween 1970-1974 were handicapped. None of the 29 infants with very severe
rhesus haemolytic disease (including 16 who were hydropic), born during the
period 1968-1974 have major handicaps.
These results, together with those of other researchers, suggest that
perinatal intensive care reduces the risk of handicaps in infants who survive
serious perinatal problems and does not itself cause serious physical or
emotional sequelae.
9﻿Boehm, John J.
1976 "Identification of the High Risk Infant". Northwestern University Medical
School — Division of Pediatrics^ prentice Women's Hospital and Maternity
Center. Chicago, Illinois.
GENERAL ABSTRACT:
This is a brief summary of criteria, standards and suggestions for action
for identifying high risk pregnancy cases as early as possible. Also, it
contains a statement about transferring these infants.
ANNOTATIONS;
Every hospital, no matter how small or how few infants are delivered chere
each year must be equipped to stabilize all infants during the transition
from intra-to-extra-uterine life. This includes the capability to perform
adequate resuscitation, treat shock, and provide an appropriate treatment
area within the hospital or transferred to a center with facilities for
managing the high risk infant.
In addition, every hospital delivering infants must have a written
procedure detailing the lines of referral for high risk infants.
Bryant, Sandra, R.N.
1976 "Nursing Aspects and Organization for Perinatal Care". Clinics in Perinatology
Organization for Perinatal Care 3:493-496. Guest Editor:	Louis Gluck, M.D.,
Philadelphia:	W.B. Saunders Company.
GENERAL ABSTRACT:
To effectively provide fetal intensive care nursing, it is important to
consider the functions and duties of each category of the nursing team:
(1) admission of the patient to the unit; (2) application of the fetal
monitor and assessment of the data; (3) care of the mother in delivery;
(4) care of the newborn at delivery; (5) postpartum recovery of the
patient; (6) types and numbers of operative procedures done on the unit
other than vaginal deliveries, such as postpartum tubal ligations and cesarean
sections;	(7)	cleaning and wrapping of instruments; (8) extent of patient
teaching;	(9) staff development and evaluation; and (10) antepartum assess-
ment, such as oxytocin challenge test.
ANNOTATIONS:
It is recommended that a hospital delivering 200 patients per month provide
three nurses per shift, preferably registered nurses, and perhaps a fourth
nurse depending upon the health status and number of patients in labor. If a
significant percentage of the patients are known to be high risk, then the
nurse to patient ratio should be increased to 1:1 or 1:2.
10﻿Butterfield, Joseph L., M.D.
1976 "Newborn Country USA". Clinics in Perinatology:	Organization for Perinatal
Care 3:281-295. Guest Editor:	Louis Gluck, M.D., Philadelphia:	W.B.
Saunders Company.
GENERAL ABSTRACT:
The Department of Perinatology of the Children's Hospital in Denver
has adopted the theme of Newborn Country USA in an effort to sublimate
corporate and professional egos to a common commitment to a regional
perinatal care system. Both a marketing concept and a working model
"Newborn Country USA", describes a region and a system.
ANNOTATIONS:
From a geographic viewpoint, Newborn Country USA is one of the largest
perinatal markets in the United States. The vast region is best described
by a line that connects Winnipeg, Minneapolis, Omaha, Kansas City, Oklahoma
City, Dallas, San Antonio, Albuquerque, Tucson, Phoenix, Salt Lake City and
Calgary. The 500,000 square mile territory that lies halfway between Denver
and the next nearest full-service perinatal center is the catchment area
from which Denver draws the bulk of neonatal and maternal referrals.
The Newborn Country USA efforts to upgrade perinatal health care in
rural communities have included:
•	Consultation in defining manpower, facility, equipment,
laboratory and procedural needs in prenatal, intrapartum,
postpartum and neonatal areas. For example, a checklist
of nursery items has been prepared to assist in the survey
of hospitals that request evaluation of their current
capabilities in neonatal care.
•	Participation in regional health planning with emphasis
on the perinatal component.
•	Advocation of the expansion of the EMS to include plans
and protocols for the emergency care of high-risk mother
and/or newborn.
•	Development of outreach perinatal education programs
throughout the region.
•	Acceptance of administrators, volunteers, laboratory
technicians, respiratory therapists, licensed practical
nurses, registered nurses, supervisors and physicians
from both the private practice and public health sectors
as short-term trainees in the regional perinatal center,
and affiliated Level II hospitals.
11﻿• Utilization of residents and fellows in perinatal training
as speakers in traveling clinics and workshops to increase
their awareness of the special health needs in rural
America.
•	Shifting of the public relations philosophy to a "perinatal
partnership" emphasis that underscores the positive role
of the rural health term in recognizing and stabilizing
the high-risk mother or newborn prior to transfer to the
regional perinatal center and the ongoing two-way relation-
ship in consultation, education and planning.
Butterfield, Joseph L., M.D.
1977 "Organization of Regional Perinatal Programs". Seminars in Perinatology:
Regionalization of Perinatal Care 1:217-233. Guest Editor:	Leo Stern, M.D.,
New York:	Grune and Stratton.
GENERAL ABSTRACT:
This article addresses manpower needs of Regional Perinatal programs in
the U.S.
ANNOTATIONS:
Based on Swyer's estimate of one neonatologist for each 2,500 live births,
1,280 neonatologists might be needed to serve the 3.2 million live births
annually in the United States; 301 were listed in the January 1977 report
of the American Board of Pediatrics.
Since the great majority of live births take place in large metropolitan
hospitals, the 534 hospitals reporting in excess of 1,500 deliveries per
year would need at least 1,048 neonatal perinatal medicine specialists.
The remaining 4,500 hospitals in the United States that report live births
would not be strong candidates for full-time neonatal-perinatal medicine
specialists except in isolated geographic areas, where low-volume perinatal
services can be justified. In metropolitan and suburban areas, the hospitals
serving as regional centers will take a variety of patterns in both personnel
and administrative areas.
It would appear timely to recommend updating residency training program
to attain more equitable didactic and clinical teaching of the preventive and
health promotional aspects of primary reproductive care pursued in an ambu-
latory setting. In this context, teaching programs should also include
increased emphasis on the emotional and mental health needs of patients,
greater reliance on the multidisciplinary health team, and more discriminating
use of available diagnostic and therapeutic modalities.
The author states that in order to strengthen the delivery of reproductive
health services greater application of the ambulatory approach is required,
To achieve this end, the obstetrician-gynecologist and the resident in training
must begin to appreciate that comprehensive surgical aspects of care traditionally
performed at the inpatient level, but also that there is an even greater need
for preventive and health promotional activities that can be performed in the
office or other ambulatory settings.
12﻿The emerging role of the obstetrician-gynecologist as a primary care
physician for women interfaces elegantly with the expansion of the ambu-
latory care concept of health care delivery.
Carrier, Charles, M.D., et.al.
1973 Perinatal Intensive Care after Integration of Obstetrical Services in Quebec:
A Policy Statement of the Quebec Perinatal Committee. Quebec, Canada.
Available from:	Quebec Perinatal Committee, Ministry of Social Affairs
Province of Quebec, 1005 Chemin Ste-Foy, Quebec, Canada, G1S 4N4.
GENERAL ABSTRACT:
In this policy statement, the Perinatal Committee of the Ministry of
Social Affairs examines the potential for Obstetrical and Neonatal Care in
Quebec with regrouped obstetrical services. In particular, the question
of minimum size of service necessary to provide a full range of perinatal
services is explored. The possibility of providing both fetal and neonatal
intensive care in each of the newly created enlarged centers is assessed,
and the improved results in perinatal mortality and morbidity that can be
expected to result from such a development are described.
ANNOTATIONS:
Predicted Perinatal Mortality After Obstetrical Integration
In 1970, it is reported here that the perinatal mortality rate in Quebec
for infants of over 1,000 grams was 17.9/1000. This rate was an average of
high rates found in the smallest hospitals, average rates found in larger
hospitals not utilizing neonatal intensive care, and low rates from larger
hospitals utilizing referral or intramural neonatal intensive care.
The 17.9 perinatal deaths per 1,000 can be broken down by cause of
death into those deaths that are potentially reducible utilizing modern
medical technology, and those for which there is little hope of reduction
with the present state of medical knowledge. If all viable births survived,
except for those with lethal malformations and those where the fetus died
before labor of unpredictable or unexplained causes, the irreducible minimum
perinatal mortality given today's level of knowledge is in the order of 8.5
per 1,000 births. This then leaves 9.4 perinatal deaths per 1,000 births in
Quebec in 1970 that were potentially preventable.
In comparing the rate of 11.6 per 1,000 obtained in a hospital where both
Intramural Fetal and Neonatal Intensive Care were available, to the irreducible
minimum rate of 8.5 per 1,000, the potential difference is 3.1 per 1,000
The rate of 14.7 obtained with Intramural Neonatal Intensive Care represented
6.2 potentially preventable deaths per 1,000 births. For those hospitals
utilizing Referral Neonatal Intensive Care facilities, potentially preventable
deaths were 8.4 per 1,000 and for those that utilized no intensive care 10.6
per 1,000.
It is possible then to predict that, after small obstetrical services are
absorbed following obstetrical integration, the number of perinatal deaths
to be expected in Quebec among approximately 100,000 births each year will
depend on intensive care facilities available, in the following manner;
13﻿1910 deaths without intensive care
1690 with Referral Neonatal Intensive Care
1470 with Intramural Neonatal Intensive Care
1160 with Intramural Fetal and Neonatal Intensive Care,
as compared with an irreducible minimum of 850 per year.
Size and Distribution of Obstetrical Services Providing Intensive Care
The authors of this report would consider today that for rational utili-
zation, obstetrical services providing Intramural Fetal and Neonatal Intensive
Care should deliver between 3000 and 5000 infants per year. Above this size,
the high delivery rate makes it difficult to maintain high standards of per-
sonalized care. Below 3000 deliveries per year, there are insufficient high-
risk pregnancies to develop a smoothly-functioning Intensive Care system,
and one which is economically justifiable.
It is noted here that the NICU service of a hospital delivering 5000
babies per year will utilize an average of 25 cots or incubators daily.
Peak load periods may increase this figure by 40 percent, to 35 per day.
Such a unit will treat 1000 patients per year with some degree of enhanced
care. For 3000 deliveries per year, the corresponding figures would be an
average of 15 patients per day with peaks of 21, treating about 600 patients
per year.
The number of obstetrical beds required for 5000 deliveries per year is
90 beds for a mean utilization rate of 85 percent, including the beds needed
for antenatal high-risk patients in addition to the post-partum mothers.
According to this report, 54 maternal beds will be needed to handle 3000
deliveries per year. These calculations are based on a 4-5 day stay for a
normal delivery, with 15% of beds occupied by antenatal high-risk patients.
Equipment needs for fetal I.C. surveillance in such a unit is estimated
to amount to $34,000 (1973).
Consolidating Services for Efficiency in Delivering NICU
Even though personnel may not be immediately available for intensive care
in all integrated obstetrical services, it is desirable to rapidly integrate
maternity services into centers delivering 3000 to 5000 infants a year where-
ever such size hospitals can be developed. In the remaining hospitals, it
will be necessary at first to refer high-risk pregnancies to Perinatal
Intensive Care Centers, and newborns requiring intensive care to Referral
Centers in children's hospitals. The very fact, from a planning view, that
these remaining hospitals, though presently lacking intensive care capability,
will have been integrated into services delivering 3000-5000 infants per year,
will act as a strong incentive for physicians to train in Perinatology and
Neonatology as they will recognize that hospitals exist of sufficient size to
require these services. The lack at the present time of maternity services of
sufficient size to size to support NICU facilities provides a strong disincentive
for doctors to specialize in these fields.
14﻿Carrier, Charles, M.D.
1975 Report of the Quebec Perinatal Committee for Perinatal Mortality for the Year
1970 (First Part):	Quebec Canada. Available from:	Quebec, Perinatal Committee,
Ministry of Social Affairs, Province of Quebec, 1005 Chemin Ste-Foy Ouebec,
Canada, G1S 4N4.
GENERAL ABSTRACT:
The paper is an annual report on mortality by weight group, for the
province, by region, and for hospitals grouped according to size and facili-
ties for intensive care. It is meant to be understood that perinatal
mortality provides a valuable index of the quality of health care for pregnant
women and their newborn babies. Therefore, in studying this mortality, there
is an attempt to find means to reduce death and morbidity rates and to improve
the health of mother, fetus, and newborn infant.
ANNOTATIONS:
Birth Weight Experience
The central committee obtained detailed reviews of individual perinatal
deaths, births, stillborns, neonatal deaths, and weights from each hospital
(Quebec Perinatal Committee).
There were 96,071 total births weighing more than 500 grams during 1970.
Of this number, only 622 weighed 501-1,000 grams, Five hundred and ninety-five
of the 2,311 perinatal deaths were included among these 622 infants. There-
fore, 25.7% of all perinatal deaths occurred in the 0.65% of the births weighing
501-1,000 grams. There were but 27 one-week survivors out of 407 live births in
this weight group, or 6.6%. In addition, only 19 of the 27 who survived one
week left the hospital alive.
For the Province, as a whole, total births over 1,000 grams (including
stillbirths and neonatal deaths) were 95,449 in 1970 as compared to 98,146 in
1969,	98, 729 in 1968, and 103,476 in 1967. The decrease in births over 1,000
grams is, therefore, 7.7% since 1967 and 2.8% since 1969. For the same year,
1970,	the Population Registry reports 92,698 total births, which includes births
of 28 weeks of gestation or more.
In all, there were 96,071 total births weighing more than 500 grams, and of
these there were 1,127 stillborn and 1,184 who died during the first complete
week of life. Total mortality rates over 500 grams were:	stillborn rate of
11.7% per 1,000 total births, neonatal mortality rate of 12.5 per 1,000 live
births, and perinatal mortality of 24.1 per 1,000 total births.
Among the 95,449 births weighing more than 1,000 grams, there were 912
stillbirths and 804 neonatal deaths. The total mortality rates per 1,000 for
infants weighing more than 1,000 grams were:	stillbirth rate of 9.6 neonatal
mortality rate of 17.9.
Size of Obstetrical Service
Perinatal mortality is in general inversely proportional to the size of the
obstetrical service measured by the number of births per year. In 1970, the
lowest rates, 16.9-17.5 per 1,000 births over 1,000 grams, were found in
15﻿hospitals performing more than 1,000 deliveries per year. The highest rates
were those of hospitals delivering 251-500 infants per year, where there were
21.0	perinatal deaths per 1,000 births.
The report shows that this increase in mortality with decreasing size
of obstetrical service occurred in spite of the fact that the incidence of
low birth weight was lower in the smaller than in the larger. The increase
in perinatal mortality was due to a sticking increase in low birth weight
neonatal mortality with decreasing size of obstetrical service, along with
a less pronounced increase in over 2,500 grams neonatal deaths as well as in
stillbirths.
It is shown that 34% of the smallest hospitals (1-250 births per year),
23 percent of those with 251-500 births per year, and 10 percent of those with
501-1,000 births per year, had over 1,000 grams perinatal mortality rates of
27.0	per 1,000 or higher, whereas no hospital delivering more than 1,000 infants
per year had mortality rates in this range.
The following are included in the "Recommendations to Reduce Perinatal
Mortality":
•	In order to produce a general improvement in perinatal care,
the number of hospitals will have to be reduced such that
each hospital retaining a maternity service will have
significantly larger numbers of births. These hospitals
could then gradually be provided with intensive care
facilities. In 1970, 40 out of 49 hospitals in the
Quebec region delivered fewer than 1,500 infants per
year.
•	Since low birth weight neonatal mortality is greater in
hospitals which refer patients after birth for intensive
care facilities (Level III) , it is preferable in labors
that begin prematurely to have the delivery take place
in a center providing obstetrical intensive care in
close relation to neonatal intensive care. If the 4%
of women going into labor more than five weeks before
term had been delivered in hospitals possessing perinatal
intensive care facilities, the perinatal mortality rate
would have been reduced by four deaths per 1,000 total
births.
•	If intramural neonatal intensive care centers are not
available in maternity hospitals. or if the problem
is not predictable by the onset of labor, infants who
are of low birth weight or sick should be transferred
to a referral center for neonatal intensive care.
16﻿Chase, Helen
1977 "Infant Mortality and its Concomitants, 1960-1972". Medical Care 15. 662-674.
GENERAL ABSTRACT:
A number of health indicators are used to examine the gap in health status
between the poor and non-poor has narrowed in the period 1960-1972.
ANNOTATIONS:
Although the terms are recognized not to be synonymous, white and all other
races are used as proxies for non-poor and poor respectively, because of un-
availability of data according to the latter characteristics. During this
period, the groups between white and all others have narrowed for fetal, neo-
natal, postneonatal and maternal mortality. With regard to fertility, the
gaps between white and all other races have narrowed for the maternal age
groups beginning with 25 years of age; chief progress in narrowing the gaps
was among higher birth orders. On a maternal age base, the gap in the propor-
tions of illegitimate births of racial groups appears to be widening.
Chez, Ronald A., M.D.; Harold E. Fox, M.D.; John C. Hobbins, M.D.; George J.
Peckham, M.D.; E. Derek Peske, M.D.; Mary Halderman
1978 "Monitor Every Patient in Labor?" Patient Care 12; 136-163.
GENERAL ABSTRACT;
In this article, a panel of perinatologists spells out the indications,
benefits, and pitfalls of fetal and neonatal monitoring in the community
hospital. They also point to telemetry as a future trend to make fetal
monitoring more acceptable to patients.
ANNOTATIONS;
Indications of Fetal Monitoring
These categories of patients in a community hospital should always be
monitored:	diabetes, hypertension, renal disease, and Rhesus factor diseases.
Other indications for high-risk patients in a community hospital are post-
maturity, premature labor, or prolonged labor. In general, any patient
(according to these physicians) who develops a risk after presenting in labor
is a specific candidate for fetal monitoring in the community hospital. Also,
a patient receiving continuous conduction anesthesia should be monitored.
Dr. Fox recommends monitoring every birth because 30% of the high-risk
labors occur in low-risk pregnancies.
17﻿There are studies that show a decrease in cerebral palsy and other forms
of brain injury with monitoring:
Dr. Hobbins comments:	A study in Australia definitely showed a decrease
in morbidity and neurologic sequelae in monitored patients, It was not a
pure study of monitoring because fetal heart rate monitoring played only a
part in the intensive care of the fetus. In other words, they had a control
group of patients who were not monitored and a series of patients whom they
followed in what they called a fetal intensive care unit, where they also
used hormonal assays and other such special tests. They did find that fetuses
subjected to this type of intensive scrutiny and monitoring did better as far
as long-term neurologic sequelae were concerned. Fetal.heart rate monitoring
played a big part but not an exclusive one in this improvement.
The article points out that there are five areas in which morbidity
potentially may be increased as a result of fetal monitoring with the invasive
techniques, such as the scalp clip:
•	Fetal scalp infections occur with about a 1:230 frequency. These have
been a straightforward, local type of infection and not of any
systematic severity.
•	Misapplication to the fetal part has the unlikely potential of
cosmic injury.
•	Increased incidence of amnionitis is a concern in some situations —
but it is the lesser risk, rather than not to monitor in the same
situations.
•	Uterine perforation is possible, but rare.
•	Placental perforation would also be a rare complication.
Dr. Fox notes that of the factors contributing to the increased caesarian
section rate, 80% can be attributed to altered concepts on breech delivery and
pathology of labor progress. Fetal distress accounts for 20 percent of the
increase.
The physicians suggest that special fetal heart rate monitoring equipment
can be placed in the community hospital with a "telephone hook-up" in order
for the fetal heart rate to be monitored by a specialist at the other end of
the phone.
The problem with this system involved a lack of use. A major role of
regional neonatal intensive care centers should be to continually educate
community hospital personnel in the uses of new equipment, In this case, it
is suggested and the community hospital personnel were not throughly educated
in the uses and advantages of the fetal monitoring telephone system.
Cost of Equipment
An individual charge of $35 to $50 will financially sustain a monitoring
system. ($35 is based on 3,000 births a year. The cost certainly varies with
the delivery base.) Most brands of monitoring equipment are equivalent; the
key variable is repair backup. Telemetry systems are expected to be used in
the near future to augment present equipment.
18﻿Telemetry
A practical alternative in monitoring fetal heart rate in the near
future.
Telemetry consists of a small FM transmitter usually taped to the
patient's thigh and a nearby receiver that can feed the impulses to the
monitor. The patient can walk around as the heart rate and uterine acti-
vity are recorded. In most cases, telemetry for fetal monitoring will
simply be an adaptation of currently existing equipment.
For Additional Information
The following physicians should be contacted if more information con-
cerning fetal heart rate monitoring and its uses in perinatal intensive care
is needed:
Ronald A. Chez, M.D.: Howard University College of Medicine
Washington, D. C.
Harold E. Fox, M.D.:	University of Rochester School of Medicine
and Dentistry
Rochester, N. Y.
John C. Hobbins, M.D.;Yale School of Medicine
New Haven, Ct.
Clifford, Stewart, M.D.
1964 "Medical Progress, High-Risk Pregnancy. Prevention of Prematurity and the
Sine Qua Non For Reduction of Mental Retardation and Other Neurologic
Disorders." The New England Journal of Medicine 271: 243-249.
GENERAL ABSTRACT:
Dr. Clifford focuses on newborn mortality rates and discusses some causes
of mortality which have been cited in the British Perinatal Mortality Survey.
ANNOTATIONS:
Mortality Rates
Generally, Dr. Clifford notes that the most important causes of mental
retardation are associated with pregnancy, particularly with premature births.
He says, further, that nationwide results achieved by many maternity institu-
tions testify that a fetal mortality rate of 15 and a neonatal mortality rate
of 6 to 8 percent should be within the reach of all. Dr- Clifford agrees that
these figures can be reduced.
19﻿Live-Born Premature Infants
The neonatal mortality rate is, in general, directly proportional to the
incidence of live-born premature infants. Under the most favorable condi-
tions the mature infants will contribute only three deaths per 1000, whereas
the premature infants, with a mortality rate of 15 percent will be responsi-
ble for 12 deaths on the basis of a relatively low premature birth rate of
8 percent. Dr. Clifford points out that 11 deaths per 1000 births at the
Boston Lying-In Hospital (fifteen-year study 1949-63) resulted in premature
stillborn infants. Hence, some 23 fetal-newborn premature deaths per 1000
births are theoretically preventable.
The British Perinatal Mortality Survey has summarized a number of findings
that should permit the identification of high-risk patients early in the
pregnancy.
Demographic data from the British survey shows that infant loss is higher
in the first than in the second delivery and in the semiskilled and unskilled
social classes. The risk to the infant is 60 percent greater if the pregnancy
is illegitimate. The infant mortality steadily rises with increasing maternal
age.
The British survey records that there is a significantly higher perinatal
mortality in a subsequent pregnancy for patients who have had an abortion or
ectopic pregnancy. In patients who have had a previous premature live-born
infant, the stillbirth and neonatal mortality is even higher in later
pregnancies, and for those who have had previous stillbirths or neonatal deaths,
the perinatal mortality is still higher.
Vaginal bleeding before the twenty-eighth week was present in 2.9 percent
of patients, with an overall perinatal mortality double the national average;
these cases were associated with a high incidence of premature delivery. If
pregnancy continued to near term, there was still an appreciable increased
perinatal risk. It was concluded that a history of toxemia and ante-partum
hemmorrhage is invariably an indication for special care.
Concerning labor and delivery, the British Survey records that the maturity
of pregnancy was of immense importance to the fetal outcome. At forty weeks,
the perinatal mortality was two-fifths the national average; the risk increased
by 50 percent at forty-two weeks, and by nearly 100 percent before thirty-eight
or after forty-three weeks. Postmature births at forty-two weeks and above
occurred in 11.5 percent of the patients, being more frequent in younger mother,
nulliparas and grand multiparas and in those of lower socioeconomic backgrounds.
Dr. Clifford concludes that there is promise that a reduction in mental
retardation and other neurologic disorders will accompany reduction in peri-
natal mortality. He believes that our strongest efforts should be directed
toward identifying high-risk births as early as possible and treating the
high-risk pregnancy appropriately.
20﻿Committee on Perinatal Health
1976 Toward Improving the Outcome of Pregnancy. White Plains, New York
The National Foundation March of Dimes.
GENERAL ABSTRACT:
This document provides recommendations and suggestions for use in planning
by the providers of obstetric and pediatric care and for agencies concerned
with maternal and infant health. With the aid of these recommendations, a
model for organizing and improving resources providing prenatal and perinatal
services can be designed for the population base under study.
ANNOTATIONS:
The Committee describes the three levels of newborn services which are
described throughout the literature and in these annotations. Hence, here
the concentration will be on personnel requirements and training.
Personnel Requirements in Level I Units
For Consultation — It is desirable to have specialists from Level II and
Level III units on the staff of the Level I unit as consultants.
In Labor and Delivery — Two or more qualified professional nurses or
certified nurse-midwives continuously available on call to be in attendance
to women admitted in labor. All deliveries should be attended by a physician,
or where necessary and where acceptable, by a certified nurse-midwife under
the supervision of a physician.
Personnel Requirements in Level II Units
Direction and Administration — Each institution should have one board-
certified obstetrician with special interest, training and experience in hiqh-
risk obstetrics and one board-certified pediatrician with special interest,
training and experience in neonatology to serve as co-directors.
There should be a Supervisor of Perinatal Nursing Services with overall
responsibility for inpatient activities in the maternity-newborn unit.
In Labor and Delivery — The obstetrician who is co-director of the
Level II unit should be the Director of the Obstetrical Department and will,
with the other physicians and the nursing staff, define and establish standardized
procedures for all obstetric patients.
Full-time nurses with specialized training and anesthesiologists should be
available at all times.
Note:	Requirements for Level III have been annotated in other "note" sets.
Preparatory and Continuing Education in a Regional System
A major function in a regional system for the delivery of perinatal and
prenatal health services is that of preparatory and continuing education.
21﻿Level III units must exert leadership in establishing educational programs
and, to a large extent, in carrying these out. The regional centers will have
full-time staff in all the health science disciplines. Their qualifications
should include demonstrated competence in teaching and training. The assign-
ment of their service responsibilities should allow adequate time for their
teaching activities throughout the region. Responsibility for planning and
providing educational services in a region should be assigned to a small,
multidisciplinary group at the Level III units.
The various types of educational services to be considered include one or
two-day didactic symposia presenting new technologic and scientific advances;
short courses, ranging from a few weeks duration, for didactic teaching and
practical application; longer courses with academic credit toward a degree or
toward meeting requirements of a certifying board.
Education for Career Advancement
Educational programs must be made available to staff at all levels of
responsibility throughout the component hospitals in the region.
Continuity of Care in Consultation and Referral
In both consultation and referral, continuity of care in meeting the patient's
needs is the basic consideration. This will be fostered to the extent that there
is abundant interaction among all components of the regional system. This
interaction should include participation of referring physicians, to the
fullest possible extent, in the diagnosis and management of problems requiring
consultation and for arranging transfer must be clearly established and under-
stood. Equipment and trained personnel for the transfer must be available at
the Level III units.
Financing
In a regional system of care, new costs will be generated due to increase in
consultation, increase in the number of infants transferred from one hospital
to another and increase in the number of patients to whom expensive technology
and expertise will be available.
Cost recovery is a major concern of the providers of care and the magnitude
of these costs for complicated prenatal and perinatal illness exceeds the ability
of most families to pay. If the financial barrier to extension and improvement
of perinatal services is to be lessened or eliminated, new sources of funds must
be created for both capital outlay and for operation of services. The benefits
in terms of reduction of infant mortality and of life-long handicapping condi-
tions will fully justify increased expenditures.
Model Budget
Model budgets for the operation of the different levels of newborn care
facilities are presented.
22﻿Level II Unit (2,000 deliveries with Neonatal Special Care Unit of
8 beds):
Total Operational Budget	$1,192,698.00
1.	Personnel
2.	Expenses
3.	Education
$ 749,068.00
$ 439,356.00
$ 5,000.00
Level III Regional Perinatal Care Center:
Total Operation Budget
$2,637,095.00
1.	Personnel
2.	Expenses
3.	Regional Education
$1,712,496.00
$ 876,092.00
$ 48,507.00
Donahue, Charles Jr., M.A., Lou Freedman, Robert Danley, Ph.D., Ann Pettigrew,
M.D., Nancy Shaughnessy, Sarah Fogerty, R.N.
1976 "The Use of Vital Events as a Data Source in the Planning of Maternity and
Newborn Services". Paper presented at the Annual Meeting of the American
Public Health Association, October 21, 1976, Miami, Florida.
GENERAL ABSTRACT:
This paper discusses some possible uses of data that are currently being
collected by most states throughout the country. These data could be used
for health systems planning in the development of regional newborn care
systems. The uses discussed by the authors were developed by the Massachusetts
Maternity and Newborn Regionalization Project working with the Office of Health
Statistics and the Massachusetts Department of Public Health.
ANNOTATIONS:
Live Births - Fluctuations and Trends
It is important to follow over time the number of births per year by resi-
dence of mother and hospital of birth. Therefore, (1) the residence of mother
and (2) the hospital of birth should be collected from the live birth certificate.
Birth Rate, General Fertility Rate, Percent of Population - Women Aged 15-44
Differences in the fertility experience of women living in different areas
can be helpful in making projections of the future need for fertility-related
health services. Future "needs" for acute in-patient beds, prenatal care visits
at family planning services should be based on a consideration of past general
fertility rates. The source of this data is the live birth certificate, U.S.
Census. The significant items are (1) residence of mother, (2) women aged 15-44,
and (3) total population by residence.
23﻿Low Birth Weight Infants
Areas with a disproportionately high percentage of low birth weight
infants should be identified for more in-depth analysis. With an effective
system for "prebooking" high risk mothers in tertiary perinatal centers,
higher percentages of low birth weight infants would be expected in facili-
ties providing this "level of care". Similarly, hospitals providing lower
"levels of care" would not be expected to care for high percentage of low
birth weight infants.
High Risk Newborn Transfer
The authors agree with Swyer's report on transfer of high risk newborns.
A transfer percent can be obtained by comparing neonatal deaths by hospital
of birth with hospital of death. Those deaths occurring in neonatal inten-
sive care units are credited toward the "transfer" percent. The authors
accept definitions of "referring" and "transfer" which are outlined by
Usher. The live birth certificate is the source of information for (1)
hospital of birth and (2) hospital of death.
Duxbury, Mitzi L., M.D.
1977 "Personnel and Staffing Needs for Perinatal Programs". Seminars in Perinatology
Regionalization of Perinatal Care. 1:267-278. Guest Editor:	Leo Stern, M.D.
New York:	Grune and Stratton.
GENERAL ABSTRACT:
Personnel issues have become critical considerations in developing region-
alized perinatal care systems. The author summarizes the functions of a region-
alized perinatal system, personnel required for these functions, concerns re-
garding present medical and nursing staffing approaches and some possible
solutions.
ANNOTATIONS:
Tasks of a Perinatal Care Center
These include service to patients (both in-house and within the region);
basic and continuing education to center and regional personnel; research;
and administration to perform these functions as well as enabling communications
and supporting evaluation of the system.
Patient-Centered Team Approach
The definition and utilization of a "perinatal health care team" may vary
considerably. A team may be defined as all personnel (medical, nursing, social
workers, therapists, nutritionists, technicians, religious personnel, administra
tors, maintenance workers) or as all the specialized health care teams within a
24﻿region. Additional concepts of a team stress the family as a team member or
by definition and/or practice exclude the family member. The definition one
selects for a "health care team" has significant implications for the staffing
requirements necessary to meet the goals of the perinatal center. The health
care team at the center can be broadly defined as composed of nurses,
nutritionists, physicians, social workers, respiratory therapists, physical
therapists, supportive services such as community health nurses, homemakers,
home health aides, laboratory, x-ray and ultra-sound technicians, religious
advisors, administrators, and the family. This team concept includes all
providers in the system and provides for referral, consultation, continuity,
and follow-up.
Staffing of the Tertiary Care Unit
With the intensively ill newborn infant, a ratio of one nurse per patient
is not unusual, or in extreme situations two nurses to one infant. Some
workable staffing methodology, based on patient care needs, to allow for
these situations must be identified to establish functional staffing patterns.
In addition to in-house service responsibilities, consideration must be given
to educational needs in the region.
Nurse Attrition in Neonatal Intensive Care Centers
Attrition rates for nurses are as high as 180% annually. Rates of
80%-90% are not uncommon. Further efforts are needed to remedy the dynamics
behind this phenomenon. Patient care is compromised when personnel are in-
experienced and in the process of learning.
Medical Staffing in Neonatal Intensive Care Centers
There is a dearth of written material in relation to staffing methodologies
necessary for adequate medical coverage. The number of staff contingent upon
many factors:	hard monies provided by institutions, the amount of soft monies
the medical staff was able to garner, the needs for research, and education of
medical personnel. Staffing approaches are somewhat different when the funds
were earmarked specifically for service rather than for education and research.
Almost all of the neonatologists interviewed by the author expressed some dis-
satisfaction with the need to provide service coverage using persons whose
primary need is for education or research experience. These divergent goals
of the training programs often detract from the primary responsibility of a
tertiary care center.
25﻿Effer, S.B., M.D.
1969 "Management of High-Risk Pregnancy:	Report of a Combined Obstetrical and
Neonatal Intensive Care Unit". Canadian Medical Association Journal 101:
55-63.
GENERAL ABSTRACT:
The methodology, equipment and personnel required to carry out an intensive
care program in the management of high-risk pregnancies have been outlined.
The perinatal mortality rate has been determined and its. etiology has been
analyzed in this section.
Here are three conditions in which the degree of high risk is such as
to warrant provision of the complete facilities of the service described:
a) severe pre-eclampsia; b) marked intrauterine growth retardation with
placental insufficiency as determined from serial measurements of uterine
growth and estriol determinations; and c) irreversible labor in premature
pregnancies where a birth weight of 2,220 grams or less is anticipated.
ANNOTATIONS:
An already existing area of 14 antepartum beds was selected to care for
high-risk infants. These are adjacent to the labor unit and to the OB and
NICU.
Nursing Personnel
The antepartum ward for high-risk pregnancies was found to require a
ratio of one nurse to four patients on each of the day and evening shifts,
and one to six on the night shift; in addition to the above, one extra nurse
on each shift was available for patients in labor. The requirements,
after pro-rating for time off, sickness and vacation time, place the total
staff requirements at approximately one nurse per 100 annual deliveries.
Medical Personnel
The obstetrical team consisted of one obstetrician director, who is oriented
as a perinatal physiologist, one full-time resident, the junior house staff
and the attending private physiciand and/or obstetrician.
Paramedics (Electronic)
The constantly available supervision of the equipment by a technician was
found essential in view of the frequency of needed electronic adjustments. It
was found that one full-time technician would be required to handle the load
of 100 patients per year monitored in labor.
26﻿Biochemistry Personnel
The carrying out of blood-gas and acid-base estimates, the calculation of
the values on line, and of blood gases for the newborn were found to require
two full-time technicians where 200 to 250 patients are monitored.
Prognostic Risk Scores
An attempt was made to find a qualitative expression of the degree of
risk of perinatal death faced by any one patient. Statistical information
used to develop this scoring was taken from two main studies:	"Supplement
to the Second Report of the Perinatal Mortality Study in Ten University
Teaching Hospitals, Ontario, Canada," and the first report of the 1958
British Perinatal Mortality Survey.
Using the methods developed, the NICU would have expected a total of
10 perinatal deaths in their study (a rate of 47.8/1,000 total births). The
actual results showed eight perinatal deaths (37,2/1,000 births). With the
introduction of a well managed intensive care unit for high risk newborns a
20% reduction has been achieved.
The article shows that 89% of random patients going into labor had a
Prognostic Risk core of 50 or less. A general estimate has frequently sug-
gested that 10 of all pregnancies have some increased risk. It is pointed
out that using the scoring system explained, any score above 50 represents
high-risk pregnancies.
Conclusions
The author concludes that a method of improved education of the undergraduate
the postgraduate and the practicing physicians should be sought. In both out-
lying communities and in university centered programs, an exchange of residents
in family-practice or specialty-training programs with practicing physicians
and specialists might prove useful, Mutual benefits may accrue, on the one
hand, by an awareness of actual problems and practices in outlying areas.
Egan, Edmund A., Richard Boothby, and E, Charlton Prather
1975 "Florida Regional Neonatal Intensive Care Program — Impact on Mental Retardation"
Journal of the Florida Medical Association 62: 36-39.
GENERAL ABSTRACT:
The disabled patient, with severely limited intellectual capacity as a
primary component of his disability, has little to be a personally or socially
independent individuals. Such patients place a major burden on their families
27﻿and communities; also they have great difficulty in achieving minimal personal
contentment for themselves. This article explains a follow-up study program
of individuals who required intensive care at or around the time of birth.
ANNOTATIONS:
In 1971, it was reported that very small prematures (less than 1,500 gram
birth weight), cared for in a neonatal intensive care unit, showed only a 10%
incidence of impairment, rather than the previously reported 50%. Such a
finding has been confirmed from other centers. These data, documenting the
social benefit available from perinatal care, generated the creation of an
organization now known as the Florida Perinatal Intensive Care Program in
1972.
An integral part of the RNICC program is the post discharge evaluation of
survivors. All infants under 1,500 gram birth weight are entered into the
evaluation program, which documents their physical and mental development after
discharge. A random sample of 20% of infants over 1,500 grams birth weight is
also followed in serial fashion. These are evaluated both by a pediatrician
and by a pediatric neurologist or psychologist. All children in the evaluation
program are examined at age six months, one year, two years, four years, and
six years. The dates of evaluations are adjusted for prematurity in the first
two evaluations, so that the survivors who were born premature are measured
against scales for cohorts at the same time, rather than those born at the
same time.
Results of Florida's RNICC Evaluation Program, 1974-1975:
Birth Wt.	Birth Wt.
1500 gm.	1500 gm.
No. evaluated	95	135
No. judged normal	88	123
No. judged abnormal	7	12
percent abnormal	7%	9%
percent normal	93%	91%
The authors feel that this lends credence to the hypothesis that over 90%
of these infants will continue to be measured normal as their evaluation
progresses. Yet, they do note that longitudinal program may tend to over esti-
mate impairment in the early evaluations.
Ellis, William C., M.D.
1975 "How Maternal and Infant Care Has Been Improved in New Jersey".
Contemporary OB/GYN 6: 73-77,
GENERAL ABSTRACT:
This article describes the development of Regional Perinatal Intensive Care
in a Level III NICU in New Jersey. Aspects of the discussion include:
28﻿•	A system of communication that is unencumbered and bidirectional
between the institutions of the region and their personnel;
•	An ongoing education program;
•	An evaluation process that has received the cooperation of
all the institutions within the region resulting in an
ever-increasing compliance in utilizing the system.
ANNOTATIONS;
Dr. Ellis discusses the regional newborn unit at Monmouth Medical Center
Long Beach, New Jersey, established in 1968.
The criteria recommended for the NICU at Monmouth are similar to those
formulated by the Ad Hoc Committee on Perinatal Health formed by the AMA,
the COG and AAFP, AAP.
Criteria for Neonatal Transfer
•	Mortality risk 25% or greater (gestational age less than 33 weeks or
weight less than 2,000 grams),
•	Mortality risk 4% to 25% with (a) placental or cord accident
(b) apgar score less than 6 at 5 minutes (gestational age
33-36 weeks or weight less than 2,500 grams).
•	Respiratory distress and metabolic acidosis persisting after
four hours of age or requiring ambient 02 in excess of 40%
after four hours of age.
•	Infant of diabetic mother.
•	Neonatal seizures.
•	Known isoimmunization prior to 36 weeks gestation with L/S ratio 2:1
•	Persistent cysnosis without respiratory distress syndrome.
•	Suspected neonatal sepsis and/or meningitis.
•	Congenital anomalies requiring observation for neonatal surgery.
Findings Reported from Evaluation
•	Intensive care techniques applied in inappropriately staffed and
organized nurseries deleteriously affect those infants whom they
are intended to help.
•	Over a five-year period, 1969-1972, a study revealed an inverse
relationship between the percentage of infants transferred by
the regional hospitals and the neonatal mortality rate.
29﻿• Another demonstration of the value of perinatal transfer for a
single user institution was at the Hunterdon Medical Center.
The rate of perinatal loss of infants with a mortality risk
greater than 25% dropped from 702.7 per 1,000 live births in
the years 1968-1970 to 479.2 during 1971-1973.
Transportation
Transportation of newborns to the Monmouth Medical Center is handled
under contract by an ambulance service that bills the families directly.
Mothers are transported through private arrangements or by local first-aid
squads.
Education
An education program provides resource personnel for conferences, lecturers,
and courses in hospitals served by the Center.
An important part of the regional perinatal program and the Center are the
perinatal workshops for physician-nurse teams at which specific problems are
discussed in depth. The program provides site visits to hospitals to advise
on staffing and equipment, on improving communications, and on policies and
procedures within labor-delivery and within nursery areas to handle certain
risk problems.
Other innovative services are provided by a group of women volunteers who
are starting to function on a regional basis. Most of them were at one time
high-risk mothers, so they have firsthand knowledge of what being at high-risk
means. These volunteers are working in three areas:	They function at the
Center as surrogate mothers, helping feed the growing babies who have been
transferred. They also will provide companionship for mothers who have been
separated from their own communities. A third area of activity is at the
referring institution, where a woman volunteer visits the mother whose baby
has been taken away for care at the Center.
Ferrara, Angelo, M.D., and Lucille Perrotta, M.D.
1976 "Infant Transport Services:	An Overview". Pediatric Annals 5:25-45.
GENERAL ABSTRACT:
New York City's experience with transporting high-risk neonates is traced
and guidelines for developing and operating a transport system are reviewed.
ANNOTATIONS:
Vehicles
If all transporting to be done is within a relatively small area, such as an
urban center, some type of ground vehicle is best. If transports are made over
long distances, (greater than 50 to 75 miles), aircraft will be fastest.
30﻿According to federal specification, there are three types of ambulances
which are adequate:	(1) conventional cab chassis with modular ambulance
body; (2) standard van, forward-control integral cab-body ambulance; and
(3) specialty van, forward-control integral cab-body ambulance. (Some
suggestions, specific for infant ambulances, are discussed in the Canadian
Pediatric Society handbook on transport of the newborn and the American
Academy of Pediatrics "Standards and Recommendations for Hospital Care
of Newborn Infants.")
Equipment
Basic to transport is the portable emergency kit. The basic kit
might include:
•	Laryngoscope with infant premature-size blades
•	Endotracheal tubes
•	Medications:
•• Calcium gluconate, 10%
•• Epinephrine, 1:10,000
•• Sodium bicarbonate
•• Dextrose, 50%
•• Normal saline
•• Sterile water
•• Vitamin K
•• Heparin
•• Isproterenol
•	Syringes:	2 1/2-cc., 10-cc, 20~cc
•	Needles, size 23, 20, 18
•	Three-way stopcock
•	Alcohol swabs
•	Airways:	sizes for infant and premature
•	Dextrostix
If the sending hospital is poorly stocked, equipment can be provided
in kits that are kept sterilized and ready and can be easily used when
needed.
Suggested kits should include:
•	Thoractomy kit
•	Tracheostomy kit
•	Lumbar puncture kit
•	Combined umbilical catheterization, vessel cut-down, and suture kit
•	Parenteral fluid kit
•	Bacteriologic sampling kit
Travel
Physicians should always accompany:	(1) infants under 1,000 grams,
(2) neonates requiring assisted ventilation, (3) newborns with chest tubes
in place, (4) infants with history of apnea spells or those with moderate
to severe respiratory distress, and (5) any infant whose referring doctor
requests physician accompaniment.
31﻿Cost
Cost factors should be considered when planning a transport system.
Simple unit cost per transport can be calculated by dividing the total
budget by the number of calls.
It is suggested that medical economists make use of the benefit-cost
ratio for program analysis. By this method, present dollar value of future
benefits and costs, direct and indirect, are compared. When the benefit-
cost ratio is greater than 1:1, the program is deemed praiseworthy.
This method may be applied in evaluating the cost of NICUs.
Ferrara, Angelo, M.D., Ph.D.
1977 "Evaluation of Efficacy of Regional Perinatal Programs". Seminars in
Perinatology:	Regionalization of Perinatal Care 1:	303-308. Guest
Editor:	Leo Stern, M.D., New York:	Grune and Stratton.
GENERAL ABSTRACT:
This article addresses the assumptions of evolving perinatal programs
(facility and manpower planning, levels of care and infant transport)
and approaches to evaluation of perinatal services. Specific components
of structural, process and outcome evaluations are discussed.
ANNOTATIONS:
Structural Evaluation
The purpose of structural evaluation is to determine bed need per
population group. A methodology for determining the number of neonatal
beds necessary for tertiary care is used as an example:
Bed Need
Newborns in Need of Care Level x ALOS
365
A total of 25-30 beds per neonatal center is suggested.
Swyer's one, full-time neonatologist/2,500 live births is used if one
attempts to determine physician needs in an unresearched area.
Nursing personnel need is better understood. For neonatal intensive
care (tertiary care), the nurse-patient ratio should be 1:1, while
special care areas (secondary care level) require a ratio of 1:3. To
calculate the number of nursing personnel necessary for a unit of 20 beds
(16 special care and 4 intensive care), the following steps would be used:
32﻿•	Five staff nurses are required to ensure a nurse on duty 21
shifts per week. (Since each nurse works five shifts a week,
21/5, or 4.2 nurses are needed. Add to this vacation and educa-
tional time, and it is clear that five nurses would be essential.)
• To calculate the nurse complement for one shift: a ratio of 1:3
for special (secondary) care for 16 babies = 5.3 nurses; a
ratio of 1:1 for intensive (tertiary) care for 4 babies = 4 nurses;
thus, a total of 9.3 nurses is found.
•	To have all shifts covered:	9.3 x 5 (see (a) above) = 46.5
nurses, the ideal complement for this unit.
Process Evaluation
This is a means-oriented assessment applying per group established
criteria as the evaluative measure. Most of the time process evaluation
is performed by chart review. The example of Respiratory Distress Syndrome
and the use of a sample table of criteria are discussed. Major classifica-
tions for criteria are:	maternal history, data from infant's physical
examination at delivery, laboratory tests performed and therapy provided
at delivery and in the special care nursery. Transport criteria used in
New York City are also discussed.
Outcome Evaluations
This is the most useful goal-oriented outcome assessment. Percentage
of mortality is readily assessed. However, long-term disability, discomfort
and social disruption of the family are not easily measured. Variables
which can predict perinatal mortality should be identified and tested.
An example of one effort involved analysis of weight, age, five-minute
Apgar score and temperature of infants at referral site pickup by the
New York City Infant Transport Service. A 20% random sample was analyzed
by weight group. Age at pickup did not differ significantly when three-
year survival rates were examined. High Apgar socres (^ 7) were good
predictors for increased survival rates, especially in the weight group
less than 2,000 grams. Higher temperature at pickup (7> 36° C) was related
to increased survival in all weight groups.
Fitzhardinge, Pamela M., M.D.
1976 "Follow-up Studies on the Low Birth Weight Infant". Clinics in Perinatology:
Organization for Perinatal Care 3:	503-516. Guest Editor:	Louis Gluck, M.D.
Philadelphia:	W.B. Saunders Company.
GENERAL ABSTRACT:
This report provides an overview of the nature of follow-up studies
on the health of high-risk neonates.
33﻿ANNOTATIONS:
In a prospective study of 179 very low birth weight infants dis-
charged from the Intensive Care Nursery at the Hospital for Sick Children,
Toronto, 10 (3.5%) died during the first year of life. Half of these
deaths were associated with systematic infections or severe developmental
defects. The remainder of the deaths were sudden and unexplained.
The healthy premature whose birth weight is appropriate for gesta-
tional age can be expected to grow at the same velocity as a full-term
infant of the same post-conceptual age. Even infants with very low birth
weight, appropriate for gestational age, born prior to 32 weeks follow
this pattern.
Growth failure occurs in two types of low birth weight infants:
those who are small for gestational age at birth, and those who, although
appropriately sized at birth, fail to grow during the first 4 to 6
weeks of postnatal life because of inadequate nutrition.
Prior to intensive care for the neonate, the incidence of major
neurological defects ranged as high as 15% for all low birth weight
survivors and up to 40% for the very low birth weight survivors (see
article for reference).
Among the major changes seen following the introduction of intensive
care have been not only a fall in neonatal mortality, but also a dramatic
reduction in the frequency of the neurological lesions in the survivors.
Various published results indicate that early recognition of the
high-risk pregnancy with delivery of the neonate in a center where
immediate care can be instituted may improve the outcomes of pregnancies
even more than has been accomplished. Such nurseries tend to have a
lower incidence of severe respiratory failure with a smaller proportion
of babies requiring ventilatory support.
Giles, Harlan R., M.D., Jerry Isaman, M.A., William J. Moore, M.D. and
C.D. Christian, M.D.
1977 "The Arizona High-Risk Maternal Transport System:	An Initial View".
American Journal of Obstetrics and Gynecology 128:	400-407.
GENERAL ABSTRACT:
This article profiles the Arizona Perinatal Program's high-risk
maternal transport system. Characteristics of the first 357 consecutive
maternal transport patients (a high-risk parturient patient specifically
referred for specialized care) to one Level III Center are discussed
and presented for comparison with other regions.
34﻿Ethnic background, age, previous pregnancy experience, frequency of
prenatal visits indications for transport, associated medical problems,
referred trends made of transport, and transport outcome are discussed
in light of the data.
ANNOTATIONS;
The principal reasons for transport among the 357 patients were:
premature labor (106 patients), pre-eclampsia (78), premature rupture
of membranes (74), inadequate local facilities (73), associated medical
problems (59), placental bleeding (38), abnormal lie (27), multiple
gestation (25), abnormal OB Hy (22), Rh sensitivity (22) and difficult
labor (9).
The maternal transport rate per 1,000 live births ranged from 0 to
18.0. The overall rate for Arizona is 4.4. Maricopa County (Phoenix)
has a rate of 0.003 per 1,000 live births.
Gluck Louis, M.D.
1970 "Design of a Perinatal Center". Pediatrics Clinics of North America.
17:	777-791.
GENERAL ABSTRACT:
This article discusses principles of planning and design of Perinatal
Care Centers. Dr. Gluck specifies maternal, labor and infant high-risk
factors which identify target groups for specialized care. The programs
of the University Hospital, San Diego, California are described. The
reference is particularly useful for its summary of space and design
standards.
ANNOTATIONS:
Specifications for a special care unit for newborns providing intensive
care, intensive observation and selected chronic and convalescent care
for all newborns (and certain young infants) who requite special care:
Number of Infants
A 40-infant capacity is ideal; minimum size should be 20-25 infants.
A 55-infant unit is too large.
35﻿PERSONNEL (FULL-TIME CIRCULATION—MAXIMAL FIGURES)
Total
Nurses	10	i per shift	42	per	24 hours
Physicians (attending)	1	or 2	4	per	day
Fellows	3		3	per	day
House staff	3		3	per	day
Nursing students	3	to 6	12	per	week
Graduate nurse trainees	2		50	per	year
Medical students	4		80	per	year
Ancillary personnel	3		15	per	day
Technicians, clergy and parents should be considered as well. There
may be up to 33 persons in the unit at one time.
Space Requirements
Infant care areas. For 40 infants—minimum (30 sq.ft, per infant) =
1,200 sq.ft.; maximum (50 sq.ft, per infant) = 2,000 sq.ft. For 25
infants—minimum = 750 sq.ft.; maximum = 1,250 sq.ft. (The minimum of
30 sq.ft, per infant probably is not a true minimum, but it allows for
exigencies without extreme discomfort; 20 sq.ft, is probably the absolute
minimum.)
Additional Space
Space	Type			Dimensions		
Exchange transfusion and minor		C		8	X	12
surgery						
X-ray room		D		6	X	6
Nurses' station		C			-	
Conference room, library, lounge		T		14	x 30	
Scrub area		C		5	X	8
Laboratory	R	+	D		-	
Sleep-in for house staff		c		8	X	12
Parent education (2 rooms)		c		5	X	7 each
Social service		c		6	X	8
Toilet		c		5	X	5
Lounge & Locker		c			-	
Utility & Preparation room	C	+	R	6	X	8
Storage		c			-	
Study Unit	C	+	R		-	
Officers—2(Residents, head nurse)		c		5	X	7
C = Service areas for care	T	=	Teaching	space		
D = Diagnostic space	R	=	Research	space		
36﻿Suggestions for Equipment for Other Rooms
•	Exchange transfusion and minor surgery (8 x 12 ft.; 4 or 5 people):
infant operating table, operating room lights, storage cabinents,
scrub sink, stools (about 4), Mayo stand, infant warmer or crib,
occasional table (monitoring).
•	X-ray room (6x6 ft.; 1 or 2 persons):	table, overhead x-ray
unit.
•	Nurses' station (4 or 5 persons):	built-in desk, chairs, table
top equipment, intercom, etc. Should include permanent space
for a secretary control point.
•	Conference room (14 x 30 ft.; up to 20 or more people):	conference
table (4 x 8 ft.) 20 chairs, projector table, screen, x-ray
viewer, blackboard, bulletin, 3 or 4 large chairs, bookshelves,
kitchenette. May also be used as library, lounge, etc.
•	Scrub area (5x8 ft.; 2 persons):	2 sinks, gown rack or shelves,
coat rack.
•	Sleep-in room for house staff (8 x 12 ft.; 1 or 2 persons):
built-in double decker bed, dresser and mirror, locker or closet,
toilet, shower, sink unit.
•	Parent education (5x7 ft.; 2 parents, 1 nurse):	table, 3
chairs, crib.
•	Social service (6x8 ft.; 3 or 4 people):	desk, 3 or 4 chairs,
file.
•	Offices (two; 5x7 ft.; 1 or 2 persons per office):	1 for head
nurse, 1 for resident physician, each with desk, 2 chairs, file.
•	Lounge-locker room (for nurses) :	couch-bed, 2 soft chairs, mirror
and sink, lockers (narrow) for 60 people.
•	Utility and prep room (6x8 ft.; 2 or 3 persons):	refrigerator-
freezer, prep bench, centrifuge, sink.
37﻿Goff, Robert, M.D.
1978 "The Bedside Microcomputer in the Intensive Care Nursery". Interface
Age 3:	65-67.
GENERAL ABSTRACT:
Software is being developed to enable pediatricians and neonatologists
to maintain bedside microcomputers in the neonatal ICU, providing in-
stant processing of, and access to, the voluminous laboratory data
and event summaries generated by each infant. The data is stored in
a problem-oriented format, and may be accessed with an inquiry to any
particular problem. The program is being written in North Star extended
disk BASIC and is implemented on a SOL/20 Terminal computer with 48k
RAM, and North Star Micro disk drives.
ANNOTATIONS;
It is recoenized that the solution to the problem of information proc-
essing in NICU development and operation is to utilize some form of
computer processing and synthesis or both laboratory data and event
description to daily input of data and, at the time of discharge, to
abstract from the patient's file those pertinent items appropriate for
inclusion in a discharge summary.
Text
Most of the infant's admission history (primarily prenatal and
maternal) is encoded, and at the time of review decoded by the "History
sub-routine", so that most of this textual material is confined to the
program disk and does not require space on the patient's (program) disk.
An additional feature of the output capability of the software is that
it can print the forms which are required by many states for each infant
who is transported from a referring hospital to an NICU center.
Diagnosis
Any diagnosis which can be made solely on the basis of laboratory
data and encoded events or encoded history will automatically appear in
the summary as discharge diagnosis. While the attending physician has
the option of deleting any of these or adding other diagnoses to the list,
it is anticipated that by far the majority of diagnoses will be accurately
made by diagnostic algorithms, and will maximize future access for
statistical study of patient care information. An additional feature
of the diagnostic algorithms is that any suggestive (but not conclusive)
diagnoses can be pointed out to the physician as possibilities which may
warrant further clinical or laboratory investigation.
38﻿Cost
The system described, including a printer, should cost approximately
$8,500 with all the necessary supplies and sales tax.
Goodwin, James and Paul R. Swyer, M.D.
1973 Regional Services in Reproductive Medicine. The Report of the Joint
Committee of the Society of Obstetricians and Gynecologists of Canada
and the Canadian Pediatric Society on the Regionalization of Reproductive
Care in Canada. Sherbrooke, Quebec.
GENERAL ABSTRACT:
This document presents a fundamentally comprehensive description of a
regional approach to newborn care for Canada. Generally, its main purpose
has been to document the need and justification for regional care, and to
depict the requirements for its implementation. The authors have clearly
defined the organizational parameters for which planning is needed.
ANNOTATIONS:
Mortality Statistics, Justification for Planning
Several Canadian studies indicate that an organized system providing
neonatal intensive care results in improved mortality statistics. The
Quebec Mortality Study has shown lower mortality rates from obstetrical
units using neonatal intensive care facilities compared to those which do
not. Further, J.P. Lucey states that "In Vermont (1966-1969) the neonatal
mortality rate (13.3%) for 637 low birth weight infants (1,500-2,500 grams)
born in the Medical Center Hospital of Vermont was approximately one-half
of that for the 1,449 infants born in hospitals without intensive care
capabilities (25.7%)."
High-Risk Births
The authors offer the following list of patients who are deserving
of the closest attention before, during and after labor:
•	Grand multipliers
•	Women aged 35 or over
•	Women giving a history of previous difficult delivery, caesarean
section, myomectomy, hysterotomy, ante-partum or postpartum
hemorrhage or manual removal of the placenta
•	A woman with coincidental major medical disease; diabetes mellitus,
hypertension, chronic renal disease, heart disease or anemia
•	Women with severe pre-eclamptic toxemia or even mild toxemia super-
imposed on pre-existing hypertension or renal disease
39﻿•	Women in excess of 200 pounds or whose pre-pregnancy weight was
30 pounds in excess of the recommended weight for her height
•	Women in labor with no prior antenatal care
•	Any multiple pregnancy
•	Any woman with an obviously large infant or hydramnios
•	Any malpresentation in or out of labor
•	Women with suspected choriamnionitis
•	Women in labor 24 hours or more
•	Multipara in labor with a fully dilated cervix and an unengaged
presenting part
•	Women upon whom several unsuccessful attempts at vaginal deliveries
have been made.
Justification of Perinatal Intensive Care by Results
The extensive experience of Priscilla White and Pederson attests to
the fact that strict regulation of blood sugar, the avoidance of diabetic
complications, toxemia anemia and infection, and the careful timing of pre-
term delivery using serial urinary 24-hour estriol determinations can
effect a significant reduction in both perinatal mortality and morbidity.
The authors report that in a recent study of infants under intensive
care 86.7% of 72 infants surviving a birth weight of less than 1,500 grams
are apparently normal and only 7.4% definitely abnormal. Survivors of
artificial ventilation have a normal distribution of I.Q. compared with
controls. Another study shows that the majority of infants asphyxiated
at birth have recovered undamaged using modern methods of resuscitation
by I.P.P.V. and intravenous alkaline buffer. Further, the authors report
that there has been a significant reduction in mortality from RDS by the
use of artificial ventilation. And lower mortality rates are being
reported in low birth weight infants following the use of intravenous
alimentation. The Report of the Medical Officer of Health, city of
Toronto, 1971 points out that the neonatal mortality rate for the city has
fallen from 20 to 10.1 1,000 births between 1961, when special neonatal
care facilities were first organized in 1971.
Notes on Implementation of a Regional Plan and Administrative Arrangements
It is suggested that planning committees should be established at
State, regional, and community levels. A national liaison committee is
possibly also necessary with strong backing and cooperation from the
health departments at national and State levels.
Policy execution would be the responsibility of Federal and State
directors of health. Local administrators and planning officials at the
appropriate levels of action would be responsible for advice and detailed
direction.
40﻿Delivery of optimal health care in the region would be facilitated
by the appointment of one obstetrician and one neonatologist as regional
directors charged with the responsibility of organizing services within
the region. These individuals should have their base in the regional
hospital center. Also, the regional director would be responsible for the
proper collection of data relating to maternal and perinatal mortality
and morbidity. They should also establish their own guidelines for the
referral of high-risk pregnancies within their own region depending on
the resources in personnel and facilities available, and the distance
of peripheral centers from the regional center.
Relations of Obstetric Units, Neonatal Units and University Health Centers
Ideally, a neonatal intensive care unit should be closely integrated
with an obstetrical unit, so that competent personnel are available to
treat emergencies in the delivery room. The location of an NICU in a major
pediatric center has many advantages since it can provide a wide range of
medical and paramedical expertise. It will also usually be advantageous
to associate the NICU with the pediatric department of a medical school.
In order to justify the expense of maintaining the staff and equip-
ment necessary for optimal perinatal care and to provide the staff with the
opportunity to maintain expertise, the annual number of deliveries in a
regional NICU would have to be large, about 4,000 to 6,000 per annum.
If the maternal population is unselected for risk, problems will arise
in only 8 to 19% of pregnancies and up to 10% of neonates of which 3%
will be of major severity.
The Ontario Perinatal Study (1967) showed that 32% of pregnancies had
some broadly defined identifiable risk factor and generated 60% of the
neonatal problems. Forty percent of the neonatal problems were unheralded
and emergent, though modern techniques have somewhat reduced this pro-
portion. Therefore, if only risk pregnancies were selected for delivery
in major centers, a system would still be necessary to care for emergency
problems in community hospitals.
Regional Organization for Perinatal Care
Regional Center for Reproductive Care:
•	Resource center for pregnancy and perinatal risk processing.
•	Staffed by specialist obstetricians and neonatologists.
•	Equipment for monitoring both mother and fetus during labor.
•	Neonatal resuscitation and location near delivery suites.
•	Available consultant specialists in pediatrics and surgery.
Community Hospitals:
•	Close affiliation with the regional center.
•	Primary responsibility for delivery of women whose pregnancies
appear normal.
•	Equipped and staffed to deal with emergencies in both mother
and infant.
•	Able to call on consultants from regional centers.
41﻿Out-Post Nursing Stations and Clinics:
•	Probably justifiable only where geographic features make access
to larger centers difficult.
•	Delivery of infants in low-risk category.
•	Nursery prepared to care for low-risk infants only.
Each hospital should provide both inpatient and outpatient services.
The use of centralized computer based in the regional hospital center
could provide a program for determining the antepartum risk score for
all women within the region.
Laboratory Service
The regional hospital may provide centralized lab services for the
region to carry out specialized tests for the community hospitals.
Telephone consultations between a peripheral unit or referring
physician and the regional center should be encouraged.
Transport
To ensure adequacy and suitability of space for housing and caring
for patients in ambulances and aircraft used for infant transport re-
gional planning should review the facilities in these vehicles, the appro-
priateness of electrical current supplies for incubators and other life-
support equipment, and specified qualifications of personnel who drive
and operate transport vehicles. Courses in techniques of neonatal trans-
port should be offered by the regional referral center.
Notes on Personnel for Regional Organization
For Canadian population, the authors have determined that for a birth
rate of 20 per 1,000 population requires on OB specialist. If the family
physician is delivering the bulk of normal patients, this figure might
be adjusted to one specialist obstetrician/gynecologist per 25,000 to
30,000 population (500 to 600 deliveries annually).
In order to remain competent, a family physician should deliver
at least 50 infants per year.
Nursing Requirements for "Normal" Patients in Labor and Delivery Units
Aside from an experienced nursing supervisor in charge of the total
unit, there should be one experienced registered nurse assigned as head
nurse to the labor and delivery unit for each 8-hour shift. In addition,
there should be a minimum of one graduate registered nurse attending
every two laboring women.
Staffing Requirement in Postpartum Units
Assuming that the recently delivered mother is monitored in the
delivery area for a period of at least one hour following delivery, the
postpartum ward should be staffed with a minimum of one registered nurse
per 10 patients per 8-hour shift.
42﻿Nursing Staff Requirements for High-Risk OB Patients
The authors report that it has been calculated that the ratio of
one registered nurse per three antepartum high-risk patients per 8-
hour shift was satisfactory coverage. During labor in which active moni-
toring is being carried, it has been found that one graduate nurse is
necessary for the laboring mother and one graduate nurse with some
expertise in the use of fetal monitoring equipment is necessary for the
unborn baby.
Beds, Facilities and Equipment
Regional Centers
It is estimated that 3.6 beds are required for the antenatal super-
vision of high-risk pregnancies for every 1,000 deliveries annually (a
pilot study in McMaster University surrounding Hamilton, Ontario).
The Perinatal Intensive Care Unit
For every 1,000 annual risk deliveries in a University or equivalent
center, 18 antepartum supervision beds is adequate.
The authors continue to describe in some detail the essential equip-
ment needed for the different levels of newborn care. Their descriptions
include:	P.I.C.U. equipment, fetal scalp blood sampling equipment, general
labor and delivery unit guidelines, utility room standards, and standards
for blood gas analysis facilities.
Pregnancy Register
The authors agree that it would be both feasible and desirable that
a pregnancy register be instituted and maintained at the regional birth
center. This would be initiated by the primary contact physician as
soon as pregnancy was diagnosed. Computer-based forms would be used
enabling rapid processing at the regional center which would initiate
a dialogue with the attending physician aimed at predicting categories
of risk to both mother and fetus within the pregnancy and offering
advice on the correct management to minimize risks.
Postnatal Risk Register
In addition, a postnatal risk register could be developed for follow-
up surveillance of infants with perinatal problems.
Such pregnancy and postnatal risk registers would form the basis
for continuing peer surveillance of performance in relationship to
morbidity and mortality statistics.
43﻿Cost Effectiveness of Optimal Reproductive Care
The cost effectiveness of optimal perinatal and postnatal care com-
pares favorably with other medical programs.
Estimates of cost for antepartum and intrapartum supervision of high-
risk pregnancies are difficult, but from pilot studies carried out at the
Henderson General Hospital in Hamilton, St. Joseph's Hospital in London,
and Women's College Hospital in Toronto (1970), the cost per patient
per day has been estimated at $80-$190 in excess of the standard hospital
per diem rate. Insofar as morbidity to the mother and her infant is
avoided by optimal care, later costs of treatment and custodial care of
a much larger order of magnitude will be avoided.
Care in the NICU at the Hospital for Sick Children, Toronto, costs
approximately $200 per patient day. Since the average stay in the NICU
is 6 days, the cost is roughly $1,200 per patient admitted. This figure
excluded care in transitional and convalescent areas of the neonatal
unit, which costs about $100 per patient day; the average stay in these
areas is 20 days and the average cost is $2,000, so that the average total
cost for a patient treated in the neonatal referral unit is $3,200.
The authors stress the humanitarian aspect itself may justify the
high operating cost of a perinatal unit, including an NICU. In addition,
these costs must be balanced against the potential lifetime earning power,
taxability, and value to the community of the individual preserved intact
who might otherwise have died or required long-term custodial care in an
institution. Total costs of supporting such a disabled individual
have been estimated at $500,000 over a lifetime, according to E.J. Quilligan
These methodologies have been suggested in order to determine bed
need levels;
• The number of maternity beds required has been calculated as follows
estimated deliveries per year X the average length
of stay in days X occupancy factor of 1.3
365 Days
number of maternity beds.
• Bassinets should be provided on the basis of well established
principles for the majority df pregnancies which are of low
risk. The formula is:
estimated live births per year X
average length of stay in days X
occupancy factor (1.3 or 1.35)
365 Days
number of bassinets.
These bassinets will mostly be situated in community hospitals,
district hospitals or exceptionally in nursing outpost units.
44﻿• Expected Patient Load in a Regional Neonatal Referral Center
and Bed Requirement
On the basis of the experience of the Neonatal Division at the Hospital
for Sick Children, Toronto, and other data available on the incidence of
neonatal disease, it is estimated 45 infants will require intensive care
and 45 further intermediate care in the neonatal periods for every 1,000 live
births. In round numbers, 15 of the 45 intensive care patients will die,
leaving 75 in all for convalescence for every 1,000 live births.
On the average an I.C. infant occupies the bed for 6 days and 1 bed
accommodates 365/6 = 60 patients/yr.
Hence, the bed requirement for 1,000 live births is 45/60 = 0.75 for I.C.
On the average, intermediate/convalescent care beds are occupied for 20
days by each patient.
Hence, 1 bed accommodates 365/20 = 18 patients in this category. There
are 75 patients/1,000 live births requiring such beds.
Thirty survivors from I.C.
Forty-five survivors from intermediate care.
Hence, beds required/1,000 live births is 75/18 = 4.2
*These figures are dependent upon the neonatal mortality rate.
Gorwitz, Kurt and Donald Smith, M.D.
"Some Implications of Declining Birth Rates for Pediatrics." Pediatrics
56:	592-597.
GENERAL ABSTRACT:
Currently, declining birth and fertility rates in the United States
are having a significant impact upon both the absolute and relative
size of the child population. The magnitude and extent of this impact
are described, and implications for the practice of pediatrics are discussed.
The timeliness of re-examining the role of the pediatrician in the delivery of
health care services is emphasized.
ANNOTATIONS:
The article focusses on these questions of concern to planners:	Are
we building and maintaining an excessive number of pediatric beds? Are we
approaching a point where the current shortage of pediatricians will be
converted to a surplus? What is an appropriate ratio of pediatricians to
children and adolescents? How many patients does a pediatrician see during the
course of a year?
45﻿Data support the assumption that the decrease in the number of live
births has been primarily concentrated in certain high-risk groups. It
has produced a drop in the number and proportion of low weight and low
gestation babies. This, in turn, has reduced the probability of mortality
during the first year of life and of chronic disabilities related to
pregnancy or delivery.
In recent years, about 8% of all infants weighed 2,500 grams or less
at birth. However, these 8% have been responsible for about 85% of all
deaths in the first four weeks of life and a majority of those dying
in the next 11 months. Both of these rates of mortality are now at
record low levels, due in part to the larger than average drop in the
number of low weight babies. This drop has been pronounced among babies
weighing 1,000 grams or less where virtually all die shortly after delivery
and survivors commonly have chronic handicaps. The need for services
and facilities that treat mental retardation, blindness, cerebral palsy
and epilepsy can be expected to decline.
The authors believe that the number of births and the birth rate
will remain close to present levels for the remainder of the present
decade and then will gradually decrease further to around 2.7 million
by 1990.
The authors say the number of pediatric beds in the nation is now
estimated to be 90,000 or an average of 1 per 800 children and adolescents.
Haasis, Patricia, R.N. and Jay P. Goldsmith, M.D.
1977 "The First Year's Experience of a Neonatal Intensive Care Unit". The
Journal of the Louisiana State Medical Society 129:	247-251.
GENERAL ABSTRACT:
A statewide need exists in Louisiana for physician and nurse educa-
tion in neonatal care and for public support of the modern concept of
regional neonatal intensive care units. Newborn infants who require
intensive care can be moved to a regional center without increased
morbidity if a special two-way transport system is used. Infants
weighing less than 1,500 grams can survive without permanent physical
or neonatal sequelae with neonatal care. Primary care at the community
hospital level needs to concentrate on improving basic infant needs,
such as control of body temperature and of blood-glucose levels. Based
on models in other states, a voluntary regionalized program of NICUs
can work in Louisiana.
ANNOTATIONS:
During the first year, the unit at Alton Ochsner Medical Founda-
tion in New Orleans, Louisiana (1976), admitted 318 patients, 160 of
46﻿whom required ventilatory assistance. From July, 1976 to July, 1977,
there were 620 inborn deliveries at Ochsner. Fifty-six of these
infants required intensive care. The survival rate of these in-
borns reflects the small number of patients in each weight category.
Of the total inborn and referred admitted, there were 54 deaths with
an overall survival rate of 83%. Of the 64 patients who weighed 1,500
grams or less, 60% survived. These infants with proper transport and
neonatal intensive care can and do live without permanent physical
or mental sequelae, and community hospitals need not relinquish these
infants as nonviable.
Hawes, Warren E., M.D., M.P.H.
1975 "A Survey of Newborn Intensive Care in California". The Western Journal
of Medicine 23.
GENERAL ABSTRACT:
There are over 450 newborn intensive care beds available in California
hospitals. Over 400 of these are in tertiary units providing total
care for sick newborns. At the present time, there is an NTCU bed short-
age in some areas of the State, the units running at an occupancy rate of
over 87%. The bed situation is particularly critical in the East Bay
and San Diego. The survey suggests that from 2 to 9% of newborns need
intensive care. The present bed availability is ample for the lower
figure but insufficient to meet the upper one.
ANNOTATIONS:
The cost of newborn intensive care is high', daily rates averaged
$198 per day and $123 per day for intermediate care, total costs
averaging $5,178 per day. The total cost of NICU care in the 20 units
is in excess of $35 million annually.
Respiratory distress syndrome was the reason for NICU admission
in more than 40% of the cases. Congenital anomalies and sepsis were
next, accounting for approximately 20% of the admissions.
The recovery rate in these units is very high, averaging 84.4%
for inborn babies and 73.5% for transferred babies.
47﻿Hein, Herman A., M.D.
1977 "Regionalization of Perinatal Care in Rural Areas Based on the Iowa
Experience". Seminars in Perinatology:	Regionalization of Perinatal
Care 1:	241-254. Guest Editor:	Leo Stern, M.D. New York:	Grune
and Stratton.
GENERAL ABSTRACT:
This article discusses the process and issues in planning a regionalized
perinatal center in Iowa.
ANNOTATIONS:
Organizational Elements of Planning
During the initial stages of planning, discussions ensued between practicing
pediatricians and pediatric faculty members of the College of Medicine
at the University of Iowa who were concerned about infant mortality
rates and need to improve neonatal care. The initial plan did
not include obstetrical care, but as the program took shape the
obstetricians became interested. Although the joint planning was
difficult and took additional time, it was justified because it resulted
in a truly perinatal program.
A perinatal educational team was hired and based at the University.
The initial team members included two nurses and a pediatrician who served
as the director. The nurses came from clinical backgrounds of high-
risk obstetrics and neonatal intensive care. The direct background
included experience in the private practice of pediatrics and a special
interest in neonatal intensive care. An obstetric consultant was avail-
able for counsel and advice.
Three basic objectives were formulated to guide the program:	(1) Routine
perinatal care should be improved in community hospitals (Level I);
(2) Level II care must be developed and maintained; and (3) Tertiary care
centers must develop close ties with Level II hospitals to maintain
the quality of care in those centers.
Realizing that proximate Level II or Level III care was improbable
for some period of time, the team felt that the greatest impact of the
perinatal program would be in the improvement of routine perinatal
services in community hospitals. To achieve this goal, the team
determined that it would create a hospital visitation program. All
parties were in agreement that the team should visit the hospitals
as professional helpers and not as judges or regulators. The idea was
to assist each hospital to provide perinatal care to the advantage of
the patient, with the expectation that the unit could provide the
quality care needed and, therefore, continue to provide such services.
48﻿Training the Team
Prior to each hospital visit, the personnel working in the hospital
were informed that a short educational program would be given immediately
following the assessment. They were told that the program would be
concerned with contemporary obstetric and neonatal care practices and would
discuss the items that were included in the teams' assessment instrument.
Since this program was to be given the same day of the visit, it was
necessary for the team members to be prepared to discuss a number of
perinatal topics. To do this, the team developed a list of anticipated
problems and a short presentation on each. They practiced these pre-
sentations with each other to develop clarity and simplicity. The training
period lasted approximately four weeks.
Central Office and Staff
The central office of the Perinatal Care Program is located in Iowa
City at University Hospital and serves as a focal point for the State's
perinatal activities. While it is important that many individuals and groups
become involved in the cause of improvement of perinatal care, there must
be a unifying thread that runs throughout the program and coordinates the
total effort. The central perinatal staff has served that purpose in
Iowa. The author suggests other rural areas, desirous of regionalizing
perinatal regionalization, consider hiring full-time professionals to guide
their program.
Positive Results
Data indicates that care practices have improved in Level I hospitals.
Furthermore, analysis of perinatal data indicates that the trend toward
lower perinatal mortality rates noted in Iowa and nationally has been
more pronounced in Iowa since the onset of the Perinatal Care Program in
1973. The decline in neonatal death rate has been more pronounced than
the corresponding fetal death rate.
Hobel, Calvin J., M.D., Marcia A. Hyvarinen, M.D., Donald M. Okada, M.D.
and William Oh, M.D.
1973 "Perinatal and Intrapartum High-Risk Screening:	Prediction of the High-
Risk Neonate". American Journal of Obstetrics and Gynecology 117:	1-9.
GENERAL ABSTRACT:
A high-risk pregnancy screening system based on a prospective analvsis
of prenatal, intrapartum, and neonatal factors in 738 pregnancies was
found to predict perinatal morbidity and mortality.
49﻿ANNOTATIONS:
Factors were assigned with weighted values according to their assumed
risk. Total scores of the prenatal, intrapartum, and neonatal period
were dichotomized to simplify the scoring system by forming a low-risk
groups (scores less than 10) and a high-risk group (scores greater
than or equal to 10). Three hundred and forty patients (46%) were low
risk during both the prenatal and intrapartum period (low/low-risk
group). In this group, the incidence of high-risk neonates and the
perinatal mortality rate was extremely low. There was no significant
increase in neonatal morbidity or perinatal mortality in 135 patients
(18%) who were identified as high risk during the intrapartum period
(high/low-risk group). However, in 144 patients (20%) at risk only during
the intrapartum period (low/high-risk group) or in 119 patients (high/
high risk) there was a significant increase in the number of high-risk neonates
and perinatal mortality. By a stepwise multiple regression analysis,
actual intrapartum scores are most predictive of neonatal risk (days
in hospital) followed by actual prenatal scores. Identifying a popu-
lation as to their risk status for both the prenatal and intrapartum
period has added depth to the understanding of the continuum of risk which
exists within the framework of the perinatal period.
*It is noted that in North America, Nesbitt and Aubry (Nesbitt,
R.E.L., Jr, and Aubry, R.H.: American Journal of Obstetrics and
Gynecology, 103: 972, 1969, and Goodwin, Dunne, and B.W. Thomas:
Canadian Medical Association Journal. Vol. 101, 1969) have developed
methods of identifying high-risk pregnancies; however, they have either
limited their evaluation to the period prior to labor or have neglected
to evaluate the neonatal period in regard to morbidity.
Honeyfield, Peter R.
1976 "Evaluation, Mortality and Morbidity Indices and Compliance in the Care
of the High-Risk Newborn Infant". Denver, CO:	Department of Pediatrics,
University of Colorado.
GENERAL ABSTRACT:
This article discusses data requirements for perinatal care systems.
ANNOTATIONS:
The following are indicators of mortality and should be collected
in a standard form:
50﻿•	Number of live births.
•	Number of fetal deaths (early, immediate, late).
•	Number of neonatal deaths.
•	Birth weight.
•	Gestational age (accurate to within two weeks).
•	Age at death.
•	Number of infants transferred to another facility.
These should be collected at each newborn facility and gathered at
the four regional Level III centers.
Inter-Hospital Care
The agency responsible for this activity (usually the referral center)
is charged with data collection relating to:
•	Number of infants transferred.
•	Hospital of origin.
•	Hospital of destination (Level II or III).
Referral Center
Data should include:
•	Birth weight (broken down by gestation age).
•	Neonatal Death Rate (broken down by age at death).
•	Perinatal death rate.
Transfer rates (per 100) should be compared with mortality rates (pre-
dicted and actual) for both hospitals and regions. Low transfer rates
and high mortality suggest a hospital which may have problems in identifica-
tion of high-risk infants, early supportive care or poor compliance with
the concept of regionalization. High transfer rates (in excess of 5%) may
suggest inability to provide minimum supportive care appropriate to a
Level I unit or indicate a high-risk population. The latter can be
determined by assessing the predicted mortality.
Inter-hospital care of sick newborns is an integral part of regionaliza-
tion of health care services. It would be desirable for all inter-hospital
transfer activity to be reported to the State Health Department for
collation and evaluation of data.
Johnson, Dorothy K., Dr.P.H. and Elizabeth A. Hefferin, Dr.P.H.
1977 "Perinatal Outcomes Among High-Risk Patients in Two Prenatal Care Programs".
Inquiry 14:	293-302.
GENERAL ABSTRACT:
The paper describes a retrospective study designed to compare peri-
natal outcomes among two groups of high-risk women. One received perinatal
care from a Maternal and Infant Care program (483 subjects); the other
from a traditional health department prenatal program (419 subjects).
Both groups were from Los Angeles.
51﻿ANNOTATIONS:
When perinatal outcomes for the two groups were compared, no signifi-
cant differences were observed in the number of infants surviving to 28
days, low or high birth weight, asphyxia status or other respiratory
problems, metabolic or infection problems, birth trauma, physical or
mental defects or Apgar scores at one and five minutes postpartum. Cost
per delivery determinations were unsuccessful.
The authors conclude that their findings do not support the efficacy
of the special Maternal and Infant Care program studied over regular
traditional programs, but note findings cannot be generalized. They
suggest that there is high probability that Maternal and Infant
Care programs may have great impact where the mortality statistics pre-
viously have been high, where general medical care has been only minimally
available, and where many patients in the past have sought no prenatal
care. Where these programs have been implemented in districts with high
previous mortality, investigators have been able to present quite well-
grounded evidence to document the reduction in poor perinatal outcomes.
This suggests that comprehensive MIC service programs will probably
prove most fruitful if established in poverty areas where poor perinatal
outcome rates are high, as compared with those in adjoining areas.
Johnson, John D., Natalie C. Malachowski, Rose Grobstein, Doris Welsh,
William J.R. Daily, and Philip Sunshine
1974 "Prognosis of Children Surviving with the Aid of Mechanical Ventilation in
the Newborn Period". Fetal and Neonatal Medicine 84:	272-276.
GENERAL ABSTRACT:
This study attempted to partially answer whether infants with severe
respiratory failure who survive with the aid of assisted ventilation have
a reasonable long-term prognosis.
ANNOTATIONS:
Fifty-five infants who survived respifatory failure in the neonatal
period with the aid of assisted ventilation were studied. Over 80%
of these infants had normal intelligence and either minimal or no neurologic
abnormalities. Only four infants had I.Q. scores of less than 85. The
incidence of neurologic and intellectual handicaps was highest in low
birth weight infants (-<1500 grams) . As experience with the use of mechanical
ventilation has been refined, the prognosis of surviving infants has also
improved. Those infants who survived beyond 7 months of age with broncho-
pulmonary dysplasia had no cariopulmonary disability when they reached
5 or 6 years of age.
52﻿Jonsen, A.R., Ph.D., R.H. Phibbs, M.D., W.H. Tooley, M.D. and
M.J. Garland, Ph.D.
1975 "Critical Issues in Newborn Intensive Care:	A Conference Report and
Policy Proposal". Pediatrics 55:	756-768.
GENERAL ABSTRACT:
The article presents several cases and certain key clinical-ethical
questions associated with the provision of neonatal intensive care.
Summaries of five papers on major considerations affecting these issues
are included. The authors propose a moral policy stating areas of
responsibility and ethical guidelines for decisions about care of
newborn infants.
ANNOTATIONS:
The following general recommendations concern allocation of medical
resources, priorities between preventive and curative medicine, and
between medical care and other forms of care affecting neonatal
intensive care regionalization plans:
•	Research in neonatology should be coordinated at the national
level in the interests of efficiency and caution.
•	Neonatal intensive care should be organized on a regional basis
so that its quality and access are relatively equal among various
communities, so that continuing information and technologies
can be shared, and so that adequate epidemiological data can be
gathered and compared.
•	On the basis of clinical experience, professionals in neonatal
intensive care should refine those clinical criteria which render
more specific the general conditions for prolonged life without
pain and the potential for human communication. These criteria
should be communicated broadly within the pediatric, obstetric
and mental health community.
•	Resuscitation criteria should be established with full awareness
of the economic and medical implications of providing this care.
Estimates should be made of the financial cost to society of
prolonging life at a humane level, depending upon the condition
at birth.
•	Delivery room policy, based on certain criteria, should state
conditions for which resuscitation is not indicated. This
policy should be made known to health professionals and to parents
who may be at particular risk (possibly to all parents).
53﻿•	Parents at risk should be counseled about the possibilities of
risks. While recognizing that parents often will be unable or
unwilling to make decisions, medical professionals must always
accept, in principle, that the responsibility should be borne
by the parents and attempt to facilitate but not force them to
make the decision.
•	Regional neonatal intensive care units should establish an advisory
board consisting of health professionals and other involved and
interested persons. The board would discuss the problems of
the unit and make a periodic retrospective review of the
difficult decisions. They would assess the criteria for diagnosis
and prognosis in terms of medical validity and social acceptability.
To implement such a procedure, a neonatal intensive care unit
could review its experience prospectively for each case in which
a decision had to be made. All data pertinent to the decision
could be recorded in a case summary, which would be reviewed
monthly, describe the current process of ethical decision-
making in the unit, and provide a basis for any changes which
might be planned.
•	Since some infants may be abandoned by their parents or, because
of their condition, be maintained in an institution, neonatology
must concern itself with the adequacy of such institutions.
The advocates for intensive care must become the advocates for
the development of humane care and for sufficient funding of
programs to support families whose children require special
attention at home or in institutions. Neonatology cannot be
developed in isolation from continuing specialized care which,
unfortunately, will be needed by some of the survivors of life-
threatening neonatal disorders.
Jonsen, Albert R., Ph.D and George Lister
1978 "Newborn Intensive Care:	The Ethical Problems". Hastings Center Report.
GENERAL ABSTRACT:
Over the past ten to fifteen years, a considerable body of data has
been developed to guide scientific and technical advances in the medical
treatment of seriously ill newborn infants. A neonatal intensive care unit
brings together sophisticated equipment and specially trained personnel
to provide the optimal environment for assisting the endangered infant
through a critical stage in the transition to extrauterine life. A list
of the ethical problems in perinatology includes:	termination of care,
research on the fetus, randomization and controls in clinical trials with
54﻿unconsenting subjects, use of genetic information, risks to fetus and
mother in prenatal diagnosis and the consequences of such diagnosis,
resuscitation and preservation of severely damaged infants, etc. The best
estimates currently available suggest that at least 60 of every 1,000
live born infants will need some form of neonatal intensive care if
death and disability are to be minimized.
ANNOTATIONS:
Iatrogenicity
"Iatrogenic disease" is commonly referred to as an abnormal condition
induced in a patient by the effects of treatment by a doctor. However,
the term does not necessarily impute blame. Also, the term can mean
"disease of medical progress... that would not have occurred if sound
therapeutic procedure had not been employed".
As a medical problem, iatrogenicity is irritating, but it entails
a particularly frustrating moral problem. This moral problem centers
on the question, "What should I do?" One common response to the moral
question is to exhort the physician to proceed with care, initiating
treatment cautiously and meticulously analyzing success and failure.
To proceed with care, implies the establishment of safety and efficacy
through laboratory and clinical research.
Experimental Enigma
No doubt, the fetus usually remains sheltered from the grasp of data-
collecting devices. Biological and medical restrictions frequently
discourage careful clinical investigation, particularly in the case of
new drugs, many of which are used despite warning of insufficient studies
in infants and children. The iatrogenic dangers of perinatal medicine
suggest that the perinatologist bears a heavy moral responsibility to
move experimentally in order to establish the safety and efficacy of
perinatal care. Having accepted this moral responsibility, the peri-
natologist finds that experimentation is, itself, an ethical enigma.
Prognostic Perplexity
The question of principle which must be joined to the foresight of
prognosis is "What sort of human being is desirable?" Unless it is
answered, prognosis is futile. Yet, the art of prognosis, which has been
overlaid with a great deal of science, still is largely a function of
clinical experience. In addition, neonatal medicine is complicated by
lack of experience with the class of patients.
55﻿Kanto, William P., Jr., M.D. and Alex F. Robertson, III, M.D.
1977 "One Year's Experience of a Regional Neonatal Intensive Care Unit".
Journal of the Medical Association of Georgia 66:	616-623.
GENERAL ABSTRACT:
Preliminary evidence suggests that regional neonatal intensive care
units in Georgia have been successful in reducing the State's neonatal
mortality. A major reason for the reduction in neonatal mortality
has been the decrease in the mortality rate of low birth weight infants,
associated with the improved care available in intensive care centers.
Since every hospital does not have the capability of developing these
complex care facilities, regional programs have been developed to extend
the service to every hospital. This report reviews the experience of
the NICU Center at Augusta, Georgia, during a single year of operation.
ANNOTATIONS:
There were 464 admissions in the NICU of Talmadge Hospital during
1975 and 326 (70%) of the neonates survived. In-born admissions were
130 (28% of total admissions) neonates of whom 93 (72%) survived. Out-
born admissions totaled 334 (72% of total admissions) and 223 (70%)
survived. The out-born admissions were delivered in 51 community hospitals
in Georgia and South Carolina prior to transfer to the NICU.
As has been reported throughout the literature, the survival rate
of all infants increased as birth weight increased. Of the infants
below 1,001 grams, 54% of the out-born infants survived, compared to
only 29% of the in-born infants. The survival of 47% of all admissions
with birth weights less than 1,001 grams is better than recent reports
and would suggest that continued aggressive treatment is justified.
Of infants weighing 1,501-2,000 grams, 83% of the in-born infants
survived, compared to 63% of the out-born infants. There is a high
incidence of the Respiratory Distress Syndrome within this birth weight
group and evidence is accumulating that the provision of intensive
respiratory support early in the infant's course may reduce the mortality
associated with this disorder.
Of the infants weighing over 2,500 grams, 90% of the in-born infants
survived compared to 79% of the out-born infants. This probably reflects
the increased severity of disease in the transplanted group which re-
quired their transfer.
This report presents tables of the most common diagnosis and
procedures.
In 1970 the neonatal death rate in Georgia was 15.5 deaths per
1,000 live births. In 1975 this rate was 12.4, a decrease of 20%.
Further improvement should occur with complete development of this
program into a statewide system.
56﻿Knuppel, Robert A., M.D., Curtis L. Centrulo, M.D., Charles J. Ingardia, M.D
Kenneth A. Kappy, M.D., Joseph L. Kennedy, M.D., Marguerite J. Herschel, M.D
Gretchen Aumann, R.N., Marian Lake, R.N. and Anthony J. Sharra, Ph.D.
1979 "Experience of a Massachusetts Perinatal Center". New England Journal
of Medicine 300:	560-562.
GENERAL ABSTRACT:
The authors discuss the experience of the St. Margaret's Hospital
for Women, the major obstetric affiliate of Tufts University School of
Medicine and a Level III Perinatal Center. This article presents data
on 200 maternal transfers treated between 1976-1978. Some findings of
a study of Massachusetts' regionalization of maternal services conducted
by the American College of Obstetricians and Gynecologists are also
discussed.
ANNOTATIONS:
The Perinatal Center has a service area which generates 15,000-
20,000 annual deliveries. St. Margaret's accounts for approximately
2,950 deliveries and 1,000 patients in its intensive care nursery. The
neonatology service has a 25 bed intensive care unit.
Referring physicians have three options in the belief that a closed
medical staff discourages free use of a regional system:	coming to the
hospital with the high-risk mother and performing the delivery, have the
patient managed by the perinatal staff but perform the delivery themselves,
or have the staff assume total care.
Findings from a retrospective study of 200 high-risk maternal transfers
(176 resulting in delivery of 186 infants) include:
•	Ninety-eight percent of transfers arrived by ambulance; the
number ranged between 2 and 6 per month. Predominating diagnoses
were related to premature labor. Mean gestational age was 31.4
weeks. The average length of maternal stay was 11.2 days.
• The overall uncorrected perinatal mortality rate was 11.8%, 37%
for newborns weighing 1,500 grams or less and 2% for those
weighing more than 1,500 grams. There were 21 neonatal deaths
and 1 stillbirth, but no intrapartum deaths. When correction
of perinatal deaths for congenital anomalies and irrevocable
fetal hypoxia or admission was made, no deaths occurred.
•	The average length of stay for 186 infants was 19.8 days.
•	The following findings from an American College of Obstetricians
and Gynecologists' study of Massachusetts' regionalization of
maternity services were cited:
57﻿•• Twenty-four percent of responding obstetricians has experienced
the closing of one hospital where they had staff privileges.
•• Fifty-eight percent of all respondents believed access to
Level II or Level III hospitals was improved.
•• Sixty-nine percent of those involved in unit closures ex-
perienced improved access to Level II or III hospitals.
•• Sixty-two percent of respondents believe better coordination
between hospitals has resulted.
•• The majority of respondents perceived regional planning
would improve perinatal mortality.
Koops, Beverly L., M.D., John T. McCarthy, M.D. and L. Joseph Butterfield, M.D.
1978 "Who Pays the Bill for Neonatal Intensive Care?:	Part III:	Outcome"
(mimeo). Reprints:	L. Joseph Butterfield, M.D., The Children's Hospital,
Department of Perinatology, 1056 East Nineteenth Ave., Denver, CO 80218.
GENERAL ABSTRACT:
This study reviews the relationship of hospital charges to medical
outcome for 174 patients of The Children's Hospital Newborn Intensive
Care Center during four months of 1976. The mean hospital charge, charge
per day and length of stay for the patients were related to the factors
of birth weight, gestational age, major medical diagnosis, and mortality
outcome. The relationship of charges to long-term morbidity was discussed.
Findings for neonatal intensive care outcome compared favorably to the
charges as well as to mortality and morbidity outcomes which have been
reported for adult intensive care. Implications for maternal-fetal
transport, upgrading of perinatal care in all centers throughout a service
region, and establisnment of long-term foilow-up programs are addressed.
ANNOTATIONS:
Hospital charges were, in general, well correlated with birth weight
gestational age and number of hospital days. Infants weighing less than
1,000 grams were 28 weeks gestation and stayed 2 1/2 months; infants
weighing 1,000 to 1,500 grams were 30 weeks gestation and stayed 1 2/3
months; infants between 1,500-2,000 grams were 34 weeks and stayed 1
month; infants between 2,000-2,500 grams were 36 weeks gestation and
stayed 3 weeks; infants between 3,000-3,500 grams were 40 weeks and
stayed 2 weeks. For survivors, the mean hospital charge was $340 per day.
The authors noted that this reflected the first few days of very high
cost during which a nurse-to-infant care ratio of 1:1 or 1:2 was
required along with the high cost of ancillary procedures, as compared to
58﻿recovery days with a nurse-to-infant care ratio of 1:4 and fewer daily
tests and services. The high per diem cost of non-survivors, $607,
reflects the early, very intensive care days without the averaging
effect of the days of recovery.
Eighty-four (50%) of the patients were admitted with the major
diagnosis of hyaline membrane disease. Their charges were correlated
with birth weight/gestational age categories, and reflected both the acute
illness and the length of time to reach 38-40 weeks gestation. Those
infants with extremely high charges which were skewed considerably above
the mean had the complications leading to bronchopulmonary dysplasia (BPD)
Since BPD is most likely to occur in the extremely immature baby and is
associated with high risk obstetrical problems, maternal transport to
a tertiary care center for prevention of delivery or optimum delivery
conditions is supported by the financial data.
For patients with the primary diagnosis of birth defect, the charges
reflected prolonged hospitalizations. Often these defects were complex
and not remediable with current medical or surgical therapeutic modalities
The long hospitalizations were associated with frustration of the health
care professionals and ineffective management. Specifically, major bills
were accrued by infants with birth defects involving the central nervous
system or multiple organs, or who had difficult courses. Amongst the
latter group were those infants who were also premature.
The data indicated that 14 (8%) of the admissions had bills which
were greater than three to six times the mean charge, and were responsible
for 35% of the total hospital charges generated during this four-month
period. All had a bad outcome. An additional 10 (6%) of the patients
had bills which were two to three times the mean charge and 40% of these
had a bad outcome. The diagnoses of these patients were birth defects,
extreme immaturity and complicated hyaline membrane disease.
Studies have shown that 20-30% of infants with birth weights of
less than 1,500 grams have serious, permanent handicaps. These include
blindness, hemiparesis, spastic quadriplagia and mental retardation.
It has been suggested that poor long-term outcome may be predicted in low
birth weight infants who have three or more complications, prolonged
or problematic ventilation therapy and/or intraventricular hemorrhage with
progressive hydrocephalus.
The outcome of neonatal intensive care may also be evaluated by
comparing it to the outcome of adult intensive care. The authors found
a favorable comparison for survival rates, hospital charges, use of blood
products and long-term morbidity. Factors to be included in cost/
benefit ratios are discussed.
59﻿Leake, Rosemary, M.D., Alexandra Loew and William Oh, M.D.
1976 "Retransfer of Convalescent Intensive Care to Community Intermediate
Care Nurseries". Clinical Pediatrics 15:	293-294.
GENERAL ABSTRACT:
As neonatal transport systems develop and the concept of regionaliza-
tion of infant care becomes established, it is increasingly important
that neonatal intensive care units be reserved for sick infants. Less
ill ones can be transferred back to the community hospital nurseries once
they have recovered from the initial acute episode which required inten-
sive care. Several advantages can be derived from the practice:
(1) parental visits can be more frequent, establishing a sounder infant-
parent relationship in a less threatening environment; (2) the expense
is less; (3) referral of such previously ill infants back to their community
hospitals fosters better relationships between institutions; and
(4) in the intensive care unit, the specialized attention can be con-
centrated on very sick infants. The purpose of this study was to investi-
gate the safety and practicality of such a transfer system by instituting
a prospective evaluation of the outcome of convalescent newborn infants
in two especially selected referring hospitals.
ANNOTATIONS:
From July, 1972 to June, 1974, 43 infants were transferred from the
neonatal intensive care unit at Harbor General Hospital to intermediate
care units at two referring hospitals for convalescent care. These
hospitals had obtained a variance of the California Public Health Law
allowing the transfer of infants back into their nurseries and cohorting
with inborn infants. The convalescent infants were transferred when the
following criteria were met:	(1) the infant was free of clinical evidence
of infection, (2) the infant had a stool culture free of pathogens,
(3) the infant had been kept in less than 25% oxygen with satisfactory
blood gas levels for at least 24 hours, (4) the infant was tolerating
oral feedings by nipple or by gavage to maintain caloric and nutritional
requirements, and (5) infants weighing less than 1,200 grams had no
heart murmur at the time of transfer.
From observations of 43 convalescent newborn infants, it is concluded
with confidence that transfer of convalescent infants from a neonatal
intensive care unit to community intermediate care nursery is feasible
and desirable for efficient distribution of health care in the neonatal
period.
60﻿Mazzi, Eduardo, M.D., Ronald Gutberlet, M.D. and Jack A. Phillips,
1977 "The Maryland State Intensive Care Neonatal Program, Transport System".
Maryland State Medical Journal 26:	86-87.
GENERAL ABSTRACT:
Concurrent with the regionalization of neonatal intensive care has
evolved the development of transport systems for optimal and rapid
evacuation of the sick infant from regional hospitals to specialized
neonatal centers. Each State has created its own appropriate and
practical transport system to improve neonatal mortality, and in
the long range, morbidity.
A key in the development of the MSICNP has been the coordination and
cooperation by the Police Aviation Division of the State of Maryland.
The majority of the infants transported to Regional Centers have been
accomplished successfully and with minimum mortality by police heli-
copters and their trained crews.
ANNOTATIONS:
Crew and Equipment
Each helicopter has a crew consisting of a Maryland State Police
pilot and an EMT-certified Maryland State Police observer/medic,
whose duties are to coordinate the missions of the helicopter and, more
importantly, care for the patient once the evacuation has been initiated.
Each helicopter is equipped with basic first aid supplies, oxygen bottle,
suction, cardiac monitor beeper, oxygen analyzer, thermometer and intra-
venous infusion pump. In addition, each of the helicopters is equipped
with a State Police FM and Emergency Medical Systems Radio, whereby
they are able to converse while airborne with the State Emergency
Communication System, doctors at referral centers, and ground ambulance
units.
Locations and Flight Request
A 30-mile radius from an operational base is the primary response
zone for each helicopter. A total of seven helicopters are eventually
planned to give the entire State this type of complete coverage.
Helicopters are dispatched according to their need and top priority
is given to life-saving missions. The aircraft transports premature
and sick neonates from outlying hospitals to the Baltimore City Regional ICUs.
Flight Restrictions
The flight minimum standards established for the helicopters are
guided by FAA regulations and are based on visual flight rules.
61﻿Landing Area
Ground personnel, whether policemen or firemen, who have radio
contact with the helicopter must advise the crew of the most desirable
landing zone.
For neonatal transport, landing areas have been set up according
to aviation regulations in all participating hospitals.
Ground personnel must check the entire landing zone for obstructions
or obstacles such as trees, poles, ditches, signs, wires, etc. and
advise the helicopter crew of these obstacles prior to landing.
Crew Training
The helicopter is manned by a two-man crew. One is a FAA-Licensed
Rotary Wing Pilot and the other and EMT-A Medic/Observer. The
Medic/Observer obtains the following training:
• Eighty hours of Basic Emergency Medical Training Course.
• Eighty hours of training at the Maryland Institute for Emergency
Medicine.
•	Two days annually at Baltimore City Hospital and University of
Maryland Hospital Neonatal Intensive Care Unit. They are
instructed in the importance of temperature and oxygen control,
perform neonatal resuscitation with bag and mask, conduct
assisted ventilation and are trained in suctioning and taking
vital signs.
•	Yearly retraining in all related emergency care treatment.
The Medic's duties are to provide the necessary life-supporting
treatment that sustain or prolongs the patients' ability to survive,
and survive intact until he/she arrives at a suitable treatment center.
Neonatal Transport Evaluation (One Year)
In the 12-month period from March, 1975-March, 1976, 341 infants
were transferred from 30 different regional hospitals to Baltimore
Neonatal Center. Most of them (71.6%) were accomplished successfully
by air, with two neonatal deaths during transport. These infants both
were under 1,000 grams and critically ill.
About 60% of the air evacuations of sick neonates have been accomplished
without a physician. The mean arrival time from the time of initial
call to the neonatal center has been 1 1/2 hours. The average flying
time from different regional hospitals to Baltimore neonatal centers
ranges from 2-57 minutes.
The most common reason for referral was prematurity and respiratory
distress syndrome, accounting for 52% of the total number of sick
infants transplanted to Baltimore Neonatal Center.
62﻿McCarthy, John T., M.D., Beverly L. Koops, M.D., Peter R. Honeyfield, M.D.
and L. Joseph Butterfield, M.D.
1978 "Who Pays the Bill for Neonatal Intensive Care. Part I:	Transport"
(mimeo). Reprints:	L. Joseph Butterfield, M.D., The Children's Hospital,
Department of Perinatology, 1056 East Nineteenth Ave., Denver, CO 80218.
GENERAL ABSTRACT:
This study examined the characteristics of infants transported by
the Newborn Emergency Service of The Children's Hospital. The service
area covers ten states (500,000 square miles). The study group consisted
of all infants who were transported by the Newborn Emergency Service to
The Children's Hospital for four months during 1976.
The report is a brief treatment of the experience of a neonatal trans-
port system. It describes service area, care provided, patient and family
characteristics and transport charges for services.
ANNOTATIONS:
The mean (N=174) transport charge for all infants was $539, the median
charge was $318, and the mode charge was $180. For the whole group, trans-
port charges represented 5.2% of the total hospital bill. The charge for
transport was directly related to the distance between the hospital of
referral and the tertiary care center. At this time, families must bear
transportation charges as insurance does not include this as a benefit.
The data suggest that further planning for Level II and III centers in rural
areas may need to consider the cost effectiveness ratio of transport
charges as well as duplication and feasibility of services in distant
and/or isolated communities.
McCarthy, John T. M.D., Beverly L. Koops, M.D. and L. Joseph Butterfield, M.D.
1978 "Who Pays the Bill for Neonatal Intensive Care: Part II: Hospitalization"
(mimeo). Reprints:	L. Joseph Butterfield, M.D., The Children's Hospital,
Department of Perinatology, 1056 East Nineteenth Ave., Denver, CO 80218.
GENERAL ABSTRACT:
This report examines hospital charges excluding physician fee and
source of payment for neonatal intensive care for a group of 174 admissions
to The Children's Hospital during four months in 1976. The study used
patient financial accounts as the data source. Each patient's account
63﻿was tracked for 22 months after discharge or until the hospital bill was
paid in full. Accounts paid in full were analyzed to determine total
hospital charges, type of third-party payors, amount of the bill paid
by the family and/or a third party and the amoung written off by the
hospital. Accounts where closure was still pending were projected to
when the bill would be paid in full.
The implications for cooperation between private and public institu-
tions, the importance of counseling families early in the hospital course
of their neonates, and the need for financial planning by the referring
and tertiary care centers are discussed.
ANNOTATIONS:
Eighty-five percent of bills were paid to some degree by third-
party payors, 4% by families and 11% were uncollectable. Fifteen percent
of the bills were not paid in full by 22 months after the last discharge
in the study group. The mean hospital charge (excluding physician's
services) was $10,513. The range was between $541 and $66,652. Seventy-
three percent of the infants had bills of less than $10,000.
The amount paid by the third-party payor was dependent on the type
of insurance coverage. For example, Kaiser, a pre-paid insurance plan,
covered 100% of the hospital bill including the transport charge. In
contrast, Medicaid paid for slightly more than 50% of the total hospital
bill and a small amount of the transport charge. Blue Cross covered
nearly all of the hospital bill but none of the transport charges.
Of the 174 infants transported during the study period, 28.7%
accounts involved a write off. There was a correlation between the type
of insurance coverage a family had and the amount absorbed or written off
by the hospital. Infants covered by Medicaid (15%) were responsible for
a disproportionate (51%) amount of write-offs.
Meyer, Belton
1974 "Transport of High-Risk Infants in Arizona". Regionalization of Perinatal
Care. Report of the Sixty-Sixth Ross Conference on Pediatric Care.
Edited by Philip Sunshine, M.D. Columbus, Ohio:	Ross Laboratories.
GENERAL ABSTRACT:
This document suggests standards for pre-hospital and intensive care
management for newborn patients and pregnant women.
64﻿ANNOTATIONS:
Perinatal Care Delivery
•	In office or clinic, the obstetric team should identify early
any factors that are threats to fetal health; approximately
60% of all fetuses destined to be sick neonates can be identified.
•	A record system should be used that is designed to systematize
the collection, screening, transmission, and analysis of pregnancy
data throughout all its phases, from office through hospital
discharge of mother and infant.
•	The neonatal care team must have full knowledge of the obstetric
data and appreciate its significance.
•	Prenatal records should be forwarded at 36 weeks from office to
hospital in routine pregnancies and subsequently updated.
•	High-risk pregnancy lists should be developed and protocols
established for arrangement of care at the appropriate level
of facility and for involvement of appropriate consultants.
•	Protocols should be established for response to specific fetal
distress indicators and for use of fetal diagnostic procedures
in chronic intrauterine distress, for elective delivery and for
conduct of artificial induction.
Resuscitation
•	A protocol should be established by the medical director in con-
sultation with the regional intensive care center for appropriate
management of neonatal asphyxia, acidosis and shock.
•	Equipment should be pre-arranged in each delivery room or in a
special resuscitation area providing an adequate treatment space
and adequate warmth, usually by means of a radiant heater.
•	Each infant at birth should be characterized through a brief
general examination and the Apgar scoring system at one and five
minutes and serially at five-minute intervals thereafter for infants
with continuing distress.
Transitional Nursery
A protocol should be established for nursing examination of the
infant upon admission to the nursery:	(1) the infant should be weighed
65﻿and the infant's weight and gestational age plotted on an intrauterine
growth chart to detect the large and small for dates, and the low birth
weight due to immaturity; and (2) A protocol should be established for
continuing evaluation of these infants providing for serial determinations
of blood sugar and for initiation of early glucose feedings of intravenous
infusion of glucose as appropriate.
Intensive Care Centers
In various regions, it will be appropriate to develop a cooperative
relationship between various Level III Centers for particular areas of
special care.
Transport
The referring center should prepare appropriate documents and needed
materials to accompany the patient during the transport. These include
copies of maternal and infant records, maternal and cord blood specimens,
laboratory results, bacteriologic studies and x-rays.
The family should be give opportunity to examine and hold or touch
their infant prior to transport and to meet the personnel from the Intensive
Care Center.
Evaluation
In the regional system, all levels of care should participate in
developing mechanisms for reporting and analyzing results to assess
function in the regional system. A major function of the record system
chosen for recording and transmitting patient care information is to pro-
vide a sequential analysis of patient care information which is compatible
with standard data processing methods.
Muirhead, Donald, M.D., Chairman
1971 Report on Perinatal and Infant Mortality in Massachusetts 1967 and 1968.
Committee on Perinatal Welfare of the Massachusetts Medical Society.
Boston, Massachusetts.
GENERAL ABSTRACT:
The report tabulates the major causes as they relate to various ages.
The perinatal group is characterized by asphyxia, prematurity of unknown
cause, abruptio placenta, congenital anomalies, cord abnormalities, Rh
disease, sepsis and placental insufficiency. Infant deaths (less than one
year) were characterized by ideopathic Respiratory Distress Syndrome,
congenital anomalies, anoxia and central nervous system hemorrhage,
sepsis and "Sudden Unexpected Death" as immediate causes.
66﻿ANNOTATIONS:
Preventability and Responsibility
The Committee felt that 35% of the deaths studied might have been
prevented. Responsibility for these preventable deaths was assigned to
maternal factors (30%); obstetrical factors (46%) and pediatric factors
(40%). The report notes that these are not mutually exclusive.
Improving Perinatal and Infant Mortality
Some of the suggested areas for concentrating upon:
Education of the obstetrician—1) The concept of postmaturity (16% of the
obstetrically related preventable deaths in the fetal age group concerned
women beyond the 42nd week of gestation). 2) Techniques of fetal monitoring,
tests of placental dysfunction and fetal well-being. 3) Diabetes, and other
maternal medical and surgical illnesses requiring special expertise. The
concept of the incompetent cervix. 4) Problems concerning early elective
caesarean section.
Education of the pediatrician—1) Sepsis; a continued review of the frequently
non-specific signs and symptoms in the newborn period. 2) Management of the
infant with the Respiratory Distress Syndrome. 3) Signs, symptoms and
management of congenital heart disease. 4) The importance of oxygen therapy,
maintenance of body temperature, hypoglycemia, and acid-base balance.
Some other recommendations include the following:
•	Combining of certain maternity services under one physical plant
unless geographic reasons dictate otherwise.
•	When a high-risk mother is identified prior to delivery, care should
be observed that the hospital of delivery be fully equipped to handle
any of her anticipated problems as well as that of her newborn.
•	Explore the feasibility of a "high-risk" maternal and infant registry.
•	The institution of a statewide transport system divided on a regional
basis for the neonate (and possibly the mother when necessary) to
provide efficient and safe transport to an established regional
intensive care nursery.
•	All maternity services and newborn units should be provided with
consultation services (i.e., neonatologists, internists, obstetrics,
hematologists, etc.) in order that they might provide increasingly
better care for more of their own problems.
•	A system for the automatic matching of birth and death certificates
should be established in our vital statistics departments.
•	The formation of a perinatal mortality committee should be required
for all maternity services. It should form the basis of peer
review.
67﻿Murphy, J. and W.A. Hodson
1974 "Neonatal Intensive Care". Postgraduate Medicine 56:	55-58.
GENERAL ABSTRACT:
Potential Patient Load:	The potential patient load must be estimated
when a neonatal ICU is planned. The total population of the area to be
served, its birth rate, and the incidence of prematurity directly influence
this figure and are in turn directly related to overall socioeconomic
status and to utilization of maternal health care facilities.
ANNOTATIONS:
In an area with a population of one million and a birth rate of 17
per thousand, 5% of newborn infants will be sick and about half of these,
or about 425 each year, will require intensive care. Experience indicates
that the average stay in a neonatal ICU is 12 days. Accordingly, in an
area with a population of one million, 425 infants require 5, 100 bed -
days and 14 beds. If we assume an 80% bed occupancy, about 18 neonatal
ICU beds are required.
Transport:	Safe, efficient transport of infants is of prime importance
in operating a referral unit. The transport vehicle must be rapidly and
efficiently mobilized, must carry a heated incubator capable of maintaining
a stable thermal environment, and must have facilities to administer and
monitor a controlled supply of oxygen. A physician, nurse, or other
medically trained person adept in resuscitation procedures should accom-
pany the infant. The infant being transported should be readily visible
to the attendant and easily accessible for care and resuscitation.
Ground transport is not practical within a 20 to 30 mile radius of the
central hospital. Air transport may be more rapid beyond this radius.
Staff:	The ratio of nurses to patients in a neonatal ICU may vary
from 1:1 to 1:3, depending on the number of infants in the unit who are
critically ill or convalescing. Consideration should be given to several
innovative staff patterns, including nursing shifts of ten hours to provide
overlap and better continuity of care. A team made up of nurses aides,
practical nurses, and highly trained registered nurses may prove more
efficient and cover a broad spectrum of duties, ranging from feeding and
bathing of infants to resuscitation and ventilator support.
All diagnostic laboratory facilities as well as pulmonary therapy
services must be accessible within the hospital 24 hours a day, which means
the appropriate personnel must be available at all times. Micromethod
determinations are essential for biochemical and hematologic monitoring
of neonates, whose total blood volume is only 80 ml/kg of body weight.
68﻿Pomerance, Jeffrey, M.D., Christine T. Ukrainski, M.D., Tara Ukra,
Diane H. Henderson, M.D., Andrea H. Nash, M.D. and Janet L. Meredith, R.N., B.S.
1978 "Cost of Living for Infants Weighing 1,000 Grams or Less at Birth".
From the Department of Pediatrics, Ceders-Sinai Medical Center and
University of California at Los Angeles. Pediatrics 61:	908-910.
GENERAL ABSTRACT:
This article reports the in-hospital cost of caring for 75 infants
weighing 1,000 grams or less at birth who were born during the two and
one-half year period between January 1973 and June 1975. Thirty infants
(40%) survived. Nineteen of 27 infants tested (70%) appear to be
neurologically and developmentally "normal" at one to three years of age.
Hospital charges were adjusted to September 1976 rates and corrected
for a 94% collection rate. Physicians' fees represented less than
5% of the total bill and were not included. The average adjusted daily
and total costs for the 45 infants who died were $825 and $14,236,
respectively. The average adjusted daily and total costs for the 30
survivors were $450 and $40,287, respectively. The average adjusted total
cost per "normal" survivor was $88,058. It is the authors' belief that
the outcome justifies this expense.' Society, however, must be the ultimate
j udge.
ANNOTATIONS:
Overall Cost
The total adjusted cost for the 45 non-surviving infants was $640,634.
The total adjusted cost for the 30 surviving infants was $1,208,582.
The overall total adjusted cost for both groups was $1,849,216. The
percentage breakdown of the total charges for the 75 infants is as
follows:	room charges, 43%; ventilation and oxygen support, 19%; blood
gases, 11%; pharmacy, 9%; miscellaneous, 1%. Fully 30% of the total
charges was for support and management of ventilation and oxygenation.
(Changes were adjusted to September 1976 rates.)
If the total adjusted cost for the 30 survivors alone is used to cal-
culate the cost per survivor, the result is $40,287. This figure
greatly underestimates the true cost, because care provided to infants
who die must surely be included in any calculations of cost per sur-
vivor. When total adjusted cost for all 75 infants is used to calculate
cost per survivor, then the figure becomes $61,641 per survivor. Survival,
however, should not be taken as the sole measure of success. A much more
meaningful definition would include evaluations of the infant's potential
to become a normal, productive member of society. In developmental/
neorological follow-up evaluations, 19 to 27 infants (70%) were apparently
functioning normally. If we assume that three infants whose parents did
not permit them to be examined had approximately the same developmental
69﻿quotient distribution as the other 27 infants who were examined, then the
figure of 21 infants (19 of 27 infants examined plus 2 of 3 not examined)
may be used as the most realistic denominator to equate cost per "normal"
survivor. This figure is $88,058.
Quebec Perinatal Mortality Committee
1967 "Analysis of Perinatal Deaths by Cause, Part Two". Quebec, Canada.
Available from:	Quebec Perinatal Committee, Ministry of Social Affairs
Province of Quebec, 1005 Chemin Ste-Foy, Quebec, Canada, GlS 4N4.
GENERAL ABSTRACT:
In this analysis the cause of death was assigned according to a
specific classification on the basis of information received in a de-
tailed (150 item) questionnaire. These questionnaires were completed for
stillbirths and first week neonatal deaths by perinatal mortality committees
organized in each hospital with obstetric facilities. This information
was reviewed by the provincial committee and the assigned cause of death
was either approved or amended according to circumstances.
During the first year of operation of the Perinatal Mortality Committee,
detailed questionnaires were received for 46% of the Province's 2,960
perinatal deaths. These deaths occurred among 104,168 births registered
in the 158 obstetrics services in Quebec. All told, 81 hospitals completed
and returned questionnaires and among these, 50 hospitals submitted analysis
on 75% or more of their perinatal deaths.
For most purposes, only deaths of infants weighing over 1,000 grams
at birth were analyzed. In fact, infants whose weight at birth is less
than 1,000 grams almost always (97%) die and analysis of the cause of
their death elicits little information of value in assessing preventive
factors.
There were 103,476 infants born in the province weighing over 1,000
grams; of this number 2,289 died, and 1,005 of these cases (44%) were
analyzed on the appropriate questionnaire by the respective hospital
committee.
Total perinatal mortality is 17.9 deaths per 1,000 births. "Asphyxia"
of various causes is responsible for 6.1 deaths per 1,000 births; "mal-
formations", for 32; "respiratory syndrome", for 2.7; the other causes
together account for 2.4 per 1,000. After these deaths have been accounted
for, 3.5 deaths per 1,000 remain explained.
ANNOTATIONS:
Asphyxia
Neonatal deaths or stillbirths were considered due to "asphyxia" if
asphyxia before or during birth, unrelated to pre-existing fetal pathology,
70﻿was thought sufficient to account for the death. Deaths were attributed
to "unexplained peripartum asphyxia" when fetuses died during labor of
uncertain cause or neonatal deaths occurred due to birth asphyxia of
unknown cause. Unexplained stillbirths dying before labor were listed as
unexplained deaths, rather than deaths due to asphyxia. Death by asphyxia
of infants who were significantly underweight for gestational age was
assigned to "fetal malnutrition" rather than to "asphyxia" per se.
Asphyxia from all causes except fetal malnutrition was responsible
for 2.2 neonatal deaths and 3.9 stillbirths per 1,000 births. The death
rates from specific causes of asphyxia are documented in this paper. Thev
are, briefly, antepartum hemorrhage, umbilical cord loops, knots or
hematomata, prolapsed cord, unexplained peripartum asphyxia, certain
maternal diseases, and meconium aspiration.
Comments:	Improved fetal heart monitoring, intensive care during
labor and delivery, and expert infant resuscitation at birth could pre-
sumably reduce the rate of death by asphyxia when asphyxia occurs during
labor or delivery.
In 1967, there were 186 such peripartum asphyxia deaths which could
have been prevented had conditions been ideal. The report presents de-
tails of this.
Congenital Malformations
This category includes any perinatal death due to a malformation,
a complication of a malformation, or to corrective surgery for such a
malformation. In addition, infants dying of other or ill-defined causes
who had life-threatening or potentially serious malformations were
assigned as malformation deaths rather than to causes such as infection,
respiratory distress syndrome, fetal malnutrition, or unexplained deaths.
Malformations, thus defined, accounted for 3.1 deaths per 1,000 births,
and 70% of them were neonatal deaths. Of all malformation deaths,
anecephaly accounted for 1.1 per 1,000 births, other central nervous system
malformations .5, heart 0.7 gastrointestinal 0.3, genitourinary 0.3,
chromosomal 0.1, and other malformations 0.1. Of the 122 liveborn infants
dying of malformations, in only 9 cases were the lesions such that surgical
correction had a fair chance of being successful.
Comments:	Only an insignificant proportion of malformation deaths
could have been avoided with better treatment. With present lack of under-
standing of the etiology of most anomalies, these losses of 3 or 4 infants
per 1,000 births must be considered inevitable. Furthermore, it must be
noted that deaths from malformations also often occur after 7 days of life
and are not, therefore, included here.
Respiratory Distress Syndrome
Respiratory Distress Syndrome accounted for 2.7 deaths per 1,000
births. Autopsies were performed in 66% of the cases, and hyaline membranes
were found in 88% of those autopsied.
Comments:	Progress in neonatal intensive care has greatly reduced
mortality from R.D.S. It is probable that at least one-half of the R.D.S.
deaths would be avoidable if neonatal intensive care were generally available.
71﻿Iso-immunization
This condition is considered the cause of death when the diagnosis of
iso-immunization is confirmed serologically or at an autopsy. Death may
occur in utero or may occur after birth from hydrops, anemia or kernicterus.
Since serological tests and autopsies are often lacking in the deaths re-
viewed across the province, the true mortality from this cause must be
somewhat higher than that obtained in this study.
This cause of death accounted for 0.4 neonatal deaths and 0.5 stillbirths
per 1,000 births. Ninety percent of these deaths were due to the Rh "D"
antibody. Of the 24 neonatal deaths from iso-immunization, 19 were due to
anemia with hydrops, 2 to kernicterus, and 3 to complications of exchange
transfusion.
Comments:	Since the incidence of iso-immunization is usually 8 per
1,000 births, a perinatal mortality of 1.1 per 1,000 births suggests that
mortality rate among affected infants was about 1.1/8 or 14%, a generally
accepted mortality rate for this disease.
As most infants with iso-immunization receive one or more exchange
transfusions, it can be estimated that 7 deaths from exchange transfusions
in perhaps 800 affected infants would mean an operative mortality rate of
1%, which compares favorably with published rates.
Deaths from Rh hemolytic disease after 32 weeks of gestation are
preventable by selective premature delivery of severely affected infants.
Hence, more than 50% of the 1967 deaths could have been prevented.
Trauma
Birth trauma was considered the cause of death only when a specific
traumatic lesion sufficient to cause death was found in an infant after a
difficult delivery. Infants who died from birth asphyxia after difficult
deliveries, without definite evidence of such specific traumatic lesions,
were assigned as having died from asphyxia due to abnormal labor and delivery
rather than from trauma per se.
Birth trauma was assigned as the cause of death in 0.1 deaths per 1,000
births. The 7 cases included 6 of trauma to the head, and 1 to internal
organs.
Comments:	Deaths from trauma are potentially preventable, given early
diagnosis of obstetrical problems and ready access to Caesarean section
delivery.
Possibility of Reducing Perinatal Deaths
In the present state of medical knowledge, certain perinatal deaths
cannot be significantly reduced. Such are stillbirths occurring before
labor, either because of antepartum hemorrhage, umbilical cord loops, or
for no apparent reason.
This section suggests that the present perinatal death rate of 22.1
per 1,000 births weighing over 1,000 grams could presumably approximate the
irreducible minimum of 8.3 per 1,000 births, which would mean that 62%
of the present perinatal mortality rate could potentially be prevented,
in the widest sense of the word.
72﻿It is noted that infants weighing 2,500 grams or more at birth,
representing 93% of the population, accounted for only one-quarter of
neonatal deaths. Infants of 1,000-2,499 grams, 7 deaths, and the 500-999
grams newborn infants, who comprise less than half of 1% of the population,
answer for the remaining one-quarter of all neonatal deaths.
Consequently, efforts to reduce neonatal mortality should be con-
centrated on application of more effective preventive measures and
resuscitative treatment of birth asphyxia, as well as toward improvement
of the management of the low birthweight infant with respiratory distress
syndrome.
Age at Neonatal Death
Of infants weighing more than 1,000 grams who died during the first
week, 42% died within 12 hours and 79% within 48 hours. Forty-five percent
of infants of birthweight 1,000-2,499 grams, and 26% of infants dying of
R.D.S. had died by the 12th hour following birth. A similar high frequency
of early death is noted among infants dying from iso-immunization or of
unexplained causes. As these are the weight groups and diseases most
responsive to neonatal intensive care, it is evident that intensive care
facilities should, wherever possible, be available within, or readily
accessible to, the maternity hospitals where such infants are born if
considerable losses during the first 12 hours are to be prevented.
Deaths (and brain damage) from birth asphyxia are only reducible by
intervention during labor and by expert resuscitation at delivery. Birth
asphyxia is not susceptible to treatment after the first few minutes of
life; it requires, therefore, intensive care facilities within the hospital
where delivery is made.
Effect of Intensive Care and Hospital Size
Overall perinatal mortality is 16.6 and 18.4 for hospitals with 1,000
deliveries a year or over with an intramural neonatal intensive care service,
or referring cases to hospitals so equipped. The mortality rate is 20.3
for the same category of hospitals not utilizing either form of intensive
care facility, and 24.3 per 1,000 in hospitals registering less than 1,000
deliveries a year. The difference is mainly due to:
•	A lower rate of mortality due to respiratory distress syndrome
(1.65 deaths per 1,000 for cases with intramural neonatal intensive
care facilities and 3.5 per 1,000 when intramural intensive care
is lacking);
•	Deaths from unexplained causes, which are approximately 50% higher
in hospitals not using neonatal intensive care facilities; and
•	Deaths from asphyxia which are approximately 35% higher in small
hospitals.
73﻿Quebec Perinatal Mortality Committee
1975 Significant Reduction in Perinatal Deaths in Quebec, 1967 to 1973. Press
Release (December). Quebec, Canada. Available from:	Quebec Perinatal
Committee, Ministry of Social Affairs Province of Quebec, 1005 Chemin
Ste-Foy, Quebec, Canada, G1S 4N4.
GENERAL ABSTRACT:
The Quebec Perinatal Committee has analyzed all perinatal deaths which
occurred in the province since 1967. This study permits an evaluation
of the perinatal health care situation in Quebec.
ANNOTATIONS:
In 1967, the perinatal mortality for infants weighing more than 1,000
grams at birth was 22.1 per 1,000. In 1973, this rate had fallen to
14.4 per 1,000, a fall of 35% in 6 years.
There were 2,289 perinatal deaths over 1,000 grams in 1967 and 1,265
in 1973, that is a difference of 1,024 of which only one-third is
attributable to the decreasing birth rate. Most deaths occur among low
birthweight infants (2,500 grams or less). It was determined that the
decrease in mortality in Quebec was not attributable to a reduced incidence
of low birthweight infants which remained unchanged over the 6-year period.
The reduced perinatal mortality rate is, therefore, considered largely
due to the improved care delivered to the mother, fetus and newborn.
Improved obstetrical care is evident by a 33.9% reduction in stillbirth
rate. Combined improvements in obstetrical and neonatal care have resulted
in a 36.4% decrease in neonatal mortality.
The authors explain that regional differences in mortality depend,
to a large degree, on the accessibility of care for the low birthweight
infant. Neonatal mortality for low birthweight infants is 86 per 1,000
for regions other than Montreal, 49 per 1,000 in the Montreal region, and
24 per 1,000 in the hospitals comprising the "optimum" group.
If the best modern medical techniques were generally available, the
irreducible minimum perinatal mortality rate could be approached. The
authors explain that each year about 3 infants per 1,000 die of lethal
malformations incompatible with life. Four other deaths occur "in utero"
during an apparently normal pregnancy, prior to onset of labor, unexpectedly.
Also, approximately 1 or 2 more fetal deaths per 1,000 occur from potentially
treatable conditions, but the fetus dies too early in pregnancy for inter-
vention to be effective. These 8 or 9 deaths per 1,000 births represent
the "irreducible minimum" perinatal mortality rate.
The following is a summary of organizational developments now taking
place in Quebec, as reported in this release:
• The largest obstetrical hospital services are developing specialized
intramural intensive care units for high risk pregnancy and neonatal
74﻿care, and are training their personnel in order to give optimum
care to mothers and babies at risk.
•	The regrouping of hospital services is making it possible to
develop larger obstetrical services with better equipment and a
better qualified personnel in order to face all situations.
•	Women who during their pregnancy present with complications that
can compromise the life of the fetus, or those who enter into labor
too prematurely, are referred by their physician to specialized
centers where mother and baby can receive all of the care that may
be required before, during and after birth.
•	Newborn infants delivered in centers not equipped for intensive
care, who present problems at or after birth, are transferred by
trained personnel to neonatal intensive care centers as quickly
and safely as is possible.
•	Public health programs in perinatality are being developed across
the province. These preventive programs are aimed to improve
prenatal care and to inform and educate the public as the need
for early care in pregnancy as well as for good nutrition and
planned pregnancy. These programs include prenatal courses
on pregnancy and delivery with an emphasis on education, physical
conditioning, as well as psychological preparation for the coming
event.
Quilligan, Edward, M.D. and Philip Sunshine
1974 "Regionalization of Perinatal Care". Report of the Sixty-Sixth Ross
Conference on Pediatric Research. Columbus, Ohio:	Ross Laboratories.
GENERAL ABSTRACT:
A variety of systems needs and goals are described through the ex-
periences of the participants in the Sixty -Sixth Conference on Pediatric
Research. Heading topics discussed here include the progress of developing
regionalized perinatal care systems in the United States, quality standards,
recommended manpower goals, continuing education programs, consumer
interest factors, and program effectiveness criteria. This comprehensive
review of the major medical, demographic, and social factors which
contribute to the trend toward the implementation of centralized community
or regionalized perinatal programs underscores the increasing responsibility
of physicians, government, and the public in planning for high-risk
birth centers.
75﻿ANNOTATIONS:
Developmental and Educational Aspects of a Regionalization Program
In the State of Wisconsin a three-year retrospective analysis was
undertaken in order to determine the status of perinatal health care.
All fetal and neonatal deaths, and all births of less than 2,500 grams were
reviewed in 35 hospitals outside the resident county of the two area
medical schools in the State.
Two-thirds of the neonatal deaths and half of the fetal loss were
determined to be preventable. Approximately 50% were determined preventable
by utilization of a center with team resources.
Three programs were then initiated:
•	Hospital perinatal surveys by a team of physicians and nurse
practitioners to interview physicians and nurses.
•	Development of perinatal institutes consisting of eight to ten
educational programs per year for nurses and physicians throughout
the State.
•	Establishment of regional neonatal centers with commitments to
education.
Since the initiation of neonatal intensive care units in 1959,
coupled with a statewide, continuing education program, there has been
a progressive decline in the neonatal mortality rate.
It is suggested that education has been the major consideration on
continuing improvement in perinatal health care in Wisconsin. Further,
peer review is cited as being essential to education, and is effective
only when all of the interested parties review the case at the same time.
As suggested, adding physicians and nurses in Wisconsin is not the
answer to improving high-risk birth care. The problem, rather, is appro-
priate location of the intensive care center in response to a clearly
defined critical need.
In Wisconsin the following guidelines were used in determining peri-
natal care service regions:
•	Design State and community needs assessment.
• Look for a medical group which has time to devote to high-risk care.
• Look for a hospital that will take risks both financially and
procedure-wise.
•	Divide regions so that each has from 7,000 to 10,000 live births.
•	Feasibility of transportation within a region.
•	Feasibility as to where physicians and families will feel free
to go for care.
76﻿•	The center personnel can be responsible for the education
within the region.
Perinatal Health Care in Mississippi, 1973
The experience with perinatal health care in Mississippi is well
representative of the results gained in neonatal and infant mortality and
morbidity. In 1968, Mississippi had the highest neonatal and infant
mortality rates in the county. The rates for non-whites in both of these
categories are higher than for whites. Contributing factors to this are:
a statewide low birthweight rate of 15%, a deficit in parental education,
a higher illegitimacy rate, and the overriding factor, the low income
status of the infant's family.
Yet, in Holmes and Hinds Counties a plan, described as a prototype
for perinatal health services, was implemented recently. The program
included:
•	Augmenting and upgrading the county health department services to
the medically indigent pregnant and postpartal women and their
infants, which essentially meant a change from the traditional
health department's role of instituting preventive health measures
to one of offering complete care to the ambulatory patients,
including intensive home visiting.
•	Providing local physicians' services to the high-risk mothers
and babies as well as community in-hospital care.
•	Training additional health workers in existing and new categories
such as nurse-midwives, obstetric technicians and health aides
to provide more health personnel for the target area.
•	Establishing a referral system to the university medical center
for high-risk obstetric patients.
Planners from the County Health Improvement Program believed that
concentrating on and augmenting care during the maternity cycle and the
first year of life would raise the level of health care of child-bearing
women and their infants, decrease the numbers of birth-defective and
damaged citizens, and ultimately improve the quality of life.
Statistics indicate the neonatal and infant mortality rates have been
reduced in Holmes County in the non-white population, the segment of the
population primarily using these services, from above the national average
in both categories in 1968 to near the national average in 1972.
The Role of the State Department of Health in Regionalized Perinatal Care
In 1970, the Massachusetts Department of Public Health began to develop
conditions of participation for newborn services under the Medicaid
program. Also, it was decided to revise completely the State licensure
standards for obstetric and newborn services in order to make licensure
77﻿and certification standards compatible and uniform, and to promote high-
quality care and to introduce and facilitate new concepts of maternity
and newborn care.
The new regulations were developed with the assistance of a multi-
disciplinary task force composed of groups and individuals involved in
obstetric and newborn care. They have had an enormous impact, and it
is reported that Massachusetts has seen substantial improvement in
maternity and newborn services. Areas of important new regulation are:
•	The establishment of specific definitions and requirements for
special-care nurseries and for high-risk newborns.
•	The requirement that every newborn service develop specific
plans for high-risk newborns. Under the new regulations, hospitals
with more than 2,000 deliveries were required to provide a special
care or transfer nursery. Two thousand deliveries were required
to develop a plan, which has to be approved by the State health
department, for the care of their high-risk newborns either by
transfer or special provisions within the hospital.
•	The establishment of more comprehensive and sensitive services
including parent education programs, social services, case follow-
up, provisions of family planning services and liberalized
visiting policies.
The Development of Infant Transport Systems
In May of 1973, a national conference on research and maternal and
child health was held in Berkeley, California. The results of the con-
ference, a comparison of transported and nontransported infants, is
outlined in this chapter.
In 1967, Phoenix was selected as a demonstration center for trans-
porting high-risk infants from rural hospitals to two metropolitan private
hospitals having a combined rate of over 8,000 births per year.
After three years of experience with the system of newborn intensive
care and transport, significant reduction in neonatal mortality occurred.
The most striking change was observed among infants weighing 1,001 to
2,500 grams who were born in the centers. Among these infants, the
mortality was reduced by half from a formerly fixed rate of approximately
100 deaths per 1,000 live births.
A later follow-up survey of 273 babies who had received care in the
centers through February, 1971, yielded data on 147 survivors, of whom
132 appear to be well, normally developing children. Fifteen have some
physical or developmental problems.
In Arizona, there are now four state transport centers. It is estimated
that greater than 70% of newborn infants needing newborn intensive care
now receive it. As the percent of babies covered has increased, the State
neonatal mortality rate has declined from 17.3 to 11.5 deaths per 1,000
live births.
78﻿Evaluation of Equipment in Transport Systerns
The variety of available transport units which have been reported
range from a fully equipped mobile intensive care unit with complete
facilities through a minimal vehicle equipped with an indoor incubator.
The electrically powered equipment which is used will often include a
heart rate monitor, along with an oscilloscope, a temperature monitor, an
oxygen monitor and, less often, a blood pressure monitor.
The function of the transporter is stated as follows:
•	The infant must be kept warm enough under all possible conditions
which may be encountered to approximate the thermoneutrality
which imposes the least drain on his scarce energy reserves.
•	He can be oxygenated conveniently.
•	He is "safe" within the transporter.
•	He is easily accessible for the delivery of intensive care during
transport.
Quilligan, Edward, M-D.
1972 "The Obstetric Intensive Care Unit". Hospital Practice 17:	61-69.
GENERAL ABSTRACT:
Neonatal intensive care requires a concentration of equipment and
testing procedures that permits fetal-maternal status to be evaluated
in much greater detail than is possible by traditional indices adequate
for normal deliveries. The article discusses these in some detail and
describes how the technology is housed and the patients for whom it is
designed.
ANNOTATIONS:
Approach
At the University of Southern California, there is a three-phase
approach to high-risk patient care for the newborn:
• A special clinic.
• A special antepartum unit.
•	A special intrapartum unit (labor intensive care).
79﻿Manpower
In the antepartum ICU the ideal nurse-to-patient ratio is 1:5. In
the labor intensive unit the nurse requirements are one for each patient.
At least two physicians, who are residents or fellows, are in the unit at
all times and the staff director is at hand. Pediatric coverage and OB
anesthesia should also be available.
Bed Need
The article suggests that for OB ICU's with an annual delivery rate
between 2,000 and 3,000 a unit of 10 beds—8 antepartum and 2 intrapartum
covers most needs.
Procedures Standard in ICU
•	Fetal heart rate monitoring.
•	External strain gauge. A diaphram device fitted over the abdomen
can document the occurrence but not the intensity of labor
contractions.
•	Ultrasonic detectors. Provide indications of the moving fetal
heart's alteration of reflected sound frequencies. This provides
qualification of fetal heart rate.
•	Measurement of urinary estriol. During the last trimester of
pregnancy, this procedure provides a highly useful method of
following the health of the fetus. Total estrogen output is pre-
empted almost entirely by the fetaplaenatal unit as pregnancy
advances; variations from normal excretory values may be associated
with fetal jeopardy.
•	Lecithin/sphingomyelin ratio. This is helpful in determining whether
the fetus has achieved pulmonary maturity.
•	Intrapartum procedure. This is the analysis of the acid-base
balance in the fetal blood obtained by scalp puncture at various
intervals during labor.
Hence, the above procedures are examples of NICU techniques that make
possible a more intimate and objective view of the fetus' progress through
pregnancy, labor, and delivery.
Reynolds, T. Thompson
1977 "The Results of Intensive Care Therapy for Neonates". Department of
Pediatrics and Division of Neonatology. University of Minnesota.
Journal of Medicine 59.
80﻿GENERAL ABSTRACT:
This paper reviews whether NICUs have reduced neonatal mortality
rates and whether improvements in the care of sick neonates has decreased
the incidence of long-term sequelae among NICU survivors. The paper
offers a comprehensive, general discussion of neonatal mortality rates
since the advent of NICUs. Also, it discusses neonatal mortality rates
and long-term prognosis for low birth weight infants. This may be recommended
as a helpful overview for planners and decision-makers who are trying to
understand the various benefits which can, indeed, be derived from
implementing a regional system.
ANNOTATIONS:
Overall Neonatal Mortality Rates
Within one to seven years after the introduction in hospitals of
NICUs for care of critically-ill neonates, overall neonatal mortality
rates in the hospitals were reduced by 17% to 76%.
Low Birth Weight (LBW) Newborn Infants
Newborns with birth weights under 2501 grams account for 67% to 75%
of neonatal deaths with the majority of first-week deaths not due to
congenital anomalies occurring among the approximately 1% of live births
in the USA and Canada weighing 500 to 1501 grams.
Usher has reported that the mortality rate for infants 1001-1500 grams
at birth was reduced by one-half with the establishment of an NICU at
Royal Victoria Hospital in Montreal. During 1974-1975, 59% of 105 live
births weighing 501-1500 grams at birth survived at Royal Victoria Hospital,
including 94% of the infants 1251-1500 grams and 74% of the infants
1001-1250 grams at birth.
Long-Term Prognosis
Prior to the advent of intensive care therapy for LBW neonates, the
incidence of neurologic and/or intellectual deficits among LBW survivors
often exceeded 50%. The incidence and severity of such handicaps in-
creased as birth weight and gestational age decreased. Drillien has
stated that the development of the neonate delivered prematurely as a re-
sult of a catastrophe or accident is totally dependent on the quality of
care delivered in the perinatal period.
In Los Angeles, California, the improved prognosis for infants under
1500 grams, birth weight has been attributed to intensive care management:
22% of 37 infants born in 1964-1965 had definite neurologic or develop-
mental deficits compared to 3.3% of 60 infants delivered in 1969-1970.
A review by the Maternal and Child Health Task Force of the State
Health Planning Advisory Council in the State of Michigan of the effective-
ness of intensive care in rendering long-term morbidity for neonates
81﻿revealed that 50% to 70% of the cases of severe mental retardation due
to perinatal factors could be prevented, resulting in a 20% to 30% re-
duction of children with severe mental retardation. Improved social,
educational and economic conditions would decrease neonatal mortality
rates by reducing the incidence of LBW infants, particularly those
weighing under 1500 grams.
Richardson, Joan C., M.D.
1976 "Principles of Organization of a Neonatal Intensive Care Unit from
Scratch". Clinics in Perinatology:	Organization for Perinatal Care
3:	329-335. Guest Editor:	Louis Gluck, M.D. Philadelphia:	W.B.
Saunders Company.
GENERAL ABSTRACT:
The essential first step in planning a unit is the evaluation of
community and regional needs for neonatal intensive care as well as the
resources and capability of the hospital that undertakes to provide
those needed services.
ANNOTATIONS:
Potential Patient Load and Size of the Unit
In planning a neonatal intensive care unit, it is necessary to pro-
ject the potential number of patients to be cared for. The total population
of the area to be used, its birth rate, incidence of prematurity, and
neonatal mortality rate directly influence this figure. This in turn
is very important because it provides a basis for determining what the
bed capacity of the unit should be, (Swyer, P.R.:	The regional organi-
zation of special care for the neonate. Pediatric Clinics in North
America, 17:	761, 1970).
The bed requirement for a neonatal intensive care unit equals:
regional neonatal mortality rate X 3 X number of live births in thousands
for the region
60
Recent data indicate that the average occupancy of a bed in a NICU
is approximately 10 to 11 days, permitting one bed to service about 35
patients per year, and thus, increasing the bed requirement for any given
region above the number suggested by Swyer (see above reference).
Economics
Construction and remodeling costs range from $60 to $80 per square foot.
Equipment expenditures approximate $12,000 to $15,000 per intensive care
82﻿bed. A minimum cost range for one day of intensive care is $200 to $300.
A facility providing service to a population with 10,000 births, delivering
about 2,000 maternal cases, and receiving 300 to 400 neonatal transfers
has an annual operation budget of $2.6 million.
Transportation
The unit must be geographically and demographically located such
that difficulties in the transportation of high-risk infants are minimized.
Design and Space
For each intensive care bed, 80 to 100 square feet are required with
40 to 60 square feet per intermediate care bed and 20 to 30 square feet
per bed for other care areas. Areas for non-patient care activities
should be included in the overall design of the unit.
Equipment and Utility Requirements
Each intermediate care bed requires 6 to 8 electrical outlets plus 2
oxygen and 1 outlet each for suction and compressed air. Equipment needs
in the intermediate care area vary according to the individual infant's
condition, but most likely equipment for temperature support, continuous
cardiac monitoring, and indirect blood pressure determination is required.
Respiratory (apnea) monitors may be indicated in some cases.
Personnel
The NICU area requires a nurse to patient ratio of 1:1 or 1:2, whereas,
a 1:3 to 1:5 ratio is necessary to staffing the intermediate care area.
All other areas can be adequately staffed at a 1:4 to 1:6 nurse to
patient ratio.
Support Services
Lab, radiology, blood bank, inhalation therapy, social service and
medical electronics.
Robert Wood Johnson Foundation
1978 Special Report:	Regionalized Perinatal Services
Princeton, NJ:	The Robert Wood Johnson Foundation
GENERAL ABSTRACT:
This report is written in an understandably journalistic style and pre-
sented in a magazine format. It is valuable as background material for
83﻿technical committee and board members. Four articles and one graphic
presentation are included.
ANNOTATIONS:
"Pregnant Women and Newborns at Risk Are the Focus of Efforts Nation-
wide to Regionalize Perinatal Services" explains the need for regionaliza-
tion, the high risk phenomenon and its relation to infant mortality, and
the emergence of perinatal and neonatal medical specialties. The article
details functions of a regional perinatal system and discusses essential
aspects as reflected in the seven regional demonstration projects funded
by the Robert Wood Johnson Foundation. System organization, (the Committee
on Perinatal Health model), communications, transport (of mother and infant),
medical records and education programs are briefly discussed. The issues
which data from the demonstration projects are expected to address are
also discussed.
"Adam and Amanda" is a human interest article following the experience
of two high-risk births.
"The Evaluation Process:	Spectrum of Morbidity" presents preliminary
data in graphic form on morbidity of a random sample of infants from the
demonstration regions by birth weight, race, family income and prior
pregnancy history.
"The Technologies for Detecting and Managing High Risk" briefly
defines eight technologies and their uses:	ultrasound (sonography),
amniocentesis, estriol measurements, lecithin/sphingomyelin ratios,
oxytocin challenge test, fetal monitoring, ventilatory support for the new-
born and neonatal monitoring.
Russel, Keith P., M.D., Sprague H. Gardiner, M.D., and Ervin E. Nichols, M.D.
1975 "A Conceptual Model for Regionalization and Consolidation of Obstetrics-
Gynecologic Services". American Journal of Obstetrics and Gynecology
121:	756-764.
GENERAL ABSTRACT:
A conceptual model for the organization of regionalization is pre-
sented with consideration of patient care and services, medical education
and research as they apply to the model.
ANNOTATIONS:
The model set forth in the article is the three-Level model suggested
by the National Perinatal Committee.
84﻿Population
In making birth projections, certain assumptions must be made. In
reviewing the population estimates and projections published by the
United States Department of Commerce, the cohort fertility method is
used. The key assumption in this method is the average number of
children a cohort of women will bear during its lifetime.
The "completed cohort fertility rate" is the number of children
born to a cohort of 1,000 women upon completion of childbearing.
Birth data collected in 1971 and 1972 showed a major drop in birth
expectations; thus, we are now at about 2.1 births per woman in the
childbearing years. Using a more conservative projection, such as 1.8
births per woman, we find that by 1985, we should expect from 3.6 million
up to 4.5 million births per year.
The 1967 ACOG "National study of maternity care" indicated that over
64% of all deliveries at that time were carried out in hospitals delivering
1,000 or more infants per year, and that another 20% were in hospitals
that delivered between 500 and 999 babies per year.
It was also shown that 16% of the deliveries were done in hospitals
which had less than 500 deliveries per year but represented 56% of
hospitals surveyed.
Another significant factor brought out in this study was the utiliza-
tion of hospital facilities. The ACOG Committee on Professional
Standards has suggested that three delivery rooms should be adequate
for 3,000 deliveries. In the ACOG study, it was shown that this level
of utilization did not occur until the delivery load reached 1,000 per
year or more. The same situation prevailed in the use of maternity beds.
Following the initiation of a neonatal intensive care unit at the
University of Tennessee Medical Units in Memphis in the middle of 1970,
the overall neonatal mortality rate fell from an annual average of 11.7
per 1,000 live births over the preceding eight years to 8.5 per 1,000
during the year starting July 1, 1970.
Neonatal Mortality Rates in Regional Programs
In Wisconsin the statewide neonatal mortality rate had been declining
very slowly from 16.9 to 14.9 per 1,000 live births between 1960 and 1968.
The continuing education program was started in 1967 and 4 intensive care
units were opened in 1968 and 1969. In 1969 the statewide neonatal mortality
rate dropped fairly sharply to 12.9 per 1,000, with a further drop to 12.6
in 1970. The full impact of the program was evident in the area served
by one of the intensive care units which had been opened in 1968. The
neonatal mortality rate in the units service area, which had been hovering
around 15 per 1,000 between 1960 and 1968 declined to 12.8 in 1969, with
further decline to 11.7 in 1970.
85﻿Ryan, George M., Jr., M.D.
1977 "Regional Planning for Maternal and Perinatal Health Services". Seminars
in Perinatology:	Regionalization of Perinatal Care 1:	255-266. Guest
Editor:	Leo Stern, M.D. New York:	Grune and Stratton.
GENERAL ABSTRACT:
This is an excellent presentation of issues and concepts in the
regionalization of maternal and perinatal care. It summarizes the
history and current problems in maternal and perinatal care, types of
patients which are considered "high risk", and describes the Committee
on Perinatal Health model of a regional system. The article details
functions and care capabilities of institutions within the system, examines
issues in developing perinatal systems and describes the outcomes of
programs in Arizona, Wisconsin, Massachusetts and Quebec.
ANNOTATIONS:
Responsibilities and special services in a regionalized system.
Level I Facilities
The Level I unit must be justified by geographic necessity. Its
responsibilities are to provide uncomplicated maternal and perinatal care
in areas not served by other units and to provide emergency management of
unexpected complications of mother or infant. This hospital deals with
an isolated population and, therefore, must conduct an active program
of early identification of high-risk patients so that appropriate arrange-
ments can be made for consultation and/or delivery at another site.
Personnel should be familiar with services available elsewhere in the system..
Electronic fetal monitoring should be available at Level I units to
allow for the earliest identification of fetal danger during the intra-
partum period. All delivery room personnel must receive ongoing education
on neonatal resuscitation, and each delivery room should have appropriate
equipment for current resuscitation techniques.
The Level I facility should have type 0, Rh-negative blood and fresh
frozen plasma in the facility at all times and blood available on call on
a 24-hour basis. Anesthesia and delivery room capability should allow
cesarian section to begin at all times within 30 minutes of a decision
for its necessity.
Level II Facilities
Level II responsibilities include complete care for the uncomplicated
obstetrical patient as well as the ability to manage most obstetrical
complications. Maternal and perinatal intensive care is not included.
86﻿Because of the increased safety offered by the capabilities of a Level II
unit as compared to Level I, it is desirable that these requirements are
promulgated throughout the United States. If this occurs, those
hospitals not meeting these standards would continue in operation
only because of a clearly demonstrated need for their presence.
Special facilities provided by Level II units should include an
operating room for cesarean sections within or in close proximity to the
obstetric suite. Isolation facilities should be available for mother and/or
child. Special facilities for neonatal care should include a transitional
nursery in which all newborn infants may be observed and stabilized
before a decision as to final disposition is made and a special care
nursery in which compromised infants can receive definitive treatment or
treatment prior to transfer, if needed, to a Level III unit. Facilities
must also be provided for ancillary services, such as required laboratory
determinations, x-ray, ultrasound, and blood bank.
Special services at this level are many. Available prenatal labora-
tory services should include urinary estriol determinations and photometric
analysis of amniotic fluid, lechitin, and sphingomyelin ratios. Ancillary
services should include ultrasound, social services, and nutritional
counseling. During the intrapartum period, continuous electronic maternal-
fetal monitoring should be available. A 15-minute start-up time for
cesarean section is recommended. There should be a 24-hour in-house
anesthesia service and 24-hour blood bank, radiology, and clinical labora-
tory services. This unit should be able to resuscitate the newborn in
the delivery room or, if necessary, in the recovery unit, including short-
term assisted ventilation, external cardiac massage, and intravenous
therapy. Appropriate equipment must be immediately available.
In the neonatal period, 24-hour service should be provided for micro-
scopic laboratory determinations of bilirubin, blood sugars, electrolytes,
hematocrit, and blood gases obtained via the umbilical artery. The special
care nursery should be able to treat infants mildly ill with respiratory
distress syndrome, unstabilized respiratory functions, hyperbilirubinemia,
hypoglycemia, and superficial and localized infections. In addition,
it should be able to perform continuous cardiorespiratory monitoring,
measure central arterial and venous pressures, and perform exchange
transfusions.
Level III Facilities
The responsibility of the Level III unit is essentially that of the
Level II unit with the addition of providing intensive care for maternity
and neonatal patients as well as special services for the entire
region. Regional responsibilities include management of the transportation
system, a consultation service for the other hospitals in the region, and
regional evaluation of process as well as outcome.
Special services provided by the Level III unit include ambulatory
consultation for prenatal patients of other physicians in the region. This
allows the patient to continue to receive antepartum care from her primary
obstetrician while benefiting from the facilities and services available.
87﻿at the Level III unit. Antepartum beds should be provided for those
patients who must be admitted for close observation for a period of time
prior to delivery. During the intrapartum period, all current techniques
for assessment of the intrauterine status of the fetus should be available.
When fetal jeopardy is indicated, immediate cesarean section must be
available. This necessitates 24-hour in-house obstetrical anesthesia
coverage. As at Level II, other critical diagnostic services, including
x-ray and clinical laboratory services, must be available on a 24-hour
basis with in-house personnel.
Care of those patients transferred to the Level III facility from
other units in the region will ultimately be the responsibility of the
director for obstetrics or neonatology. Nevertheless, when possible, the
referring physician should continue to participate in the patient's care.
For example, an obstetrician at a Level II unit with a diabetic patient who
was transferred to a Level III unit for in-hospital intensive observation
and care during the late prenatal course could still perform the delivery
at the Level III unit when the time arises. This would enhance continuity
of patient care as well as acceptance by patients of the advantages of the
regional system approach.
Regional services provided by Level III units include transportation
and data collection and analysis. Standardized data must be analyzed at
regular intervals. Analysis should be a joint responsibility of repre-
sentatives from throughout the region. Regional maternal and perinatal re-
view committees allow for a mutual feeling of responsibility for solving
problems that all have participated in identifying.
Financing of regional programs such as transportation, data acquisition
and analysis, and continuing education has not as yet been solved. The
Level III unit alone cannot finance these regional programs in addition to
meeting its responsibility for care of its own patients.
Functions Common to all Facilities
Certain functions will be common to all facilities within the region.
These include outreach and prenatal care. Pregnancy outcome is related to
early registration of the pregnant patient, her continuation within the
system, and a standardized course of prenatal care. One of the regional
goals will be to establish an outreach program to insure early registration
of every pregnant patient. This is an urgently needed program in the best
tradition of public health. An outreach program must utilize all patient
contact points, such as neighborhood health.centers, family planning
centers, abortion clinics, and pregnancy testing laboratories to encourage
early registration of patients. Widespread media campaigns can be helpful
in the process of public education, and efforts must be made to reach
pregnant women, especially teenagers, in places they frequent. Volunteer
organizations can be most helpful in these undertakings. These programs
must be seen as an essential component throughout the regional system and
must receive a high priority in funding.
88﻿The
Schlesinger, Edward R., M.D.
1973 "Neonatal Intensive Care Planning and Outcomes Following Care".
Journal of Pediatrics 82:	916-920.
GENERAL ABSTRACT:
Recent findings strongly suggest that neonatal intensive care has
resulted in a decrease in the neonatal mortality rate and in a lower
prevalence of serious handicapping conditions among the survivors. This
article concisely summarizes several of the studies showing that neonatal
mortality has been reduced significantly in programs of neonatal intensive
care. Moreover, it points out the benefits that have been derived from
regional systems, including means for transferring infants between
different levels of care facilities.
ANNOTATIONS:
Neonatal Mortality Rates in Individual Centers
At the Johns Hopkins Hospitals where changes in newborn care were
introduced as they occurred over a period of years, there was a reduction
of nearly 32% in the neonatal mortality rate among single live-born
infants with birth weights of 2,500 grams or less who were born in the
hospital between 1961 to 1962 and 1967 to 1968.
At the University College Hospital in London, the survival rate to
28 days of age among infants born in the hospital or admitted from other
institutions increased to about 73% in the 1,000 to 1,500 gram group in
1967 to 1968 compared to an average of about 55% during the preceding five
years. Proportionate gains in survivorship among 500 to 1,000 gram infants.
The exact timing of the introduction of intensive care techniques at this
hospital is not clear.
At Vanderbilt University Hospital, a neonatal intensive care nursery
was opened in early 1960's. The average overall neonatal mortality rate
of 23.3 per 1,000 live births during the 1950's declined to 18.6 in the
1963 to 1968 period. In the weight group between 1,000 and 2,500 grams, the
average rate declined from 188 to 65 per 1,000 live births between the
two time periods.
At the Mount Zion Hospital in San Francisco, following the introduction
of perinatal intensive care services late in 1966, the overall neonatal
mortality rate dropped from 33.5 per 1,000 live births in 1966 to 15.2
in 1967 and a low of 8.8 in 1970. Among infants weighing 2,500 grams or
less at birth, the rate fell from 218 per 1,000 in 1966, to 129 per 1,000
in 1967, and to 66 per 1,000 in 1970. Among very low birth weight infants
(1,500 grams or less at birth), an impressive decline was also recorded
from 792 to 577 to 375 deaths per 1,000 live births in the 3 years,
respectively.
89﻿Scott, K.E., M.D.
1970 "It's Only Statistics:	An Approach to Perinatal Mortality in Nova Scotia".
The Nova Scotia Medical Bulletin 81-83.
GENERAL ABSTRACT:
The relatively high perinatal mortality rate in Nova Scotia is
analyzed. Particular reference is made to the regional variations
existing within the province, the factors in perinatal mortality, and to
the ways in which the perinatal mortality rate might be reduced.
Recommendations are made for a program of improved perinatal care, so
that both perinatal mortality as well as morbidity might be radically
reduced.
ANNOTATIONS:
The report recommends the following guidelines which may be helpful
to planners, administrators and physicians:
•	Screen all pregnant women for high-risk factors present
before labor.
•	Send all high-risk pregnancies to a high-risk referral
center.
•	Promote with conferences the concept of a single, conveniently
located hospital in the area for care of the women in un-
expected premature labor, and for the transfer and care of
the premature infant born unexpectantly.
•	Take steps to educate the public on the use of the one
hospital in the area for any woman who goes into premature
labor.
•	Promote the idea of one physician being in charge of the
policies and overall quality of care in the nurseries.
•	Promote the obtaining of autopsies and the setting up of a
perinatal committee to review all hospital stillbirths and
neonatal deaths, not for the purpose of assigning blame, but
to improve care of the next patient in the same circumstances.
90﻿Stetson, John, M.D. and Paul R. Swyer, M.D.
1976 Neonatal Intensive Care. St Louis, Missouri:	Warren H. Green, Inc.
GENERAL ABSTRACT:
Sections one and two of this book describe a definition of intensive
care of the newborn and discuss some organizational problems in supplying
services for a diverse community. This might be recommended for the
reader interested primarily in discovering the experiences of certain
doctors who have been involved with delivering care in a system of
intensive newborn care, especially in Canada. Several pertinent articles
are abstracted here.
Minkowski, A.
Section I, "Definition of Intensive Care of the Newborn".
Dr. Minkowski addresses the issues of defining Neonatal Intensive
Care and articulating standards and criteria for NICUs.
ANNOTATIONS:
Definition of Intensive Care of the Sick Newborn
"The care which provides careful watching and treatment to a certain
category of very ill newborns." This facility also requires having a
specific capacity or style of delivery:
•	Twenty-four hour coverage and therapy.
•	The ability to deal momentarily with emergencies.
•	One nurse for each infant.
•	Constantly available qualified medical staff.
•	Constant monitoring of vital signs.
•	Available biochemistry around the clock.
•	Continuously available radiological service.
The authors, unlike others, do not consider that sophisticated
ventilators are necessarily included in the requirements for intensive care.
One important question, valuable for an understanding of appropriate
utilization of services, is "Who deserves intensive care in the newborn
period?" In the authors view, intensive care is necessary for the
following conditions:
•	Hyaline membrane disease as defined by clinical and radiological
signs in grades 2, 3 and 4 of the disease. Occasionally patients
with grade 1 hyaline membrane disease require intensive care when
the PaO^ falls below 50 mm of mercury or when the apnea or other
clinical incidents suggest that there are deteriorations.
•	Extremely severe cases of amniotic fluid aspiration.
91﻿•	Other severe respiratory difficulties such as pneumothorax with
dyspnea, mediastinal and interstitial emphysema, or pneumonia.
•	Severe infection, i.e., septicemia, or meningitis.
•	Severe convulsive stages and coma.
•	Marked dysmaturity with hypoglycemia.
•	Certain operative and post operative conditions, mainly abdominal
and cardiac.
•	Cardiac anomalies with failure.
The authors make it clear that the following should not be included:
•	The normal premature, even the smallest.
•	Infants with mild infection.
•	Infants with intermittent convulsions
•	Transient hypoglycemia
•	Normal infant of a diabetic mother.
Further, they say that for milder degrees of hyaline membrane disease,
the case for intensive care is less clear. However, they specify if an
intra-aortic catheter has been placed, the infant should be in an intensive
care unit. It is recommended that a convalescent unit should be attached
to the intensive care center so that this center can concentrate only on
the acute severe cases.
The medical staff other than residents and interns for a thirty-bed
intensive care unit should be at least four specialized neonatologists
operating by team (two per twenty-four hours). The laboratory should
also be staffed by four technicians specially trained in micromethods
twenty-four hours a day.
The authors also advocate the availability of a doctor and nurse to
maintain close contact with the parents, who should be allowed to visit
within the unit as soon as possible. A mother is encouraged to take care
of her baby as soon as he is convalescent. Conflicting if insufficient
information to the parents has a major deleterious effect. This fact requires
strong emphasis. The authors suggest that a thirty-bed unit is required
to provide up to two weeks of intensive care and an attached unit for post
intensive or convalescent care. In the experience of the authors, the
attachment of a research unit has proved to be extremely useful, since
electroencephalographers, neurologists, cardiologists, bacteriologists,
and biochemists have access to all the important clinical material pre-
senting research problems and in turn contribute to improvement in
clinical care.
Equipment
This list of equipment is suggested as being standard for a neo-
natal intensive care units
•	Fixed apparatus including facilities for cine with image
amplification within the department.
92﻿•	Portable x-ray apparatus.
•	Monitoring for heart rate, ECG and blood pressure, respiration
and temperature.
•	Ventilator, positive pressure; ventilator, negative pressure.
•	Equipment for constant positive and negative pressure breathing.
•	Equipment to measure intra-vascular pressures.
•	Equipment to measure intracardiac pressures.
•	Autoclave with computed program.
•	Infusion pumps, catheters.
•	Laboratory equipment for emergency service in biochemistry,
bacteriology, including a scintillation counter, micro-
coagulation techniques.
In terms of maintaining a unit with only minimal requirements, the
authors say that a good premature unit could include a four-bedded unit
for intensive care in which aseptic catheterization of the aorta and manual
resuscitation can be performed. Such a unit should have 24-hour coverage
by trained physicians, nurses and laboratory technicians.
In order to provide for continuity in care, long-term follow-up
of the child who has undergone intensive neonatal care is valuable both for
research and the understanding of the individual's health. The authors
stress two points for follow-up study:
•	The subsequent pulmonary function should be evaluated. Broncho-
pulmonary dysplasia is frequent when oxygen is applied directly
to the alveolar wall. Whether or not this is a direct effect of
oxygen toxicity, or a function of the seriousness of the case now
kept alive by treatment but which might have otherwise have died,
is debatable.
•	The accurate evaluation if the neurological status is still
difficult. This is the process of being evaluated on coded
forms and it has been noticed that many of the severely involved
infants improve within six months. For instance, one medical
doctor reports that the EEG's of prematures show diminishing
spike wave activity with time. However, studies need to be
continued until the preschool age.
Cost
In the United States and Canada, the approximate cost per day per
infant for neonatal intensive care has been evaluated to be between $150-
$200 a day. In one neonatal intensive care unit in Canada, the social
security system reimburses per day per infant $100 (1971). The real
cost is approximately $600/patient/day. Certainly, these costs are high
for the system, and it is more efficient to practice preventive prenatal
care.
Comments
The author mentioned that it might be necessary for a convalescent
unit to be attached to a neonatal intensive care unit. Dr. Leo Stern,
Professor of Medical Sciences at Brown University, contends that it is
93﻿better to practice reverse referral—back to the initial hospital—in
order to free the intensive care unit for its proper function of intensive
care. Further, Dr. Stern notes that some health professionals in lesser
developed countries feel that the developed countries should not have a
monopoly on strongly technical, research-oriented neonatal intensive
care centers. In fact, this type of facility might serve as the base
for the future development of healthy living in countries such as Kenya.
Perhaps as knowledge about neonatology and the science of newborns
continues to grow, an international information-sharing network should
be developed.
The author agrees that a nurse or midwife trained to handle common
emergency situations on the spot will prove to be very valuable. However,
we also need to train neonatologists for evaluation of highly technical
medical and biochemical problems.
Dr. Paul Swyer, Associate Professor from the Department of Pediatrics
at the University of Toronto, notes the problem of the most highly trained
personnel being placed at the most intensive care level of facility. He
explains that the selection of patients occurs through the lower level
facilities in a regionalized system. Of course, those who are most highly
trained in identifying high-risk patients must be concentrated at the
regional center. He suggests that the solution to this problem will
involve education and improved methods of communication with the periphery.
A further requirement will then be a highly developed transport service.
Finally, Dr. J.F. Lucey, from the University of Vermont College of
Medicine, comments about the possible misuse of neonatal intensive care
units. It is important to adhere to a specific definition of neonatal
intensive care unit by improving perinatal care in lower level facilities.
Burnard, Eric D.
Section II, "Problems in Supplying Intensive Care for a Diverse Community".
pp. 12-20
Dr. Burnard has written a paper which examines the obstetrical care
in the State of New South Wales. He studied patterns of birth deliveries,
maternal related health status, demography, and health facility management
in this area of the Great Britain Continent. The elucidating discussion
which follows a brief summary of Dr. Burnard's paper provides the reader
with some perspectives on problems involved with operating a system with
neonatal intensive care.
ANNOTATIONS:
Dr. Burnard notes the problem of immediately going to the public and
press in order to obtain funds for construction of intensive care units.
He suggests that this should be done through proper channels after being
able to present data to support the certainty of substantially improved
clinical results when an elaborate plan for newborn care is implemented.
94﻿Dr. M. Klaus, from the Department of Pediatrics at Case-Western
Reserve University, points out that our level of economic input into
treating a 75-year old man with severe emphysema is perhaps an inefficient
manner of allocating health resources. Instead, perhaps there should be
a shift in emphasis whereby more of the resources are concentrated on the
infant who has a potential life span of 70 years. Certainly, our attitudes
are products of prevailing culture and it is necessary to change them
where appropriate.
Dr. Swyer points out the major problem in beginning negotiations
between local hospital obstetric facilities and those who would like to
centralize delivery of care under the regional concept. The problem,
he maintains, is how to educate the local hospital to appreciate the
necessity for centralization of specialized care, while continuing to
maintain first class primary care for the emergency situation as well
as for the healthy neonate. Further, it is important that the morale
of the lower level hospital does not drop when the facility is asked to
participate in a regionalization process. There must be an appreciation
of the complementary nature of their roles. In this case, the results
in the central unit will be as good as the peripheral hospital can.
Karlberg, Peter
"An Evaluation of Intensive Care of the Newborn:	The Necessity for,
and Results of Therapy", pp. 21-31.
Dr. Karlberg, from the University of Goteborg, answers two primary
questions about the quality of neonatal intensive care. First, he wants
to know, "Is the system optimal for meeting the needs of the region?"
Second, Dr. Karlberg is concerned with the success physicians are having
with neonatal intensive care. He is most familiar with neonatal intensive
care in the Goteborg region of Sweden. In the region there are two
maternity hospitals of equal size in Goteborg as well as one children's
hospital (261 beds). Each maternity hospital contains a neonatal unit
for high-risk babies. These units are to 20-30 beds. The staff of the
pediatric department of the Children's Hospital is in charge of the medical
care of all newborn infants. The pediatrician supervises the medical care
of the infant from the delivery room until discharge from the hospital.
The pediatricians and the obstetricians, it is reported, have a stimulating
association and complete collaboration is enjoyed. Dr. Karlberg discusses
some issues involved in evaluating the operation of delivering neonatal
intensive care and shares his views with other physicians from the
United States and Canada.
ANNOTATIONS:
Evaluation
The value of keeping neonatal mortality and morbidity figures is for
follow-up purposes and the discovery of new trends or events. There may
be some disagreement among physicians as to how certain information can
95﻿be obtained and how the obtained information can later be recalled and
monitored. Generally, there seems to be agreement as to what data needs
to be gathered. In the evaluation of neonatal care, mortality and
morbidity must be combined. There must be some statistical analysis
provided as to the severity of morbidity, especially in reference to the
type of treatment performed and in reference to the obstetrical history.
There should be data on the social background of the family.
Dr. Karlberg would recommend what he calls a "Medical Birth Record"
system for data recording. He explains that the neonatal relevant infor-
mation is stored on the first page of the hospital record for each mother
and infant. This information can then be relayed between maternity and the
child health service, as well as being used for risk grouping of the new-
borns for directed health control. Information can also be sent from the
maternity unit to the maternal postnatal care service. Certainly, in our
record keeping systems, there is computerized processing for statistical
data of various kinds and the possibility exists that this information can
be used as a base for individual health records from birth onward.
Moreover, Dr. Karlberg comments that this medical birth information
must become an integral part of the clinical record and record the outcomes
of delivery. It must be regarded by the clinician as very valuable for
the patient. If the collection of information is considered to be a
chore, the quality of the information gathered will not be sufficient to
aid in administration of medical care.
Stern, Leo, M.D.
"Neonatal Mortality and Intensive Care in the Province of Quebec", pp. 33-37.
Dr. Stern views neonatal mortality as a means to understand how the
neonatal intensive care facility justifies its reason for existence. He
considers the magnitude of demonstrable improvement in survival of the
newborn, the impact of referral patterns on the results of neonatal health,
and the need for comprehensive regional planning based on available facility
and population distributions, as well as on realistic geographic lines.
ANNOTATIONS:
Dr. Stern speaks of a system combined of Group A hospitals which have
their own in-service facilities for neonatal intensive care. Group B
hospitals do not, but transfer the greatest proportion of their problems
(at least 75% of subsequent deaths) to one of the two referral centers;
while Group C hospitals in the same geographic locale as A and B have
neither their own facilities as defined above nor do they transfer to
the referral centers although they could easily have arranged to do so.
Dr. Stern points out that hospitals who either have their own facility
or utilize one of the available referral intensive care centers have
very significantly better survival rates for their infants. It is
reported that while there are no real differences between Group A and B
96﻿hospitals, hospitals in Group C show a much higher neonatal mortality
for both low birth weight and full-term infants (30% higher in the 1,000-
2,500 gram group and almost double in the greater than 2,500 gram group).
Therefore, we may conclude that health status benefits are certainly
derived from maintaining close referral relations between hospital
centers which have high level intensive care capacities and hospitals
which have lower care capacities.
Stewart, A.
1972 "Prognosis for Infants of Very Low Birth Weight". London, England:
University College Hospital.
GENERAL ABSTRACT:
The object of this study is to assess the extent to which intensive
care improves survival rate and affects the incidence of abnormality
among the survivors. To make an assessment of this kind, it is necessary
to have some measure of intellectual development in addition to data
concerning physical fitness and behavioral response.
ANNOTATIONS:
Ten percent of a group of 98 very low birth weight infants have
some physical or medical handicap likely to affect their mode of life,
detectable within the first five years of life. Only 4% of the group
have definite evidence of mental handicap. These figures are based
on assessment made at ages ranging from 8 months for the youngest child
to over 5 years. Subdivision into those aged over or under 3 years does
not make a significant difference to the prognosis.
Prior to the introduction of intensive care, other studies have
reported incidences of handicaps of 33-70% in very low birth weight
survivors and expressed concern that if survival rates improved, the
incidence of handicap might be even greater. Since the introduction of
modern methods of care in University College Hospital, 10% of very low
birth weight survivors have a physical or mental handicap. The incidence
of mental handicap is only 4%. These results have been obtained at a
time when the survival rate for infants weighing 1001-1500 grams has
improved from less than 50% to over 70%. Improvements of this kind
could only have been achieved if the majority of the handicaps previously
observed had been acquired after birth. Furthermore, these recent re-
sults show that prognosis for infants of very low birth weight is related
to perinatal events rather than to prenatal factors associated with the
etiology of very low birth weight.
97﻿Stewart, Ann.. M.D. and E.O.R. Reynolds, M.D.
1974 "Improved Prognosis for Infants of Very Low Birth Weight".
Pediatrics 54:	724-735.
GENERAL ABSTRACT:
Following the introduction of a program of intensive care for in-
fants of very low birth weight, 197 infants who weighed 1,500 grams or
less at birth were cared for in the Neonatal Unit of University College
Hospital, London, in the five years from 1966 to 1970.
ANNOTATIONS:
During this period, the neonatal survival rate for infants weighing
501 to 1,000 grams was 23% and for those weighing 1,001 to 1,500 grams
was 69%. The principal immediate cause of death was hypoxia around the
time of birth or hyaline membrane disease. Ninety-five of the 98
surviving children, aged 2 years, 10 months to 7 years, 10 months (mean
5 years, 2 months), were followed up. They all had repeated physical
examinations and developmental assessments. Sixty-five of the 86 (90.5%)
of the children had no detectable handicap, 4 (4.2%) had physical
handicaps only, and 5 (5.3%) had mental handicaps. The incidence of
handicap was very much greater among infants presumed to have been
severely hypoxic than among the remaining infants.
It is concluded that intensive care can both increase the chance of
survival for infants of very low birth weight and reduce the incidence
of serious handicap in survivors.
Swyer, Paul R., M.D.
1975 "The Organization of Perinatal Care with Particular Reference to the Newborn".
Neonatology:	Pathophysiology and Management of the Newborn. Edited by
Gordon B. Avery, M.D., Ph.D., Philadelphia:	J.B. Lippincott Company.
GENERAL ABSTRACT:
Swyer suggests how a systematic organization for the delivery of
reproductive medical care can contribute to the success of women giving
birth. Further, he suggests how to broaden the concepts of neonatology
and perinatology in order to achieve a wider concept of reproductive
medicine.
98﻿ANNOTATIONS:
There has been a historical change in attitude toward the newborn.
In the past, health professionals and the public accepted neonatal
mortality rates of the order to 100 to 200 per 1,000 births. More
recently, however, most neonatal mortality rates have been much lower,
and there has been greater concern for the quality of survival.
The article points out that marked reductions in mortality can be
achieved by coordinating rural and urban maternity services and by providing
special hospital facilities for identified risk pregnancies. In addition,
in countries with better neonatal and perinatal mortality statistics,
there tend to be well-developed regional systems for neonatal and
maternity care.
In beginning to develop a regional system for caring for neonates
and pregnant women, it is first important to have a standardized system
for statistical evaluation of infant care or health status. Common
definitions for "birth", "live birth", "stillbirth", "abortion", "neonatal
period", "low birth weight", etc., and mortality recording practices
would provide a firmer basis for regional, national, and international
comparisons. This is especially important to have in order to accurately
determine the health care needs of the populations.
As much of the literature now suggests, if derangements in internal
homeostasis of the pregnant mother are treated in special care high-
risk units, it not only reduces mortality but prevents subsequent morbidity
in survivors. Hence, having the standardized recording practices and
definitions can contribute to successfully identifying high-risk cases.
The need for new intensive care technology must also be provided by
a central newborn intensive care facility. This type of facility is
drastically improving the infant mortality rate for service populations.
One recent study, according to Swyer, shows that 86.7% of 72 infants
surviving birth weight of less than 1,500 grams are apparently normal and
only 7.4% definitely abnormal. For example, Swyer says that survivors
of artificial ventilation have a normal I.Q. compared with control groups.
The majority of infants asphyxiated at birth have recovered undamaged through
using modern methods of resuscitation by IPPV and intravenous alkaline
buffer.
Further, Swyer explains that modern techniques of resuscitation of
experimentally asphyxiated animals have resulted in a high proportion of
normal survivors compared with control animals. This suggests there are
specific techniques of neonatal intensive care needed in order to provide
optimal levels of care. For instance, a lower mortality rate was reported
in low birth weight infants following the use of' intravenous alimentation,
particularly in neonates with severe surgical problems.
Swyer stresses that three main foci of operation should be identified
in a health care system for reproductive medical care. They are as
follows:
Level I facilities. These facilities can be outpost nursing stations
or rural district hospitals. They can take the responsibility for delivering
normal pregnancies as well as pregnancy supervision and looking after
normal newborn infants. They should probably operate only in relatively
isolated communities.
99﻿Level II facilities. These can be fairly large general hospitals
with maternity services. They can have available specialist obstetricians
and perinatologist-pediatricians and have the necessary staff and facilities
to deal with the common obstetric and neonatal problems, referring only
those of a complicated nature to the Level III center. Informed estimates
of neonatal disease anticipate 45/1,000 live births with serious problems
requiring special care, of whom perhaps less than 10/1,000 need eventual
transfer to the regional Center, the remainder being adequately cared
for in the community hospital ICU.
Level III facilities. These can constitute the regional centers for
reproductive medicine. They can be situated in major general hospitals.
They can take responsibility not only for the supervision and delivery of
normal pregnancies in their immediate area, but also accept a large pro-
portion of the risk pregnancies from the region as well as function as the
resource center for consultation with and support of physicians at Level I
and II hospitals. They can offer advice and diagnostic facilities for
risk pregnancies and special care facilities in both obstetrics and
pediatrics for the delivery and subsequent care of the infant. The
center can be responsible for coordinating, training and educational
services and for conducting basic, clinical and operational research within
the region. They can provide support to the Levels I and II units in the
region in case of emergency, and be fully equipped with appropriate
transport services to move the patient into the facility or the consultants
out to the patient.
Estimate of Bed Need
Swyer believes that the requirement for neonatal intensive care unit
beds is .75/1,000 live births. Thus, for a range of deliveries of 2,000
to 4,000 per year, the requirement of ICU beds would be 1.5 to 3; and for
convalescent beds 4.2/1,000 or 8.4 to 16.8 beds. If a minority of the more
severe or special problems are transferred to the regional center as
suggested, this number can be diminished by approximately 20%. For a
delivery rate of 2,000 per year, the above considerations suggest a patient
load of approximately 60 to 80 neonatal ICU patients a year. In addition,
experience suggests that approximately 200/1,000 live births can advantageously
spend up to 24 hours postnatally in a special observation nursery.
Necessary Ancillary Services
Laboratory services should be organized within the regional center or
provide specialized tests for the service of community and district
hospitals. Essential services include the genetic, cytologic, chemical
and spectrophotometric analysis of amniotic fluid and estriol determination.
The blood bank at the regional center should also be well equipped to
handle referral problems. It can provide a typing laboratory and depot
for the storage of donor blood. Delivery of mothers in hospitals that do
not carry donor blood is not advised, according to Swyer. At least one
unit of 0 Rh-negative blood should be kept in each community birth facility.
The facility can arrange to have replacement blood supplies cycled to them
from the regional center within 72 hours.
100﻿The regional center is the logical place to locate the majority of
the transport services for the system. These services might include care
to bring patients to and from clinics, ambulances and flying vehicles with
adequate head room and services for monitoring and life-support care.
A comprehensively developed transport system can also provide mobile
units for carrying public health personnel into the community for the
purpose of delivering maternal and child health care, preventive medicine,
and educational programs.
Swyer points out that there is an inverse relationship between the
number of prenatal examinations and the perinatal mortality and morbidity.
In other words, the more neglect in educating women about pregnancy,
the greater will be the likelihood for death of the baby or that some develop-
ment problem will occur. Hence, the regional center should encourage a
sustained effort to instruct the public in sex education, family planning,
genetics, antenatal, natal and postnatal care.
The team concept of health care delivery can ensure that both mother
and child receive optimal care from a variety of medical and paramedical
personnel. Appropriate functioning of the personnel organization is
facilitated if established lines of communication and patterns of referral
are developed.
Finally, because the patient's interest should always be the first
priority, prenatal risk records or registers for the mother and postnatal
records for the infant can be maintained at the regional center. These
can be used as tools to ensure that risk patients are identified and pro-
vided with care which is appropriate for their needs. By keeping in mind
that the patients' needs come first, factional and jurisdictional conflicts
may be resolved with fewer conflicts.
Work Volume Note
Dr. Swyer and Dr. Hardie have recommended that approximately 200
intensive care admissions annually is a reasonable workload for one full-
time Specialist Perinatologist assuming time for attendance at prenatal
clinics, follow-up clinics, research and teaching.
Swyer, Paul R., M.D.
1970 "The Regional Organization of Special Care for the Neonate". Pediatric
Clinics of North America 17:	761-776.
GENERAL ABSTRACT:
Swyer's discussion emphasizes the pediatric aspects of regional plan-
ning for optimal neonatal care of high-risk infants. However, such planning
must be integrated with obstetrics and other disciplines to deliver total
interconceptual, antenatal, natal and postnatal care.
101﻿ANNOTATIONS:
The Ontario Perinatal Study (1967) showed that 32% of pregnancies
had some broadly defined identifiable risk factor and generated 60% of
the neonatal problems. Forty percent of the neonatal problems were
unheralded and emergent, though modern techniques have somewhat reduced
this proportion.
The decision to transfer a patient from a community hospital depends
on four criteria:	(1) the nature of severity of the patient's criteria;
(2)	the relative capabilities of the delivery and referral centers;
(3)	the type of transport facility available and the distances involved;
and (4) the climatic conditions.
Bed Requirements for Regional Neonatal Centers
The mortality rate in the NICU is about one-third of all admissions.
The potential admissions to an NICU would be the regional neonatal
mortality rate (per 1,000 live births) X 3/1,000 live births, and the
requirement for NICU beds would be:
regional neonatal mortality rate X 3	_ n .	, . ,,
no. of live births in
------------------------------------- X
thousands for the region
Only about half of all admissions to a neonatal referral unit require
the highest grade of intensive care; provision should be made for the
remaining infants at a somewhat lower, but still relatively advanced level
of care. In addition, patients recovering from NIC will require
transitional and convalescent care. It had been found that six transitional
and convalescent care beds are required for each NICU bed.
Methods for Measuring Fulfillment of Objectives
•	Demonstration of ability to correct clinical abnormalities, such
as a low PH or hypocalcemia, and to treat effectively asphyxial
respiratory failure.
•	Controlled trials of techniques or procedures using clearly
defined end points such as survival.
•	Long-term assessment of the growth, development, and mortality
of survivors.
•	Demonstration of a sequential improvement in the overall
mortality rate.
•	A comprehensive medical audit or morbidity and mortality should be
compared with other regional, national, and international
indices.
102﻿Cost
Care in the NICU at the Hospital for Sick Children, Toronto (1970)
costs approximately $200 per patient day. Since the average stay in the
NICU is 6 days, the cost is roughly $1,200 per patient admitted. This
figure excludes care in transitional and convalescent areas which costs about
$100 per patient day or an average cost of $2,000 for 20 days so that the
average total cost for a patient is $3,200.
Teberg, Annabel, M.D., Joan E. Hodgman, M.D., Paul Y.K. Wu, M.D. and
Robert L. Spears, M.D.
1977 "Recent Improvement in Outcome for the Small Premature Infant:	Follow-
up of Infants with a Birth Weight of Less Than 1,500 Grams". Clinical
Pediatrics 16:	307-313.
GENERAL ABSTRACT:
The authors describe their experience with 176 infants cared for in
the Special Care Nurseries of the Los Angeles County-USC Medical Center
and followed in the Medical Center's Premature Clinic. Data was
analyzed according to birth weights of under 1,000 grams and between
1,000 and 1,500 grams. Demographic factors are examined. Neurologic
and developmental outcomes were assessed during the follow-up visits at
the Premature Clinic.
ANNOTATIONS:
Advances in perinatal care correlate with improved developmental and
neurologic outcomes in the infants. Developmental and neurologic outcome
was determined for 176 premature infants of birth weight less than 1,500
grams, born during 1965-1970. No significant difference in outcome was
found when infants of birth weight 1,000 to 1,500 grams were compared
with those weighing less than 1,000 grams. The overall outcome is
generally favorable, with 67% of the entire group now deemed developmentally
and neurologically normal. Improvement in outcome in the small premature
infant seems related to aggressive perinatal care, particularly monitoring
of biophysical, biochemical and metabolic factors in the fetus and neonate.
103﻿Thogmartin, Bruce and Michael Tyne
1969 "Essential Design Elements of Neonatal Intensive Care Unit". Problems
of Neonatal Intensive Care Centers. Edited by Ross Hospital Planning
Associates, Ross Laboratories:	Columbus, Ohio.
GENERAL ABSTRACT:
This is a planning document used to assist planners, administrators
or physicians in designing a NICU.
ANNOTATIONS:
Environment
A stabilized room environment is necessary. A temperature of 74-76
degrees F. and a relative humidity of 50% is deemed most favorable.
Incoming clean air should pass through medium to high efficiency filters
immediately prior to being introduced into the nursery with room air
changed 12 or more times per hour. A positive pressure (outward flow
of air) with adjacent areas should be maintained to reduce the possibility
of germ-laden dust entering the area.
Ancillary Space and Functions
•	Procedure Room
Designing a maximal care unit takes into consideration the
fact that most diagnostic and treatment procedures are per-
formed at an individual incubator location, precluding the
need to remove a sick neonate from its stabilized environment.
Where a separate procedure room is required, it should adjoin
a maximal care unit.
•	Clerical Center
Though most charting activities take place at a patient's
location, a distinct area should be designated as a clerical
center, primarily for a ward se’cretary or unit clerk. Here,
requisitions may be initiated and telephone and intercom
communications handled.
•	Clean and Soiled Utility Rooms
Separate rooms, contiguous to a maximal care unit for clean
and soiled utility functions are needed. Base and wall
cabinetry and built-in sinks are appropriate in each room.
Ideally, the utility room has dual access to speed outward
flow of supplies.
104﻿Storage
Two storage requirements need to be considered:
a.	In-unit storage of routine and special infant care supplies.
b.	Storage of idle equipment.
•	Laboratory
Most maximal care units include a micro-laboratory capable
of blood gas and bilirubin determinations. Countertop work
space and some storage are required. A micro-laboratory is
sometimes combined with a procedure room or utility area.
Additional space programmed at the start will provide for
expanding tests.
•	X-Ray
A portable x-ray unit, assigned exclusively to the nurseries,
will usually satisfy the maximal care unit requirements.
A separate NICU x-ray center is also a possible consideration.
Thompson, Theodore, M.D. and J. Reynolds
1977 "The Results of Intensive Care Therapy for Neonates:	I. Overall Neonatal
Mortality Rates; II. Neonatal Mortality Rates and Long-Term Prognosis
for Low Birth Rate Neonates". Journal of Perinatal Medicine 5:	59-75.
GENERAL ABSTRACT:
This paper is divided into two sections. The first reviews the
evidence on neonatal intensive care units' impact upon neonatal mortality
rates among low birth weight infants. The second examines whether improve-
ments in the care of sick neonates have decreased the incidence of long-
term sequelae among neonatal intensive care unit survivors.
ANNOTATIONS:
The first section presents data which documents the improvements in
neonatal survival with the utilization of NICUs by comparing overall
neonatal mortality rates before and after the introduction of NICUs
in hospitals and the development of regionalized NICUs.
Data comparing low birth weight neonatal mortality rates for various
areas are presented. The greatly improved survival rates for low birth
weight infants have been a major factor in reduced neonatal mortality
rates. Current survival overall rates approach 75% to 95% for LBW infants
managed within NICUs. Survival rates exceed 75% for neonates weighing
between 1001-1501 grams who have received intensive care.
105﻿Survival rates for AGA infants weighing under 1001 grams at birth
have improved and are currently below 30%.
The second section examines the long-term prognosis for LBW infants.
There has been a reduction of the incidence of disabilities among LBW
survivors. The majority of survivors do not have defects requiring
institutional care and the minor defects do not often preclude a useful
life.
Neonatal intensive care has been responsible for improved physical
and developmental status for survivors. Measures responsible for pre-
venting death also prevent central nervous system damage. Physicians
are more knowledgeable of the antenatal and perinatal factors con-
tributing to mortality and morbidity of the fetus and the neonate and
have enhanced to prevent or correct detrimental conditions.
The article reviews the findings of the Maternal and Child Health
Task Force of the State Health Planning Advisory Council in Michigan
on the effectiveness of intensive care in reducing long-term morbidity
for neonates. Fifty to 70% of the cases of severe mental retardation
due to perinatal factors could be prevented, resulting in a 20% to 30%
reduction of children with severe mental retardation.
Improved social, educational and economic conditions would decrease
neonatal mortality and morbidity rates by reducing the incidence of
LBW infants, particularly those weighing under 1,500 grams. However,
since most premature deliveries currently cannot be prevented, one
immediate method of further reducing both neonatal mortality rates
and long-term physical and developmental sequelae among LBW infants is
to make available optimal care to the fetus during the prenatal and peri-
partum periods and to the LBW neonate, the latter in NICUs where bio-
chemical, metabolic, and physiologic disturbances can be anticipated and
prevented or at least promptly detected and treated.
The provision of optimal perinatal care requires that high-risk
pregnant women, particularly those in premature labor under 33-34 weeks
gestation, be transferred to a high-risk perinatal center for delivery
so that the LBW infant can be assured optimal care from birth. The least
traumatic and most dependable transport incubator remains the pregnant
woman's uterus.
Usher, Robert H., M.D.
1971 "Clinical Implications of Perinatal Mortality Statistics". Perinatal
Biology. Leo Stern, M.D. (guest editor). Clinical Obstetrics and
Gynecology 14:	885-925.
GENERAL ABSTRACT:
Usher presents a discussion of perinatal mortality which deals
with three aspects. Background factors in the social and maternal
history on which differences in medical care are superimposed are
106﻿briefly reviewed from data made available from the British Perinatal
Mortality Study. The current status of perinatal mortality across a wide
geographic area is analyzed using data obtained from the Perinatal
Mortality Committee of the Province of Quebec, Canada. Finally, the
recent perinatal mortality experience of a university teaching maternity
hospital which has been active in the development of neonatal intensive
care is presented to indicate the results currently obtainable under
optimal circumstances of perinatal care.
ANNOTATIONS:
Data Gathering Standards
In order to obtain accurate and meaningful perinatal mortality
statistics, a firm definition of viability must be established based
upon birth weight. All stillbirths and first-week neonatal deaths of
infants delivered in an institution must be included, and neonatal deaths
must be attributed to the hospital of birth, irrespective of where the
infants die. Neonatal death rates must be analyzed separately for each
weight group. Stillbirths and neonatal deaths must be analyzed by cause.
Completeness of reporting, standardized definitions of causes of death,
and accuracy in weighing infants and assigning causes of death are pre-
requisite to obtaining useful figures.
Clarifying Differences Between Neonatal Mortality Rates Across Racial
Lines May Be Helpful for Refining Our Definitions of High-Risk Births
Neonatal mortality rates should be grouped by weight group. It
has been suspected that low birth weight black infants are stronger
than their similar-weight white counterparts. Figures which show this
trend have been collected by Dr. Malcolm Freeman at the Grady Memorial
Hospital, Atlanta, Georgia. This hospital delivers some 7,000 infants
per year from an indigent population which has a low birth weight
incidence of 13.2% among white patients and 15.4% among its black
patients (1971).
White infants of low birth weight, for each 500 grams weight group,
had neonatal mortality rates 27, 80 and 102% higher than black infants
of similar weight. Low birth weight infants (1001-2500 grams) as a total
group had neonatal mortality rates of 123 per 1,000 among whites and 72
per 1,000 among black infants treated with the same,neonatal care.
(Statistics for each weight group are provided within the article.)
Therefore, it may be suggestible for planners to analyze differences
in neonatal mortality rates for different racial groups.
Differences Between Neonatal Mortality Rates, Maternity Hospitals That
Utilize NICU vs. Facilities Without Special NICU
Figures indicate that perinatal mortality, when region and size of
hospital are held constant, is 33% higher in maternity hospitals that do
107﻿not utilize intensive care facilities than in those with intramural
facilities. It was calculated that all infants delivered in the Province
(Quebec) during 1967 and 1968 have been delivered in hospitals containing
neonatal intensive care facilities, 774 of the 2,045 first-week deaths
that occurred could have been prevented.
Potentially Preventable Deaths
According to statistics in the article, the "Irreducible Minimum",
perinatal mortality rate over 1,000 grams would consist of the following:
inoperable lethal malformation (3.3 per 1,000); asphyxic fetal deaths
prior to admission to hospital from abruptio placentia (0.8), or cord
loops (1.1); deaths before 32 weeks' gestation from erthroblastosis (0.8);
maternal diabetes or toxemia (0.3); and unexplained stillbirths prior to
labor (2.7).
The total irreducible minimum perinatal mortality rate is 9.0
per 1,000.
Maintaining Mortality Rate Standards, A Reflection of Quality
The statistics which most closely reflect adequacy of perinatal
care are the following, with rates from an intensive care center
indicated as standards:
• Total Mortality
•• Perinatal mortality over 1000 g
•• Neonatal mortality: 1001-2500 g
over 2500 g
Total over 1000 g
•• Stillbirth rate over 1000 g, total—
in labor—
• Mortality By Cause
•• Deaths from Respiratory Distress
Syndrome
Asphyxia
Fetal Malnutrition
Isoimmunization
Unexplained Cause
15.9 per 1,000 births
55.8
2.5
6.3
9.7
2.8
1.2	per 1,000 births
5.3
1.1
0.8
2.8
If all deaths from potentially treatable causes could be prevented,
perinatal mortality over 1,000 grams would be reduced to 9.0 per 1,000
births.
108﻿Usher, Robert, M.D.
1977 "Changing Mortality Rates With Perinatal Intensive Care and Regionalization".
Seminars in Perinatology:	Regionalization of Perinatal Care 1:	309-319.
Guest Editor: Leo Stern, M.D. New York: Grune and Stratton.
GENERAL ABSTRACT:
The author reviewed perinatal mortality data by weight group from
three sources:	(1) the Province of Quebec (1961-1974);	(2) the obstetrical
services associated with McGill University; and (3) the Royal Victoria
Hospital, a perinatal referral center in Quebec.
Infants weighing 1,000 grams or less were excluded from the analysis
because:	deaths are mainly unavoidable and usually the result of
immaturity; often go unreported as stillbirths; or are inaccurately
reported. As mortality rates among infants weighing more than 1,000
grams have decreased, those born with lower birth weights (0.5%) create
a disproportionate influence in perinatal mortality statistics and de-
tract from their usefulness in reflecting the adequacy of perinatal care.
The minimum perinatal mortality from causes that are not yet potentially
preventable can be calculated to be 5.6/1,000.
Analysis of perinatal mortality statistics provides a sensitive
measure of quality of obstetric and neonatal care. Regionalized perinatal
intensive care can markedly reduce perinatal mortality. The challenge
for the future is to make intensive care services available to all
pregnant women and newborn infants at risk. An effective monitoring
system for perinatal deaths is a prerequisite to assure that optimum
obstetric and neonatal intensive care is being provided to those in need.
ANNOTATIONS:
Province of Quebec
•	There was a 41% decrease in perinatal mortality of infants of
greater than 1,000 grams birth weight, to a rate of 13.1/1,000.
This was due to a 39% decrease in the stillbirth rate to 7.0/1,000
and a 43% drop in neonatal mortality to 6.1/1,000. This still
left the Province higher than the optimal rate of 9.7/1,000
that obtained in the best third of the larger obstetric services.
•	The 43% decrease in neonatal mortality was mostly due to de-
creased mortality rates within weight groups rather than to de-
creased incidence of low birth weight infants. Within each 500
gram weight group (from 1,001 to 2,500 grams), there was a
32%-53% decrease in neonatal mortality from 1968 to 1974, the
greatest improvement being in the 2,001-2,500 gram weight group.
Among full-size infants (over 2,500 grams), neonatal mortality
fell by 31% between 1968 and 1974 and by 42% between 1967 and 1974.
109﻿•	"Optimal" neonatal mortality rates are below the provincial rates
for the 1,001-1,500 gram infants indicating that most improvement
can be sought in future by provision of modern methods of care
to the few very small infants (under 1,500 grams, 0.8% of live
births) by regionalizing deliveries that occur very prematurely
(1,500 grams is equivalent to 30-31 weeks of gestation).
Regionalization of Obstetric Services Associated With McGill University
•	There has been a decrease in perinatal mortality coincident with
the regionalization of McGill-associated obstetric services in
1973. After a plateau period of five years with perinatal
mortality averaging 14.0/1,000 from 1968 through 1972 (prior to
regionalization), mortality fell to an average of 10.1/1,000
between 1973 and 1975, with a further decrease to 8.2/1,000
in 1976. The decrease in mortality rates cannot be attributed
to a change in incidence of low birth weight infants.
•	After regionalization, there was a 34% decrease in neonatal
mortality from 5.3 to 3.5/1,000 between 1972 and 1974, compared
to a fall in provincial rates from 7.7 to 6.1/1,000 in the same
period. This decrease occurred during, and was in part related to,
the development of antenatal referral of women in premature labor
for delivery in the perinatal intensive care center at the
Royal Victoria Hospital. The greatest emphasis in this program
was placed in the referral of pregnancies threatening to deliver
very prematurely. The greatest benefit was obtained among infants
of very low birth weight.
Regionalization with antenatal referral of high-risk pregnancies
cannot be held to be responsible for all of the improvement in
neonatal mortality rates in the McGill Hospitals. Much credit
must be given to developments in the technology of neonatal
intensive care.
•	Analysis of 1972 and 1973 data showed regional differences in
perinatal mortality between the large metropolitan Montreal
region (13.9/1,000), the smaller metropolitan Quebec City region
(15.9/1,000), and the remaining regions of the province lacking
large cities (17.4/1,000). These differences could not be related
to population differences in maternal characteristics or incidence
of low birth weight infants or of malformations. This difference
in mortality between Montreal and the non-metropolitan regions was
primarily due to neonatal mortality in low birth weight infants
(52/1,000 versus 93/1,000). Stillbirth rates were very similar
(7.9 versus 8.3/1,000).
110﻿•	An analysis of the 0-7 day neonatal mortality rate in 1974 for
very low birth weight infants according to the type of hospital
in which the infants were delivered. Hospitals providing neonatal
intensive care facilities in the hospital of birth, or "intramural"
services, had a neonatal mortality rate for infants weighing
1,001-1,500 grams of 15%. Hospitals that systematically trans-
ferred such infants to pediatric referral centers had a mortality
rate of 25%. Those that provided neither intramural nor post-
natal referral neonatal intensive care lost 48% of their 1,001-
1,500 gram infants. These comparisons were made among each of the
34 large hospital services delivering more than 1,000 infants
per year. Smaller hospitals, some of which referred their infants
postnatally for intensive care, had a 43% mortality rate.
Together, these four types of hospitals gave the province, as
a whole, a 36% mortality rate for this weight group.
•	From these data, it can be concluded that if all 428 infants
weighing 1,001-1,500 grams who were delivered in the province in
1974 could have been delivered in hospitals providing intra-
mural neonatal intensive care, 92 (59%) of the 156 deaths in this
weight group alone could have been avoided.
•	The potential for preventing neonatal deaths with antenatal
transfer of very few high-risk pregnancies is, therefore, great
since the 0.5% of live births weighing 1,001-1,500 grams accounted
for 30% of the province's> 1,000 grams neonatal deaths in 1974.
Changing Mortality Rates in a Perinatal Center
•	Causes of death have changed over the 1966-1975 decade. This is
to be expected from the nature of advances in therapy. In all,
there was a decrease in the frequency of the "potentially pre-
ventable" causes of death for two periods selected for comparison:
1966-1970 (76.1/10,000) and 1971-1975 (40.1/10,000). This de-
crease could in most instances have been related to developments
in prevention or treatment of perinatal diseases instituted
during the period under study.
•• Asphyxia During Labor and Delivery
Whether due to mechanical abnormalities or to unexplained
cause, they are preventable with modern monitoring techniques.
These two causes of death should ultimately be completely
preventable.
•• Maternal Diabetes or Toxemia of Pregnancy
These were rarely associated with otherwise unexplained
stillbirths, with rates of 4.5/10,000 in the first period
and 4.2/10,000 in the second for the two conditions together.
Ill﻿Fetal Malnutrition
Deaths of growth-retarded infants from chronic fetal depriva-
tion were reduced in frequency from 11.3 to 7.6/10,000 with
the aid of ultrasound, oxytocin challenge tests and monitoring
during labor. Neonatal deaths from this cause were eliminated
completely by improved obstetric care.
•• Respiratory Distress Syndrome
This decreased in frequency as a cause of death from 12.1
to 8.5/10,000. Comparing 1970-1975 births before and after
steroids, the mortality rate fell from 12 to 4/10,000 following
the introduction of steroid prophylaxis in mid-1973. This was
due to a decrease in incidence rather than to improvement in
therapy. It seems likely that with the aid of corticosteroids
the toll from respiratory distress syndrome can be reduced by
two-thirds to about 4/10,000 live births.
•• Isoimmunization Deaths
These deaths dropped from 8.3/10,000 births to zero due to
immunoglobulin prophylaxis rather than to better therapy.
•• Infections
Infection killed 9.8 infants per 10,000 in the first period
and only 1.8/10,000 in the second. Most infections in the
first period, and both cases in the second, were intra-
uterine in origin.
• Certain causes of death are usually not potentially preventable.
These include from lethal malformations, most from antepartum
hemorrhage (usually abruptio placenta, with fetal death occurring
before arrival of the mother in hospital), most stillbirths from
cord loops or knots (again, with fetal death usually occurring
before hospitalization of the mother), and all unexplained
antepartum stillbirths, where an apparently normal fetus of normal
size dies unexpectedly, prior to admission of a healthy mother
to hospital. Together these accounted for 55.5 deaths per 10,000
births in 1971-1975, which can be considered as the "minimum
possible perinatal mortality rate" if all potentially preventable
perinatal deaths could have been avoided. This minimum comprises
17.8 neonatal deaths and 37.4 stillbirths per 10,000.
112﻿Westbers, Jimmie A., M.D., Dayton W. Clark, M.D. and Gilbert A. Webb, M.D.
1973 "An Evaluation of High-Risk Maternity Care in a Community Hospital".
American Journal of Obstetrics and Gynecology 116:	557-563.
GENERAL ABSTRACT:
A program of high-risk maternity care was established in 1967 at
Children's Hospital and Adult Medical Center of San Francisco. This
is a 363 bed community hospital with approximately 2,000 births per year.
ANNOTATIONS:
The following goals for planning regional systems for the provisions
of neonatal care are reiterations of statements which have been made
throughout the literature:
•	Every high-risk pregnancy should be treated at a high-risk
hospital center.
•	If the high-risk patient cannot be adequately identified in
advance, there are two solutions. One way is to establish
regional high-risk centers to which those patients obviously
at high-risk will be referred at an early date and to which
patients with sudden complications can be quickly transferred.
A second solution is to upgrade as many maternity and nursery
units as possible to a high-risk level of care, thus avoiding
the necessity of the transfer of sick patients except from those
areas where the population density will not support the expense
of intensive maternal and newborn care.
•	Finally, wherever it is possible to consolidate obstetric services
in a fewer number of hospitals to create network high-risk units,
this should be a high priority.
Zamansky, Harriet and Kathleen Strobel
1976 "Care of the Critically 111 Newborn in An Infant Care Center". American
Journal of Nursing 76:	566-568.
GENERAL ABSTRACT:
This article describes the operations of the Infant Care Center at
the Queen of Valley Hospital, West Covina, California. It offers a clear,
113﻿understandable treatment of typical NICU activities and would be valuable
as background material for committees. It describes infants cared for
and the levels of care provided.
ANNOTATIONS:
The categories of infants are cared for in the ICC unit:
•	all those weighing less than 1,500 grams at birth
•	premature or full-term infants with respiratory distress
•	infants with surgical conditions requiring immediate treatment
•	infants with neonatal sepsis, erythroblastosis, or other
causes of jaundice requiring transfusion
•	infants with neonatal seizures.
The infant receiving critical care:
•	needs mechanical respirator assistance
•	has an endotracheal tube or tracheostomy
•	requires oxygen with respirator assistance greater than 50%
•	has indwelling umbilical catheters
•	is on monitors for vital signs
•	must have his intake and output strictly measured
• needs care on a one-nurse to one-infant ratio.
This area is closed to admissions when the census is four.
The infant receiving intensive care:
•	has returned from surgery
•	has chest tubes
•	requires high oxygen concentration without respirator
assistance
•	has an indwelling arterial or venous catheter for frequent
blood gases and IV therapy
•	needs hyperalimentation
•	needs care on a one-nurse to two-infant ratio.
For intermediate care, the infant:
•	needs oxygen greater than 30%
•	has peripheral intravenous infusions
•	must be on isolation for possible sepsis or infectious disease
•	needs vital signs monitored every two to three hours
•	needs gavage feedings
•	needs care on a one-nurse to four-infant ratio.
For observational care, the infant:
•	needs to be fed every three to four hours
•	needs to be observed for weight gain
•	needs vital signs monitored every four hours
•	needs care on a one-nurse to six-infant ratio.
114﻿CHAPTER 2
HEALTH PLANNING LITERATURE﻿﻿Bergen-Passaic Health Systems Agency
1977 Report on the Task-Force on Regional Perinatal Services in the Bergen-Passaic
Health Service Area to the Project Review Committee. Rochelle Park, N.J.
GENERAL ABSTRACT;
On March 31, 1976, the Bergen-Passaic HSA received a Certificate of Need
(CON) application from St. Joseph's Hospital and Medical Center in Patterson,
Passaic County, New Jersey, to be designated as the Regional Perinatal Center
for the Bergen-Passaic health service area. This report describes the
chronology of events and the decision-making constituting the planning process
which the HSA embarked upon to formally evaluate the St. Joseph's application.
It details the task force meetings, the use of the medical and financial con-
sultants, and the HSA's staff activities as they interacted and enabled appropri-
ate planning decisions to be made.
ANNOTATIONS:
Generally, this report is very good for the purpose of examining the case-
oriented approach to planning a Level III Perinatal Center. Although many of
the elements of the document are too detailed to explain here, the application
lists the following as some of the components of service which the hospital
plans to perform.
•	Identification of patients at-risk during their pregnancy.
•	Provision of antepartum diagnostic techniques and consultative
advice.
•	Increase intrapartum surveillance and provide an interpretive
facility available on a twenty-four hour basis.
•	Intensified neonatal care for normal newborns who unexpectedly
develop problems and sick newborns anticipated to develop
problems.
•	Development of a means to collect and analyze perinatal data.
•	Formal education programs for the public and nursing and physician
personnel based on the experience of the area.
•	An "Open Staff" in order to develop a qualified regional staff
that will utilize the Hospital when problems arise with a
pregnant patient or newborn infant. Qualified physicians will
be able to admit and follow their patients in the hospital.
Finally, an effective method for evaluating the quality of delivery in a
Level III center should always be an integral part of its operation. In
the St. Joseph's plan, daily rounds are conducted in the antenatal unit
116﻿﻿and the labor-delivery suite as well as in the neonatal intensive care and
regular newborn nurseries. Interdisciplinary conferences are held on a
regular basis along with quarterly perinatal morbidity and mortality con-
ferences which are open to all members of the perinatal team. Hence, a
similar method for continual evaluation should be standard.
The Task Force Report includes a background report on regionalization
of perinatal services, a discussion of the Task Force process, the HSA
staff's need assessment analysis, a report from the site visit and Task
Force Recommendations.
Appendices include St. Joseph's CON application, the HSA's staff
summary, "Levels of Care Criteria for the Regionalization of Maternal and
Neonatal Services in New Jersey" (New Jersey State Dept, of Health), Perinatal
Task Force and Review Committee meeting minutes, correspondence from consul-
tants, additional material submitted by the hospital and a pre-site visit
questionnaire.
Blackmon, Lillian, M.D. and Audrey K. Brown, M.D.
1973 Recommended Standards for Hospital Nursery Services:	Plan for a Statewide
System of High-Risk Newborn Care. Atlanta, GA:	Georgia Regional Medical
Program.
GENERAL ABSTRACT:
A central point of this paper is that in organizing a statewide network
of centers for care of the high-risk newborn, it is necessary to depart from
the single hospital concept. All hospitals with facilities for delivery do
not also have facilities for care of the high-risk newborn, nor is it
reasonable to plan that they will develop such facilities because of limited
resources or limited potential patient population. Recommendations for the
organization and operation of the various units in the three hospital cate-
gories are presented in the paper. The thrust is toward the best possible
circumstances and not the minimally acceptable.
ANNOTATIONS:
Primary Hospital
• Personnel Requirements
The number of patient care personnel should be calculated
on the basis of capacity as well as usual census, and on
the age from birth of the infants. For infants of less than
24 hours, the staff: patient ratio is 1:4-6 and for infants
of greater than 24 hours, 1:6-8. At least one R.N. should
be present on each shift. They should be assigned to the
nursery area permanently and not serve as substitutes in
other units in the hospital except in a rare and urgent
situation. Clerical and housekeeping personnel should also
117﻿be assigned as needed to free the professional nurse for patient
care duties.
•	Space Standards
Admission Area — 30 square feet per patient station.
Special Procedures Area — 30 square feet.
Normal Newborn — 20 square feet per patient station.
Secondary Hospital
•	Nursery Service
The nursery service of the Secondary Hospital shall be a distinct
unit in hospital organization, nursing administration, and medical
staff responsibility. It shall consist of at least two separate
parts: the Normal Newborn Nursery and the Intermediate and
Special Care Units. A committee of physicians appointed by the
medical staff shall have the responsibility for reviewing the
operation of the nursery service and establishing routine
policies and procedures.
•	Personnel Requirements
In the Normal Newborn Nursery the staff: patient ratio is 1:4-6.
At least one R.N. should be present on each shift. Her work
assignment should be such that she is free to observe each new
baby admitted. The remaining staff may be composed of aides and/or
L.P.N.s.
In the Intermediate Care Unit, one R.N. should be assigned as a
supervisory person for each ten patients on each shift. She may
also have patient care responsibilities. The remaining staff
shall be composed of aides and L.P.N.s at a ratio of 1:3-4 personnel
to patients. R.N.s should be assigned for patient care in the
Special Care Unit as a nurse to patient ratio of 1:2. They should
be freed from all non-patient care duties.
Assignment to the Intermediate and Special Care staff should be
considered permanent and members of the staff should not be
utilized as substitutes in other hospital areas except in a rare
and urgent situation.
•	Space Standards
•• Intermediate Care
30-40 square feet per patient station
Mothering Area — 80-100 square feet
•• Special Care
60-80 square feet per patient station
Special Procedure Area — 80-100 square feet
118﻿Tertiary Hospitals
•	Personnel Requirements
The patient population of the Special (Intensive) Care Unit
shall consist of infants at greatest risk for morbidity and
mortality and thus shall require the highest level of nursing
skill and time. R.N.s should be assigned at a nurse to patient
ratio of 1:1-2 according to the patient care demands. They
should be freed of all other duties when assigned to patient
care. Requirements for other care units remain the same as for
secondary hospitals.
•	Space Standards
Intermediate Care — 30-40 square feet per patient station.
Mothering Area — 80-100 square feet.
Special (Intensive) Care — 60-80 square feet per patient station.
Special Procedure Area — 100-150 square feet.
Special Capabilities (Secondary and Tertiary Hospitals)
•	Secondary Hospital
A pediatrician with special interest and training in neonatalogy
is essential for both administrative and patient care reasons
in the Secondary Hospital. A part-time hospital employed pediatri
cian-neonatologist is desirable.
•	Tertiary Hospital
A full-time neonatologist and a pediatric surgeon are essential.
A suggested clustering of other sub-specialists is:
Pediatric Cardiologist
Pediatric Hemotologist
Pediatric Radiologist
Pediatric Pathologist
Pediatric Neurologist
Limited to a few specializing centers:
Thoracic and Cardiac Surgeon
Pediatric Urologist
Orthopedic Surgeon
Plastic Surgeon
Neurosurgeon
Pediatric Anesthesiologist
Opthalmologist
Professional Skills
The physician who assumes responsibility for resuscitation of the new
born should be skilled in all aspects of managing the depressed newborn
at birth. The following are essentail skills:
• Laryngoscopy and endotracheal intubation of all size newborns
119﻿•	Trachial suctioning
•	External cardiac massage
•	Bag-to-tube and bag-to-mask assisted ventilation
•	Umbilical vein catheterization
•	Administration of appropriate drugs, dosage and route
Nursing Skills for resuscitation of the newborn include:
•	Oropharyngeal and nasopharyngeal suctioning with catheter
•	Bag-and-mask assisted ventilation
•	External cardiac massage
•	Administration of appropriate drugs, dosage and route
Comprehensive Health Planning of Northwest Illinois.
1977 Planning Policies for Perinatal High Risk. Rockford, IL.
GENERAL ABSTRACT:
This reference is a chapter from the third edition of the CHP of North-
west Illinois' Planning Policies document. It articulates the agencies'
policies for Perinatal/High Risk Services on: service and determination,
affiliation agreements, consolidation, levels of care, communications,
educational networks, faculty liaison relationships, staffing and bed addi-
tions and bed need determination.
ANNOTATIONS:
Accessibility
Inaccessibility for NICU systems denotes conditions in a defined area
for provision of perinatal/high-risk services are or will be lacking as
evidenced by such quantifiable factors which adversely affect care and the
health status of the area as follows:	(a) unfavorable infant mortality rate
and other health status indices, (b) size of indigent population, (c) non-
availability of physician or hospital care for the poor, (d) socio-economic
barriers to care, (e) unfavorable time and" distance factors to nearest
physician and/or hospital resource accessbile and acceptable to this popula-
tion.
Staffing
When a NICU project is proposed by a planning agency the plan must show
documentation of the availability and numbers of staffing to include but not
be limited to the following personnel possessing the proper credentials:
•	Neonatalogists.
•	Obstetrician — a physician certified by the American Board
of Obstetrics and Gynecology and who specializes in high
risk maternal care.
120﻿•	Pediatrician — a physician certified by the American Board
of Pediatrics.
•	Nurses — including a person with an advanced degree in obstetric
or newborn nursing qualified to be a designated Clinical Nurse
Specialist if that person is the unit's nursing service supervisor.
•	Administrative Director — a physician who may be a neonatologist
with extensive training and experience in perinatal medicine and
in administration of health services.
Bed Need
To determine bed need in an area the procedure is as follows:
•	Determine the number of live births in the area.
•	Multiply the number of live births by .10 to determine the
number of neonates requiring intensive and/or intermediate
care. (A standard estimate is that 10% of all live births
will require intensive and/or intermediate perinatal care.)
•	Multiply the number of neonates by 15 (length of stay) to
determine the projected patient day total.
•	Divide the projected patient day total by 365 (days) to
determine the average daily census.
•	Divide the average daily census by an 85% optimum occupancy
figure to determine the bed need in an area.
Comprehensive Health Planning Agency of Southeastern Wisconsin, Inc.
1976 Guidelines for Perinatal Services. Milwaukee, WI.
GENERAL ABSTRACT:
This is an outline or guide for the development of perinatal services
within a comprehensive, quality system. It presents explanations of the
purpose for perinatal and neonatal care systems, rationale for systemization,
and various designations of responsibility in a regional system. Standards
for transferring newborn cases with the system are presented.
ANNOTATIONS:
Every attempt should be made to transport certain high-risk cases to a
facility able to meet all their needs. If it seems necessary to treat the
121﻿patient at an intensive care center, the patient should be transferred directly
to a Regional Intensive Care Center without stop at a Perinatal Center. In
order to maintain continuity of care it is essential to establish a coordinated
program for continued participation of the referring physician in the care
of the partient. The following list is standard for the decision to transfer
a case to the Regional Intensive Care Center:
•	Gestation of less than 32 weeks; birth weight less than 1,500
grams or 3 lbs. 6 oz.
•	Respiratory distress and metabolic acidosis persisting after
2 hours of age or requiring ambient oxygen in excess of 40%
after 2 hours of age.
•	Infant of a diabetic mother.
•	Neonatal seizures.
•	Suspected neonatal sepsis and/or meningitis.
•	Congenital anomalies requiring observation for neonatal surgery.
•	Meconium aspiration.
The following is a listing of complications of pregnancy and may suggest
consultation and possible transfer of the mother. Previous History:
•	Two or more previous premature labors of history of low
birth weight infants (less than 2,500 grams).
•	Excessively large infants (greater than 4,000 grams).
•	Previous caesarean section.
•	Previous significant dystocia.
•	Two or more previous abortions.
Federal Register
1978 National Guidelines for Health Planning:	Neonatal Special Care Centers
43 (March 28), Section 121.204:	13040-13050.
GENERAL ABSTRACT:
•	Neonatal services should be planned on a regional basis with
linkages with obstetrical services.
•	The total number of neonatal intensive and intermediate care
beds should not exceed 4 per 1,000 live births per year in
a defined neonatal service area. An ajustment upward may be
justified when the rate of high-risk pregnancies is unusually
high, based on analyses by the HSA.
•	A single neonatal special care unit (Level II or III) should
contain a minimum of 15 beds. An adjustment downward may be
justified for a Level II unit when travel time to an alternate
unit is a serious hardship due to geographic remoteness, based
on analyses by the HSA.
122﻿ANNOTATIONS:
For this standard, the Department has adopted the widely endorsed
concept of regionalization, involving various levels of care. Under this
concept, Level III units are staffed and equipped for the intensive care
of newborns as well as intermediate and recovery care. Level II units
provide intermediate and recovery care as well as some specialized services.
Level I units provide recovery care.
Neonatal, special care is a highly specialized service required by only
a very small percentage of infants. The Department believes that four
neonatal special care beds for intensive and intermediate care per 1,000
live births will usually be adequate to meet the needs, taking into account
the incidence of high risk pregnancies, the percentage of live births re-
quiring intensive care, and the average length of stay. ("Bed" includes
incubators or other heated units for specialized care, and bassinettes.)
In addition, the Department has established a minimum of 15 beds per unit
for Levels II and III as the minimum number necessary to support economical
operation for these services. Both standards are supported and recommended
by the American Academy of Pediatrics.
The American Academy of Pediatrics has noted that "the best care will
be given to high risk and seriously ill neonates if intensive care units are
developed in a few adequately qualified institutions within a community
rather than within many hospitals. Properly conducted, early transfer of
these infants to a qualified unit provides better care than do attempts
to maintain them in inadequate units". This regionalized approach is re-
flected in the minimum size standard which is designed to foster the loca-
tion of specialized units in medical centers which have available special
staff, equipment, and consultative services and facilities.
Since perinatal centers which include neonatal units will serve the
patient load resulting from a representative population of more than one
million, a defined neonatal service area should be identified by the rele-
vant HSAs in conjunction with the State Agency. Special attention should
also be given to ensure adequate communication and transportation systems,
including joint transfers of mother and child and maintenance of family
contact. Hospitals with such units should have agreements with other facili-
ties to serve referred patients. The regional plan should include a struc-
tured ongoing system of review, including assessment of changes in health
status indicators.
Florida Department of Health and Rehabilitative Services, Children's
Medical Services
1977 Statewide Program for Perinatal Intensive Care Centers:	Obstetrical Com-
ponent. Tallahassee, FL.
GENERAL ABSTRACT:
A comprehensive set of standards for delivery of newborn services in
Florida are outlined with the purpose of meeting these goals:
123﻿•	To provide a Statewide Perinatal Intensive Care Center Program
to women with high-risk pregnancies in order to save the lives
and preserve the minds of their infants and to reduce maternal
morbidity that may result from sub-optimal care during the peri-
natal period.
•	To provide access to Perinatal Intensive Care Center Program
services for all high-risk maternal patients requiring such
care in Florida.
•	To provide assurance that the program method of service delivery
represents the most efficient use of funds in providing optimal
professional and comprehensive care and treatment for high-risk
maternal patients and their babies.
The sources for the developed standards were the Committee on Perinatal
Health's Toward Improving the Outcome of Pregnancy; the American College
of Obstetricians and Gynecologists, Standards for Obstetric—Gynecologic
Services; the Nurses Association of the American College of Obstetricians
and Gynecologists, Obstetric Gynecologic and Neonatal Nursing Functions
and standards; and Florida statute 383.15-21. It is recommended that the
document be referred to in order to obtain the entire listing of recommended
standards and criteria.
ANNOTATIONS:
These standards include the following sample:
Facilities/Location
•	Institutions accepting grant-in-aid funds shall meet space
standards as stated in the DHRS Rules for Hospital Licensure,
within three years of designation of a hospital as a Regional
Perinatal Intensive Care Center and shall meet the physical
plant requirements of the American College of Obstetricians
and Gynecologists as defined in Standards for Obstetric-Gyneco-
logic Services.
•	Each delivery room shall be a minimum of 324 square feet.
•	Delivery rooms shall be located in close proximity to labor
rooms but away from the labor/delivery entrance.
Personnel
•	The obstetrical service shall have 24-hour coverage by a
consultant obstetrician for patient care and for communica-
tion with other physicians in other hospitals.
124﻿•	An anesthesiologist, with special training or experience in maternal
fetal anesthesia, should direct anesthesia services.
•	Subspecialists in pediatrics, internal medicine, surgery, and
genetics shall be available to provide consultation.
Florida, State of, Office of Comprehensive Planning
1978 Health Problem Analysis, A Phase in the Development of the Florida State
Health Plan, Infant Mortality. Tallahassee, FL.
GENERAL ABSTRACT:
Florida is trying to reduce the rate of infant loss through a number of
specific programs. A variety of interventions are considered along with
thorough analysis of infant mortality trends and projections.
ANNOTATIONS:
The equation for the calculation of the infant mortality rate for year
x is:
# of deaths to those under one year old for year x X 1,000
# of live births occurring during year x
The formula used to calculate the perinatal mortality rate is:
#	of late fetal deaths + # of neonatal deaths X 1,000
total number of births
The remaining infant mortality measures used are cause-specific rates.
These measures are used to provide probabilities of infants dying from
specific diseases or conditions. The formula for these measures is:
X of infant deaths due to cause Y in year x X 1,000
total number of live births in year x
It should be noted that the vast majority of causes of infant death are
not easily classifiable under the four major factors discussed above. Fifty-
one percent of all infant deaths are due to the broad spectrum of diseases
known as "diseases of early infancy". Similarly, the category "all other
causes" account for 11.5 percent.
Regional Neonatal Intensive Care Centers
Infants/expecting mothers must be referred to the Regional Neonatal
Intensive Care Centers by local physicians. Usually, these local physicians
attempt to identify high-risk mothers before delivery in order to assure
125﻿that the infant is born within the center. This technique is thought to
improve the chances of the infant's survival.
Since admission is based upon the criteria of physical need and avail-
able space, local physicians are supplied with a statewide telephone line to
an information center which provides data concerning the availability of
space at all centers, arranges for transportation and provides immediate
consultation.
Currently, there is a statewide discharge developmental evaluation
occurring. Included in this evaluation program are all infants discharged
from a center whose birth weight was less than 1501 grams and a 20% random
sample of all other discharged infants. The focus of this evaluation is
to ascertain the impact that these centers have on the survivability of
patients and the reduction of physical and/or mental impairments to these
infants.
Possible Health System Intervention Modes to Reduce the Level of Infant
Mortality
Community Health Promotion and Protection Services
These programs would be primarily educational in scope. They would
attempt to inform the public, particularly those who are in or are approaching
the primary child-bearing period of life (18-35 years old), of the associated
danger of behaviors which are known to be harmful to the unborn. The
settings of such programs could vary from public schools to television/
radio/newspaper advertisements.
Prevention and Detection Service
Their purpose is early detection of causes of disease and/or diseases
themselves. Included under this rubric would be genetic counseling, pre-
conception examinations, prenatal examination, post-partum examinations
and other services to try and identify potential high-risk parents.
Diagnostic and Treatment Services
The further development of such services designed to lower infant loss
should be based on the outcome of new programs developed under the preven-
tion and detection services.
Health System Enabling Services
These services would include provisions to enhance the condition under
which other health services would be delivered. Programs developed within
this area could include those which continually monitor the trend in infant
mortality and make recommendations to the appropriate governmental bodies.
On needed action, the awarding of grants that would enable the conducting
of research which might have a positive effect on infants' life chances,
and the development of ancillary services, such as transportation, which
would increase the utilization of preconception, prenatal and post-partum
services to those who require such.
126﻿Florida, State of, Office of Comprehensive Planning
1978 Health Problem Analysis, A Phase in the Development of the Florida State
Health Plan, Obstetric Services. Tallahassee, FL.
GENERAL ABSTRACT;
Both in Florida and elsewhere in the nation, experience with the neonatal
centers has shown an approximate reduction from 50% to less than 10% in
retardation, cerebral palsy, and other results of prematurity, when com-
pared with impairment rates of low birth weight babies prior to the avail-
ability of neonatal intensive care services.
Efforts to improve pregnancy outcomes have resulted in the development
of several programs. Three major programs are operational in Florida. They
are the Maternal and Infant Care Program, the Regional Neonatal Intensive
Care Centers and the Special Supplemental Food Program for Women, Infant and
Children. The Improved Pregnancy Outcome project will begin operation soon.
ANNOTATIONS:
There are data which indicate that those infants who are born in a hospital
with a neonatal intensive care center, have a better chance of survival
than those infants who must be transferred. The following data from the
1975-76 Evaluation Report by Children's Medical Services, Department of
±i and Rehabilitative	Services,	give a comparison:	
	PATIENTS	DEATHS	MORTALITY RATE
BORN IN RNICU	2,288	210	9.2%
TRANSPORTED IN	1,232	217	17.6%
TOTAL	3,520	427	12.1%
Also, there are other data which indicate those infants born in hospitals
with a RNICU may also have lower impairment rates.
According to the Florida information, the average cost per patient (NICU)
center) in fiscal year 1975-76 for both hospital and neo-natologist fee was
$2,270.33. However, the cost of an individual infant can vary significantly.
For example, an infant with respiratory distress syndrome can cost about
$17,000 while a very small sick infant can run a bill up to $50,000 or
$60,000.
Maternal and Infant Program (MIC)
MIC is a program offering general obstetric and infant care services to
indigent families. Services including nutrition, personal hygiene and
general health counseling as well as medical and nursing care are provided
in-house and referrals are made to other programs for medical treatment
of high-risk patients.
127﻿Regional Neonatal Intensive Care Centers (RNICC)
Florida will soon have a statewide network of RNICC's. Since national
standards recommend one RNICC for every 10,000 live births per year, these
new centers should allow Florida to more adequately address the needs of
its newborns requiring early intensive care. Within the RNICC's, services
for maternal-fetal, and neonatal care must all be located in the same
complex and within close proximity. The purpose of the program is to
reduce morbidity and the chances for permanent impairment.
Special Supplemental Food Program for Women, Infants and Children (WIC)
The WIC provides high nutritional supplements to high nutritional risk
population groups which include low income pregnant women, post-partum
(up to six months) and nursing (up to one year) mothers, infants, and
children under age five. In order to be eligible for program benefits,
applicants must meet three criteria:	nutritional risk, financial risk,
and geographic location.
Improved Pregnancy Outcome Project (IPO)
In an attempt to reduce the incidence of fetal, neonatal, post-neonatal
and maternal mortality, the IPO project will employ two general intervention
strategies:
•	identify and treat the factors associated with high-risk
pregnancy, and
•	develop a sophisticated and comprehensive medical program to
provide care to mothers and newborns at high risk.
General findings of the Florida Office of Comprehensive Health Planning
include:
•	Government sponsored programs of prenatal care should develop
casefinding and outreach capabilities to the extent that these
programs achieve at least 50% client enrollment during trimester
1 of pregnancy.
•	The State should continue to fund the development of regional
perinatal intensive care centers in order to facilitate the
development of "levels of care" in obstetric care.
•	Fuller utilization should be made of Certified Nurse-Midwives
in the delivery of obstetric care. Evidence indicates that
these professionals provide high quality care and enjoy a
high degree of consumer acceptability.
•	Changing prevailing assumptions employed in determining obstetric
bed need can result in greater cost containment than efforts to
consolidate services. For this reason efforts should be made:
to encourage more OB physicians and physician groups to maintain
128﻿privileges at more than one hospital and investigate the
potential for decreasing the length of postpartum hospital
stays.
Grand Rapids Emergency Medical Services Symposium on Critical Care
Patients
1975 Grand Rapids EMS Symposium on Critical Care Patients:	Part III. Systems
Approach to the Care of the High Risk Perinatal/Neonatal Patient.
Grand Rapids, Michigan. Available from: Emergency Medical Services
Clearinghouse, Box 911, Rockville, MD 20852.
GENERAL ABSTRACT:
This document provides an overview of ten topics from the EMS perspec-
tive. Relevant material is annotated below. In addition to a selected
bibliography, each chapter references several specialized documents.
ANNOTATIONS:
Systems Conceptualization and Design for Pre-Hospital, Hospital and
Critical Care Phases
This chapter covers topics of: state and regional organizational
responsibilities for development of a perinatal system, regional relation-
ships, three phases in the development of a system (from 0 to 5 years),
historical development of perinatal care systems and brief descriptions
of existing systems noting staffing and other characteristics. A section
on special issues discusses: rationale for location and structure of
the high risk unit, physical structure, communications, use of a central
clearinghouse for patient census and referral information and new adapta-
tions (resuscitation teams, increased nursing staffing at peak times,
subsidies for regional high risk care, medical cross-staffing within a
region, research and tracking of care for EMS patients going through the
system).
Identification of the High Risk Infant
Approximately 50%-60% of all fetuses destined to become sick neonates
can be identified. Early identification of high risk infants and delivery
of proper care has decreased perinatal mortality and improved long-term
morbidity. After Wisconsin initiated regional care centers in 1968-1969,
there was an overall statewide improvement in morbidity which was much
greater in areas served by NICUs compared to those without them. A partial
list of high risk factors, organized by the time sequence in which they
occur is abstracted here.
129﻿• Prior to Pregnancy
•• Anemia
•• Diabetes
•• Age under 15, and over 40
•• History of previous fetal or neonatal loss, low
birth weight infants or genetic defects.
•	During Pregnancy
•• Medical illnesses, e.g., hypertensive cardiovascular
disease, hyperthyroidism, renal disease, venereal
disease, hepatitis, herpes simplex, and other infectious
diseases.
•• Narcotic addiction
•• Drugs
•• Marked nutritional abnormality
•• Abnormal fetal presentation
•• Abruptio Placentae
•• Eclampsia, pre-eclampsia
•• Malignancy
•• Premature Labor
•• Prolonged pregnancy
•• Multiple pregnancy
•	During Labor
•• Fetal heart rate aberrations
•• Fever or other signs of infection
•• Meconium in amniotic fluid in other than breech
presentations.
•	At Birth
•• Birthweight less than 1500 grams
•• Severe blood loss
•• Low Apgar score especially at five minutes.
•• Congenital malformations e.g., choanal atresia,
cleft lip and/or palate, diaphragmatic hernia,
cardiac and pulmonary malformations, tracheo-
esophageal fistula, omphalocele, myelomeningocele
and renal anomalies.
•• Meconium staining
•• Severe hemolytic disease
•	In the Nursery
•• Respiratory distress, e.g., respiratory distress
syndrome, pneumothorax.
•• Thermal instability, fever or maternal fever.
•• Convulsions.
•• Shock, asphyxia, or metabolic acidosis persisting
beyond two hours of age.
130﻿•• Cyanosis
•• Sepsis and/or meningitis
•• Apnea
•• Hypoglycemia
Areawide Planning for an Emergency Neonatal Transport System, Facilities
and Manpower Resources
Facilities and manpower planning are discussed under the assumption
that a viable transport service is a necessary component of a three level
system.
A method for determining NICU bed need is proposed:
• Calculate population in need on the basis of number of
live births in an area and its population density,
socio-economic factors, and percent of pre-term
infants.
•	Determine average length of stay.
•	Calculate number of patient days.
Patient days = Number of Newborns in Need
of Secondary or Tertiary Care
Average Length
of Stay
Determine bed need
,	, , Number of Patient Days
Beds Needed = ----------—------------
365
Sample staffing standards are also presented:
•	5 staff nurses are required for 1 nurse for each
of 21 shifts per week when vacation, education
and sick time are considered.
•	Ratios of 1 nurse to 3 infants for special care
and 1:1 for intensive care are suggested.
Treatment Protocols for Pre-Hospital and Critical Care Phases
There are at least four approaches used to develop treatment protocols.
•	Early referral (at 20-28 weeks gestation) of the high
risk mother to the center obstetrical staff for defini-
tive management protocols for specific conditions with
the mother being delivered at the hospital or at the
center according to her response to management.
•	Early arrival at referring hospital and stabilization of
patient by center personnel rapid transport to the Center.
In some instances a team from the center may arrive prior
to delivery for resuscitation of the newborn at delivery
or to transport the mother to the center.
131﻿•	The center team begins definitive treatment of the newborn
or mother and then transports the stabilized patient
to the Center in a deliberate manner.
•	Stabilization and treatment primarily provided by the
referring hospital staff in phone consultation with
the center staff prior to transport of the patient by
the referring hospital or center personnel.
Protocols for functions performed at Levels I, II and III are briefly
discussed as are education, transport and communications. The need for
system-wide protocol standards is discussed.
Technological Advances and Adoptation Now Applicable
This chapter briefly examines antepartum evaluation of fetal well-being
(amniocentesis, ultrasonography, oxytocin challenge test); intrapartum
monitoring of fetal status (continuous fetal monitoring) and neonatal
monitoring (neonatal respiratory assistance, apnea monitors, measurement
of ambient oxygen concentration, blood gas assessment, ultrasonography,
photo therapy, infusion pump, environmental control and biochemical moni-
toring .
Training Programs for Spectrum of Para-Professionals and Professionals
The staff components of an EMS system are discussed as they relate to
high risk newborn care.
Evaluation of Mortality and Morbidity Indices and Compliance in the
Care of the High Risk Newborn Infant
Evaluation is performed at 4 levels: the hospital of birth, inter-
hospital care (transport), the Level II or III referral unit and at the
State agency (Health Department).
Data collection at the hospital of birth should include, for all
infants:
•	number of live births
•	number of fetal deaths (early, intermediate, late)
•	number of neonatal deaths
•	birthweight (in grams)
•	gestational age (accurate to within two weeks)
•	age at death
•	number of infants transferred to another facility for
special care
Care practices should be reviewed in light of the standards of care of
newborn infants as outlined in Hospital Care of the Newborn Infant. Parti-
cular attention should be paid to resuscitative and supportive care practices,
surveillance for the infant at risk, management of special conditions (pre-
maturity, respiratory distress, the small for gestational age and large for
132﻿gestational age infant), and criteria for transfer. Autopsies should be
obtained wherever possible and reviewed on a regular basis to identify pre-
ventable causes.
Inter-hospital care data is usually collected by the referral center.
Such information relates to:
•	number of infants transferred
•	hospital of origin
•	hospital of destination (Level II or III)
Inter-hospital transfer activity is a measure of the degree to which an
institution or region complies with the concept of regionalization. Level I
and II units should transfer patients according to their high-risk newborn
rate. Transfers will also vary with the number of high-risk pregnancies
transferred to an appropriate facility prior to delivery (which will reduce
the number of high-risk infants) and the level of care able to be provided
locally (which may modify the criteria for transfer). Such figures can give
an estimate of the number of newborns in a region which might require
transfer for special care and this can be compared with the actual volume.
The disparity between the figures (calculated versus actual) could provide
an index of compliance. Poor compliance may be assessed at the patient
(parent), physician, hospital administration or community levels.
Referral center data should include (in addition to that recorded for
any inborn infants under hospital of birth):
•	Birth weight (in grams)
•	Gestational age (accurate to within two weeks)
•	Neonatal deaths
•	Age at death
Transfer rates (per 100 live births) should be compared with mortality
rates (predicted and actual) for both hospitals and regions (counties,
states, etc.). Studies have shown that the transport of high-risk newborns
can significantly alter mortality rates. Low transfer rates and high
mortality suggest a hospital which may have problems in identification of
high-risk infants, early supportive care or poor compliance with the concept
of regionalization for whatever reason. High transfer rates (e.g., in excess
of 5%) may suggest inability to provide minimum supportive care appropriate
to a Level I unit or indicate a high-risk population.
The referral center is charged with the greatest responsibility for
evaluating the qualitative aspects of high-risk newborn care. While mor-
bidity in infants between 1500 grams and 2500 grams birth weight has decreased
dramatically, attempts to produce survivors from infants less than 1500 grams
(particularly less than 1000 grams) may still be associated with significant
morbidity in the form of CNS handicap, retrolental fibroplasia and chronic
lung disease. Centers should carry out frequent audit of the short and
long-term outcome of intensive care practices.
The state's Health Department should play the major role in the collation
and evaluation of mortality indices. Information should be compiled on:
•	fetal death rate (broken down by gestation: early, intermediate
and late; the latter being further sub-divided antepartum,
intrapartum),
133﻿•	neonatal death rate (broken down by age at death: under 24
hours; under 72 hours; under 7 days; under 28 days),
•	perinatal death rate
All deaths should be charged to the hospital of birth. Deaths occurring
outside hospital should be recorded by scrutiny of death certificates.
Neonatal death rates can be grossly misleading for comparative purposes
if birth weight and gestational age characteristics of the population are
not taken into account. Neonatal mortality is markedly affected by the
number of low birth weight and premature infants in a newborn population.
These effects can be normalized by using "predicted mortality" for a given
population based on a reference standard. This can then be compared to the
actual mortality of that population.
When comparisons are made between populations, the predicted mortalities
are used to determine their comparability with respect to birth weight and
gestational age.
Educational Programs in Perinatal Health to the Consumer and the Community
Health Systems Agencies are encouraged to develop consumer education
subcommittees which can facilitate regionalization efforts.
Federal Program on Neonatal High Risk Infants
The major support for services for high risk infants are appropriations
ynder Title V of the Social Security Act. The title is described.
The Impact of P.L. 93-641 on Perinatal Manpower
A major concern from the perinatal perspective is how manpower can be
developed to meet health service needs. Approaches to predicting national
requirements are discussed.
Health Planning Council, Inc.
1978 Perinatal Care Center Standards (Draft). Madison, WI.
GENERAL ABSTRACT:
This document discusses agency-developed standards and criteria for
Perinatal Care Centers.
ANNOTATIONS:
Determination of need projections for perinatal care centers shall be
calculated in the following manner:
134﻿•	In a service area, the five-year projected need for neonatal
care beds shall be computed as follows:
•• Neonatal Intensive Care Beds =
0.9 x 5 year projected average annual number of births
1,000
•• Neonatal Intermediate Care Beds = 1.13 x Intensive Care Bed Need.
• The number of additional neonatal ICU beds needed in a service
area shall be determined by subtracting existing beds from the
five-year projected bed need.
•• The neonatal ICU bed need shall be adjusted to reflect
the volume of patients who may appropriately be served
in neonatal intensive care facilities located out of
state or in adjacent service areas and the volume of
patients from adjacent areas who may appropriately be
served in the NICU facility.
Emergency Transport for High-Risk Infants
These standards agree with those explained by authors such as Swyer.
Generally, a twenty-four hour emergency transport capability for high-risk
infants, either through agreements with ambulance services outside of the
perinatal care center or through a transport capability operated by the
perinatal care center should be available.
Health Systems Agency of Kane, Lake and McHenry Counties
1977 Perinatal Services Support Document. Lake Zurich, IL.
GENERAL ABSTRACT:
The specific services that are included within this Perinatal report
include:	(1) family planning services; (2) pregnancy testing and counseling;
(3) patient education on high-risk factors; (4) prenatal medical evaluation
and monitoring; and (5) inpatient obstetric and newborn care.
Several significant problems are identified. These include:
• The neonatal mortality rate for Kane County and the perinatal
mortality rates for non-whites in the Kane, Lake and McHenry
population are significantly higher than the overall figures
indicating health status problems particularly to those
segments of the population.
135﻿•	There is strong emphasis placed on providing more counseling
and family planning services.
•	Other improvements in the perinatal system call for alternative
medically coordinated settings for delivering both prenatal
and obstetric services as a means to provide more acceptable
care and to restrain costs. The increased use of allied
health manpower—nurse practitioners in particular—was
emphasized to increase the productivity of existing physician
manpower.
ANNOTATIONS:
Systems Goals
•	There should be a multi-level system of delivery for all
inpatient facilities servicing the maternal and infant
obstetric needs of the HSA residents.
•	Increase the availability of subsidized prenatal medical
evaluation and monitoring for low and moderate income
women in the HSA.
•	The cost of delivering inpatient obstetrical and newborn
services in the HSA should be restrained, consistent with
the delivery of quality medical care.
•	Inpatient obstetrical units, capable of providing at Level
I care, should be accessible to all pregnant women within
the HSA.
•	The incidence of perinatal loss per individual hospital should
be reduced.
•	Continuity of perinatal care should be improved by the avail-
ability in all inpatient obstetric facilities of partial or
complete rooming-in programs and family-oriented services.
•	To improve the availability of perinatal care by allowing
for increased utilization of alternative medically coordinated
delivery settings and services.
•	There should be a sufficient number of neonatal, Level III,
intensive care beds accessible to the HSA population.
136﻿Health Systems Agency of San Diego and Imperial Counties
1978 Newborn Intensive Care Study Group, Report to the Planners Committee.
San Diego, CA.
GENERAL ABSTRACT:
The HSA of San Diego and Imperial Counties recognizes that neonatal
intensive care is the provision of comprehensive and intensive care for all
contingencies of the newborn infant. Infant transport is an integral part
of a regional NICU system. Three integrated levels of newborn care must
be provided.
ANNOTATIONS;
In a letter from the San Diego/Imperial HSA to the Federal Office of
Planning, Evaluation and Legislation, some comments were made on Proposed
Guidelines for Neonatal Intensive Care Units. Some of these are useful for
investigating in the process of systems development.
First, the letter suggests that the age limitation on patients served
is too restrictive. Although newborns are defined for statistical purposes
as babies less than 29 days old, newborn intensive care units serve somewhat
older babies as well. The HSA recommends that the age limitation be eliminated
from the guidelines.
Also, it is stressed that Guidelines should distinguish between tertiary
and intermediate level neonatal intensive care facilities. In particular,
standards that require all neonatal intensive care units to serve populations
of a million or more and that do not allow exceptions for travel time are
not appropriate for the intermediate level neonatal intensive care units.
It is recommended that the National Guidelines (and regulations adopted
by planning agencies) be revised to allow consideration of travel time in
determining how intermediate level units should be distributed.
Finally, it is recommended that any reference to a study of efficiency
of NICUs should be specific as to the bed composition and the occupancy
rates maintained in the units studied.
Some Recommended Guidelines
•	Existing intensive care newborn nursery services shall have an
annual occupancy rate of 85 percent or more.
•	In consideration of economic feasibility and high quality of
care, any hospital with an intermediate intensive care unit
should have at least 1,000 to 2,000 births per year. (Toward
Improving the Outcome of Pregnancy, Committee on Perinatal
Health.)
•	Physicians referring sick newborns for intensive care should
be kept informed of and involved in the care of the newborn
at all stages in the care process. (Toward Improving the Outcome
of Pregnancy.)
137﻿•	There should be an established plan in the newborn NICU
whereby infants no longer requiring acute or sub-acute
intensive care are transferred back to the referring hospi-
tals. (as stated by George Ryan, M.D.)
•	Intermediate newborn intensive care facilities should not
provide care for newborns who require assisted ventilation
for more than 12 hours.
Illinois Central Health Systems Agency
1977 Health Systems Plan (Perinatal Services Component). East Peoria, IL.
GENERAL ABSTRACT:
These standards and recommendations are based on the 1974 State of
Illinois Plan for Perinatal Health in Illinois, developed by the Compre-
hensive State Planning Agency and Illinois Committee on Perinatal Health.
ANNOTATIONS:
Definition of High-Risk Pregnancy:
•	Occurring to women older than 40 years;
•	Occurring to women 19 or younger;
•	Occurring to women who have had 5 or more previous pregnancies;
• The birth of an infant of less than 5 pounds, 8 ounces;
• The birth of an infant weighing more than 9 pounds, 15 ounces;
•	Multiple births;
•	Those occurring to women with a previous fetal or infant death;
•	Those occurring to single women;
•	Those occurring to women with previous abnormalities;
•	Those occurring to women with a medical history of hypertension,
diabetes, or other medical or surgical conditions, and those
that are born during a delivery that had complication.
The above criteria/conditions should be served at a Level II or III
facility.
Some General Standards:
•	Illinois suggests that there be 1 perinatal intensive care
center per 15-20,000 births and the following guidelines
for Perinatal Medicine Intensive Care Center:
•• A geographic location reasonable related to the region
to be served by the center;
138﻿•• A record of perinatal care demonstrating high quality,
progress, and experience in high-risk perinatal medicine;
•• A well-qualified medical staff, continuously available
and capable of providing care and consultation in a
maximally efficient manner;
•• On-going educational program at many levels of experience,
not only as a measure of a hospital's ability and prepared-
ness to mount new educational programs, but also to pro-
vide continuous care of high-risk perinatal medicine;
•• Qualified full-time personnel in key positions to evaluate
and help upgrade programs in regional hospitals.
Acceptability:
Perinatal care centers and follow-up services should seek feedback
regarding satisfaction with services from both consumers and providers.
Continuity:
The different levels of perinatal care services should be closely linked
with each other and with follow-up programs for continuing care.
Maine, State of, Comprehensive Planning Agency
1973 Intensive Care of High Risk Newborns in Maine. Augusta, ME.
GENERAL ABSTRACT:
The concerns expressed by Maine's pediatricians and medical practitioners
on the issue of developing a neonatal intensive care capacity prompted the
State Comprehensive Health Planning Agency to bring together interested
physicians and concerned groups. The meetings resulted in the Planning
Agency's contracting with a consultant neonatologist, Dr. Joseph Kennedy,
Jr. of St. Margarette's Hospital and Tufts University. Dr. Kennedy agreed
to review background material supplied by the Planning Agency, visit
selected Maine hospitals and pediatricians and write a report summarizing
his findings on neonatal care in Maine. It is suggested that its statistical
findings and recommendations for program development are comparable and
applicable to situations in other states.
ANNOTATIONS:
In Maine for the years 1969-1971 there were 771 neonatal deaths reported.
The following categories (I.C.D.A. codes) describe the primary cause of
death of nearly 775 neonatal deaths in these three years. This report contends
139﻿that 225 were probably preventable if an intensive care facility were
established and if a transportation referral system were operative.
Neonatal deaths for Maine
by selected causes 1969-1971
Cause (I.C.D.A. code)	Explanation	Number
776.1	Hyaline membrane	101
776.2	Respiratory distress	37
776.9	Asphyxia of newborn	183
777	Immaturity	93
Other		357
Total	771
Size of Neonatal I.C.U. for Maine
An intensive care nursery must be prepared to handle a minimum of 1/3
of the infant deaths in a region. It is recommended that a fifteen to
twenty bed nursery seems to be ideal size in terms of optimum care and
management. With current mortality experience, a fifteen bed I.C.U. nursery
can serve 20-25,000 annual deliveries. Maine would supply approximately
two hundred and fifty infants yearly if the region served is the entire
State- This would require a ten to twenty bed I.C.U. nursery and would also
depend upon the amount of care given in the referring hospital and whether
or not convalescing infants were transferred back to the hospital of birth.
Transportation
As described in the statistical section of the report, an important
component of any proposed regional care system is the safe and rapid
transfer of the infant from outlying facilities to the intensive care
nursery. An effective transfer system must include one or more infant
transfer incubators with a battery capable of coping with low temperatures
over prolonged periods of time. It must be able to derive power from
either helicopter or ambulance. A heated ambulance will be necessary in any
event for the small infant. Methods for the provision of oxygen, suction
and incubation must be available when needed. A physician or nurse
experienced in neonatal techniques must accompany the sick infant. Copies
of all pertinent records, together with a sample of the mother's blood
must be provided.
Dr. Kennedy recommends that two hours in transit by land is a maximum
for the sick newborn. Air transport should be considered for longer
trips.
Financial Considerations
The consultants' investigation determined that care in a N.T.C.U. costs
$150 to $200 per day. Average stay in an I.C.U. is 6 days. Cost per
individual is $900 to $1200. Care in a convalescent nursery at the center
or referring hospital is $100 per day. Cost per individual $2,000 plus
$1,000 = $3,000. Transportation costs may add $100 to $200 = $3,200 total
costs.
140﻿With 20,000 deliveries per year, Maine will generate two hundred and
fifty referrals per year, generating fifteen hundred patient days at $100
per day or an additional $350,000. This is a total bill of $525,000 per
year, plus transportation charges ($35,000 annually).
Insurance Coverage of Neonatal Care
A survey of Maine newborn insurance in late 1972 reveals the following
•	Blue Cross and Blue Shield
In the event the mother is eligible for maternity care
benefits, the newborn child is provided with routine
nursery care benefits in accordance with the existing
contract and is also provided with the identical level
of benefits available to the member parents under the
parents contract.
•	Union Mutual
Covers a flat rate for maternity. Whatever is left from
the mother's hospitalization will be carried over to
routine nursery care. Should the baby have a defect,
the claim has to be submitted and the insurance company
will decide whether they will pay.
•	Aetna and Travelers
These insurance companies will write policies that either
pay for the child's care from day one or from day fourteen.
Evaluation
After a regional neonatal intensive care delivery system had been set
in operation, the following methods for measuring the level of fulfillment
of objectives may be applied:
•	Demonstration of ability to correct clinical abnormalities
(e.g. hypoglycemia, hypocalcemia, anemia, hemolytic diseases,
bleeding asphyxial respiratory failure).
•	Demonstration of ability to identify and manage high-risk
pregnancies by interceding with the appropriate amount of
medical care.
•	Demonstration of improvement of overall mortality rate, compari-
son of these statistics with other area indices and notification
of institutions with significantly different rates than comparable
institutions.
•	Long term assessment of the growth, development, and morbidity
of services including evaluation through school age.﻿Maryland, State of, Comprehensive Health Planning Agency
1978 Perinatal Services. Baltimore, Maryland.
GENERAL ABSTRACT:
This paper outlines a regional approach to planning for the provision
of perinatal services, which implies a coordinated and cooperative sub-
system of health care for pregnant mothers and neonates within the Metro-
politan Baltimore area. The complexity of services, new developments in
the sub-specialties about outcomes and results require a coordinated,
planned approach to perinatal care appropriate to the needs of the mother,
fetus and neonate. For a regional perinatal system to function, attention
must be given to coordination and communication among all the components
of the system, so that appropriate levels of care are sought to improve
the outcomes of all pregnancies.
ANNOTATIONS:
Cost
According to estimates prepared by the National Committee on Perinatal
Health, the operating budget for a perinatal intensive care center with
the capacity to provide intensive care to 1,000 mothers and newborns annually
is 2.6 million dollars per year. However, since the numbers of very
high risk patients are not large, most complications of pregnancy and
abnormalities of newborn infants can be properly managed in perinatal high-
risk centers, equipped and staffed to provide moderately complex care.
These perinatal high-risk centers still cost an estimated $1.2 million
per year to operate.
Perinatal Intensive Care Center
•	Centers shall have three types of nurseries: the newborn
normal; an intermediate care area which can be used for
acutely ill and convalescent infants; and a neonatal I.C.U.
•	In order to maintain competence, the Perinatal Intensive
Care Center should be able to identify a service area
that had 8,000-12,000 deliveries annually based on
present birth rates and has 1,000 or more deliveries
in that institution.
Capabilities
•	performance of a Caesarean section within 30 minutes.
•	a blood bank service on a 24-hour basis with blood
and fresh frozen plasma available at all times.
•	in-house obstetrical anesthesia service on a 24-hour basis.
142﻿•	radiology service on a 24-hour basis, providing diagnostic
and evaluation techniques for the prenatal period. Ultra-
sonography should be available at the High-Risk Center or at
a nearby Perinatal Center (ICU). X-ray technicians shall
be available on a 24-hour basis.
•	continuous electronic external-internal maternal-fetal
monitoring with appropriate interpretation on a 24-hour
basis.
•	availability of a skilled dietician for nutritional manage-
ment of high-risk pregnant patients and moderately ill
newborn infants.
•	the NICU Center should provide education and orientation
to other public and private agencies and personnel who
provide services to the family and infants.
Massachusetts, Commonwealth of, Department of Public Health,
State and Regional Task Force on Neonatal Intensive Care and
Technical Advisory Group of the Perinatal Welfare Committee.
1977 Neonatal Intensive Care Standards and Criteria (Draft) Boston, MA
GENERAL ABSTRACT:
This group developed neonatal intensive care standards and criteria
for use in Certificate of Need and the development of plans to establish
regional multi-institutional arrangements. The Committee on Perinatal
Health Model is used as the basis for a regionalized perinatal system.
The document discusses:	effectiveness of neonatal intensive care in pre-
venting neonatal mortality and morbidity, prenatal screening of high-risk
mothers, levels of care, status of neonatal intensive care in Massachusetts,
the Committee on Perinatal Health's hypothetical budget model of a Level III
Regional Perinatal center and planning issues and methodology. The annota-
tions of standards and criteria presented here were developed from this
latter section. Appendices on a queuing approach to bed need estimation,
special care nursery standards, psychosocial aspects of neonatal intensive
care and selected references are included.
ANNOTATIONS:
Standard
Neonatal intensive care programs consist of Intensive, Intermediate
and Growing/Recovery beds which are housed within a single institution or
within separate but affiliated institutions within a 2.5 mile radius of each
other. A hospital providing 3 levels of neonatal intensive care must pro-
143﻿vide appropriate support services, maintain formal affiliation agreements
and comply with the other criteria to be part of a Neonatal Intensive Care
Program. A hospital providing three levels of care without meeting these
requirements is designated as a NICU.
•	Criteria
•• A Neonatal Intensive Care Program or NICU must:	(1) have
a 24-hour respiratory therapy service (including neonatal
ventilatory equipment); (2) have laboratory, diagnostic
and evaluative procedures for maternal, fetal and neonatal
care as stated in the current standards of the American
College of Obstetricians and Gynecologists and the American
Academy of Pediatrics; (3) maintain adequate space, equipment
and personnel for maternal and fetal intensive care in ante-
partum, labor, delivery, recovery and postpartum periods;
(4) be staffed and equipped to treat critically ill infants
including those requiring prolonged assisted respiratory
support, intravenous therapy, hyperalimentation, major
surgery, blood pressure monitoring, heart rate and respira-
tory monitoring as well as treatment of sepsis; and (5) have
affiliations directed toward formal agreements that include
transfer agreements, cooperative staffs sharing, rotation
of students and a clear definition of how return transfers
are to occur between the hospitals in a Neonatal Intensive
Care Program and Level II community hospitals.
Standard
Neonatal Intensive Care Units must be part of a NICU Program that
provides for efficient and effective affiliations with other ICUs and
linkages with community hospitals.
•	Criteria
•• A NICU should develop and/or maintain: (1) cooperative
educational programs covering all aspects of neonatal
and perinatal care (not just for returning infants);
(2) 24-hour consultation services; (3) prenatal screening
of high-risk mothers and newborns; (4) clear referral
networks for the transfer of high-risk patients when
indicated and return transfer to the community hospital;
and (5) mechanisms monitoring mortality and morbidity
within regional systems.
Standard
A NICU Program or Unit must care for at least 2% of its service area's
total births in Intensive Care, 3% of its service area's total births in
Intermediate Care, and 5% of its service area's total births on Growing/
Recovery care.
144﻿• Criteria
•• Level II hospitals with Special Care Nurseries should
have a minimum of 2,000 births and meet standards for
Special Care Nurseries set out by the Public Health
Department (in an Appendix). Level II hospitals without
Special Care Nurseries must be supported by at least
1,000 births. Exceptions are made on the basis of
geographic isolation.
Standard
A NICU Program should have a total of at least 24 and no more than 30
combined intensive and intermediate care beds and 8 to 10 growing/recovery
beds. (These growing/recovery beds may be regular nursery beds).
• Criteria
•• All these beds should maintain an annual occupancy rate
of 85% in already existing beds, and a corresponding
average length of stay of 15 days for all three levels
combined.
These levels are broken down	in the following way:
Level of Care	Average Length of Stay
Maximum/Intensive	5
Intermediate	5
Growing/Recovery	5
	15
Standard
Bed Need
• Criteria
Two formulas for determining bed need are used:
•• Queuing formula
Bed Need = (K X ADC) + (K2 X ADC)
Where = .84 and = 2.1
This formula is based on a queuing theory which is
developed on the principle that a bed will be available
95% of the time. These beds should project an annual
occupancy rate of 85% and a corresponding average length
of stay of 15 days for all three levels combined.
145﻿•• Utilization rate formula:
Standard	Total Births in Service Area X _ , „ % Infants Needing Care X ALOS Bed Need = 		. ■ ■ \ 		 365 X Occupancy (85%) This formula allows for a regional bed need projection and a bed need projection for individual programs.
A Neonatal Intensive Care Program or Unit should be located in or
affiliated with a general hospital with a maternity unit.
• Criteria
••	Optimal placement of additional Neonatal Intensive Care Unit beds should be in perinatal centers.
•• If the Neonatal Intensive Care Beds are located in a
hospital without an obstetrical unit, it must have a
Standard	formal affiliation with a hospital providing this service.
Service area: A Neonatal Intensive Care Program must be supported by
a minimum of 8,000 births per year. (A 90 minute travel time is possible
within Massachusetts' 3 designated service areas).
• Criteria
••	Adjustments to numbers of births will be made if it can be clearly demonstrated that there is a change in population projections within a given service area, not accounted for in general population pro- jections.
•• Two or more Neonatal Intensive Care Programs with
Standard	common service areas should not exist if it can be demonstrated that consolidation of those programs can effectively produce fewer Neonatal Intensive Care programs that: increase the quality of care with no significant increase in cost or reduce cost with no decrease in quality of care.
Hospitals must clearly document: (1) the number of intensive, inter-
mediate and/or growing/recovery beds in their existing unit (if any) and
in the proposed unit; (2) the manner in which the unit fits into or performs
146﻿the functions of a Neonatal Intensive Care Program; (3) the possibility
that existing or proposed beds may be used to fulfill nursery and/or
medical services other than Neonatal Intensive Care; and (4) the average
number of admissions originating strictly from within the Program or
Unit's service area, in the two most recent years.
Massachusetts Maternity and Newborn Regionalization Project
1976 The Final Report. Boston, MA. (Available from the Massachusetts
Department of Public Health).
GENERAL ABSTRACT:
Regionalization of perinatal care has received increasing support and
acceptance over the past decade. Regional perinatal programs are an effort
to organize efficient and effective systems for the delivery of perinatal
services within a defined geographic area so that all mothers and newborns
have adequate access to a full range of high-quality services appropriate
to their needs. The present emphasis on regionalization has evolved from
a variety of concerns and developments in maternal and newborn care and
has been shaped by important trends in our health care system. The
Massachusetts project is the broadest examination of the various health,
social and economic factors affecting the outcomes of birth and the
practices of newborn and maternity care.
ANNOTATIONS:
Standards and regulations for maternity and newborn services are
recommended:
•	Recommended immediate revisions of hospital licensure
regulations for perinatal services;
•	Recommended classification of high-risk maternity and
newborn patients;
•	Recommended criteria for designation of: Perinatal Centers;
•	Recommended uniform perinatal terminology; and
•	Revision of Department of Public Health Statistical Forms and
Live Birth Certificate.
Travel Time
It is necessary that all mothers and newborn infants have access to high-
risk intensive care. The "travel time" should be equitable for all high-
risk mothers throughout the State who are prebooked for their delivery in
147﻿a perinatal center and for those newborns needing to be transferred to a
neonatal intensive care unit.
A number of studies have made recommendations of 1-2 hours regarding
the maximum amount of time allowable for the ambulance transfer of a high
risk baby (Maine State Comprehensive Health Planning Agency, 1973, Ross
Conference, 1974).
Average Length of Stay
•	The calculation of average length of stay should be sensitive
to patient mixture and level of care provided. These calcula-
tions should include the four basic ICDA categories:	delivery
without complications, deliveries with complications, complica-
tions of pregnancy and induced abortions.
•	The Project recommends the following average length of stay
levels be used in projecting future utilization rates:
•• Deliveries without complications - 3.22
•• Deliveries with complications - 4.34
•• Complications of pregnancy - 2.26
•• Abortions - 1.68
•	These average lengths of stay should be updated on a yearly basis
Birth Projections
•	It is recommended that the birth projections developed by
the Project be used as a basis for the development of
birth projections for hospital maternity service areas.
A range of projected births should be calculated and
adjusted over time in accordance with actual fertility
experience.
•	Future maternity service construction and renovation
should take into consideration anticipated long-range
fluctuations in births.
Bed Need
•	Multi-institutional systems should be’ developed for maternity
services throughout Massachusetts. Two or more maternity
services, when operating together in a given multi-institu-
tional system, can realize significant cost savings when
coordinated with the queuing maternity bed need approach
recommended by the Project.
•	The Project recommends use of the queuing bed need formula
approach.
148﻿Transfer Systems
•	High-risk mothers should be identified as early as possible
in their pregnancy. Appropriate consultation should be
obtained and delivery arranged in a high-risk center when
indicated. This is particularly important for mothers with
potentially high-risk newborns.
•	Community hospitals should identify high-risk newborns as
early as possible. Appropriate consultation should be
obtained and transfer arranged to neonatal intensive care
units when indicated. Early identification and transfer
is important in order to obtain maximum benefit.
Michigan, State of, Department of Public Health, Bureau of Health Facilities
1977 Hospital Nurseries and Newborn in Care Services, Minimal Criteria and
Guidelines. Lansing, MI.
GENERAL ABSTRACT:
The Michigan Department of Public Health has developed a document which
outlines planning policies , and criteria formulas for NICU bed need require-
ments based upon methods suggested by Dr. Paul Swyer are presented.
ANNOTATIONS:
Some general considerations for planning:
•	NICU centers should serve a minimum of 6,000 infant
deliveries per year.
•	NICU should be separate from the newborn nursery since
it will usually be admitting patients born elsewhere
than in the hospital's own delivery suite.
Follow-up Evaluation
Inherent in meeting the objectives for regional intensive care centers
is the need for long-term follow-up of the patients who have been cared
for in a unit. Careful prospective follow-up shall be done for critical
evaluation of the modes of therapy as well as to assure that necessary
continuing care is being received. This long-term evaluation shall be a
function of the regional intensive care unit's staff.
149﻿Potential Patients
In planning for the development of a regional center a projection of
the potential number of infants to be cared for on a yearly basis shall
be made. Roughly, ten percent of all deliveries can be classified as high-
risk; with two to four percent of all deliveries needing intensive care.
Nursing Manpower
Neonatal intensive care consists primarily of intensive nursing. In
the highest level of intensive care it may be necessary to provide nurse
staffing on a 1:1 or 1:2 nurse patient ratio.
Medical Supervision
A qualified physician, responsible for the unit, shall be available in
the hospital 24 hours a day to assure the delivery of intensive care. This
means a large staff of trained nurses and physicians is essential for
24-hour coverage.
Physical Standards
•	To accommodate the daily or hourly changes in care require-
ments, units should be contained in one large, open space
providing a minimum of 80 square feet of usable floor
space per infant.
•	A service module situated at least 30 inches above the
floor shall be available for each infant bassinet station.
This should contain:
•• Six 110-volt single phase duplex receptacles with
proper grounding.
•• Oxygen and compressed air outlets.
•• Compressed air and oxygen storage and distribution
systems shall be installed in accordance with proper
safety codes.
Intermediate Care
An intermediate care area should be adjacent to the NICU and provide
space for 40 square feet per patient with a minimum of three feet of
clearance between bassinets.
Convalescent Area
The convalescent area is for prematurely born or low weight infants
who require frequent feeding and more nursing hours per day than normal
infants at term and those who no longer need to be in the intermediate
care area but require more nursing hours than normal infants because of
150﻿slow feeding or other conditions. This area should be adjacent to the
intensive and intermediate care areas. Thirty square feet per infant
shall be provided. There shall be four electrical, one oxygen and one
suction outlet available at each infant station in the convalescent area.
Transportation
•	The pattern of transport shall follow the natural and
logical lines of referral, as established by a regional
plan of intensive care nurseries.
•	An adequate and trained staff shall execute the transfer
in a well-equipped vehicle — either ambulance or aircraft.
The vehicle shall be equipped for total life support of the
infant.
•	The infant should not be moved until optimal stabilization
of his condition is established by the response team.
During transfer he should receive intensive therapy and
monitoring as indicated by any change in his clinical
status.
•	The establishment of a central state agency is desirable
to coordinate transport efforts and to develop avenues
for the funding of such a system. Such an agency could
also monitor the patient loads in the various nurseries,
suggesting proper primary transport to a less busy
nursery, particularly in high density population areas.
•	The referring hospital shall in all instances provide
all necessary and pertinent information concerning the
infant, parents, referring physician, medical history
of the mother, the pregnancy, labor and birth informa-
tion, diagnosis and medical detail causing the referral,
as well as blood, fluids or other therapy given. This
detail will be best provided on a specially designed
referral and transfer sheet properly executed and
accompanying the infant at the time of transfer.
Minnesota State Health Planning and Development Agency
1977 Perinatal Services. St. Paul, MN.
GENERAL ABSTRACT:
The State Planning Agency of Minnesota has developed a document which
outlines in guideline/criteria form plans for further developing and
improving the state's system of perinatal health services.
151﻿ANNOTATIONS:
Cost
The report notes that the operating budget for a perinatal intensive
care center with the capacity to provide intensive care to 1,000 mothers
and newborns annually is $2.6 million per year. However, the very high-
risk patients can be serviced in a regional perinatal high-risk center
at an estimated $1.2 million per year.
Infant Mortality Goal
Minnesota's infant mortality rate was 13.7 in 1975, below the national
average of 16.1; Minnesota proposes to lower it to fewer than 12 per
1.000	live births.
Financial Feasibility
The financial viability of an inpatient perinatal service should be
demonstrated by illustrating that the unit can be operated without financial
support from other institutional services or units. As a general rule,
the revenue generated by the unit should be able to support its own direct
operating cost, the depreciation costs, and its share of allocated overhead
costs. The following should be considered when applying this criteria:
•	The proposed schedule of prices to be used compared to
prevailing prices and costs;
•	The anticipated sources of payment;
•	A projection of anticipated utilization; and
•	Income and expense statements for each year of operation until
breakeven.
The breakeven point (when revenue exceeds all of the direct operating
costs) should be passed within three years of operation and proposed prices
should reflect actual costs.
Accessibility to High-Risk Center
The high-risk center should serve an area providing not fewer than
5.000	to 8,000 live births per year. An area of at least 2,000-8,000 live
births per year may be justified on the basis of geographic isolation.
Such centers should have a minimum of 500 deliveries per year. There should
be a minimum of 1,000 in-hospital deliveries per year in the Metropolitan
Urban Statistical Area.
Outcome of Service Use for High-Risk Centers
The intensive care center, in cooperation with other perinatal services,
should assume responsibility for developing reporting mechanisms to evaluate
152﻿the results of the regionalized perinatal system. Such evaluations
should be done annually and results made available to all components of
the system.
Input Resources
For all perinatal health facilities which are community based:
•	The service should possess the capability to perform
a cesarian section within 30 to 60 minutes.
•	The service should have blood available, including
fresh frozen plasma.
•	Anesthesia service should be available on call.
•	The service should have radiology service on call,
with the ability to perform portable basic radiologic
studies in the nursery.
•	Clinical laboratory service should be available on call,
with the capability to perform studies on blood glucose,
hematocrit and bilirubin. Capability to perform sem-
micro studies on blood gases, electrolytes and calcium
is advisable.
Nassau-Suffolk Health Systems Agency, Inc.
1977 A Plan for Regionalization of Obstetrical Services in Nassau-Suffolk.
Melville, N.Y.
GENERAL ABSTRACT:
The Plan provides a comprehensive look at what needs to be accomplished
in regionalization of newborn services in Nassau-Suffolk areas of New York.
ANNOTATIONS:
Standards for three levels of newborn care are outlined. Here some
standards for the most intensive level of care are presented.
Consultation
The Level III unit must have the capability for providing consultation
for prenatal, intrapartum and neonatal problems of patients at Level I
units and at Level II units in the region. This must be an around-the-clock
capability, and the consultant must be highly qualified and have recognized
competence. Clearly defined responsibility and procedures for obtaining
consultation must be understood by all.
153﻿Transport
Equipment and a trained team of personnel for the transport of
maternal-fetal and newborn patients must be available at all times at
the Level III unit. Optimal management and treatment before and during
transport will be possible if the transport team goes to the referring
hospital with all necessary supplies and equipment. This also provides
opportunity for direct education of personnel at the referring hospital.
Perinatal Data Collection/Evaluation
The Level III unit must take an active role in developing mechanisms
for reporting results and analyzing these to assess accomplishments or
deficiencies of function in the regional system. All components of the
region must participate in planning these mechanisms and in sharing and
analyzing the data. There will be periodic evaluation of Level III care.
Teaching
Most Level III units will be actively involved in the teaching and
training of medical students, interns, residents, and fellows, as well
as of nurses and other health professionals.
The Level III units must also be responsible for developing and
implementing programs of continuing or refresher education throughout
the regional system.
Research
Most Level III units will be actively engaged in research to develop
new knowledge in reproductive biology and in related aspects of behavioral
science. This must be considered in relation to the extent of responsi-
bility for patient care assigned to staff members.
Personnel Requirements
• Physicians
•• Administration
Administration of a Level III unit should be the responsibility
of a physician with extensive training and experience in
perinatal medicine and in administration of health services.
Responsibilities will include financial and budgetary matters,
planning and development of regional programs, coordination
of program elements, and evaluation of effectiveness. The
Administrative Director may also function as the Associate
Director for Obstetrics or Neonatology.
•• Associate Directors for Obstetrics and Neonatology
The Associate Director for Obstetrics and the Associate Director
for Neonatology must have extensive training and experience in
154﻿high-risk care and in interspecialty function in providing
this care.
•• Anesthesia
Perinatal anesthesia services must be directed by an
anesthesiologist with special training in maternal-fetal
and neonatal anesthesia and be on call 24 hours a day.
•• Physician Function
The physician should examine the mother shortly after
admission and at appropriate intervals during labor to
assess her condition and the status of the fetus. He
should be appropriately available during the later stages
of labor or when there is evidence of fetal distress or
maternal complications. In cases of emergency cesarean
section or unexpected fetal distress, a second physician
should be available to assist the primary physician with
the care of the mother and subsequent resuscitation and
care of the newborn.
In situations where this is not possible, the physician
should be assisted by a surgical technician and the
professional nurse serving the delivery room must be
competent in resuscitation of the newborn.
•• High-Risk Delivery
All deliveries of high-risk mothers should have an
obstetrician and pediatrician in attendance.
• Other Personnel
•• Nursing
The Level III unit must have a highly trained Supervisor of
Nursing Services for inpatient nursing services. She
should have advanced training in obstetrics and newborn
nursing.
•• Social Worker
One or more medical social workers will be required in
Level III units.
•• Nutritionists
Nutritionists with special knowledge of prenatal dietary
management must be available.
155﻿New Jersey, State of, Department of Health
1978 Standards and General Criteria for the Planning, Certification of Need
and Designation of Perinatal Services. Trenton, N.J.
GENERAL ABSTRACT:
The New Jersey Department of Health adopted standards and criteria
based upon the earlier work of the Maternal and Infant Care Services
Committee of the State Health Planning Council with supplemental standards
developed by the Health Plan Development Services Division. This work
employs the Committee on Perinatal Health model.
ANNOTATIONS:
Level I
•	Functions
A Level I Perinatal Unit provides services primarily for un-
complicated maternity and newborn patients and supportive care
for infants returning from Level II or Level III units. There
shall be capabilities and equipment for: immediate resuscitation,
cesarean sections on short notice (24 hours), on-call anesthesia
service (24 hours), oxygen administration, intravenous therapy
and electronic fetal monitoring.
•	Utilization
At least 1,000 annual deliveries. Waivers granted in cases of
severe geographic isolation or when there is demonstrated
medical necessity created by accessibility barriers for certain
population groups. A hospital may not have fewer than 500
deliveries per year.
•	Size
No neonatal unit size is stated. The minimum size of a Level I
obstetric unit is 20 beds, waived in the above-mentioned cases
or where cost-effectiveness considerations indicate otherwise.
In no case should the unit be less than 10 beds.
•	Physical Facilities
Neonatal units shall contain a normal newborn nursery, an observa-
tion area and a resuscitation area where an infant can be
immediately treated.
•	Staff
The neonatal area shall have:
156﻿•• A Board-eligible or Certified pediatrician on active staff
and in charge of the service.
•• A Board-eligible or Certified pediatrician on call on a 24-
hour basis.
•• A Clinical Nurse Coordinator, Supervisor or Head Nurse who
has completed an approved continuing education course in
neonatal nursing and eligible for ANA Board Certification
in neonatal nursing.
•• A nursing service staffing ratio of one R.N. per shift with
one additional member of the nursing staff for each six-
eight infants.
•• Trained personnel for infant resuscitation, on a 24-hour
basis.
Level II
•	Functions
In addition to Level I functions, a Level II Perinatal Unit
provides intermediate care services for the majority of women
and newborns with complications. A unit must be able to provide
care for newborn respiratory problems up to the point when pro-
longed ventilatory assistance is required of newborns with other
complications, up to the point at which transfer to a Level III
facility is necessary.
There shall be capabilities for short-term assisted respiratory
therapy to enable stabilization and preparation for transfer,
where indicated, including:	continuous monitoring of inspired
oxygen, continuous monitoring of body core temperature and altera-
tion of environmental temperature, control of intravenous therapy
with infusion pumps, provide assisted ventilation on a short-term
basis with bag and mask and continuously monitor cardio-respiratory
function. There shall also be resuscitation equipment (including
laryngoscopes, endotracheal tubes, a resuscitation table with
temperature control and infusion equipment), provision of 24-hour
clinical laboratory services (including, but not limited to,
microbilirubin, blood sugars, electrolytes, blood gases and
calcium) and 24-hour radiology and blood banking services for the
newborn.
•	Utilization
A minimum of 1,500 annual deliveries. Waivers granted in cases
noted under Level I. A hospital may not have fewer than 500
deliveries per year.
157﻿• Size
No neonatal unit size is stated. The minimum size of a Level
II obstetric unit is 20 beds, waived in cases discussed under
Level I.
•	Physical Facilities
Neonatal units shall contain a normal newborn nursery, an
observation area, an intermediate care nursery and an area
for the immediate resuscitation of the newborn.
•	Staff
The neonatal area shall have:
••	A Board-eligible or Certified pediatrician on active staff and in charge of the service.
, M	A Board-eligible or Certified pediatrician on call on a 24-hour basis.
••	A Clinical Nurse Coordinator, Supervisor or Head Nurse who has completed an approved continuing education course in neonatal nursing and eligible for ANA Board Certification in neonatal nursing.
••	A physician or R.N. in-house 24 hours who is qualified to perform ongoing assessment and emergency treatment including, but not restricted to, endotracheal intubation, umbilical catheterization, and biochemical resuscitation of a distressed neonate.
••	Physician anesthesiologist coverage on call 24-hours.
•• ••	All registered nurses shall have completed an approved continuing education course in neonatal nursing within one year of designation as a Level II unit. A nurse staffing pattern for the Normal Nursery of one R.N. per shift with one additional member of the nursing staff for each six-eight infants.
• •	A nurse staffing pattern for the Intermediate Care Nursery of at least one R.N. per shift with a ratio of one nurse for each three infants.
Level III	
• Function
In addition to care capabilities for Level I and II patients, the
158﻿Regional Perinatal Center must have: a designated NICU 24-hour
fetal scalp and neonatal arterial blood gas analysis; ultrasonic
(A and B scan) and radio isotope capabilities; continuous
neonatal blood pressure, heart rate and respiration monitoring;
the capability to maintain respiration on a long-term basis;
the capability to regulate temperature and monitor oxygen flow,
a permanent monitor in the delivery/cesarean section room for
monitoring capability until delivery and genetic counseling
services.
The Level III center's regional responsibilities include establish-
ing a regional laboratory for estriol determinations and
amniotic fluid analysis with 24-hour or same day service, data
collection and analysis, continuing education, and developing
a well-defined regional system of communication, consultation
and transport between all three levels of care.
Children's hospitals, or NICUs in hospitals with no or small
obstetric services, are desirable as referral centers only when
they have the capability to provide certain highly subspecialized
care, otherwise unavailable.
•	Minimum Population Base
The service area shall have a minimum of 10,000 deliveries per
year including those of the applicant. No waivers are possible.
•	Utilization
At least 2,000 deliveries annually. No waivers are possible.
•	Size
No neonatal unit size is stated. The minimum size of a Level III
obstetric unit is 20 beds, waived in cases discussed under Level
I.	No waiver conditions are stated.
•	Physical Facilities
In addition to the Level II unit requirements, a designated NICU.
•	Staff
The NICU shall have the same requirements as stated under Level
II.
159﻿New Mexico Health Systems Agency.
1976 Toward Improving the Outcome of Pregnancy—Perinatal Care in
New Mexico. Sante Fe, N.M.
GENERAL ABSTRACT:
The New Mexico HSA accepts the philosophy set forth by the National
Committee on Perinatal Health.
ANNOTATIONS:
The following are New Mexico's guidelines for referral from a Level
I Unit to a Level II or Level III newborn intensive care unit:
Previous History
Consultation is recommended for the following, determined by history:
•	Two or more previous premature labors or history of low birth-
weight infants (less than 2,500 grams).
•	Excessively large previous infants (greater than 4,000 grams).
•	Previous Cesarean section.
•	Previous significant dystocia.
•	Two or more previous abortions.
•	Previous stillbirth or neonatal loss.
•	Suspected previous incompetent cervix.
•	Medical indication for termination in previous pregnancy.
•	Previously diagnosed abnormalities of genital tract.
•	Previous history of need for special neonatal care.
•	Previous infant with a known or suspected genetic or familiar
disorder.
•	Previous severe emotional problems associated with previous
pregnancy or delivery.
Early Pregnancy
Consultation is recommended in early pregnancy for:
•	Maternal diabetes mellitus.
160﻿•	Patient of less than 15 or more than 40 years.
•	Psychiatric disorder.
•	Marked nutritional abnormality (obesity, abnormal stature,
low weight for height, etc.).
•	Malignancy.
•	Unresponding urinary tract infection.
•	Suspected ectopic pregnancy.
•	Suspected missed abortion.
•	Severe hyperemesis.
•	Exposure to teratogens (radiation, infection, chemicals).
•	Positive serologic test for syphilis.
•	Pregnancies complicated by medical disease (endocrine, renal,
cardiac, hypertensive, etc.).
•	Anemia not responsive to iron therapy.
•	Rh isoimmunization.
Late Pregnancy
Consultation is recommended in late pregnancy for:
•	Vaginal bleeding.
•	Toxemia.
•	Hydramnios or oligo-hydramnios.
•	Antepartum fetal death.
•	Thromboembolic disease.
Newborn
The following conditions are recommended as the basis for seeking
consultation and, when necessary, arranging transport of a newborn from
a Level I Unit to a Level III Unit. A coordinated program for continued
participation of the referring physician in the care of the patient is
essential.
•	Gestation of 35 weeks or less; birthweight less than 2,000
grams.
161﻿•	Respiratory distress and metabolic acidosis persisting after
2 hours of age or requiring ambient oxygen in excess of
40% after 2 hours of age.
•	Infant of a diabetic mother.
•	Neonatal seizures.
•	Suspected neonatal sepsis and/or meningitis.
•	Congenital anomalies requiring observation for neonatal
surgery.
•	Meconium aspiration.
New York, State of, Department of Health
1978 Memorandum—Neonatal Centers for "High Risk" Babies.
Albany, N.Y.
GENERAL ABSTRACT:
This is a summary of a research investigation into the basic question
of whether or not the high risk pregnancy is being identified at the
community hospital level (Level I) and pre-booked for delivery at the
larger hospitals. The data show that most high risk infants are being
delivered at community hospitals and are not pre-booked.
ANNOTATIONS:
The following recommendations are derived from this study:
•	Since postpartum neonate transfers carry added risks for the
infant compared to pre-booking, we should as a minimum
establish prenatal consultative, monitoring, and pre-booking
standards for all hospitals of all sizes to eliminate duplica-
tion of high cost services in the larger hospitals and stimulate
the smaller more numerous maternity services to detect and
pre-book.
•	All high risk pregnancies should be identified during the pre-
natal period or the intrapartum period by those larger hospitals
in a single city and transported to the single hospital with
the intensive neonatal care resource.
•	A general principle should be that it is much better to pre-book
the high risk mothers at the NICU regional center than it is to
transfer their high risk babies after birth to the neonatology
162﻿centers. This principle is set forth in the national committee
report "Improving the Outcome of Pregnancy."
• No woman should be delivering a baby weighing less than 2,500
grams in a Level I hospital in the central city if such systems
are installed and transfers during the prenatal and the intra-
partum period are planned for in advance.
New York, State of, Department of Health
1977 Memorandum — Measuring Value Received for High-Priced Care of Premature
Infants. Albany, N.Y.
GENERAL ABSTRACT:
This report raises some serious health systems management questions as
to why three times as much is paid for the care of a "premature" infant
(birthweight under 2,500 grams) as for other infants. Moreover, an analysis
of the survivorship of 29,781 white "premature" infants for the years
1973-76 in upstate New York shows that the infants born at "approved"
premature care centers did not show any evidence of benefit that was
commensurate with the three-fold price paid.
ANNOTATIONS:
This study is aimed at two conclusions:	(1) The health systems manage-
ment process for designation of approved centers and pricing practices for
care of premature infants are uncoordinated.; and (2) the inputs required for
premature care center designation are probably irrelevant or at least
unassociated with the output results sought — improved life expectancy
for "premature infants."
The study suggests that the results of this analysis create a need to
conjecture as to what might be producing this paradox of highest risk
infants doing better at unapproved centers and lowest risk at approved
centers. It should be recommended that careful evaluations of our pricing
practices for care of premature infants be conducted.
North Central Health Systems Agency
1978 Specialized Services Review Criteria: Neonatal Special Care Centers (Draft).
Atlanta, GA.
GENERAL ABSTRACT:
The North Central Georgia HSA had defined financial standards and
criteria which should be helpful for monitoring the financial conditions of
163﻿NICUs as well as contribute to efforts of rational and feasible construction
and development of systems.
ANNOTATIONS:
Statements regarding financial feasibility:
•	There would be a two year set of pro forma financial statements
(including revenue and expense) and their statements would
indicate financial feasibility.
•	Rates for the service would be comparable within the region for
similar services and would relate to cost and operating expenses.
•	The service would be cost effective when compared to existing
or other alternatives.
Financial Information to be Included:
•	Estimated Annual Service Expenditure
•	Estimated Annual Capital Expenditure
•	Estimated Annual Cost to the Community
•	Equipment Cost per Unit of Service
•	Construction Cost per Unit of Service
• Other Costs per Unit of Service
• Total Costs per Unit of Service
•	Percent Occupancy
•	Return on Investment
•	Return on Revenue
Ohio Department of Health, Division of Maternal and Child Health,
Bureau of Maternal and Child Health.
1977 State Perinatal Guidelines.
GENERAL ABSTRACT:
The Ohio Perinatal Consultant's Group has accepted, in principle, the
national recommendations for the Regional Development of Maternal and
Perinatal Health Services (Toward Improving the Outcome of Pregnancy),
developed by the Committee on Perinatal Health.
ANNOTATIONS:
Communication in the Regionalized System
Communication and coordination can be strengthened through the pro-
vision of regular and frequent opportunities for education within the region.
164﻿Workshops, seminars and regional conferences are excellent ways to establish,
improve, and secure lines of communication.
Emergency Transportation
An effective emergency transportation service for high-risk maternal
and newborn patients should:
•	Make available emergency transportation to all sick mothers
and newborns from home to hospital or from one center to
another within the system.
•	Make that transportation as efficient, rapid, and safe for
the patient as possible.
•	Provide care to a patient during that transportation by a
specially trained team of personnel skilled in maternal
and newborn emergency medical care.
•	Provide proper preparation for any patient who needs
transportation.
Education, Regional Level Objectives
With each region, regional perinatal education coordinators will:
•	Identify the scope of the provider target group by number
and specific type of personnel, e.g., pediatrician, obste-
trician, etc.
•	Determine the needs/desires of each identified group. This
determination can be made through a survey of a representative
sample of each group and/or through consultation with profes-
sional organizations and through the Neonatal Log.
•	Develop educational program and curricula tailored to suit
the educational requirements of each group.
•	Provide assistance to professional groups in the development,
organization, and presentation of perinatal education activities.
•	Assist hospitals and ambulatory care facilities in organizing
developing, and implementing in-service education programs.
•	Identify existent perinatal education resources.
•	Develop and assist in developing printed materials and other
educational aids.
•	Work toward the establishment of a regional learning resource
center to function as a source for perinatal education
materials.
165﻿•	Assist in promoting the use of the Neonatal Log as a mechanism
for hospitals to identify areas suitable for educational
programs.
•	Develop an education component designed to foster the team
concept of patient care, i.e., understanding and coordination
of the capabilities of each member of the perinatal care team
by all team members.
Ross Laboratories
1977 Planning and Design for Perinatal and Pediatric Facilities.
Columbus, Ohio:	Ross Laboratories
GENERAL ABSTRACT:
This report, developed for physicians, administrators and architects,
is an excellent discussion of the planning considerations and design
requirements of perinatal, obstetric and pediatric units.
Chapters are devoted to: perinatal care planning and regionalization,
obstetrical facilities, Neonatal Special Care Units (annotated below),
normal newborn nurseries, pediatric units (including pediatric ICUs),
psychological aspects of care for pediatric patients and pediatric ambula-
tory care planning. Sample floor plans, functional programs and equipment
specifications are included.
ANNOTATIONS:
Levels of Care in Neonatal Special Care Units
Maximal care is for neonates requiring nearly constant observation,
highly specialized support services and frequent medical interventions.
Intermediate care infants do not need intensive management but do require
frequent observation and/or special interventions. Minimal care is
convalescent care.
Population in Need
The American Academy of Pediatrics estimate is referenced:	10% of
all live births require intermediate or maximal care. Two to four percent
of all newborns will require maximal care although socio-economic differences
can raise this figure as high as eight percent. Six to eight percent
of all newborns will require intermediate care.
Length of Stay
For planning purposes, a base of 20 day ALOS for maximal care infants
and 12 days for intermediate care are suggested. The authors note that
166﻿actual lengths of stay in most Special Care Nurseries are lower.
Bed Ratios
Depending upon the adequacy of local Level II perinatal care centers,
the ratio of intermediate and minimal to maximal care beds could range
from 4:1 to 1:1 within a Level III facility.
Cost
In 1974 the annual cost of operating a 32 bed regional Special Care
Nursery was estimated at $1.5 million (including equipment, personnel,
supplies overhead, transport and education). Using a 10% annual inflation
rate yields a $2 million figure for 1977 average daily room charges are
over $250. The initial cost of one maximal care bed is over $15,000.
Maintenance and replacement of equipment costs between $2,000 and $3,000
annually.
Minimum Volume
A minimum of 2,000 annual births (or combined in-house births and
transport referrals) is necessary to justify at least two physicians
trained in high-risk neonatal care.
Personnel Requirements
A minimum of two neonatologists is suggested. Ideal nurse/patient
ratios are 1:1 or 1:2 for maximal care infants, 1:2 to 1:4 for intermediate
care and 1:4 to 1:6 in minimal care. Paramedical personnel required for
support include inhalation therapists, laboratory technicians, biomedical
engineers, social service representatives and nursing subspecialists.
Functional Planning and Programming
The objective of a functional program is to provide the architect with
information which enables him to prepare high quality, detailed plans. A
functional program identifies each space required for the medical program
and to specify size and dimensions of these spaces.
An example of a detailed functional program developed for a 50 bed
(10 maximal, 20 intermediate, 20 minimal) unit in a regional referral center
is included (10,000 births in region). The authors stress that each proj-
ect is unique and space planning must be individualized.
Equipment
There are four major categories of equipment. Environment control
includes incubators, open care platforms and radiant heaters. Life support's
respiratory equipment, positive or negative pressure respirators, ventila-
tion bags and oxygen mist hoods, along with oxygen and compressed air
sources. The diagnostic category equipment includes patient monitors,
portable X-ray machines, electrocardiograph and electroencephalograph
167﻿machines. Treatment equipment includes infusion pumps, dupcounting devices,
bilirubin lights, phototherapy lights, heaters, heat lamos and suction
apparatus.
Texas, State of, Department of Health
1976 Guidelines for the Texas Regional Perinatal Care System. Austin, TX.
GENERAL ABSTRACT:
The Task Force decided that workable guidelines could be developed for
regional perinatal care in Texas by taking the recommendations of the
National Committee on Perinatal Health for the development of a regionalized
system for perinatal care and modifying them as necessary for use in
Texas.
ANNOTATIONS:
Nursing Personnel, Level III
The overall responsibility for the perinatal nursing activities should
be vested in a director of perinatal nursing services or in two directors,
one of obstetric nursing services and one of neonatal nursing services,
who coordinate their activities. These directors should have experience
and training in the nursing care of patients at high-risk including, where
possible, an advanced degree in obstetric-newborn nursing.
Other Personnel
•	Nutritionists with social knowledge of perinatal dietary-
management must be available.
•	One or more medical social workers will be required.
•	An administrative coordinator should be provided. Clerks
are essential for efficient operation of the unit.
Intensive Care Nursery Personnel
•	There should be a Head Nurse with experience and special training,
and where possible an advanced degree, in the care of critically
ill newborns and in relating to and communicating with parents.,
•	All nurses working in the Intensive Care Nursery should be highly
skilled in the observation and nursing treatment of critically
ill infants. The nurses must be highly skilled in cardiorespiratory
monitoring, techniques and equipment for assisting ventilation
and for administering intravenous fluids, preoperative and post-
168﻿operative care of newborns, and emergency treatment of conditions
such as apnea and seizures.
•	These nurses should be skilled in infant transport and may be
part of the transport team bringing infants to the Level III
Unit. These nurses usually will have additional advanced
didactic and practice training.
•	A recommended minimum for nursing staffing pattern on all shift
assignments in an intensive care nursery is one professional
registered nurse for every 1 or 2 critically ill infants. In
addition, two qualified nurses are recommended for every 5
moderately ill infants who are early in the recovery phase of
their illnesses.
Communication, Education, Evaluation
•	The Level III Units must be responsible for ensuring that programs
of continuing in-service education are developed and implemented
throughout the Regional Perinatal System. The Level III Units
should have an active, formalized, ongoing relationship with the
Level I and Level II Units of the region for the support of in-
service education.
•	The Level III Unit must assume the responsibility for ensuring
that systems of communication and transport are devised and
implemented but the other Units must be fully participatory in
formulating and agreeing with the Regional Perinatal Plan.
•	The Level III Units must provide the leadership in planning the
mechanisms for reporting results and in collecting, sharing and
analyzing the data to assess accomplishments and possible
deficiencies of function in the Regional System. The other Units
in the area must participate in planning these mechanisms and
in collecting, reporting, sharing and analyzing the data.
West Virginia Planning Committee for Perinatal Health
1976 A Plan to Improve Perinatal Health Care in West Virginia.
Charleston, W. VA.
GENERAL ABSTRACT:
In order to develop a system of care which will enable every mother
and newborn to receive optimal care in West Virginia, the committee recommended
adherence, whenever possible, to standards established by the professional
specialty societies which are regularly updated, published and circulated
among hospitals and physicians. According to the committee, the standards
for hospitals policies, services, personnel, equipment, facilities and
169﻿record keeping which are outlined in this document should be emphasized
in all hospitals providing maternity and newborn care.
ANNOTATIONS:
General Policies
•	An active program for identification and appropriate
consultation for the high risk maternity patient at the
earliest possible date in the pregnancy.
•	Provisions for a copy of the prenatal record to be avail-
able in the labor and delivery unit 4 weeks prior to
the established expected date of delivery.
•	A protocol of observations and procedures to be carried
out during the postpartum period.
Level I
It should be the intent of planners and administrators to encourage
both improvements in perinatal care in these hospitals and early consulta-
tion for mothers and prompt referral of newborns whenever appropriate.
Level II
The following services should make up the Level II newborn hospital.
•	Prenatal, intrapartum and postpartum care of uncomplicated
patients consistent with standards of the American College
of Obstetricians and Gynecologists; including capability
for fetal monitoring.
•	Specialized prenatal services for the mother and fetus identified
as high risk, including laboratory facilities for the determina-
tion of urinary estriols and amniotic fluid analysis. B-scan
ultrasound capability is highly desirable.
•	An intensive care area.
•	Caesarian section capability within 15 minutes.
•	Tubal litigation for purposes of sterilization.
•	A specifically designated special care nursery for the care of
the neonate with complications. This nursery must be prepared
to accept long-term patients transferred from other hospitals
because of prematurity, sepsis, etc.
•	Capability for continuous electronic monitoring of heart rate
and respiratory rate of infants in the special care nursery.
170﻿•	Exchange transfusion and phototherapy capability.
•	Capability for controlled full range oxygen administration.
•	Capability for emergency ventilatory assistance until transfer
to Level III can be effected.
•	Twenty-four hour inhalation therapy services on call.
•	Transport assisted by professional personnel on call.
Level III
The following services should be provided by the Level III tertiary
care newborn center.
•	Be capable of providing advanced life support in critical
care units specially designed for obstetrical and newborn
patients respectively.
•	Provide 24-hour consultation service for the region.
•	Coordinate and provide regional educational programs for
physicians and nurses.
•	Provide a regional transport system for critically ill
patients within the region, and coordinate transport
with Level II hospitals.
•	Data collection for evaluation and analysis.
•	Conduct high risk infant follow-up clinics.
•	Conduct research in perinatal medicine.
Transportation
Approximately 90% of the 129 patients brought to the West Virginia
University Medical Center by the neonatal transport team in 1975 required
some form of supportive care and/or medication during transport. It is
highly significant that prior to the formation of this transport service
one-half of all admissions to the neonatal ICU died, whereas mortality
dropped to 27% after its inception. There is overwhelming data from other
countries, the United States and West Virginia indicating the marked
decrease in neonatal mortality and morbidity in areas where such services
are available.
Communications
Telephone copiers are commercially available to transmit copy through
the wire. Such an instrument with terminals stationed at optimal points
throughout the region, could be used to transmit chart and test results
171﻿almost immediately to referral centers for consultation. The following
recommendations pertain to necessary activities prior to the selection
of any major technological undertaking in the area of communication
concerning patient status.
•	A minimum data set of patient information needs to
be agreed upon.
•	All data collection should pertain to private and
service patients alike.
•	All data should be computerizable and a system of
confidentiality should be built in as to the use
of these data.
•	The transmittal of information must be interactive,
or two-way.
•	A technical review of available systems should be
undertaken with the intent of selecting a uniform
transmittal system for the entire state.
•	Any system must be acceptable to the Emergency Medical
Service System.
Western Center for Health Planning
1976 Certificate of Need Planning and Review Criteria, Intensive Care
Newborn Nursery. San Francisco, CA.
GENERAL ABSTRACT:
Each area health planning agency is encouraged to develop a plan for
optimal neonatal care of high risk infants which is integrated with a
comprehensive plan for the delivery of maternal and child health care.
The following annotations are some criteria which are to be used for
evaluation and review.
ANNOTATIONS:
Standard
For any project to be located within a standard metropolitan statistical
area of 500,000 or greater:
• Travel time for surface infant transport should not
exceed one hour.
172﻿•	For planning purposes a base of 8000 deliveries per
year is required to support a tertiary care newborn
intensive care nursery.
•	For planning purposes, a base of 2000 deliveries per
year in the area is required to support a secondary
(intermediate care) newborn ICU.
Standard
For any project to be located within an SMSA of less than 500,000
or in an area not included in an SMSA:
•	Travel time for surface infant transport should not
exceed 2 hours. Helicopter and/or fixed wing trans-
port should be arranged for service areas beyond
2 hour driving time.
•	For planning purposes, a base of 1500 deliveries
per year in the area is required to support a NICU.
Average Length of Stay:
The average length of stay is 28-42 days.
Estimation of Demand for NICU:
Demand for NICU services may be estimated by projecting past utiliza-
tion against the future population requiring services.
Estimated No. of _ expected no. of births, 1978 X NICU patient days
NICU beds needed	live births, 1975	______1975 _______
in 1978	365 X .85
The previous formula contains these assumptions:
•	The incidence of conditions requiring intensive care
will remain constant.
•	The average length of stay of each neonate will remain
constant.
•	85% occupancy is the desired efficiency level.
Capacity of NICU Facilities
•	Standard
For any project to be located in a SMSA of 500,000 or
greater:
•• Eight beds for infants requiring 1:1 to 1:2 nurse
to patient ratio.
173﻿• Standard
For any project to be located in a SMSA of less than
500,000 or in an area not included in an SMSA:
•• Two beds for infants requiring 1:1 to 1:2 nurse
to patient ratio and 6-8 additional beds with
1:3 to 1:4 nurse to patient ratio.
Wisconsin, State of, Division of Health Policy and Planning
1976 Planning Standards for Perinatal Care Centers. Madison, WI.
GENERAL ABSTRACT:
These Standards are provided in order to establish an objective basis
for review of proposed plans to construct or expand newborn services
using the regional concept of delivery.
ANNOTATIONS:
Standards
Accessibility-Time/Distance
The location of a perinatal care center shall allow reasonable access
to the perinatal services by patients in the region served. Reasonable
access shall be interpreted as a maximum of two hours of driving time one
way.
Needs Estimation
The five-year projected need for neonatal intensive care unit beds in
a service area shall be computed as follows:
•	Intensive care beds:	.9 x the 5 year projected average
_____annual # of births __________
1,000
•	Intermediate care	1.13 x the intensive care bed needs
beds:
The number of additional neonatal ICU beds needed in a service area
shall be determined by subtracting existing beds from the five year pro-
jected bed need using the above method. In- and out-migrations of
patients should be a factor taken into account when determining appropriate
bed levels.
174﻿Obstetric Facilities at Perinatal Care Centers
In a service area with a single perinatal care center, the high-risk
obstetrics service at the perinatal care center shall have facilities
for high-risk deliveries sufficient to accommodate 5% of the deliveries
in the service area of the center. In addition to high-risk deliveries,
the obstetrics unit at a perinatal care center shall provide services
for normal deliveries which amount to at least another 5% of the
deliveries in the service area of the center.
In a service area where there is more than one perinatal care center,
each shall have obstetrics facilities for high-risk and normal patients
sufficient to accommodate at least 5% of the deliveries in the service
area of the center.
Minimum Delivery Base
In a service area where the five year projected annual number of
births is under 20,000, this projection shall constitute the minimum
delivery base of the perinatal care center in that service area. In a
service area where the five year projected average annual number of births
is 20,000 or more, the minimum delivery base of each perinatal care center
shall be at least 10,000.
Affiliation Agreements
Hospitals proposing to establish a new or to expand an existing
perinatal care center, where the neonatal ICU and high-risk obstetrics
components of the perinatal care center are located in separate institu-
tions, shall (plan for) affiliation agreements between the institutions
which are proposing to establish the center.
The Affiliation Agreement shall include:
•	Provisions for a perinatal care center coordinating board
involving medical and administrative staff from partici-
pating institutions.
•	A description of the perinatal care center system which
clearly distinguishes the respective roles, functions,
and responsibilities of the institutions in this system.
•	Provisions for the effective coordination between the
institutions of the perinatal care center designed to
assure continuity of care to perinatal care center
patients and to prevent the unnecessary duplication of
staff in the perinatal care center.
•	Provision for the effective coordination of perinatal
facilities and equipment between the institutions
designed to minimize the duplication of these facilities
and equipment in the perinatal care center.
175﻿Staffing Standards
Applicants proposing to develop a new or to expand an existing
perinatal care center shall document that they meet or have plans to
meet the following minimum staffing requirements:
• Direction
Nursing and medical care aspects of the perinatal care center
shall be under the combined direction of a professional
team including an obstetrician experienced in high-risk
care, a neonatologist/pediatrician, and a registered nurse
with special training and experience in high-risk perinatal
care.
•	High-Risk Obstetrics Specialist
The high-risk maternal/fetal and intrapartum intensive care
segment of a perinatal care center shall be under the
direction of a high-risk obstetrics specialist. This
obstetrician shall have at least one year of experience
in high-risk maternal intensive care in an organized high-
risk program beyond the standard residency in obstetrics.
Consultation from a high-risk obstetrics specialist shall
be available for all high-risk patients.
•	Neonatologist/Pediatrician
The high-risk fetal/neonatal intensive care segment of the
perinatal care center shall be under the direction of a
neonatologist who is board eligible or board certified in
neonatology, or a pediatrician with at least one year of
experience in an organized high-risk neonatal intensive
care program. Consultation from a high-risk neonatal
specialist shall be available for all high-risk patients.
•	Nursing Care Coordinator - Perinatal
Nursing care aspects of both the obstetric and neonatal
components of the perinatal care center shall be coordinated
by a registered nurse. This nurse shall have obtained
specialized nursing knowledge and skills by successfully
completing an organized educational program in maternal
or neonatal intensive care, and shall have at least one year
of experience in an obstetrics unit associated with a
neonatal intensive care unit, or in a recognized perinatal
care center.
•	Registered Nurses - Staff - Obstetric
A registered nurse shall be available to assist with the
antepartum, intrapartum, and postpartum care of each
176﻿high-risk mother. These RNs shall have obtained specialized
nursing knowledge and skills by successfully completing an
organized educational program in maternal intensive care.
•	Registered Nurses - Staff - Neonatal
Registered nurses shall be on the nursing staff of the neonatal
intensive care units on each shift. At least one RN on each
shift shall have obtained specialized nursing knowledge and
skills by successfully completing an organized educational
program in neonatal intensive care.
•	Nurse/Patient Ratio - Neonatal ICU
The neonatal intensive care units shall maintain a minimum
ratio of one RN for every two intensive care infants. The
intermediate care unit shall be staffed with one RN for every
four patients or one RN and additional nursing personnel at
a minimum ratio of one RN or LPN for every four patients.
•	Anesthesiologist
An anesthesiologist shall be on call 24-hours per day.
Staffing Guidelines
•	Anesthesiologist
Anesthesia services should be directed by an anesthesiologists
with special training in maternal-fetal and neonatal anesthesia.
•	Opthamologist
An optamologist should be available on a routine basis to the
neonatal intensive care unit.
•	Respiratory Therapist
A respiratory therapist should be continuously available to
the neonatal intensive care unit.
•	Social Workers
The application of skills and methods of medical social work to
the specific needs of the high-risk perinatal patients in the
region should be under the direction of a social worker with
a master of social work degree, who should supervise the social
work staff.
•	Support Staff
The following support staff should be available to the perinatal
care center for consultations within a 24-hour period:	Cardiologist
177﻿(pediatric and adult), Pediatric Surgeon, Radiologist,
General Surgeon, Neurosurgeon, Hematologist, Cardiothoracic
Surgeon, Endocrinologist, Internist, Pathologist, Urologist,
Orthopedist, and a Nephrologist (pediatric and adult).
In addition, the following support staff should be available
to the perinatal care center for periodic consultation:
Medical Geneticist, Bio-Medical Engineer, Pediatric Neurologist,
Psychiatrist, and Psychologist.
Staffing Privileges Arrangements
Perinatal care centers should establish or have plans to establish a
policy of granting appropriate staff privileges to qualified physicians
to deliver high-risk services at the perinatal care center under the
supervision of the appropriate members of the medical staff of the insti-
tution.
Physician Consultation Arrangements
There should be a 24-hour telephone consultation service from physician
members and other high-risk unit staff to physicians and other professionals
and hospitals in the service area of the perinatal care center to provide
advice on approaches to patient care management and problems.
Infant Transport
•	Transport team
The treatment team for high-risk neonatal patients shall
include a physician and a nurse who are active in newborn
intensive care at the perinatal care center, and who
have undergone a program of in-service training in the
preparation and transport of high-risk infants. A transport
team as above must be available 24 hours per day.
•	A physician need not accompany an infant who requires special
care but is not in distress at the time of transfer, but at
least one trained individual whose only responsibility shall
be the care of the infant shall accompany the infant.
•	The transport team for high-risk maternal patients shall
include a physician, eligible for certification in obstetrics
and a nurse with basic knowledge of obstetric services. A
transport team as above shall be available 24-hours per day.
•	Transport Vehicle
Transport vehicles used for transport of high-risk infants
should meet the provisions of General Services Administration
EMS Standards "Federal Specifications - Ambulance and Emergency
Medical Care Vehicles" (KKK-A-1822).
178﻿•	The transport vehicle shall be designed for standing room
and shall carry appropriate self-contained equipment for
intensive treatment and life support of newborn infants
and maternal patients while in transit.
•	Equipment of high-risk infant transport should include,
but is not limited to: infant transport incubator; infant
resuscitation; and monitoring equipment for temperature
if infant, heart rate, and environment oxygen.
Facilities
Applicants proposing to develop a new or to expand an existing peri
natal care center shall document that they provide, or have plans to pro
vide the following minimum facilities and services:
•	A perinatal care center shall be based in a single, or for
this purpose, affiliated acute care facilities, with
obstetrics and pediatrics units.
•	The neonatal ICU and high-risk obstetrics components of a
perinatal care center shall be located in close physical
proximity to each other in a single, or for this purpose,
affiliated acute care facilities in the same or in contiguous
communities, to encourage continuity and efficiency of care to
the mother and her newborn, effective intrapartum care of the
fetus, and to foster early infant/mother togetherness.
•	A perinatal care center shall provide family planning services
as part of the perinatal services.
•	Medical services do not need to include abortions, but peri-
natal center staff shall be knowledgeable of, and shall
provide information regarding the availability of abortion
services and problem pregnancy counseling.
Facilities for High-Risk Neonatal Services Shall Provide For:
•	Neonatal intensive care area equipped for continuous
intensive monitoring and life support therapy.
•	Intermediate care area or areas for infants po longer re-
quiring continuous monitoring, but still needing frequent
and special observation and treatment. This area may also
admit infants directly from delivery requiring intermediate
level care initially.
•	Capabilities continuously available in-house for: micro
lab methods; blood gas and PH studies; hematology; in-unit
radiology; blood bank, with blood and fibrinogen, platelets,
and fresh blood; and, other capabilities as appropriate.
179﻿• Follow-up clinic for the continued evaluation and care of the
infant following discharge, including education for the
parents in the proper care and management of the infant.
180﻿BIBLIOGRAPHY OF CLINICAL LITERATURE
Alden, Errol R. , M.D., Ted Mandelkorn, M.D., David E. Woodrum, M.D., Richard
P. Wennberg, M.D., Colby R. Parks, M.D. and W. Alan Hodson, M.D.
1976 "Morbidity and Mortality of Infants Weighing less than 1,000 grains in an
Intensive Care Nursery". From the Division of Neonatal Biology, Department
of Pediatrics, School of Medicine, University of Washington, Seattle,
Washington.
American College of Obstetricians and Gynecologists
1978 Health Care for Mothers and Infants in Rural and Isolated Areas. Chicago,
IL:	American College of Obstetricians and Gynecologists.
American College of Obstetricians and Gynecologists
1973 Memorandum to the Committee on Health Care Delivery from Ervin E. Nichols, M.D.
FACOG, Associate Director; Maternity and Neonatal Services and Facilities
Guidelines for Levels of Complexities of Care. Chicago, IL:	American
College of Obstetricians and Gynecologists.
American College of Obstetricians and Gynecologists
1971 National Needs in Obstetrics and Gynecology. Chicago, IL:	American
College of Obstetricians and Gynecologists.
Berger, Gary S., M.D., Dennis B. Gillings, Ph.D., and Earl Siegel, M.D.
1976	"The Evaluation of Regionalized Perinatal Health Care Programs". American
Journal of Obstetrics and Gynecology 125:	924-932.
Berger, Gary S., M.D., J. Richard Udry, Ph.D., and Charles H. Hendricks, M.D.
1975	"Regionalized Perinatal Care:	An Estimate of Its Potential Effect on Racial
Differences in Perinatal Mortality in North Carolina". North Carolina
Medical Journal 36:	476-479.
Blake, Anthea, Ann Stewart and Diane Turcan
1977	"Perinatal Intensive Care". Journal of Psychosomatic Research 21:	261-272.
Boehm, John J.
1976	"Identification of the High Risk Infant". Northwestern University Medical
School — Division of Pediatrics Prentice Woman's Hospital and Maternity
Center. Chicago, Illinois.
Bryant, Sandra, R.N.
1976 "Nursing Aspects and Organization for Perinatal Care". Clinics in Perinatology
Organization for Perinatal Care 3:	493-496. Guest Editor:	Louis Gluck, M.D.,
Philadelphia:	W.B. Saunders Company.
Butterfield, Joseph L., M.D.
1976	"Newborn Country USA". Clinics in Perinatology:	Organization for Perinatal
Care 3:	281-295. Guest Editor:	Louis Gluck, M.D., Philadelphia:	W.B.
Saunders Company.
Butterfield, Joseph L., M.D.
1977	"Organization of Regional Perinatal Programs". Seminars in Perinatology:
Regionalization of Perinatal Care 1:	217-233. Guest Editor:	Leo Stern, M.D.
New York:	Grune and Stratton.
181﻿1973
1975
1977
1978
1964
1976
1976
1977
1969
1975
1975
Bibliography of Clinical Literature -
(Continued)
Carrier, Charles, M.D., et al.
Perinatal Intensive Care after Integration of Obstetrical Services in Quebec:
A Policy Statement of the Quebec Perinatal Committee. Quebec, Canada.
Available from:	Quebec Perinatal Committee, Ministry of Social Affairs
Province of Quebec, 1005 Chemin Ste-Foy, Quebec, Canada, G1S 4N4.
Carrier, Charles, M.D.
Report of the Quebec Perinatal Committee for Perinatal Mortality for the Year
1970 (First Part): Quebec Canada. Available from: Quebec Perinatal Committee,
Ministry of Social Affairs Province of Quebec, 1005 Chemin Ste-Foy, Quebec,
Canada, G1S 4N4.
Chase, Helen
"Infant Mortality and Its Concomitants, 1960-1972". Medical Care 15:	662-674.
Chez, Ronald A., M.D., Harold E. Fox, M.D., John C. Hobbins, M.D., George J.
Peckham, M.D., E. Derek Peske, M.D. and Mary Halderman
"Monitor Every Patient in Labor?" Patient Care 12:	136-163.
Clifford, Stewart, M.D.
"Medical Progress, High-Risk Pregnancy. Prevention of Prematurity and the
Sine Qua Non for Reduction of Mental Retardation and Other Neurologic
Disorders". The New England Journal of Medicine 271:	243-249.
Committee on Perinatal Health
Toward Improving the Outcome of Pregnancy. White Plains, New York.
The National Foundation March of Dimes.
Donahue, Charles, Jr., M.A., Lou Freedman, Robert Danley, Ph.D., Ann Pettigrew,
M.D., Nancy Shaughnessy, Sarah Fogerty, R.N.
"The Use of Vital Events as a Data Source in the Planning of Maternity and
Newborn Services". Paper presented at the Annual Meeting of the American
Public Health Association, October 21, 1976, Miami, Florida.
Duxbury, Mitzi L., M.D.
"Personnel and Staffing Needs for Perinatal Programs". Seminars in Perinatology
Regionalization of Perinatal Care 1:	267-278. Guest Editor:	Leo Stern, M.D.
New York:	Grune and Stratton.
Effer, S.B., M.D.
"Management of High-Risk Pregnancy:	Report of a Combined Obstetrical and
Neonatal Intensive Care Unit". Canadian Medical Association Journal 101:
55-63.
Egan, Edmund A., Richard Boothby and E. Charlton Prather
"Florida Regional Neonatal Intensive Care Program — Impact on Mental
Retardation". Journal of the Florida Medical Association 62:	36-39.
Ellis, William C., M.D.
"How Maternal and Infant Care Has Been Improved in New Jersey". Contemporary
OB/GYN 6:	73-77.
182﻿1976
1977
1976
1977
1970
1978
1973
1975
1977
1975
1977
Bibliography of Clinical Literature -
(Continued)
Ferrara, Angelo, M.D., and Lucille Perrotta, M.D.
"Infant Transport Services:	An Overview". Pediatric Annals 5:	25-45.
Ferrara, Angelo, M.D., Ph.D.
"Evaluation of Efficacy of Regional Perinatal Programs". Seminars in
Perinatology:	Regionalization of Perinatal Care 1:	303-308. Guest Editor:
Leo Stern, M.D., New York:	Grune and Stratton.
Fitzhardinge, Pamela M., M.D.
"Follow-up Studies on the Low Birth Weight Infant". Clinics in Perinatology:
Organization for Perinatal Care 3:	503-516. Guest Editor:	Louis Gluck, M.D.
Philadelphia:	W.B. Saunders Company.
Giles, Harlan R., M.D., Jerry Isaman, M.A., William J. Moore, M.D. and
C.D. Christian, M.D.
"The Arizona High-Risk Maternal Transport System:	An Initial View".
American Journal of Obstetrics and Gynecology 128:	400-407.
Gluck, Louis, M.D.
"Design of a Perinatal Center". Pediatrics Clinics of North America 17:
777-791.
Goff, Robert, M.D.
"The Bedside Microcomputer in the Intensive Care Nursery". Interface
Age 3:	65-67.
Goodwin, James and Paul R. Swyer, M.D.
Regional Services in Reproductive Medicine. The Report of the Joint
Committee of the Society of Obstetricians and Gynecologists of Canada
and the Canadian Pediatric Society on the Regionalization of Reproductive
Care in Canada. Sherbrooke, Quebec.
Gorwitz, Kurt and Donald Smith, M.D.
"Some Implications of Declining Birth Rates for Pediatrics". Pediatrics
56:	592-597.
Haasis, Patricia, R.N. and Jay P. Goldsmith, M.D.
"The First Year's Experience of a Neonatal Intensive Care Unit". The
Journal of the Louisiana State Medical Society 129:	247-251.
Hawes, Warren E., M.D., M.P.H.
"A Survey of Newborn Intensive Care in California". The Western Journal
of Medicine 23.
Hein, Herman A., M.D.
"Regionalization of Perinatal Care in Rural Areas Based on the Iowa
Experience". Seminars in Perinatology:	Regionalization of Perinatal
Care 1:	241-254. Guest Editor:	Leo Stern, M.D. New York:	Grune and
Stratton.
183﻿Bibliography of Clinical Literature -
(Continued)
Hobel, Calvin J., M.D., Marcia A. Hyvarinen, M.D., Donald M. Okada, M.D.
and William Oh, M.D.
1973	"Perinatal and Intrapartum High-Risk Screening:	Prediction of the High-
Risk Neonate". American Journal of Obstetrics and Gynecology 117:	1-9.
Honeyfield, Peter R.
1976	"Evaluation, Mortality and Morbidity Indices and Compliance in the Care
of the High-Risk Newborn Infant". Denver, CO:	Department of Pediatrics,
University of Colorado.
Johnson, Dorothy K., Dr.P.H. and Elizabeth A. Hefferin, Dr.P.H.
1977	"Perinatal Outcomes Among High-Risk Patients in Two Prenatal Care Programs".
Inquiry 14:	293-302.
Johnson, John D., Natalie C. Malachowski, Rose Grobstein, Doris Welsh,
William J.R. Daily and Philip Sunshine
1974	"Prognosis of Children Surviving with the Aid of Mechanical Ventilation
in the Newborn Period". Fetal and Neonatal Medicine 84:	272-276.
Jonsen, A.R. , Ph.D, R.H. Phibbs, M.D., W.H. Tooley, M.D. and
M.J. Garland, Ph.D.
1975	"Critical Issues in Newborn Intensive Care:	A Conference Report and
Policy Proposal". Pediatrics 55:	756-768.
Jonsen, Albert R., Ph.D. and George Lister
1978	"Newborn Intensive Care:	The Ethical Problems". Hastings Center Report.
Kanto, William P., Jr., M.D. and Alex F. Robertson, III, M.D.
1977	"One Year's Experience of a Regional Neonatal Intensive Care Unit".
Journal of the Medical Association of Georgia 66:	616-623.
Knuppel, Robert A., M.D., Curtis L. Centrulo, M.D., Charles J. Ingardia, M.D.,
Kenneth A. Kappy, M.D., Joseph L. Kennedy, M.D., Marguerite J. Herschel, M.D.,
Gretchen Aumann, R.N., Marian Lake, R.N. and Anthony J. Sharra, Ph.D.
1979	"Experience of a Massachusetts Perinatal Center". New England Journal of
Medicine 300:	560-562.
Koops, Beverly L., M.D., John T. McCarthy, M.D. and L. Joseph Butterfield, M.D.
1978	"Who Pays the Bill for Neonatal Intensive Care?:	Part III:	Outcome"
(mimeo). Reprints:	L. Joseph Butterfield, M.D., The Children's Hospital,
Department of Perinatology, 1056 East Nineteenth Ave., Denver, CO 80218.
Leake, Rosemary, M.D., Alexandra Loew and William Oh, M.D.
1976	"Retransfer of Convalescent Intensive Care to Community Intermediate
Care Nurseries". Clinical Pediatrics 15:	293-294.
Mazzi, Eduardo, M.D., Ronald Gutberlet, M.D. and Jack A. Phillips
1977	"The Maryland State Intensive Care Neonatal Program, Transport System".
Maryland State Medical Journal 26:	86-87.
184﻿1978
1978
1974
1971
1974
1978
1967
1975
1974
1972
1977
Bibliography of Clinical Literature -
(Continued)
McCarthy, John T., M.D., Beverly L. Koops, M.D., Peter R. Honeyfield, M.D.
and L. Joseph Butterfield, M.D.
"Who Pays the Bill for Neonatal Intensive Care. Part I:	Transport"
(mimeo). Reprints:	L. Joseph Butterfield, M.D., The Children's Hospital,
Department of Perinatology, 1056 East Nineteenth Ave., Denver, CO 80218.
McCarthy, John T., M.D., Beverly L. Koops, M.D. and L. Joseph Butterfield, M.D
"Who Pays the Bill for Neonatal Intensive Care:	Part II:	Hospitalization"
(mimeo). Reprints:	L. Joseph Butterfield, M.D., The Children's Hospital,
Department of Perinatology, 1056 East Nineteenth Ave., Denver, CO 80218.
Meyer, Belton
"Transport of High-Risk Infants in Arizona". Regionalization of Perinatal
Care. Report of the Sixty-Sixth Ross Conference on Pediatric Care.
Edited by Philip Sunshine, M.D., Columbus, Ohio:	Ross Laboratories.
Muirhead, Donald, M.D., Chairman
Report on Perinatal and Infant Mortality in Massachusetts 1967 and 1968.
Committee on Perinatal Welfare of the Massachusetts Medical Society.
Boston, Massachusetts.
Murphy, J. and W.A. Hodson
"Neonatal Intensive Care". Postgraduate Medicine 56:	55-58.
Pomerance Jeffrey, M.D., Christine T. Ukrainski, M.D., Tara Ukra,
Diane H. Henderson, M.D., Andrea H. Nash, M.D. and Janet L. Meredith, R.N.,B.S
"Cost of Living for Infants Weighing 1,000 Grams or Less at Birth".
From the Department of Pediatrics, Ceders-Sinai Medical Center and
University of California at Los Angeles. Pediatrics 61:	908-910.
Quebec Perinatal Mortality Committee
"Analysis of Perinatal Deaths by Cause, Part Two". Quebec, Canada.
Available from:	Quebec Perinatal Committee, Ministry of Social Affairs
Province of Quebec, 1005 Chemin Ste-Foy, Quebec, Canada, G1S 4N4.
Quebec Perinatal Mortality Committee
Significant Reduction in Perinatal Deaths in Quebec, 1967 to 1973. Press
Release (December). Quebec, Canada. Available from:	Quebec Perinatal
Committee, Ministry of Social Affairs Province of Quebec, 1005 Chemin
Ste-Foy, Quebec, Canada, G1S 4N4.
Quilligan, Edward, M.D. and Philip Sunshine
"Regionalization of Perinatal Care". Report of the Sixty-Sixth Ross
Conference on Pediatric Research. Columbus, Ohio:	Ross Laboratories.
Quilligan, Edward, M.D.
"The Obstetric Intensive Care Unit". Hospital Practice 17:	61-69.
Reynolds, T. Thompson
"The Results of Intensive Care Therapy for Neonates". Department of
Pediatrics and Division of Neonatology. University of Minnesota.
Journal of Medicine 59.
185﻿Bibliography of Clinical Literature -
(Continued)
Richardson, Joan C., M.D.
1976	"Principles of Organization of a Neonatal Intensive Care Unit from
Scratch". Clinics in Perinatology:	Organization for Perinatal Care
3:	329-335. Guest Editor:	Louis Gluck, M.D. Philadelphia:	W.B.
Saunders Company.
Robert Wood Johnson Foundation
1978 Special Report:	Regionalized Perinatal Services, Princeton, NJ:
The Robert Wood Johnson Foundation
Russel, Keith P., M.D., Sprague H. Gardiner, M.D. and Ervin E. Nichols, M.D.
1975	"A Conceptual Model for Regionalization and Consolidation of Obstetrics-
Gynecologic Services". American Journal of Obstetrics and Gynecology
121:	756-764.
Ryan, George M., Jr., M.D.
1977	"Regional Planning for Maternal and Perinatal Health Services". Seminars
in Perinatology:	Regionalization of Perinatal Care 1:	255-266. Guest
Editor:	Leo Stern, M.D. New York:	Grune and Stratton.
Schlesinger, Edward R., M.D.
1973	"Neonatal Intensive Care Planning and Outcomes Following Care". The
Journal of Pediatrics 82:	916-920.
Scott, K.E., M.D.
1970 "It's Only Statistics:	An Approach to Perinatal Mortality in Nova Scotia".
The Nova Scotia Medical Bulletin 81-83.
Stetson, John, M.D. and Paul R. Swyer, M.D.
1976	Neonatal Intensive Care. St. Louis, Missouri:	Warren H. Green, Inc.
Minkowski, A.
Section I, "Definition of Intensive Care of the Newborn".
Burnard, Eric D.
Section II, "Problems in Supplying Intensive Care for a Diverse
Community", pp. 12-20.
Karlberg, Peter
"An Evaluation of Intensive Care of the Newborn:	The Necessity for,
and Results of Therapy", pp. 21-31.
Stern, Leo, M.D.
"Neonatal Mortality and Intensive Care in the Province of Quebec",
pp. 33-37.
Stewart, A.
1972 "Prognosis for Infants of Very Low Birth Weight". London, England:
University College Hospital.
Stewart, Ann, M.B. and E.O.R. Reynolds, M.D.
1974	"Improved Prognosis for Infants of Very Low Birth Weight". Pediatrics
54:	724-735.
186﻿Bibliography of Clinical Literature -
(Continued)
Swyer, Paul R., M.D.
1975	"The Organization of Perinatal Care with Particular Reference to the Newborn
Neonatology:	Pathophysiology and Management of the Newborn. Edited by
Gordon B. Avery, M.D., Ph.D., Philadelphia:	J.B. Lippincott Company.
Swyer, Paul R., M.D.
1970	"The Regional Organization of Special Care for the Neonate". Pediatric
Clinics of North America 17:	761-776.
Teberg, Annabel, M.D., Joan E. Hodgman, M.D., Paul Y.K. Wu, M.D. and
Robert L. Spears, M.D.
1977 "Recent Improvement in Outcome for the Small Premature Infant:	Follow-
up of Infants with a Birth Weight of Less Than 1,500 Grams". Clinical
Pediatrics 16:	307-313.
Thogmartin, Bruce and Michael Tyne
1969 "Essential Design Elements of Neonatal Intensive Care Unit". Problems
of Neonatal Intensive Care Centers. Edited by Ross Hospital Planning
Associates, Ross Laboratories:	Columbus, Ohio.
Thompson, Theodore, M.D. and J. Reynolds
1977 "The Results of Intensive Care Therapy for Neonates:	I. Overall Neonatal
Mortality Rates; II. Neonatal Mortality Rates and Long-Term Prognosis
for Low Birth Rate Neonates". Journal of Perinatal Medicine 5:	59-75.
Usher, Robert H., M.D.
1971	"Clinical Implications of Perinatal Mortality Statistics". Perinatal
Biology. Leo Stern, M.D. (guest editor). Clinical Obstetrics and
Gynecology 14:	885-925.
Usher, Robert, M.D.
1977 "Changing Mortality Rates with Perinatal Intensive Care and Regionalization"
Seminars in Perinatology:	Regionalization of Perinatal Care 1:	309-319.
Guest Editor:	Leo Stern, M.D. New York:	Grune and Stratton.
Westbers, Jimmie A., M.D., Dayton W. Clark, M.D. and Gilbert A. Webb, M.D.
1973 "An Evaluation of High-Risk Maternity Care in a Community Hospital".
American Journal of Obstetrics and Gynecology 116:	557-563.
Zamansky, Harriet and Kathleen Strobel
1976	"Care of the Critically 111 Newborn in An Infant Care Center". American
Journal of Nursing 76:	566-568.
187﻿BIBLIOGRAPHY OF HEALTH PLANNING LITERATURE
Bergen-Passaic Health Systems Agency
1977 Report on the Task-Force on Regional Perinatal Services in the Bergen-Passaic
Health Service Area to the Project Review Committee. Rochelle Park, N.J.
Blackmon, Lillian, M.D. and Audrey K. Brown, M.D.
1973 Recommended Standards for Hospital Nursery Services:	Plan for a Statewide
System of High-Risk Newborn Care. Atlanta, GA:	Georgia Regional Medical
Program.
Comprehensive Health Planning of Northwest Illinois.
1977	Planning Policies for Perinatal High Risk. Rockford, IL.
Comprehensive Health Planning Agency of Southeastern Wisconsin, Inc.
1976	Guidelines for Perinatal Services. Milwaukee, WI.
Federal Register
1978	National Guidelines for Health Planning:	Neonatal Special Care Centers
43 (March 28), Section 121.204:	13040-13050.
Florida Department of Health and Rehabilitative Services, Children's Medical
Services
1977	Statewide Program for Perinatal Intensive Care Centers:	Obstetrical Com-
ponent. Tallahassee, FL.
Florida, State of, Office of Comprehensive Planning
1978	Health Problem Analysis, A Phase in the Development of the Florida State
Health Plan, Infant Mortality. Tallahassee, FL.
Florida, State of, Office of Comprehensive Planning
1978 Health Problem Analysis, A Phase in the Development of the Florida State
Health Plan, Obstetric Services. Tallahassee, FL.
Grand Rapids Emergency Medical Services Symposium on Critical Care Patients
1975 Grand Rapids EMS Symposium on Critical Care Patients:	Part III. Systems
Approach to the Care of the High Risk Perinatal/Neonatal Patient. Grand
Rapids, Michigan. Available from:	Emergency Medical Services Clearinghouse,
Box 911, Rockville, MD 20852.
Health Planning Council, Inc.
1978	Perinatal Care Center Standards (Draft). Madison, WI.
Health Systems Agency of Kane, Lake and McHenry Counties
1977	Perinatal Services Support Document. Lake Zurich, IL.
Health Systems Agency of San Diego and Imperial Counties
1978	Newborn Intensive Care Study Group, Report to the Planners Committee.
San Diego, CA.
Illinois Central Health Systems Agency
1977 Health Systems Plan (Perinatal Services Component). East Peoria, IL.
188﻿Bibliography of Health Planning
Literature - (Continued)
Maine, State of, Comprehensive Planning Agency
1973 Intensive Care of High Risk Newborns in Maine. Augusta, ME.
Maryland, State of, Comprehensive Health Planning Agency
1978 Perinatal Services. Baltimore, MD.
Massachusetts, Commonwealth of, Department of Public Health, State and
Regional Task Force on Neonatal Intensive Care and Technical Advisory
Group of the Perinatal Welfare Committee.
1977 Neonatal Intensive Care Standards and Criteria (Draft) Boston, MA.
Massachusetts Maternity and Newborn Regionalization Project
1976	The Final Report. Boston, MA. (Available from the Massachusetts
Department of Public Health).
Michigan, State of, Department of Public Health, Bureau of Health Facilities
1977	Hospital Nurseries and Newborn in Care Services, Minimal Criteria and
Guidelines. Lansing, MI.
Minnesota State Health Planning and Development Agency
1977 Perinatal Services. St. Paul, MN.
Nassau-Suffolk Health Systems Agency, Inc.
1977	A Plan for Regionalization of Obstetrical Services in Nassau-Suffolk.
Melville, N.Y.
New Jersey, State of, Department of Health
1978	Standards and General Criteria for the Planning, Certification of Need and
Designation of Perinatal Services. Trenton, N.J.
New Mexico Health Systems Agency.
1976	Toward Improving the Outcome of Pregnancy—Perinatal Care in New Mexico.
Sante Fe, N.M.
New York, State of, Department of Health
1978 Memorandum—Neonatal Centers for "High Risk" Babies.
Albany, NY. .
New York, State of, Department of Health
1977	Memorandum—Measuring Value Received for High-Priced Care of Premature
Infants. Albany, NY.
North Central Health Systems Agency
1978	Specialized Services Review Criteria: Neonatal Special Care Centers (Draft).
Atlanta, GA.
Ohio Department of Health, Division of Maternal and Child Health, Bureau
of Maternal and Child Health.
1977 State Perinatal Guidelines.
Ross Laboratories
1977 Planning and Design for Perinatal and Pediatric Facilities.
Columbus, Ohio:	Ross Laboratories.
189﻿Bibliography of Health Planning
Literature - (Continued)
Texas, State of, Department of Health
1976 Guidelines for the Texas Regional Perinatal Care System. Austin, TX.
West Virginia Planning Committee for Perinatal Health
1976 A Plan to Improve Perinatal Health Care in West Virginia.
Charleston, W. VA.
Western Center for Health Planning
1976 Certificate of Need Planning and Review Criteria, Intensive Care Newborn
Nursery. San Francisco, CA.
Wisconsin, State of, Division of Health Policy and Planning
1976 Planning Standards for Perinatal Care Centers. Madison, WI.
190﻿INDEX
Accessibility (Travel Time), 120, 143,
147, 152, 172, 173, 174
Average Length of Stay, 44, 45, 57,
135, 145, 148, 166, 173
Bed Need Determination Methods
Obstetrics, 14, 44
Neonatal Intensive Care Units, 32, 45,
68,	80, 82, 100, 101, 121, 123, 131,
145-148, 173, 174
Bed/Population Ratio, 43
Birth Projections, 25, 26, 45, 85, 148
Birth Volume, Minimum for Neonatal Unit
Service Area Determination, 43, 57,
67, 78, 80, 122, 128, 137, 138, 144,
145, 146, 149, 152, 157, 167, 174
Budget Models
Level II, 25
Level III, 25
Communications, 19, 164, 169, 171
Computers, 38
Consolidation of Services, 3, 13, 14,
58
Continuity of Care, 22
Cost
Fetal Monitoring Equipment, 18
Neonatal Intensive Care, 44, 58, 59
69,	93, 127, 140, 141, 142, 152,
163
Neonatal Intensive Care Units, 47,
58, 59, 69, 103, 140, 167
Transfer, 32
Cost-Benefit, Cost-Effectiveness of
Neonatal Intensive Care, 44
Data, 3, 50, 51, 107, 108, 132,
133, 134
Education in a regionalized
Perinatal System, 21, 30, 76,
101, 134, 165, 169
Emergency Medical Services, 30, 31,
62, 63, 129, 132
Equipment, 36, 37, 92, 93, 167, 168
Ethical Issues, 54, 55
Evaluation
Of Neonatal Intensive Care, 46, 95,
107, 108, 113, 141, 171, 172
Of Regional Perinatal Systems, 7, 8,
33, 50, 51, 107, 108, 111, 112,
132, 133, 134, 141, 154, 169
Fetal Monitoring, 18
Financial Feasibility, 151, 164
Financing, 6, 7, 63, 64
High-Risk Infant, 10, 20, 49, 52, 122,
129, 130, 162
High-Risk Mother, 39, 49, 90, 122,
138, 162
Laboratory Capabilities, 68, 100
Level I Facility Neonatal Intensive
Care Unit
Care Capabilities, 3, 5, 86, 91, 99,
153, 156, 170
Personnel Requirements, 3, 21, 117,
153, 157
Referral to Level II or Level III
Units, 160
Space Standards, 117
Level II Facility Neonatal Intensive
Care Unit
Care Capabilities, 3, 5, 86, 91,
99, 157, 170, 171
Components, 100, 158
Personnel Requirements, 21, 118, 158
Space Standards, 118
Level III Perinatal Center, 100, 142,
153, 159, 174
Level III Facility Neonatal Intensive
Care Unit
Care Capabilities, 5,	00	91, 119,
144, 159, 169, 171,	175,	179
Equipment, 36, 37		
Personnel Requirements	;, 25	, 33, 36,
119, 144, 159, 169,	170,	175,
176, 177
Space Standards, 36, 37, 119, 142,
143
Low Birth Weight, 1, 15, 20, 28, 33,
34, 8;, 97, 107, 111
Manpower
Needs of Perinatal Programs, 12, 25,
42, 43
Morbidity, 20, 39, 90
Mortality Rate
Fetal, 19
Infant, 17, 18, 39, 67, 125, 152
Neonatal, 19, 20, 81, 85, 89, 96,
105,	106, 109, 134, 139, 140
Perinatal, 9. 13, 14, 15, 16, 19,
57, 67, 70, 71, 72, 73, 74, 90,
106,	109, 110, 111
191﻿Neonatal Intensive Care
Clinical Indication for Care, 91,
92, 122
Impact Upon Morbidity and
Mortality, 9, 16, 26, 28, 39,
40, 56, 67, 73, 81, 89, 107,
109, 110, 111, 127
Prognoses of Infants Receiving
Care, 9, 27, 28, 52, 56, 81,
97, 98, 103
Neonatal Intensive Care Units
Care Capabilities, 114, 115
Components, 166
Design, 83, 104, 167
Equipment, 167, 168
Personnel Requirements, 167
Planning, 167
Space Standards, 36, 37, 83
Obstetric/Gynecologic Care, 175
Occupancy Rate 137, 145, 173
Perinatal Systems
Planning, 11, 41, 42, 48, 98
Regionalized System Relationships, 2
11,	39, 41, 42,	CO	53, 75, 77, 78
83,	84, 98, 99,	101,	109, 110,
136,	171, 175		
Population in Need, 68, 82, 166
Transport/Trans fer
Arrangements, 66, 78, 135, 140,
151, 154, 165, 171
Cost, 63, 140
Criteria	for	Maternal,		, 35,	64,	65,
66, 102	, 122					
Criteria	for	Newborn,		17,	18,	24,
35, 60,	64,	65,	66,	78,	122,	178
Vehicles,	30,	31,	42,	61,	62,	178
Unit Size,	35,	122,	140,	, 145	, 173	
Utilization Rates, 101
*U.S. GOVERNMENT PRINTING OFFICE: 1979 0— 629-947/2439 REGION 3-1﻿BUREAU OF HEALTH PLANNING
Health Planning Series
Health Planning Bibliography Series:
1.	Selected Bibliographic References for Community Health Status Assessment
2.	HRP-0300101 Consumer Participation in Health Planning: An Annotated Bibliography
3.	HRP-0300201 Services Shared by Health Care Organizations: An Annotated Bibliography
4'.	HRP-0300301 Women and the Health System: Selected Annotated References
5.	HRP-0300401 Selected Bibliographic References on Computerized Axial Tomography
6.	HRP-0300501 Certificate-of-Need/1122 Project Reviews: A Bibliography
7.	HRP-0300601 Community Nutrition in Preventive Health Care Services HRP-0300701
8.	Methods for Setting Priorities in Areawide Health Care Planning: An
9.	Annotated Bibliography HRP-0300801 Guidelines for Planning Health Services: An Annotated Bibliography
10.	HRP-0300901 Emergency Medical Service Systems and Health Planning: A Bibliography
11.	HRP-0301001 Mental Health Planning: An Annotated Bibliography HRP-0301101
12.	Certificate of Need Programs: A Review, Analysis, and Annotated Bibliography
13.	of the Research Literature HRP-0301201 Planning for Outpatient Surgery Services: An Annotated Bibliography
14.	HRP-0301301 Health Maintenance Organizations and Health Planning: An Annotated
15.	Bibliography HRP-0301401 Hospices and Related Facilities and Services for the Terminally 111:
16.	Selected Bibliographic References HRP-0301501 A Review of Planning Methods and Criteria for Neonatal Intensive Care Units:
	An Annotated Bibliography of the Clinical and Health Planning Literature
	HRP-0301601﻿US. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
Public Health Service
Health Resources Administration
Bureau of Health Planning
National Health Planning Information Center
DHEW Publication No. (HRA) 79-14038﻿******** iMitiMV
Health Planning
Bibliography Series ) n
Methods to Determine
Geographic/Population
Boundaries for Specific
Health Services:
An Annotated Bibliography﻿HEALTH PLANNING SERIES
The Bureau of Health Planning is a primary resource for current information
on a wide variety of topics related to health planning. To facilitate the
dissemination of this information to health planners, the Bureau issues
publications in the following series.
Health Planning Methods and Technology
This series focuses on the technical and administrative aspects
of health planning. Included are such areas as methods and
approaches to the various aspects of the health planning process,
techniques for analyzing health planning information and problems,
and approaches to the effective dissemination and utilization of
technical information.
Health Planning Information
This series presents information on the analysis of issues and
problems relating to health planning including trend data, data
analysis, and sources of data to support health planning
activities.
Health Planning Bibliography
Bibliographies on specific health planning subjects are published
in this series. Subject areas are selected by the frequency of
inquiries on specific topics and from suggestions by Bureau staff
and health planners throughout the Nation.
International Health Planning
This series presents information and case studies relating to
health planning methodologies and experiences of other nations
which are potentially applicable by U.S. health planners.
Health Planning in Action
Case studies of selected activities of State and local health
planning agencies are presented to highlight their achievements in
providing "equal access to quality health care at a reasonable
cost."
"Methods to Determine Geographic/Population Boundaries for Specific Health
Services:	An Annotated Bibliography" is the seventeenth publication in the
Health Planning Bibliography Series. A list of all the publications in this
series appears on the inside of the back cover of this publication. Copies may
be purchased from the National Technical Information Service (NTIS), Department
of Commerce, 5285 Port Royal, Springfield, Virginia 22161.﻿Methods to Determine
Geographic/Population
Boundaries for Specific
Health Services:
An Annotated Bibliography
Prepared under
Contract No. HRA 231-77-0055
by
| Government Studies and Systems, Inc.)
Philadelphia, Pennsylvania
July 1980
HRP-0301 701
o
i
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Public Health Service
Health Resources Administration
Bureau of Health Planning \
DHHS Publication No. (HRA) 80-14013﻿﻿PREFACE
In seeking to identify methods to determine geographic/
population boundaries for health services in the health
field, in public services, and in the private sector,
Government Studies Systems, Inc., conducted an exhaustive
survey of the literature during the research phase of
Contract No. HRA 231-77-0055. Over 200 relevant refer-
ences were identified. This annotated bibliography is
being published as a separate volume for the convenience
of health planners. The full report is based on
extensive analysis of 25 of the most promising methods
for solving location/allocation problems. It is entitled:
Methods to Determine Geographic/Population Boundaries for
Specific Health Services:	Part I. State-of-The-Art
Report and Part II. Alternative Methods.
Both publications will be made available from the
National Technical Information Service (NTIS), Springfield,
Virginia 22161, (703) 487-4650. The accession number
for the bibliography is HRP-0301701; and for the report,
HRP-0101901.

iii﻿﻿1.
Abernathy, William J. and John C. Hershey. "A Spatial-Allocation
Model for Regional Health-Services Planning." Operations Research
20(3), May-June 1972, 629-642.
In planning for health services, the need arises to determine the
location, capacity, and number of health centers for a geographi-
cally defined region. The present paper formulates this problem in
a form convenient for solution and presents results from the model
to clarify some important aspects of this allocation decision. The
planning region is assumed to consist of geographically defined
subareas or census blocks, of known location. The population is
stratified in such a way that each stratum exhibits relatively
homogeneous patterns of health care utilization. The model charac-
terizes the effects of center locations upon aggregate utilization
and utilization of individual centers and gives optimal locations
of centers with respect to several alternative criteria. An
example illustrating computational feasibility and the implications
of various criteria for the location decision is presented.
2.	Abler, R., J. S. Adams, and P. Gould. Spatial Organization,
Englewood Cliffs, New Jersey:	Prentice-Hall, 1971. Chapter 14 and
pp. 330-1.
Chapter 14 presents the L-A problem as "Geographers Unsolved
Problem." Emphasizes complexity of problem.
3.	Barber, Brian K., Michael H. Branson and Riad Mahayni. "Minimizing
Travel Time to Hospital Beds in Rhode Island Department of Health,
Technical Report No. 10, August 1977 (unpublished).
Project was undertaken to answer the question:	Given projected
total hospital bed need in Rhode Island, maximum and minimum
numbers of beds for efficient hospitals, and tertiary level care,
what would be the ideal distribution of hospital beds in the State?
The problem was formulated as a mixed zero-one linear programming
problem, with total travel time for residents to be minimized. The
State was subdivided into 25 areas, and certain heuristic assumptions
were made about patient travel to reduce the problem to a computation-
ally feasible size. Data are required on population, transportation,
and hospital bed needs. Results are discussed, as well as possible
improvements and refinements.
4.	Bell, Colin E., and David Allen. "Optimal Planning of an Emergency
Ambulance Service," Socio-Economic Planning Sciences 3(2), 95-101.
Queuing theory models are developed for computing the size of
emergency ambulance fleet required to meet specified standards of
service. As a function of the demand for ambulance service and the
distribution of duration of service calls, tables are given for the
number of ambulances required to respond immediately to 95 (or 99)
percent of service requests. By analyzing a queuing system with
1﻿with unlimited server (ambulance) availability, approximate results
to overcome the difficulties of computing steady-state results for
a multi-server (ambulance) queuing system are provided. This
results in guidelines for determining the size of an ambulance
fleet based at a single garage. The extension of the results to
more complex situations is also discussed.
5.	Bergwall, David F., et al. Introduction to Health Planning
Washington, D.C.:	Information Resources Press, 1974.
This book is written as an introductory textbook and is rather
elementary in its approach. Chapter 6 covers "Spatial Aspects of
Health Planning." Under "regionalization" it discusses patient
origin studies, indexes of relevance and commitment, the "areawide
planning" model of Schneider, etc.
6.	Berlin, Geoffrey N. and Jon C. Liebman. "Mathematical Analysis of
Emergency Ambulance Location," Socio-Economic Planning Sciences,
8(6), December 1974, 323-328.
Proper ambulance location is crucial in saving lives and reducing
injury. To determine efficient and effective locations, it is
necessary to evaluate both the spatial and temporal distributions
of demand. Because of the complex nature of the ambulance response
and transport activities, the problem is subdivided into a facility
location problem and a vehicle allocation problem. A set covering
model is modified to solve the location problem, while a simulation
is used to solve the allocation problem. Combining these two
models provides a powerful methodology for systematically finding a
satisfactory solution to the ambulance location problem.
7.	Berlin, Geoffrey N., C. S. ReVelle, and D. J. Elzinga. "Deter-
mining Ambulance-Hospital Locations for On-Scene and Hospital
Services," Environment and Planning A, 8(5),(August 1976), 553-
561.
Network models, incorporating several new impedance measures for
the location of ambulances and hospitals are introduced. Although
the minimization of average impedance is frequently stated as a
criterion in location models, the impedance does not usually
embrace more than a single link, such as the time taken from a
facility and demand, but also the link from the point of rescue to
the point of ultimate service. The return link between the service
point and dispatching station is considered as well. Models for
the simultaneous location of ambulance dispatching points and
hospitals are framed.
8.	Berry, B. J. L. (1961) "A Method for Deriving Multifactor Uniform
Regions," Przeglad Geograficzny, 33:	263-82.
After factor analysis is applied to characteristics of small area
units, the factor scores are plotted on a dimensional graph, and
2﻿observations are grouped hierarchially by sequentially joining
the two closest points and replacing them by their centroid.
The actual measure is the average of squared distance between
all possible pairs of points - one for each group.
9.	________ (1967) "Grouping and Regionalizing:	An Approach to the
Problem Using Multivariate Analysis," in Quantitative Geography
Part I, W. L. Garrison and D. F. Marble (eds,), Northwestern
University, Evanston, Illinois.
Suggests a three stage procedure including factor analysis,
distance scaling of similarities, and grouping. Several methods of
grouping are viewed and their difficulties stated. An extensive
bibliography is provided.
9A. Bertolazzi* Paola, Lucio Bianco, and Salvatore Rlcciardelli.
"A Method for Determining the Optimal Districting in Urban
Emergency Services," Computers and Operations Research, 4 (1977), 1-12.
In this paper a method is suggested for dealing with one of the
main allocation problems often appearing in urban emergency
services, that of districting. A mathematical model is developed
in terms of constrained optimization with binary variables. An
algorithm, derived from Balas' filter-method, which seems to be
quite efficient for this class of problems, has been utilized. The
objective function considered here is the overall travel time in
the region under consideration. The optimal travel time in the
case of the Rome fire service turns out to be meaningfully lower
than that derived from the actual situation. Also included is a
way of applying the method to the assignment of response areas to
the different stations as a function of the current state of the
system.
10.	Brown, Lawrence A., and John Holmes, "The Delimitation of
Functional Regions, Nodal Regions, and Hierarchies by Functional
Distance Approaches," Journal of Regional Science 11(1), 1971,
57-72.
Following a brief discussion of the concept of region, some of the
technical approaches to delimiting functional and nodal regions are
reviewed. The approach adopted here is based on the notion of
'functional distance' which, it is suggested, can be directly
measured using the mean first passage times (MFPT) of a Markov
chain analysis of a flow matrix. The method is described clearly
and in detail. Application of the method to the Journey-to-work
data (1951) for 107 urban places in Derbyshire, Nottinghamshire,
and the West Riding results in the identification of three main
functional regions 1) North Midlands containing Nottingham, Derby,
Matlock, and Mansfield 2) South Yorkshire with Chesterfield,
Sheffield, Doncaster, Barnsley, and Worksop and 3) the West Riding
region of Leeds, Bradford, Huddersfield, Halifax, and Wakefield.
3﻿Subsequent analysis of flows within each functional region
enables nodal regions to be identified for the major towns.
These conform closely to the authors' intituitive knowledge of
the nodal structure of the area, a structure which is strongly
influenced by the hierarchical organization of settlement and
physical features.
11.	Bureau of the Census, DIME Geocoding System, Washington, D.C.,
1970, 50.
The development by the Census Use Study of the DIME (Dual Inde-
pendent Map Encoding) geographic base file is described as part of
a small-area data research study sponsored in New Haven, Connect-
icut, by the Bureau of Census. The report is intended for three
audiences:	(1) individuals interested in technical developments in
geographic coding; (2) organizations within standard metropolitan
statistical areas participating in or eligible to use the products
of the Census Bureau's address coding guide (ACG/DIME) program; and
(3) organizations interested in creating their own DIME file. The
introductory chapter describes the Census Bureau's development of a
computerized geographic system for use in conducting the 1970
census. The use of census and local data with the DIME file for
planning and analysis in the public and private sectors is described.
Included are discussions of computer mapping; adding local area
codes to the DIME file codes to local records through use of the
ADMATCH computer package; network and node analysis; adjacency
analysis (including aggregation of areas and clustering of similar
areas); and other potential uses, such as monitoring of programs by
geographic area, development of parcel files and larger-grained
files, information systems, area sampling, geographic grouping of
census summary tape data, time series studies, spatial comparisons,
and market analysis. The conceptual origins and technical aspects
of the DIME system are discussed. The process of creating a DIME
file from clerical coding through computer processing is described
as are coordinate insertion, updating and maintenance research.
Included are cost and staffing considerations. Information on
availability of Census Bureau geographic files, the Metropolitan
Mapping Series, and data from 1970 census is appended.
12.	Bureau of Health Resources Development, Delineation of Economic
and Health Service Areas and the Location of Health Manpower
Education Programs. A Summary, Bethesda, Md., Division of Manpower
Intelligence. February 1974, 36 p.
Findings of a study to determine whether the economic system could
be used effectively as a surrogate in health planning for the
health-care delivery system, and to estimate the interrelationships
of the health manpower education system with the economic system
and the health-care delivery system are summarized. The methodology
used 74 economic variables and 24 health care delivery variables
for the same intermountain geographic region characterized by rural
conditions and low population density. It was found analytically
4﻿feasible to identify a close relationship between the economic and
health care systems in the region. Although the study did not
involve the estimating of health needs and the use of time-series
data, the empircal results indicated that:	(1) the methodology
provides meaningful models of the economic and health systems in
the intermountain region; (2) the economic system data are easier
to obtain than the health-care system data; and (3) given the close
relationship of the two systems, the economic system has the
potential of being a surrogate for the health-care delivery system.
The potential also is suggested for using the economic system as a
method for improving the quality of health planning, health man-
power planning, and health manpower program and facility planning.
13. Burt, 0. R. and C. C. Harris (1963) "Apportionment of United States
House of Representatives: A Minimum Range Integer Solution Alloca-
tion Problem," Operations Research, 11:	648-52.
A method is proposed for the problem of allocating representatives
of States which minimizes the maximum inequality of representation
using dynamic programming but suggests a non-integer solution be
used as an approximation. An example is given and compared to a
standard method of 'equal proportions.'
14.	Byfuglien, J. and A. Nordgard, "Region-Building—A Comparison of
Methods," Norsk Geografisk Tidsskrift, 27(2), 1973, pp. 153-161.
This study deals with the use of quantitative methods to build up
uniform or formal regions, the construction of farming-type regions
in eastern Norway being used as an example. To achieve reasonable
uniformity in the size of the basic areal units, some of the
original administrative areas are first combined; the resulting
areal units are then described by four orthogonal variables
extracted by principal components analysis. Six different methods
of aggregating the basic units to form regions are discussed, each
method using the same measure of similarity (Dz) between the basic
units. The homogeneity of the regions produced by these methods is
compared, the Ward's method shown to give the best (although not
necessarily the optimal) solution.
15.	Calvo, Alberto B., and David H. Marks, "Location of Health Care
Facilities:	An Analytical Approach," Socio-Economic Planning
Sciences 2(5), (October 1973), 407-422.
The problem of health facilities location is explored under a
mathematical optimization approach. Several models are developed
for the location of a generalized health facility system in a region
in a manner that the selected criteria are optimized. Locational
criteria are structured as a hierarchy of social, economic and
political attributes and defined analytically using economic and
utility theory. A planning framework, based on the models developed,
is presented to aid the health planner in the formulation of health
facility location decisions.
5﻿16.	Carbone, Robert, "Public Facilities Location Under Stochastic
Demand," Infor, 12(3), (October 1974), pp. 261-270.
This paper extends the current state of public facilities location
problems on a road network to cover situations in which the number
of users at each node may be a random variable. The basic location
model is reformulated as a chance-constrained programming problem
with fractile criterion. A computational procedure for solving the
non-linear deterministic equivalent problem derived is presented.
Finally, a hypothetical numerical example illustrates the possi-
bility of changes in the location decision when the stochastic
nature of the problem is taken into account.
17.	Carr, W. J. (1970) "Economic Efficiency in the Allocation of
Hospital Resources:	Central Planning vs. Evolutionary Develop-
ment," in Empirical Studies in Health Economics, H. Klarman (ed.),
Baltimore:	Johns Hopkins Press, pp. 195-221.
Empirical hospital sizes and locations with respect to demand are
compared to central planning solutions with hexagonal market
areas. Monte Carlo simulation is used. L-A analysis per se does
not enter.
18.	Carter, Grace M., Jan M. Chaiken, and Edward Ignall, "Response
Areas for Two Emergency Units," Operations Research, 20(3) May-June
1972, 571-594.
This paper gives a model in which two urban emergency service units
(such as fire engines or ambulances) cooperate in responding to
alarms or calls from the public in a specified region of a city.
Given the home locations of the units and the spatial distribution
of alarm rates, it is possible to specify which unit should respond
to each alarm by defining a response area for each unit. The
average response time to alarms and the workload of each unit are
calculated as functions of the boundary that separates their
response areas. The boundaries that minimize average response time
and the ones that equalize workload are determined. Some bound-
aries can be dominated, in the sense that another boundary improves
both workload balance and response time. The set of undominated
boundaries is found.
19.	Casetti, E. (1964) Classifactory and Regional Analysis by Discriminant
Iterations, Technical Report No. 12, Computer Applications in the
Earth Sciences Project, Department of Geography, Northwestern
University.
Uses multivariate discriminant procedures which will improve upon
an initial classification of regionalization. The heuristic pro-
cedure used is the basic 'locate-allocate' heuristic used by many
others. The centroid of the cluster is used in the locate phase.
The procedure is suggested for evaluating classifications or
regionalizations and identifying cores.
6﻿20.	Castellan, J. W. (1967) "Political Reappointment by Computer,"
Brown University Computing Review, 1:	5-24.
A readable statement of the districting problem and its formulation
for algorithm solution. Discusses the Weaver-Hess method and
presents the problem as a (0, 1) integer program and as a modified
Knapsack problem which is successful on large problems.
21.	Chaiken, Jan M., and Richard C. Larson, "Method for Allocating Urban
Emergency Units," Management Science, 19(4 Part II) December 1972,
p. 110-130.
An urban emergency service system provides mobile units (vehicles) to
respond to requests for service which can occur at any time and any
place throughout a city. This paper describes the common character-
istics and operational problems of these systems and surveys the
various methods, both traditional and recently developed, which may
be used for allocating their units. Aspects of allocation policy
discussed include:	(1) determining the number of units to have on
duty; (2) locating the units; (3) designing their response areas or
patrol areas; (4) relocating units; and (5) planning preventive-
patrol patterns for police cars.
Typical policy changes which may be suggested by the use of quantita-
tive allocation models include selective queuing of low priority calls,
varying the number of units on duty (and their locations) by time of
day, dispatching units other than the closest ones to develop certain
incidents, relocating units as unavailabilities begin to develop, and
assigning police cars to overlapping patrol sectors. As a result of
making such changes, it is often possible to reduce queuing and travel
time delays, improve the balance of workload among units, and enhance
the amount of preventive patrol where needed.
22.	Chapelle, R. A. (1970) Location of Central Facilities, Heuristic
Algorithms for Large Systems, unpublished Ph.D. thesis, University
of Oklahoma, Norman, Oklahoma.
Provides a discussion of the history of the continuous space L-A
problem noting that current algorithms are prohibitively expensive
in time and storage and develops two heuristics - one variable grid
method and one variable discrimination method which can cope with
very large problems. Cluster analysis is suggested for n dimensional
space. Applies algorithms to post office location. The Fortran
listings of 6 algorithms are included in the appendices.
23.	Cherniack, H. D. and J. B. Schneider (1967) A New Approach to the
Delineation of Hospital Service Regions, Discussion Paper No. 16,
Regional Science Research Institute, University of Pennsylvania,
Philadelphia.
A different approach is used for regionalization in an application
to a real problem. One of the relevant indices is based on medians
7﻿which are found by the Newton-Raphson method. A method is proposed
for relocating hospitals for a more efficient covering of the
population. The approach has been criticized for approximating the
demand for services or discrete points (postal zone centroids).
24.	Christianson, Jon B., "Evaluating Locations for Outpatient Medical
Care Facilities." Land Economics, 52(3), 1967, pp. 299-313.
The paper adapts the travel cost approach normally used in evaluating
recreation benefits in order to measure the benefits associated
with alternative locations for a medical center. An empirical
study is described in which demand equations for two different
types of medical problems are estimated:	first, treatment of a
specific illness; and second, preventive medical care taken at the
consumer's discretion. The total number of visits to the existing
facilities is taken to be the product of the probability of
experiencing a particular medical problem and the expected number
of visits to the clinic, which in turn is a function of travel cost
(including time costs) and a number of other variables including
travel cost to alternative facilities, age, income, sex, and family
size. Willingness to pay for a new facility is estimated by
integrating the mean individual demand for visits in price space
between the old and new levels of travel cost.
25.	Christofides, N. and P. Viola (1972) "The Optimum Location of
Multicentres on a Graph," Operational Research Quarterly, 22:	145-
154.
An iterative algorithm is presented for solving the p- centres of
a graph, to minimize the maximum distance travelled; and a similar
problem of minimizing the number of centres for a critical distance
or time. In the absolute p- centre problem the centres need not be
at the vertices. The algorithm generates optimal solutions but can
also produce suboptimals quickly.
26.	Church, Richard, and Charles ReVelle, "The Maximal Covering Location
Problem," Papers of the Regional Science Association, 32 (Fall
1974), pp. 101-118.
The problem is defined as, "Maximize coverage (population covered)
within a desired service distance S by locating a fixed number of
facilities." The problem is formulated mathematically, and two
algorithms for solving it are described, as is a linear programming
solution. A variation of the original problem is stated, "Locate a
fixed number of facilities in order to maximize the population
covered within a service distance S, while maintaining mandatory
coverage within a distance T (T_>S)." This is called the maximal
covering location problem with mandatory closeness constraints; its
solution is described.
27.	Cliff, A. D., and P. Haggett (1970) "On the Efficiency of Alternate
Aggregations in Region Building Problems," Environment and Planning,
2:	285-94.
8﻿Discusses the regionalization problem in geography as aggregating
small areal units with no continuity constraint and with a complete
continuity constraint. The combinatorial size of the problem is
stressed and a measure is devised to assess the efficiency of the
grouping. They suggest a random partitioning schema where the
probability of finding a better grouping is worked out and, if
sufficiently small, the procedure terminates.
28.	Comprehensive Health Planning Agency of Southeastern Wisconsin,
Inc. Health Facilities Geographic Subareas, Milwaukee, March
1973, 41 p.
Patterns of utilization of 33 general hospitals in the seven-
county southeastern Wisconsin planning area are examined, and
geographic subareas are defined which relate groups of general
hospitals with the geographic areas containing the population
they primarily serve. The planning area incorporates both urban
(Milwaukee, Wisconsin) and rural populations. Purposes of
identification of geographic subareas for health facilities
planning are enumerated, and the use of hospital patient origin
data in identifying subareas is described. Factors which limit
the applicability of the subarea study are noted. The geographic
identifications apply only to inpatient services, allowing service
areas for emergency and ambulatory patients to be determined
separately. No hospitals outside the planning area were con-
sidered in the study, and the influence of area hospitals beyond
the boundaries of the planning area was also excluded. Criteria
used in selecting optimum subarea configurations from the alter-
natives developed in the study are given. Definitions used to
distinguish general hospitals from specialty hospitals are noted,
and area hospitals are placed into one of these two categories.
For each general hospital in the area, the degree of dominance
in that hospital's service area, i.e., the percentage of patients
in each service area served by each hospital, is discussed.
Relationships among hospitals serving similar areas also are
discussed. A final subarea grouping, selected on the basis of
the degree of compatibility between the service areas of various
groups, is described in detail. A profile of each of these nine
subareas is provided. Uses of the subareas as basic planning
tools are described. Maps and supporting tabular data are
included.
29.	Corley, H. W., Jr., and S. D. Roberts, "A Partitioning Problem
with Applications in Regional Design," Operations Research,
20(5), Sept. - Oct. 1972, 1010-1019.
Many problems in two-dimensional location analysis can be
formulated as one of optimally dividing a given region into n
subregions with specified areas. Examples are problems involving
districting, facility design, warehouse layout, and urban plan-
ning. This paper contains a study of such a partitioning problem.
9﻿Theoretical results are presented for a problem of optimally
partitioning a given set of points in k-dimensional Euclidean
space into n subsets, where each subset has a specified Lebesgue
measure. The existence of an optimal solution is established,
and necessary and sufficient optimal conditions are proved.
Models are then formulated in terms of this partitioning problem
for specific districting and warehouse-layout problems.
30.	Cox, K. R. (1965) "The Application of Linear Programming to
Geographic Problems", Tijdschrift Voor Economische en Sociale
Geographie, 56:	228-36.
Outlines the usefulness of the transportation problem of linear
programming to geography. Presents the stepping-stone solution
procedure and discusses the dual. Mention is made of the
application of the technique to delimit regions.
31.	Dean, Robert D., David McKee, and William H. Leahy. Spatial
Economic Theory. New York:	Free Press, 1970, 365 pp.
This book of 20 readings is organized into six parts. An intro-
duction presents the partial and general equilibrium approaches.
Part I is concerned with least cost theory; Part II with locational
interdependence (spatial competition); Part III with market area
analysis; Part IV with locational equilibrium analysis; and Part V
with general equilibrium theory. Papers included are:
"The partial equilibrium approach" - E. M. Hoover;
"The general equilibrium approach" - Leon N. Moses;
"Distance inputs and the space economy - the locational
equilibrium of the firm" - W. Isard;
"Location and theory of production" - Leon N. Moses;
"A theoretical framework for geographical studies of
industrial location" - D. M. Smith;
"Stability in competition" - H. Hotelling;
"Optimum location in spatial competition" - A. Smithies;
"A dissenting view of duopoly and spatial competition" -
Nicos E. Devletoglou;
"The economic law of market areas" - Frank A. Fetter;
"The economic law of market areas" - C. D. Hyson & W. P.
Hyson;
"The size and shape of the market area of the firm" -
Melvin L. Greenhut;
"A model of market areas of a firm" - Edwin S. Mills &
Michael R. Vav;
"Equilibrium of a household" - William Alonso;
"Integrating the leading theories of plant location" -
Melvin L. Greenhut;
"The equilibrium of land use patterns in agriculture" -
Edgar S. Dunn, Jr.;
"A general location principle of an optimum space economy" -
W. Isard;
10﻿"Equilibrium among spatially-separated markets - solution
by electric analog" - Stephen Enke;
"Spatial price equilibrium and linear programming" - Paul A.
Samuelson;
"Equilibrium among spatially-separated markets - a reform-
ulation" - T. Takauama & G. G. Judge;
"Towards a united theory of spatial economic structure" -
Edwin Von Boventer.
32.	Dearing, P. M., and R. L. Francis, "A Network Flow Solution to
a Multi-Facility Minimax Location Problem Involving Rectilinear
Distances," ORSA Bulletin, 20, Sup. 2, Fall 1972, B-469.
The problem considered is to locate new facilities with respect
to existing facilities so as to minimize the maximum cost, where
the costs are either proportional to the rectilinear distances
between new and existing facilities or proportional to the
distances among new facilities. The problem also includes upper
bound constraints on the rectilinear distances among facilities.
The problem is decomposed into two independent one dimensional
minimax location problems. Each subproblem is formulated as an
equivalent network weighted flow problem, which is solved by
determining a simple chain of maximum cost to weight ratio.
33.	Dee, Norbert and Jon C. Liebman, "Optimal Location of Public
Facilities," Naval Research Logistics Quarterly, 19(4),
December 1972, 753-759.
Location of both public and private facilities has become an
important consideration in today's society. Progress in solution
of location problems has been impeded by difficulty of the fixed
charge problem and the lack of an efficient algorithm for large
problems. In this paper a method is developed for solving large-
scale public location problems. An implicit enumeration scheme
with an imbedded transportation algorithm forms the basis of the
solution technique.
34.	Deighton, D. M. (1971) "A Comment on Location Models," Communi-
cation to the Editor, Management Science, 18:	113-15.
Deighton is distressed that L-A models are static. He suggests
tracing an optimal path through optimal solutions to static
problems over the planning horizon. Moves, closings and openings
are possible by looking at discounted costs. He also suggests
adding confidence intervals on uncertain future demand estimates.
35.	Disker, Robert A., William C. Park, Jr. A Method for Delineating
Health Service Areas, Bowman Gray School of Medicine, Winston-
Salem, NC.
Paper presents a method for determining the service area of a
health care facility. Patient addresses gleaned from any data
11﻿source which consistently includes this information are plotted
by geographic unit of residence. The size of the geographic unit
chosen—county, census tract, city block—is dictated by the
sophistication of the health service being analyzed. The crude
service area obtained by plotting patient addresses against
geographic units is refined according to the criteria magnitude of
demand and repetition of demand. A geographic unit must provide
either 10 percent of the total patients to service or five or more
patients to the service during the specified time period to be
included in the final trade area. Maps of crude and refined
service areas of a 500 bed North Carolina medical center are
included to illustrate results of the method.
36.	Drosness, Daniel L. and Jerome W. Lubin, "Planning Can be Based
on Patient Travel," Modern Hospital, 106, April 1966.
Improved planning in hospital services and facilities requires
that each institution know what community area it is serving.
New research now indicates that it is potentially possible to use
routinely collected birth certificate data which identify the
movement of patients to hospitals as an estimator of the movement
of all inpatients, and for those hospitals not providing obstet-
rics that a relatively simple sampling of total patients will
suffice.
35A. Donabedian, Avedis, "Capacity to Produce Service in Relation to
Need and Demand," Aspects of Medical Care Administration Cambridge,
MA:	Harvard University Press, 1969.
This chapter reviews numerous studies which deal with the rela-
tionship between geographic assessibility and the use of services
and resources, including those that attempt to define more
precisely the nature of the mathematical function that describes
this relationship. Locational efficiency is defined as the manner
in which the location of health resources influences the costs
and benefits of receiving and failing to receive health care.
Donabedian acknowledges the difficulties in accounting for and
measuring different components of locational efficiency, and that,
in general, the usually positive relationship between severity
of condition, complexity of the resource, and willingness to
travel suggests the appropriateness of a centralized or decentral-
ized location pattern. The results of several studies dealing
with the issue of locational efficiency are also reviewed.
37.	Earickson, Robert, A Behavioral Approach to Simulation Models
of Patient Use of Physicians and Hospitals:	I. Spatial and
Social Variables of Hospital Utilization, Chicago Regional Hospital
Study, Working Paper #111.11, April 1969.
First working paper of a series of three about the effect of
spatial, economic, social and perceptual variables on patient
utilization of medical services. It is contended that spatial
12﻿interaction patterns of patients and physicians with hospitals
cannot be satisfactorily explained by a simple proximity-to-care
construct. Besides racial segregation and welfare factors in the
health care system, perceived differences in quality of doctors'
offices and hospitals, based on architectural factors and religious
affiliation, are considered important influences on patient utili-
zation. Author outlines data base and methodology of models
discussed in the third paper which* described spatial behavior of
hospital patients in the Chicago metropolitan area.
38.	Earickson, Robert, A Behavioral Approach to Simulation Models of
Patient Use of Physicians and Hospitals:	II. Spatial and Social
Variation in Physician and Hospital Location, Chicago Regional
Hospital Study, Working Paper #111.12, May 1969.
In this second working paper, the author describes aspects of the
location of medical services which affect interactions of patients
with physicians and hospitals, and discusses how variations in
hospital size, and physician-hospital affiliations affect patient
travel, using data concerning 110 Chicago metropolitan hospitals.
Findings of this analysis are incorporated in the model presented
in the third paper. Some of the major findings are:	1) Hospitals
and physicians tend to be closer to each other than either is to
the patient - the hospital is often near a cluster of physicians
in the business center; 2) A majority of patients travel only
three to five miles to a hospital; 3) There is evidence of a
hierarchical spatial structure of hospital distribution in the
Chicago area according to hospital size and speciality type.
39. ___________, A Behavioral Approach to Simulation Models of Patient
Use of Physicians and Hospitals:	III. Deterministic and Behavioral
Simulation Models of Patient Use of Physicians and Hospitals,
Chicago Regional Hospital Study, Working Paper #111.13, June 1969.
Third paper of the series presents a two-stage deterministic and
behavioral simulation model of patient use of physicians and
hospitals. Demographic, social and economic data for 1965 were
compiled and disaggregated for 206 "health care areas" in Chicago.
Also, data for a 1965 patient discharge survey in 119 Chicago
hospitals were used. Metropolitan physicians' office were arbi-
trarily grouped into 250 clusters, and straight line distances
between community areas, physicians' clusters and hospitals were
computed. In phase one, the model took into account behavioral
and economic factors and a random element of choice to replicate
the health care system. The model allocated patients to physician
attraction. Doctors were shifted from one cluster to another to
achieve an average patient-per-doctor ratio, realizing patient
travel savings. In phase two, a more realistic replication of the
system, physician offices were held constant, and excess patients
shifted to the nearest cluster. Differences in travel costs
13﻿between phase one and two are a measure of the inefficiency of
the present system. Model then repeats the process of phases one
and two for patient-to-hospital patterns, also allowing for reli-
gious preferences rerun, and relaxing racial and charity constraints,
which results in even greater patient travel savings. Findings
indicate there is a need for some shift in physician and hospital
capacity and a relaxation of income and racial barriers.
40.	Eddleman, B. R., "Areawide Planning for Optimum Location of
Hospital Facilities for Rural People," Southern Journal of Agri-
cultural Economics, 4(1), July 1972, 29-95.
Paper presents a conceptual model permitting minimization of
combined patient travel costs and health service production costs
with economies of size affecting the level of production costs.
Model was applied to a nine-county area surrounding Gainesville,
Florida, to estimate the optimum number, size, location of the
hospital facilities and total regional access costs for projected
additional health service requirements. It was concluded that
the model has possibilities for aiding in areawide planning of
present and future developments in the health services industry.
41.	Ellwein, Leon B., and John T. Kalberen, Jr., Optimal Location of
Cancer Centers on the Basis of Population Access." Federation
Proceedings, 34 (8), May 1975, 1411-1416.
The National Cancer Center Institute initiated a project in the
early 1960s in response to a need for interdisciplinary cancer
research and coordinated care for cancer patients. The Institute
considered a large number of factors regarding the establishment
of comprehensive cancer centers and believed that the geographical
location should ensure that the maximum benefits of the centers
should be received by the largest proportion of people. The
approach used to assess the number and location of centers is based
upon the adaptation of a mathematical model and a computer program
designed for the solution of general "location-allocation" problems.
The objective of the analysis was to minimize costs expressed in
terms of transport distance. The population of the United States
was divided into distinct areas based on the boundaries of the 435
congressional districts, together with Washington, D.C., and
Puerto Rico. The total number of potential centers was calculated
by establishing a series of additional centers to complement
existing ones. Sixteen sites had already been selected and the
remaining series of 21 potential sites was analyzed. Travel distance
was determined so that centers should not be more than 1,000 miles
away from each congressional district, to be able to serve a
population effectively. Since accessibility increases directly
with the total number, an analysis was made to establish the point
at which population access is offset by the cost of setting up an
additional center.
14﻿42.	El-Shaieb, A. M., "A New Algorithm for Locating Sources Among
Destinations," Management Science, 20(2) (October 1973), 221-231.
The problem considered is the location of m sources among n
locations (m <T n) and the assignment of the location requirements
to the sources in order that the total transportation cost is
minimized. A branch and bound algorithm is developed to achieve
an optimal solution. Two lower bounds are developed for the
algorithm; one is efficient for small problems, while the other is
more efficient for larger problems. An illustrative example is
given, and a detailed application of the algorithm to the thirty
largest metropolitan centers in the United States is presented for
various number of destinations (10, 20 and 30) and sources (2, 4
and 6). The branch and bound algorithm was coded in Fortran IV
and tested on a number of problems. Some of the example problems
that were tested are included in this paper.
43.	Elzinga, D. Jack, and Donald W. Hearn, "The Minimum Covering
Sphere Problem," Management Science, 19(10), September 1972,
96-104.
The minimum covering sphere problem, with applications in location
theory, is that of finding the sphere of smallest radius which
encloses a set of points of En. For a finite set of points, it
is shown that the Wolfe dual is equivalent of a particular
quadratic programming problem and that converse duality holds.
A finite decomposition algorithm, based on the Simplex method of
quadratic programming, is developed for which computer storage
requirements are independent of the number of points and computing
time is approximately linear in the number of points.
44.	Erlenkotter, Donald "Economic Integration and Dynamic Location
Planning," Swedish Journal of Economics 74(1) (March 1972):	8-18.
Theories of economic integration are based in part on concepts
from location theory. Since static models are generally inadequate
for the dynamic context of integration, a dynamic version of Losch's
static location model is formulated, and its implications for
economic integration through a dynamic programming approach are
examined. Although the potential advantages from integration of
expansion and location policies are substantial, marginal returns
diminish rapidly as the size of the integrated group is enlarged.
45.	Fetter, Robert B., "Interactive Modeling of Health Care Delivery
Systems," Proceedings of the Colloquium on Systems Theory and
Scientific Management of Public Utilities, Center for the Study of
Health Services, Institution for Social and Policy Studies, Yale
University, New Haven, Connecticut, 1975.
Paper describes a modeling system to aid in the design and evalua-
tion of health care delivery systems on a national or regional
basis. It is the thesis of this paper that dealing explicitly with
15﻿the problem of resource allocation leads to more effective and
efficient delivery of health care and that a necessary information
structure for use in planning can be constructed based on a class
of models whose structure includes both macro and micro issues
and criteria in planning. The modeling system described provides
a language for making parameters. The models permit extensive user
interaction, and information generated can be displayed inter-
actively. Also, the models' own structure can be used to generate
detailed results. Examples of the models and results generated
are included.
46.	Fisher, W. D. (1956) "Economic Aggregation as Minimum Distance,"
paper presented to the Econometric Society Meeting (Detroit).
Abstract in Econometrica, 25 (1957):	363.
There is interesting early recognition that location problems and
grouping problems are two sides of the same coin. The author
states that the median lies in the convex bull of points in con-
tinuous space and markets are non-overlapping. He considers the
problems of scale economies and elasticity of demand with delivery
cost. Solution procedures are not given except for special cases.
47.	___________ (1969) Clustering and Aggregations in Economics,
Baltimore:	Johns Hopkins University Press.
Although not dealing with the L-A problem, this report summarizes
and presents several heuristic clustering routines in Chapter 5.
A 'Progressive Merger Routine' and a 'Lockstep Method' are described
and advantages cited. Cites many references.
48.	Fitzsimmons, James A., "A Methodology for Emergency Ambulance
Deployment," Management Science, 19(6) (February 1973), 627-636.
An important feature of emergency ambulance systems is that the
probability of any given unit being busy is dependent upon which
of the other units are busy. An analytical model incorporating
this feature is developed and found to be reliable in predicting
the entire response time distribution for an actual operating
system. A computer version of the model coupled with a pattern
search routine constitutes the ambulance deployment methodology
referred to as CALL (Computerized Ambulance Location Logic).
This methodology may be used to find the deployment of ambulances
that minimizes mean response time, as illustrated by an application
for the City of Los Angeles. CALL was used to select, from a
possible set of 34 firehouses, 14 for the location of emergency
ambulances. Although the objective was the minimization of mean
response time, the final deployment substantially reduced the
probability of excessive response times as well. Another ancillary
benefit from the use of CALL was a substantial smoothing of work-
load among the 14 ambulance crews.
16﻿49.	Francis R. L. , and J. N. Goldstein, "Location Theory:	A Selective
Bibliography, Operations Research, 22(2), (March-April 1974),
400-410.
This bibliography lists references to much of the work done largely
during the last ten years involving normative approaches to solving
location problems. Although it contains more than two hundred
references, the authors make no claims of completeness, as limit-
ations of many sorts precluded the development of a complete
bibliography.
The focus of the bibliography is primarily on work involving
normative approaches to solving location problems, and on work with
something of an "OR flavor." Further, references are limited
almost entirely to those written in the English language. For the
most part, references are to works appearing within the past ten
years; much of the earlier literature is either discussed or
references in readily available books. Also, there is an annotated
bibliography which complements this bibliography to some extent.
Most of the items listed tend to involve either discrete optimi-
zation (a facility is located at a site or it is not), continuous
planar problems (often convex optimization problems involving
normed distances), location problems on graphs, location-allocation
problems, or case studies. However, these categories are by no
means exclusive. Also, it appears that the discrete problems often
capture more of the reality of the situation than the continuous
problems do but are not generally as well solved. Heuristic
approaches abound for solving discrete problems and for location-
allocation problems.
50.	Frank, H. (1966) "Optimum Locations on a Graph with Probabilistic
Demand," Operations Research, 14:	409-21.
Deals with the network L-A problem and the minimax network L-A
problem (medians and absolute centres) under probabilistic con-
ditions represented by node "weights" which are random variables.
Only the case of single centres is dealt with, but the cases of a
minimum variance absolute median and optimal locations with sampled
populations are also considered. Many theorems are proved.
Several suggestions are made as to how to solve various problems
given the theorems.
51.	_______ (1967a) "Optimum Locations on Graphs with Correlated
Normal Demands, " Operations Research, 15:	552-57.
An extension of earlier work which again uses the example of traffic
centres in a network. This time the demands are not independent
random variables, but rather given by a joint normal distribution.
Again, theorems are given to allow solution techniques to locate
only single maximum probability absolute medians and centres, and
minimum variance medians.
17﻿52.	Fries, Brant E. "Bibliography of Operations Research in Health
Care Systems," Operations Research, 24(5), Sept. - Oct. 1976, 901-8.
This source lists a number of relevant studies, most of which
appear as separate listings in this bibliography.
53.	Gabriel, K. Reuben, Morris Hill, and Hubert Law-Yone, "A Multi-
variate Statistical Technique for Regionalization," Journal of
Regional Science, 14(1), 1974, pp. 89-106.
The bi plot of the singular value decomposition of matrices gives
a joint plot of row and column effects. Thus, the variation
among regions can be shown simultaneously with the variation
between variables. This technique was applied to the problem of
defining planning regions in Israel on a systematic basis. The
results allow key variables for regional models to be identified
and indicates the directions in which data definition and
acquisition can be improved.
54.	Gannen, C. A. "Fundamental Properties of Loschian Spatial Demand,"
Environment and Planning, 3(3), pp. 283-306.
Spatial demand is the aggregate demand expressed at a particular
location, under given supply conditions, by a set of individual
consumers whose location are dispersed over a geographical market
area. Although the concept of spatial demand has played a central
role in location and spatial economics, since its development by
Losch and his incorporation of it into his model of spatial com-
petitive equilibrium, a comprehensive rigorous characterization of
spatial demand does not presently exist. Accordingly, the contri-
bution of this paper is a quite general formulation of spatial
demand and a rigorous specification and evaluation of some of its
most important properties. Spatial demand is found to consist of
two components:	free spatial demand which corresponds to a
situation in which the geographic market area of suppliers do not
interact, and competitive spatial demand which corresponds to a
situation in which they do. This distinction allows the derivation
of conditions for, among other things:	1) existence of these two
components; 2) continuity of the spatial demand function and its
first derivative; and 3) convexity or concavity of the spatial
demand function. Such properties of spatial demand are basic to
an evaluation of the existence, stability, and general nature of
spatial competitive equilibrium.
55.	Garfinkel, R. S. (1970) Set Covering:	a Survey, paper presented
at the International Conference of the Institute of Management
Sciences (London).
Discusses form, applications, and some algorithms for the three
types of set covering problems:	the general covering problem, the
set partitioning problem, and the graph covering problem. An
extensive bibliography is included.
18﻿56.	Garfinkel, R. S., A. W. Neebe, and M. R. Rao, "An Algorithm for the
M-Medium Plan Location Problem," Transportation Science, 8(3),
(August 1974), pp. 217-236.
The problem is to locate a given number of plants along a road
network so as to minimize the total distance between plants and
warehouses assigned to them. It is modeled as an integer
programming problem, and the relaxed LP is solved by decomposition.
In cases of noninteger termination, the ILP is attacked using
group theoretics and a dynamic programming recursion. Computa-
tional results are given, which contrast this procedure to a branch-
and-bound approach.
57.	___________, "The m-Center Problem:	Minimax Facility Location,"
Management Science, 23(10), (june 1977), 1133-1142.
The m-Center Problem is to locate a given number of (emergency)
facilities anywhere along a road network so as to minimize the
maximum distance between these facilities and fixed demand loca-
tions assigned to them. Fundamental properties of the m-Center
Problem are examined. The problem is modeled using integer pro-
gramming, and is successfully attacked using a binary search
technique and a combination of exact tests and heuristics.
Computational results are given.
58.	Garfinkel, R. S. and G. L. Nemhauser (1969) "The Set-Partitioning
Problem:	Set Covering with Equality Constraints," Operations
Research, 17:848-56.
The problem is shown to be a member of the general set covering
problem but where each member of the set must be covered exactly.
Applications to problems including political districting are
discussed, and Garfinkel's Balas type branch-and-bound method is
described. Computational experience is given and compares
favorably with other codes.
59.	___________ (1970) "Optimal Political Districting by Implicit
Enumeration Techniques," Management Science, 16.
The first step uses a branch-and-bound search algorithm to identify
feasible (contiguous) solutions. Then the program scans these
solutions to find the optimal partition which features only a small
variance in population size. An example is given. Computational
results are favorable on small problems.
60.	Gass, S. I. (1968) "On the Division of Police Districts into Patrol
Beats," Association for Computing Machinery, Proceedings of the
23rd Conference (Las Vegas), pp. 459-73.
This paper applies the Weaver-Hess algorithm which was originally
a warehouse algorithm to the police beat problem. Five measures of
workload are used as the basic demand to be met. The actual popula-
tion is discussed at length.
19﻿61.	Gearhart, B. C. and J. M. Llittschwager (1969) "Legislative
Districting by Computer," Behavioral Science, 14:	404-17.
An extension of Thoreson and Liittschwager, this presentation
again assumes that equality of population contiguity and compact-
ness are most important. This heuristic is able to cope with some
more constraints imposed upon it in actual use in Iowa where
evidently many of its recommendations were actually accepted. It
uses small existing population units as its base. Several options
and the complete algorithm and extensions are described.
62.	Geoffrion, Arthur M. "A Guide to Computer-Assisted Methods for
Distribution Systems Planning," Sloan Management Review, 16(2),
Winter 1975, (no paging).
A great many computer-based methods have been devised for distri-
bution planning problems, especially those involving facility
location. This article develops a framework of criteria by which
to judge the practical usefulness of any such method and applies
it to obtain critical evaluations of the principal available
approaches.
63.	Giles, Robert H., Analyses of Populations, Health, and Environment
Relations in Virginia, Virginia Polytechnic Institute and State
University, Department of Fisheries and Wildlife Sciences. 1973,
Blacksburg, 225 p.
An overview of human populations, environmental factors, and the
ecology of health in Virginia is presented. Literature reviews,
surveys and analyses of existing data were incorporated in the study,
the results of which included three computer programs and four
data books, in addition to this report. It is concluded that
Virginia and Virginians are now experiencing the consequences of
crowding. The analyses presented suggest improved groupings of
data, reducing data categories and thus data collection and stor-
age problems. Equations developed provide the basis for a computer-
based mathematical model of the populations of the State and for
developing a simulator for determining consequences of changes in
systems from allocating resources or adopting policy. It is noted
that average annual mortality rates are increasing on a parallel
with population increase. Factors affecting natality and mortality
in planning districts are examined; it is observed that great
differences exist between mortality, births and other phenomena
from region to region. Dynamic regions (e.g., changing cancer-
specific, tuberculosis-specific, or infant-mortality reduction
districts) are suggested, and computer programs for determining
the boundaries of such regions are described and listed. It is
noted that birth rates exceed death rates in all planning districts.
Trends in childbearing are for younger mothers; delayed child-
bearing, a significant population control, seems unlikely to occur
20﻿soon. Based on data from the urban corridor of Virginia, it is
concluded that problems of interactive crowding and air pollution
will increase. An increase of 0.1 person per acre leads to
0.02 percent increase in the mortality rate. An increase of 0.2
person per acre is expected in Virginia's urban corridor by 1980.
Extensive discussions of population stresses and disease are
included. Data analysis techniques and study results are reported
in detail. A 16-page bibliography, supporting tabular data, and
computer programs and printouts are included.
64.	Goldsmith, Harold F. and Elizabeth L. Unger, Differentiation of
Urban Subareas:	A Reexamination of Social Area Dimensions. National
Institute of Mental Health, Population Research Section, Adelphi,
Md., November 1970, 54p.
Demographic and ecological dimensions that are useful in differ-
entiating among residential subareas of American cities and that
can be measured using available census data are identified from
relevant sociological and ecological literature, with explicit
reference to particular data from the 1970 Census which are
considered to be indicators of key dimensions discussed, i.e.,
social rank, life style, and ethnicity. Each of these standard
social area dimensions is evaluated in terms of its contribution
to an understanding of the association between behavior and resi-
dence. It is suggested that the standard components of the social
rank dimensions, i.e., economic class, social class, and information
status be considered separately. Lifestyle or degree of urbani-
zation is viewed as a composite dimension including the following
components:	family status, family life cycle stage (childbearing,
childrearing, childlaunching, post-childlaunching), residential
lifestyle, and perhaps, familism (degree of value placed on the
concept of family living). The ethnicity dimension is maintained.
Also included are discussions of residential instability,
feminine careerism, and area homogenity; high risk populations and
additional factors assumed to influence residential behavior are
mentioned. Census measures suggested as indicators of the various
dimensions are described in detail. Computer programs have been
designed to produce certain catchment area descriptions and age-sex
pyramids (illustrated in tables and illustrations) from the
second or fourth count census tapes for standard census areas.
The next step is viewed as a description of the specific behaviors
associated with a typology based on the dimensions presented, such
an analysis would include separate and comparative consideration
of local area, neighborhood, and household enironments. A six-
page bibliography is included.
65.	Goodchild, M. F. and B. H. Massam (1969) "Some Least Cost Models
of Spatial Administrative Systems in Southern Ontario," Geo-
grariska Annaler, 52:	86-94.
21﻿The capacitated transportation problem was used (Vogel's approxi-
mation) to solve the optimal regionalization under various capacity
assumptions given eight existing centers. Centers were relocated,
given the regions, using Kuhn and Kuenne's Newton-Raphson method
and moments of inertia compared to assess efficiency. The L-A
problem was then solved by the locate-allocate algorithm using
the transportion problem under different capacity constraints.
66.	Graves, R. J. and W. H. Thomas (1971) "A Classroom Location-Allo-
cation Model for Campus Planning,' Socioeconomic Planning Sciences,
5:	191-204.
A different approach to location and not really of the typical
L-A variety, the model presented in this paper begins with costs,
budgeted quantities, and academic locational preferences (ranked),
and maximizes the latter at a minimum cost. Only two general
locations are considered for each classroom. The model has been
used on a real problem, part of which is summarized. Sensitivity
analysis was performed. The approach is similar to many layout
models.
67.	Griffith, John R., Quantitative Techniques for Hospital Planning
and Control, Lexington, Mass:	D. C. Heath and Company, 1972.
(Chapter 4:	"Determining Population Service Areas and Calculating
Use Rates," pp. 65-91.)
Defines and discusses the concept of service area and explains
several methods of calculating it:	the Poland-Lembke Procedure
(equal likelihood service areas), relevance and commitment indexes,
and the use of birth and death rates as service area indicators
(the Drosness-Lubin technique). The application of these methods
in the urban setting is discussed. Also discusses the problem of
forecasting use rates and population trends for the area of interest.
68.	Grigg, D. (1967) "Regions, Models and Classes," in Models in
Geography, R. J. Chorley and P. Haggett (eds.), London:	Methuen,
pp. 461-509.
A general discussion of regionalization in geography. Nonstatis-
tical computer approaches are not discussed.
69.	Gross, P. F., "Urban Health Disorders, Spatial Analysis, and the
Economics of Health Facility Location," International Journal of
Health Services, 2;63-84, February 1972.
An analysis of a number of contemporary bed planning methodologies,
either in use or under development, reveals a number of limitations.
In particular, the linkage between the optimal size of health
facilities and the optimal size of their service areas has often
been ignored. An alternative approach to the size-location problem
22﻿is proposed. Working from health economics-medical sociology
concepts, a behavioral model of health services utilization is
proposed on the demand side. The interface to the supply side is
through a model of the consumer's perception of the total costs
to be faced in hospital care.
70.	Gupta, S. K. and C. Maier-Rothe (1969) "A Note on the Partitioning
of a Single Product Market into Territories of Outlets," Journal
of Marketing Research, 6:	232-36.
Derives circular and hyperbolic boundaries under linear costs and
demand elasticity, and generalizes results to n centers and
arbitrary cost functions. No pertinent references are given, and,
in fact, there is a great deal of literature on trade size and shape.
71.	Haggett, Peter, Locational Analysis in Human Geography, London:
Edward Arnold Publishers, LTD., 1965.
This book is a comprehensive discussion of the theory of locational
analysis. Part One of the books"Models in Locational Structure,"
discusses the geographically significant aspects of the main
theoretical locational models—both classical and modern (including
central place theory). Part Two, "Methods in Locational Analysis,"
reviews the tools available for attacking the problem of locational
structure, with a special chapter on the critical problem of
region-building.
72.	Hakimi, S. L. (1964), "Optimum Locations of Switching Centres and
the Absolute Centres in Medians of a Graph," Operations Research,
12:	450-59.
A very important paper giving rise to a list of further work too
numerous to list. Absolute medians and centres refer to points
on weighted graphs. The absolute centre is the minimax location;
the median will always be at a vertex whereas the centre may not
be. Cites previous use made of the property that optimal medians
must be at nodes in a network. This is a very important and
useful property of network problems. The centre is related to
the radius of a graph. Solution procedures are given and applica-
tions suggested.
73.	__________ (1965) "Optimum Distribution of Switching Centres in a
Communication Network, and some Related Graph Theoretic Problems,"
Operations Research, 13:	462-75.
Generalizes the author's earlier work. Shows that vertices need
only be considered in finding the p-median of a graph. The problem
is solved by enumeration. Then the author deals with "vertex
covering problem" which is the problem of placing the smallest
number of centres at vertices so that no part of the graph is
23﻿beyond a certain distance from a centre. It is solved by taking
all vertex coverings by the construction of Boolean functions of
the variables attached to the vertices. This is actually enumer-
ation of all feasible solutions. Several other related graph
theoretic - Boolean theorems are proved. The paper ends by
suggesting the p-centre problem (not confined to vertices) - where
p is given will be quite difficult to solve.
74.	hale, M. (1966) "Computer Methods of Districting," in Reappor-
tioning Legislatures, H. Hamilton (eds.), Columbus, Ohio:
Charles Merill Co.
Good historical background outline. Discussion of the Weaver-
Hess method (CROND), the Forrest method, and the Kaiser-Nagel
method. In addition to an evaluation, the paper discusses some
of the uses and abuses of the programs and their implications for
political values and government machinery. The book itself pro-
vides an interesting perspective.
75.	Hall, W. K. (1972b) "The Application of Multi-Function Stochastic
Service Systems in Allocating Ambulances to an Urban Area,"
Operations Research, 20:	553-70.
An interesting empirical study of the emergency system in Detroit.
Location was not handled in a sophisticated manner. Three policies
of allocation (to regions of heavy demand, uniformly over whole
area, or to one fixed station) were investigated along with many
other aspects of the system.
76.	Hamilton, C. Horace, "Procedures Used in the Preparation of a
Medical Service Area Map for North Carolina," Rural Sociology,
17(4):	367-371, 1952.
This paper describes the technique that was used to divide the
State of North Carolina into medical service areas. Four levels
of aggregation were used:	(1) local service areas; (2) community
hospital service areas; (3) regional hospital service areas; and
(4) medical education areas. The size of each service area varied
in direct proportion to the population of its center. A simple
formula was used to determine the breakpoint between two centers.
A second formula was then used to delineate service area boundaries
that were not on a straight line between the two centres.
77.	Harris, B., A. Farhi and J. DuFour (1970) Experimental Investi-
gation of a Combinatorial Problem, unpublished paper, University
of Pennsylvania Institute for Environmental Studies.
Deals with the problem of partitioning (optimally) a point set
in two space. A new method is devised for two groups. The
observation points are converted by duality into observation lines -
defining a tessellation of the dual plane. Sets of lines of the
original figure are converted by duality into the domains of the
tessellation. Very complex but interesting.
24﻿78.	Harvey, D. (1970) "Social Processes, Spatial Form and the
Redistribution of Real Income in an Urban System," in Regional
Forecasting, M. Chisholme, A. E. Frey, P. Haggett (eds.),
pp. 267-300. (The Colston Research Society) Handen, Connecticut:
Archon Books.
In the latter part of the paper, Harvey discusses the implications
interdependence in the location of multiple public facilities in
an urban area. Impure public goods - goods which are provided
for society but because of discrete locations are not equally
available to all - are discussed using notions of welfare economics.
79.	Harvey, M. E., "The Identification of Development Regions in
Developing Countries," Economic Geography, 48(3), 1972, pp. 229-243.
A methodology for deriving functional areas is outlined. Because
of the lack of interaction data in the developing world, it is
suggested that transport networks may be used as a surrogate for
such data. The method makes use of an adjacency matrix in which
nodes are weighted by the importance of their central functions.
Taking the case of Sierra Leone, an accessibility surface for the
whole country is derived, and five "functional" regional components
are isolated by principal components analysis. A nested regional
landscape (20 city regions within 5 functional regions) is depicted,
which has many advantages for development planning:	this landscape
is used in discussion of the problems of the transmission of growth
from core of periphery.
80.	Health Planning Research Services, Inc., A Manual for Evaluating
the Location of a Health Service. Vol. I: Introduction and Sample
Methodology, Vol. II: A Report on the State of the Art, Fort
Washington, Pa. Unpublished draft, 1976.
Volume I describes a method to determine current service areas and
service gaps, based on the use of a combination of travel time,
service capacity, and service area population. It then describes
the changes that a proposed service would make in accessibility,
defined in terms of either physical distance or travel time.
Volume II reviews basic concepts and the literature, and it includes
an annotated bibliography. Pages 34-49 focus on "Methods of
Delineating Service Areas."
81.	Heffley, Dennis R., "Efficient Spatial Allocation in the Quadratic
Assignment Problem" Journal of Urban Economics, 3(4), October 1976,
pp. 309-322.
Koopmans and Beckmann found that the competitive mechanism cannot
sustain an assignment of indivisible plants to spatially distinct
sites when intermediate product flows are required and transportation
costs between sites are positive. Hartwick claims to have found
25﻿valid counterexamples to this result. This paper demonstrates
that while counterexamples to the Koopmans-Beckman result do exist,
Hartwick's examples are inappropriate. A correct example of a
price-sustainable integer assignment is provided.
82.	Helbig, R. E., P. K. Orr, R. R. Roediger, and D. I. Steinberg
(1972) An Application of Linear Programming to Legislative
Redistricting Problems, paper presented at the Operations Research
Society of America Meeting (New Orleans).
A good summary of criteria and the problems associated with initial
areal (population) units. The algorithm is practically identical
to previous work of others in the field and, although each is
referenced, no credit at all is given to any of them. Slight
modifications in the heuristics are used to adjust size of district
differences at each iteration. An application of the algorithm to
redistricting in Missouri is presented.
83.	Heller, N. B., R. E. Markland and J. A. Brockelmeyer (1971)
Partitioning of Police Districts into Optimal Patrol Beats Using
a Political Districting Algorithm:	Model Design and Validation,
paper presented at the Operations Research Society of America
Meeting (Anaheim).
As the same criteria are important to police districts as political
districting, the algorithm due to Garfinkel and Nemhauser was
modified. The maximum deviation from the average of any district's
work-load is minimized. As in a similar approach by Gass several
demands for service variables were used. The empirical test of
the algorithm is to police districts in St. Louis.
84.	Helley, Walter, Urban Systems Models. New York:	Academic Press,
1975. (Chapter 7:	"Location of Public Facilities," pp. 133-154.)
Chapter considers methods of locating fixed public facilities for
customers who are dispersed about an urban area, so as to minimize
travel times or costs. Section 7-1 discusses the location of a
single facility, using calculus as the solution method. Section
7-2 discusses the districting problem. Section 7-3 discusses the
location of fire stations, including optimal district boundaries
and station sites, and interactions among districts. Section 7-4
is concerned with optimal transportation routing between given
origin and destination points. Section 7-5 discusses optimal
location of transit stations along a given route.
85.	Helms, B. P. and R. M. Clark (1971a) "Locational Models for Solid
Waste Management," American Society of Civil Engineers:	Journal
of the Sanitation Engineering Division, 97:	1-31.
They suggest there are four plant location approaches:	linear
programming, locational equilibrium models, trial and error, and
fixed charge giving a brief description of each. (There are errors
26﻿in the fixed charge formulae later corrected via errata). Several
approaches to L-A problems are summarized, all having dealt with
solid waste systems. They conclude that none of the approaches are
comprehensive enough although Skelly's comes close. The authors
are apparently unaware of the work of Marks (1969).
86.	__________ (1971b) "Selecting Solid Waste Disposal Facilities,"
American Society of Civil Engineers:	Journal of the Urban Planning
and Developmeent Division, 97:	443-51.
The authors elaborate on Skelly's approach to the fixed cost
capacitated discrete space problem. A clear outline of assumptions,
data requirements, and difficulties is given. A small sample
problem using realistic parameters is solved, but the actual
algorithm is not described. Straight line distance was used as an
approximation to actual routes.
87.	Hess, S. W. (1969a) "Realigning Districts by Computer," Wharton
Quarterly, 3:	25-30.
The basic locate-allocate algorithm is described and the similarity
of political districting and sales territory problems is stressed.
Applications of the districting programs as well as computational
experience is cited. Similar paper to Hess et al.
88.	Hess, S. W. and S. A. Samuels (1971) "Experiences with a Sales
Districting Model, Criteria and Implementation," Management Science,
18:	41-54.
Notes the similarity of legislature districting and sales territory
problems and presents a slightly modified form of REDIST (the basic
locate-allocate redistricting heuristic) called GEOLINE. Again
they use squared distances and centroids. They suggest starting
the algorithm with the present district allocation. Various demand
(or activity) measures were tested. Two actual applications are
briefly described.
89.	Hess S. W., J. B. Weaver, H. J. Siegfeldt, J. N. Whelan, and P. A.
Zitlau, "Nonpartisan Political Redistricting by Computer,"
Operations Research, 13(6), (November-December, 1965), 998-1006.
OR volunteers developed a compactness measured and a 'warehouse
location' heuristic to draw nonpartisan, constitutional political
districts. The heuristic maps compact and contiguous districts of
equal population. The minimization criterion and compactness
measure is population moment of inertia - the summed squared
distances from each person to his district's center. The districting
method is particularly useful when legislative impasse or indifference
forces courts to intervene. Federal courts have received a computer
plan for possible use in Delaware and have asked for computer
districts in Connecticut.
27﻿90.	Holmes, J., F. B. Williams, L. A. Brown (1972), "Facility Location
Under Maximum Travel Restriction:	An Example Using Day Care
Facilities," Geographical Analysis 4:	258-66.
The authors find that not all people need be served. An objective
of maximum utilization is solved by a linear program where the
problem deals with discrete areal units perhaps represented by
centroids. The model can locate new facilities given old ones but
the number of facilities (budget) must be specified. A case study
is presented. It is mentioned that capacities can not easily be
incorporated in the model.
91.	Ingram, D. R., "The Concept of Accessibility:	A Search for
Operational Form," Regional Studies, 5(2),, 1971, pp. 101-107.
The importance of the concept of accessibility in the literature
of urban studies requires that a method be found of describing
quantitatively the accessibility at a point. The paper is concerned
with, firstly:	a set of definitions related to the concept of
accessibility. A distinction is made between the relative
accessibility between two points and the integral, or total access-
ibility at a point. Secondly, various operational forms of these
definitions are illustrated with reference to the Hamilton, Ontario,
urban area. The derivation of the various measures that are
developed is discussed. A measure based on the normal, or
Gaussian, curve is recommended as the most suitable form for
determining the integral accessibility at a given point.
92.	Jensen, P. A. (1971), "Optimal Network Partitioning." Operations
Research, 19:	916-32.
The problem is seen as a partitioning problem within the class of
set covering problems. The algorithm is not designed for any
given problem and should be modified to take into account any
special properties. First the restricted sets of the network are
generated and then a dynamic programming formulation is given to
solve the recursive equation for a restricted set generated. The
algorithm is more efficient than several implicit enumeration
procedures.
90A. Hospital Survey Committee, Summary Report:	1974-75 Patient Origin
Study, Philadelphia/South Jersey Metropolitan Area (Philadelphia:
Hospital Survey Committee, 1977).
A summary report which describes the patterns of patient dependence
on hospitals in the Philadelphia/South Jersey metropolitan area as
derived from a study of the community-to-hospital patient flow
patterns in inpatient, outpatient clinic and emergency services
during a sample period. The data enable service area populations
28﻿to be defined and used in the determination of population-based
facility or service demands. In addition, the report presents a
method of determining institutional service areas through the
calculation of indices of commitment and relevance in each of
the communities served.
93.	Johnston, R. J. (1970), "Grouping and Regionalizating - Some
Methodological and Technical Observations," Economic Geography
46 (Supplement):	293-305.
An interesting discussion of the uses, misuses, problems and
advantages in using current techniques for grouping (taxonomy)
and regionalizing. He stresses using a method with reference to
a hypothesis. It is also pointed out that, while current methods
appear to guard against acceptance of a false conclusion, they do
not assure that a true hypothesis will be selected. An extensive
bibliography (largely relating to geography) is given.
94.	Juel, H., and R. F. Love, "Efficient Computational Procedure for
Solving the Multi-facility Rectilinear Facilities Location Problem,"
Operational Research Quarterly, 27 (3, Part II), 1976, pp. 697-703.
The multi-facility location problem with rectilinear distances is
considered. A necessary and sufficient optimality condition is
stated and proved. An algorithm is developed and computational
results are given. A new lower bound for multi-facility problems
with lp distances is also given.
95.	Lankford, Philip M.; and Gerald F. Pyle. Toward a Locational
Efficiency Model for Regional Medical Programs. Chicago Regional
Hospital Study, Working Paper VI.8, October, 1969, 20 pp.
The basic premise of this proposal is that there are geographical
variations in the prevalence of cancer, heart disease and stroke in
the Chicago area and that knowledge of present and projected spatial
patterns of morbidity will in turn be of primary importance in the
development of the locational efficiency model. The paper relates
methods for analysis of these three medical phenomena to the overall
locational efficiency model. This model for use in the Illinois
Regional Medical Program has four elements:	(1) a sub-model which
projects cancer, heart disease and stroke morbidity to 1980
(actual numbers of cases to be expected); (2) behavioral considera-
tions which have been added as a result of past studies of the
Chicago hospital system; (3) a hierarchical classification of the
various hospitals is utilized to determine different distances
which will be traveled by patients; and (4) on the basis of all
the above, actual allocations of patients from various communities
to hospitals.
96.	Keeney, Ralph L. "A Method for Districting Among Facilities,"
Operations Research, 20(3), May-June, 1972, 613-618.
29﻿A problem shared by many municipal agencies and many large
industries is:	Given n service facilities whose locations have
been set in a region, how should that region best be apportioned
among the facilities if each facility is to have primary
responsibility for service in one district of the region? This
paper presents a simple and intuitive procedure for determining
the boundaries of the service districts, using the criterion that
each facility serves all points to which it is closest in either
distance or travel time. The procedure is designed to be computer
rapidly, either by hand or on a computer.
97.	Khumawala, Basheer M.: "An Efficient Algorithm for the p-Median
Problem with Maximum Distance Constraints," Geographical Analysis,
4(4), October, 1973, pp. 309-321.
This paper presents an efficient heuristic method for a location
problem concerned with such facilities as medical clinics, schools,
parks, and libraries. The modified p-median problem can be stated
as follows:	given n communities requiring service, determine p of
these where facilities should be established, such that (1) the
average distance travelled per person from his community to his
nearest facility is a minimum, and (2) the distance a person has
to travel does not exceed a specific amount.
98.	Lankford, P. W. (1969 ( "Regionalization:	Theory and Alternative
Algorithms," Geographical Analysis, 1:	96-112.
A review and discussion of the merits of selected algorithms for
clustering and regionalizing. The distinction is made between
"single-linkage" algorithms (the most important) and neighborhood
limited algorithms-represented by an algorithm due to Neely which
is not published. Specifically the centroid based heuristic, and
Wards and Neely's heuristics are compared on several problems
and some advantages are noted of Neely's which attempts through
a definition of "relatively close neighbors" to pick the natural
grouping. Computational capability is also superior.
99.	Larson, Richard C. "A Hypercube Queuing Model for Facility Location
and Redistricting in Urban Emergency Services,” Computers &
Operations Research, 1 (1), March 1974, 67-95.
This paper develops computationally efficient algorithms for
studying the analytical behavior of a multi-server queuing system
with distinguishable servers. The model is intended for analyzing
problems of vehicle location and response district design in urban
computation of several point-specific as well as area-specific
performance measures.
30﻿100.	__________ "Approximating the Peformance of Urban Emergency
Service System," Operations Research, 23 (5), Sept.-Oct. 1975,
845-862.
This paper presents an approximate procedure for computing selected
performance characteristics of an urban emergency service system.
Based on a recently developed hypercube queuing model, the procedure
requires for N servers solution of only N simultaneous equations,
rather than 2N as in the exact model. The procedure relies on the
theory of M/M/N queues in which servers are selected randomly
and without replacement until the first available (free) server is
found. The underlying model is intended for analyzing problems of
vehicle location and response district design in urban emergency
services, includes interdistrict as well as intradistrict responses,
and allows computation of several point-specific as well as
area-specific performance measures.
101.	Larson, Richard C., and Keith A. Stevenson, "On Insensitivities
in Urban Redistricting and Facility Location," Operations Research,
20 (3), May-June 1972, 595-612.
This paper considers one class of problems associated with urban
service systems that dispatch vehicles from fixed facilities.
Given the limited resources available, one important issue is the
location of the facilities and the design of their response
districts to minimize average response time in the face of spatially
distributed demand patterns. Calculations with spatially homo-
geneous demands suggest that the mean travel time resulting from
totally random distribution of facilities in the region served is
reduced by only 25 per cent when the facilities are optimally
distributed. This apparent insensitivity of mean travel time to
facility location is pursued in detail by analyzing two classes of
systems involving a pair of facilities. In the first case, a
procedure is outlined for determining the optimal location of a
second facility, given a position for the first facility, when no
interfacility cooperation is allowed. In the second case, the
same region is examined, but allowing a form of interfacility
cooperation. These simple models suggest that, because of
insensitivities, it may not be necessary to quantize geographical
data finely and then to try laboriously to find the 'optimal'
solutions to redistricting and facility-location problems.
Redistricting and facility location based on rather crude assump-
tions and an awareness of some of the heuristic properties
illustrated by simple analytical models may yield mean travel times
very near the minimum possible.
31﻿102.	Lea, A. C., "Location-Allocation Systems:	An Annotated
Bibliography." Discussion Paper No. 13, Department of Geography,
University of Toronto, May 1973.
A very useful annotated bibliography, stated by the author to
include all the important English language literature on the topic
through 1971 and reasonably complete up to September 1972. The
headings of particular value for this project are found under
No. 10, "Other Variants and Applications of Location-Allocation
Analysis." They are "Public Facility Systems" and "Districting
and Regionalizing Problems."
103.	Learner, E. E., (1968) "Locational Equilibria," Journal of Regional
Science, 8:	229-42.
A locate-allocate heuristic is used to locate breweries in the
U.S. The continuous L-A problem with uniform and continuous demand
is next considered and an expression is derived for transport
costs as a function of the number of optimally located sources
based on a lower bound given by hexagonal regions. Indices of
efficiency are given for locating m facilities in a square, circle,
and triangle. An expression relating number of facilities to
demand and costs is derived. Non-uniform (continuous) demand
distributions are considered and it is indicated that political
and other boundaries necessarily increase transport costs. Some
interesting points and suggestions are made.
104.	Leyes, John M., J. Stuart Miller, Joyce Lofgren, Jeffrey White,
and Sara Goetz, "Delineation of Economic and Health Service Areas
and the Location of Health Manpower Education Programs." Wyoming
Univ., Laramie Div. of Business and Economic Research, November
1973, 240 p.
Interrelationships between the economic system and the health care
delivery system in the intermountain region (Wyoming and parts of
Montana, South Dakota, Nebraska, Colorado, Utah, and Idaho) are
explored in the context of the central place theory of using the
economic system as surrogate for the health care delivery system.
Seventy-four economic variables are identified to define the
economic system of the sparsely populated rural region, and 24
variables delineate the health care system. Seven community
groups of 538 communities selected to minimize communal differences
were tested against the economic and health care variables. It
was not possible to identify interrelationships between the economic
and health care systems; but, by using regression analysis, it
was possible to identify a close relationship between the two
systems. The use of the economic system as a surrogate has two
advantages:	data on the economic system are regularly collected
by the Federal Government and health system data are not; and
economic data have been collected over a long period of time,
allowing for trends to be observed. Because of a lack of manpower
education program data, it was not possible to test the
32﻿interrelationship of the health manpower education system with
the economic and health care systems. It is recommended that
because significant similarities exist between the health care
delivery and economic hierarchies, hierarchical service areas be
used for health planning rather than political units. Appendices
contain the economic and health care variables, the 538 communities
studied, health manpower education programs, health manpower
graduate distribution, health manpower hierarchies, and a
bibliography.
105.	Lowe, Timothy J. and Arthur P. Hurter, Jr., "The Generalized
Market Area Problem," Management Science, 22 (10), June, 1976,
1105-1115.
This paper considers the production-transportation problem of a
firm with a finite number of production (supply) facilities of
known location, facing demand distributed over a subset (market
region) of K-dimensional Euclidean space. The demand for a single
good is described by a demand density function which is continuous
almost everywhere on the market region.
Production cost functions exist at each production facility and
transport cost functions refer to delivering the good from the
production facility to the consumer. The objective is to set
production at each production facility and to select a supply
policy such that total production and transportation costs are
minimized.
The problem is formulated in its entirety and later simplified
through suboptimization. Necessary conditions for an optimal
solution are derived and economic interpretations are provided.
106.	Marks, D. H., C. S. ReVelle and J. C. Liebman (1970b) "Mathematical
Models of Location:	A Review", American Society of Civil Engineers:
Journal of the Urban Planning and Development Division, 96:	81-93.
A distinction is made between private sector and public sector
location models. Various concerns in location modelling are
discussed at length. Continuous space models are briefly mentioned
(including the Newton-Raphson interation method). The discrete
space problem is dealt with.
107.	Massam, Bryan H. (1972) The Spatial Structure of Administrative
Systems, Commission on College Geography, Resource Paper #12,
Association of American Geographers, Washington, D.C.
A paper with a reasonably broad scope. Much is a summary of
previous work especially by author and Goodchild. Part 3 discusses
various forms of the "Allocation-Location problem." Again the
locate-allocate algorithm is recommended for the capacitated
continuous space problem.
33﻿108.	Massam, Bryan H., and A. H. Burghardt (1968) The Administrative
Subdivisions of Southern Ontario:	An Attempt at Evaluation,"
Canadian Geographer, 12:	125-34.
Several administrative regionalizations are compared with respect
to their 'moments of inertia' about their present administrative
centres. Then given the regionalizations the medians were found
and compared to the actual centres using the measure of compactness
about the centre. The present systems are shown inefficient.
See subsequent work by Massam, Goodchild.
109.	Massam, Bryan H., and Michael Goodchild, "Temporal Trends in the
Spatial Organization of a Service Agency." Canadian Geographer,
15(3) 1971, pp. 193-206.
The rural operating areas of the Ontario Hydro Electric Power
Commission provide the data for a study of trends in spatial
organization through time. Locations of boundaries and service
centers have been changed each year during the study period
1948-67. Three aggregate measures have been defined to measure
the degree to which the system as a whole approaches an ideal
state of organization:	a measure of the efficiency of boundary
location, of service center location within the boundaries, and
of uniformity of scale between the operating areas in the system.
All three measures show a single trend toward greater organization
through the 20-year period. Discriminant analysis was used to
determine the weighting of the criteria used by the decision-makers
in the organizational process. The administrators involved in
these decisions show a considerable sensitivity to derived measures
of spatial efficiency in addition to their concern for the direct
scale quantities, as is indicated by the size of the loading of
the two discriminant roots on these variables.
110.	Meade, James N., "A Mathematical Model for Deriving Hospital
Service Areas." International Journal of Health Services, 4(2),
Spring 1974), pp 353-364.
The basic premise explored in this paper is that patient flow in
rural areas is based on the proximity to medical care. The hospital
is defined as the center of care, and hospital catchment areas are
defined by patient movements. A methodology is described to
analyze patient flow among an assemblage of hospitals. Finally, a
model which mathematically replicates patient movement is introduced
to act as an aid in the hospital planning process.
34﻿111.	__________ "Validity of Repeating Patient Origin Studies for Rural
Hospitals," Public Health Reports, 91(1), January-February 1976,
62-66.
Recent research in southwest Idaho and southeast Oregon indicates
that hospital service areas in this rural locality have not changed
over time. An examination of some variables that customarily
influence hospital service areas, such as number of physicians,
number of hospital beds, and size of population, revealed that,
despite sizable changes in all these variables, no appreciable
changes were noted in the spatial patterns of the hospitals'
service areas.
The 12-county study area includes approximately 42,000 square
miles inhabitated by only 265,491 people. The focal point of
hospital care in this region is Boise, Idaho, with adjacent smaller
centers.
112.	Mennes, L. B. M., "Planning for Regions and Centres," Economics
of Planning, 9(1-2) 1969, pp. 43-70.
National economic planning in the Netherlands has been framed
spatially within the context of a hierarchy model initially
elaborated by Tinbergen. This paper criticizes the model in
respect to:	1) the locational optimality it can achieve, and
2) its utility as a model for planning purposes. It is found that
the models of Tinbergen and Bos contain basic contradictions
between the spatial dispersion of industries and the allocation
industrial types among centers of different hierarchical order.
Current location patterns of production are serious constraints on
the utility of the hierarchical model for planning purposes. New
spatial models for allocating industrial and agricultural produc-
tion over space and among regions and centers, accounting for
transport costs, are developed and may feasibly arrive at a set
of optima (alternatives) using mixed integer programming methods.
113.	Meger, Carl F., "A Long-Range Selection and Timing Analysis
System for Facility Location," Management Science, 20(3),
November 1973, 261-273.
This paper describes the application of a mixed-integer program
to the problem of providing strategic decisions for the selection
and timing of a set of proposed facilities to serve distinct
geographic regions. Solutions to this selection-timing process
are obtained by the system presented so as to maximize the ratio
of total net revenues to total assets. In addition, a series of
optimal integer solutions to a large-scale problem is included
with discussion and analysis.
35﻿114.	Minnesota Systems Research, Inc., Planning Area Delineation:	A
Self-Contained Manual; Identification and Measurement of Catchment
Area Effects; Support Materials; Instructor’s Guide, March 1976.
The manual, Planning Area Delineation, contains several useful
chapters. Chapter IV, "Grouping Services," describes a method
for grouping medical services based on their frequency, risk,
intensity, and sophistication. Chapter VI, "Describing Catch-
ment Areas," explains ten methods for such description of existing
areas. Choice of method is to be based on urban or rural nature
of area, type of data available, and type of service being examined.
The methods described are basically different types of patient
origin studies. Chapter VIII, "Putting it All Together," leads to
a delineation of planning areas through examining existing catch-
ment areas, constraints, and proposals for change. The manual,
Identification and Measurement of Catchment Area Effects, describes
a method for assessing the effects of proposed changes in service
on the proposer's catchment area and on neighboring catchment areas.
115.	Manmonier, Mark S., "Maximum Difference Barriers on Alternative
Numerical Regionalisation Method," Geographical Analysis, 5(3),
1973, 245-261.
Most regionalisation methods successively group areas so as to
preserve maximum internal homogeneity at each stage. An alternative
objective might be to achieve the maximum difference across the
boundaries between regions. An algorithm has been developed to
reach this objective by first selecting the boundary which
separates the two most different areas and then extending this in
both directions along other abrupt boundaries until the map is
divided into two regions. Further subdivision is made using the
same method until the desired number of regions (measured by the
ratio of within-group variance to between-group variance) is
reached. Comparison with a more conventional hierarchical grouping
method (with continguity constraint) shows that the maximum
difference partition methods produces more significant barriers
but less homogeneous regions as expected. A program is available.
116. Morrill, Richard L., and Robert Earickson, "Variation in the
Character and Use of Chicago Area Hospitals," Health Services
Research, 3(3), Fall 1968, 224-238.
This report summarizes briefly three of the analyses of variations
made by the Chicago Regional Hospital Study in order to reduce a
large number of demographic and ecologic variables to a small
number of controlling variables. The correlations determined are
intended to be used in the construction of a descriptive simula-
tion model, from which can be generated a predictive model and,
eventually, a prescriptive model capable of testing alternative
future plans for the hospitals of the region.
36﻿117.
118.
119.
120.
__________ "Variations in the Character and Use of Hospital
Services," Health Services Research, 3:	391-399, 1968.
Two studies are briefly reported here:	(1) Estimation of the
hierarchy of hospital services - that is, the level of service
offered; (2) General classification or grouping of hospitals
based on hospital characteristics, service areas, and relative
location. The first study reveals significant distinctions in
the level of service offered; facilities and services, kinds of
resident and intern programs, size of medical staff, and overall
size are the major differentiating characteristics. The second
undertakes a principal-components analysis, in which many variables
involving hospitals and their patients are reduced to a few
major dimensions of variation and hospital groups are then
classified on the basis of these dimensions. Groups reflect
especially variation in volume of service offered, relative
location of hospitals and patients, and scope of service.
__________ Evaluating the Distribution of Hospitals—An Experimental
Model of the Locational Efficiency of Chicago Hospitals—Chicago
Regional Hospital Study, Working Paper #1.18, December 1968.
The study aimed at finding ways for cities and regions to evaluate
the adequacy of the present distribution of hospitals, and to
locate or relocate facilities to meet future needs. To create a
normative model several variables were identified:	the variability
of patients, hospitals and policies; and existing patient use of
the hospital system. The model developed was applied to the
metropolitan Chicago hospital system in the hope that it could
measure imbalances in the present distribution, and suggest
location shifts to better meet the needs of both the patients and
hospitals.
__________ "Hospital Variation and Patient Travel Distances,"
Inquiry, 5(4), December 1968, 26-34.
The paper explores whether the pattern of distances patients
travel to hospitals varies according to kind of hospital. A
principal components analysis is used to reduce 99 variables to
nine dimensions of highly interrelated variables. The analysis
of distances traveled reveals varying patterns for different
kinds of hospitals. Five classes of hospitals and one exception
were identified by the analysis, each with distinctive travel
patterns.
__________ "Locational Efficiency of Chicago Hospitals:	an
Experimental Model," Health Services Research, 4:	128-41,
Summer 1969.
An experimental simulation model is described by which imbalances
in the distribution of hospitals may be evaluated and location
shifts suggested to meet future needs. The model, partly deter-
ministic and partly probabilistic, is here used to project the
37﻿effects on patient travel of shifting capacity and of shifting
demand. Applied to the metropolitan Chicago hospital system,
the model results indicate that relocation of hospital beds would
considerably decrease patient travel, but that the same improvement
in patient travel and in hospital utilization could be achieved
with a far less radical and costly shift of beds, by relaxing
existing constraints of income and race.
121.	Morrill, Richard, Robert Earickson, and Phillip Ross, "Factors
Influencing Distances Travelled to Hospitals," Economic Geography,
46:	161-71, 1970.
Patient movements to hospitals reveal a rather regular pattern.
The rate of use of hospitals declines with distance, or more
precisely, with intervening opportunities. However, patients do
frequent hospitals farther away than they would if the distances
traveled were minimized.
Movement beyond intervening hospitals occurs for institutional,
social, and other reasons. Of course, one obvious reason may be
that hospitals cannot provide needed specialized care. In the
first section the travel impact of the fact that in most cases the
patient's physician, not the patient, chooses the hospital is
explored. In the second section the effect of racial, religious,
and income variations on the hospital trip is discussed. Analyses
are based on 1965 hospital discharge dates for Chicago.
122.	Morrill, R. L. and M. B. Kelley, "The Simulation of Hospital Use
and the Estimation of Locational Efficiency," Geographical Analysis,
2:	283-300, 1970.
A simulation model is developed for allocating patients to
physicians and hospitals.
Given patient and physician locations, patients are allocated to
physicians by a gravity-type formulation. The initial allocation,
however, represents a desired pattern of use and may result in
unacceptable excess demands on some physician locations and
deficit demand for others. A descriptive option of the model
employs and iterative allocation, shifting the excess patients from
overdemanded to underdemanded physicians, until demand and supply
at locations are in reasonable balance. A prescriptive option
uses a similar procedure but shifts physicians from initially
underdemanded to overdemanded locations, again until demand and
supply are in reasonable balance at all locations.
The same description applies to the allocation of patients to
hospitals, except that the outcome of the hospital allocation
depends on the results of the allocation to physicians, since the
physician as well as the patient influences the choice of hospital.
38﻿123.	__________ "Optimum Allocation of Services:	An Hospital Example,"
Annals of Regional Science, 3:	55-66, 1969.
A simple computer program is developed to determine the optimal
location of hospitals. The program first allocates services to
all communities which have a population in excess of a predefined
minimum.
Next, in descending sequence of population below the minimum,
communities are grouped with up to two adjacent communities if
these communities do not already have hospitals, and if they are
not closer to communities with hospitals. If the minimum threshold
is exceeded, a hospital is allocated, proportional in size to the
joint population of the communities. When no further combination
of two or three eligible communities can reach the minimum
threshold, all remaining communities are allocated to the closest
hospital, and the capacity of that hospital adjusted accordingly.
124.	Morrill, Richard L., and John Symons, "Efficiency and Equity
Aspects of Optimum Location," Geographical Analysis, 9(3),
July 1977.
Despite problems of defining efficiency and equity, it is useful
to assess the equity and efficiency implications of various
optimal location strategies. Most location criteria have been
efficiency oriented, and if there is are variability in density
or income, or very strong returns to scale, equity may be well
sacrificed.
125.	Morris, D. (1972) "Inclusion of Social Values in Facility Location
Flanning," American Society of Civil Engineers:	Journal of the
Urban Planning and Development Division, 98:	17-32.
Deals with the discrete space single location problem where the
area is gridded. Weighted objective decision theory is used to
create a contour map of social values which is converted to dollar
equivalent and is added to a contour map of costs including costs
to feeder lines and adverse terrain costs. It is assumed the
locator is willing to spend some money toward social costs incurred
An example of a small but realistic noxious facility is given in
an appendix.
126.	Morse, N. (1972) The Use of Facility Selection Location Algorithms
as Costing Devices Within Urban System Analyses, paper presented
to the Operations Research Society of America Meeting (New Orleans)
Stresses the importance of using location algorithms (including
his uncapacitated brute force approach) under a variety of cost
and other assumptions to get a feel for how the system behaves
with respect to service quality, system configuration, pollution
39﻿levels, etc. If costs are minimized in each case comparisons
between systems relate to service levels and can be evaluated
by decisionmakers on grounds other than wise use of funds.
127.	Morse, N. and E. W. Roth (1968) Systems Analysis of Regional
Solid Waste Handling, U.S. Public Health Service Pub. #2065, 1970,
pp. 62-123. Cornell Lab Report No. V.Y.-2485-G-1-1968.
(Especially Chapter 4 "A Facility Choice Model").
Several simple location models are dealt with initially and an
elementary discussion of service area boundaries and the allocation
phase ensues. A discrete space, uncapacitated, hierarchical model
where processing plants and disposal sites are chosen is presented.
It is noted, however, that a mixed integer program requires a
prespecification of disposal sites used by each processing facility.
So a complete enumeration approach was developed which, of course,
overcomes this difficulty. Evidently, reasonable computer times
are reported with the program (which has a printout listed in the
Public Health Service Publication) for problems with less than
10 integer variables. Problems relating to Buffalo are presented.
The chapter takes many pages to present a simple model. They note
that they have been unsuccessful in developing a good multi-time
period model but present a heuristic approach identical to that
of Marks and Jirka. Again they suggest complete enumeration in
tracing the path through the series of static optimal solutions.
128.	Navarro, V., "Methodology on Regional Planning of Personal Health
Services. A Case Study:	Sweden." Medical Care 8(5):	386-94,
1970.
The methods used by health planners in Sweden to regionalize
personal health services are presented. In the first part, the
methodology used to estimate future demand is described, and
special mention is made of the consumption unit index, by which
health planners estimate the effect of demographic development
upon demand for personal health services. The resources required
to meet this demand for each regional level are also defined, as
well as the methods by which that functional distribution is
reached. In the second part, the methods used to determine the
size of a region as well as the geographic distribution 'of resources
are presented. These methods, which take into account the national
and local demographic and socioeconomic development, consider
accessibility to the regional center as the determinant variable
to decide among different place alternatives for these centers.
According to a cost-efficiency approach, measuring efficiency in
terms of accessibility, the chosen alternative is the one which
minimizes aggregate traveling times and costs.
40﻿129.	Nemhauser, G. L., L. E. Trotter, Jr., and R. M. Nauss, "Set
Partitioning and Chain Decomposition," Management Science,
20(11), July 1974, 1413-1423.
There is given a finite set I and a family of subsets of I. The
problem is that of determining a minimum cardinality subfamily
that is a partition of I. A branch-and-bound algorithm is presented.
The bounds are obtained by determining chain decompositions of
directed acyclic graphs. The computation time required to determine
a bound is bounded by a polynomial in the cardinality of I. Some
computational experience is reported and relationships with other
methods are discussed.
130.	Nutenko, L. Y. (1970a) "A Mathematical Method for Partitioning
a Territory," (translated from Russian), Soviet Geography, 11:
48—89.
Measures of contiguity and compactness are developed and an integer
programming formulation of the political districting problem is
given. The centroid is used for the location phase. The algorithm
can cope with up to eight socioeconomic variables associated with
each basic population unit whereas he notes the Hess formulation
usually deals with population equality and can cope with 3 criteria
at most. His algorithm has been used to solve small problems with
up to 35 initial subareas. The compactness criterion is optimized
subject to linear constraints but the number of resulting regions
need not be prespecified. He notes that his approach is useful in
forming hierarchical regions.
131.	___________ (1970b) "An Information-Theory Approach to the
Partitioning of an Area" (translated from Russian), Soviet
Geography, 11:	71-75.
The districting problem is seen as one of general hierarchical
classification and formulated as a set partitioning problem or
hierarchical clustering problem. In graph theoretic terms a
directed "antisymmetric" and nonreflexive graph is formed and
partitioned. In the partitioning process information is gained
about the system but lost with respect to individual elements;
this, the criteria of information theory, can be used to evaluate
partitions and to help solve regionalization problems.
132. O'Brien, R. J. (1969) "Model for Planning the Location and Size
of Urban Schools," Socioeconomic Planning Sciences, 2: 141-53.
A rather complete spatial model of an urban school system.
Various submodels are described which are concerned with social
values, racial balance, staff arrangements, etc. Locations and
sizes of schools are handled as a complete enumeration of the
planners inputs. Various capacities and configurations would be
fed in as inputs in search of a distance minimizing system under
prevailing social constraints. Linear programming is used. Both
a cost and an effectiveness submodel are described.
41﻿133.	Orange County Health Planning Council, Framework for an Area
Health System, Tustin, Calif. 1975, 36p.
A framework for comprehensive health planning in Orange County,
California, is presented. The background of comprehensive health
planning in general, in California, and in Orange County is
outlined, and recent Federal developments, i.e., the Health Systems
Agency concept, are discussed. The structure, purposes, and
functions of the Orange County Health Planning Council are explained,
and the process of developing the health plan in outlined. In a
presentation of the elements in an area health system framework,
the major kinds of health services are described according to the
patient flow patterns which typically occur as individuals seek
treatment. Levels of specialization, primary, secondary, and
tertiary care, are defined and examples of services comprising each
level are given. The designation of Orange County's health service
areas is explained. The fundamental purpose of comprehensive health
planning in Orange County is reviewed, health goals are stated,
and long-range health system and environmental objectives are
specified.
134.	Palmer, D. S. "The Placing of Service Points to Minimize Travel,"
Operations Research Quarterly 24(1), March 1973, 121-124.
We wish to site a given number of offices or other service points
in a district, so that the mean traveling distance between clients
and their nearest office is as low as possible. It is shown that
the density of the offices should be proportional to the two-thirds
power of the population density. This theoretical result does not
take account of local conditions, but it may be made the basis
for a study on that ground.
135.	Parker, Barnet R., and V. Srinivasan," "A Consumer Preference
Approach to the Planning of Rural Primary Health-Care Facilities,
Operations Research, 24(5), September-October 1976, 991-1025.
A problem in planning the expansion of a rural primary health-care
delivery system is to determine the set of facilities to be added
to an existing system so as to maximize the incremental benefit
to the community subject to a cost constraint. The proposed
approach involves the following five steps:	(1) identification of
facility attributes relevant to patients in their choice of
health-care facilities, (2) modeling of an individual's overall
preference for alternate facilities as a weighted linear function
of these facility attributes, (3) transformation of each consumer's
preference model into a benefit function expressing the individual's
benefit in dollars/year for an existing or potential facility,
42﻿(4) provision of a method for determining the total incremental
benefit to the community from a set of proposed health-care
facilities, and (5) use of heuristic procedure to determine a set
of facilities that yields near-optimum total incremental benefit
subject to the cost constraint. A practical application of the
proposed approach reveals that the consumer preference model has
substantial reliability and predictive validity.
136.	Pernarowski, Leszek, "Regionalization Methods:	A Set-Theoretical
Evaluation," Geographia Polonica, 28 (1974), pp. 77-85.
In this paper the theory of sets is used for developing a method
of dividing territories into definite areas and subsequently
compared to previous qualitative and mathematical methods to
provide a final evaluation of them.
137.	Peterson, Robert A. "Trade Area Analysis Using Trend Surface
Mapping," Journal of Marketing Research, 11 August 1974,
338-342.
The purpose of this article is to approach the problem of
quantitatively defining the delineating trade areas from a
slightly different perspective, that of trend surface mapping.
In this manner, an objective, parsimonious map of consumer spatial
behavior can be constructed.
138.	Plessas, Demetrius J., and Eugenia S. Carpenter, "Empirical
Designation of Health Service Areas," Health Services Research,
10(4), Winter 1975, 333-348.
A method for identifying viable health service areas for a State
is described. A computer program was developed that evaluates
combinations of contiguous counties using a set of 56 variables
strategic to the construction of health planning areas and the
spatial context of health care delivery, in keeping with the
structural requirements of the National Health Planning and
Resources Development Act of 1974. The objective of the evaluation
is to minimize differences among planning regions and between
planning regions and the State.
139.	Pocock, D. C. D. and D. Wishart (1969) "Methods of Deriving
Multi-Factor Uniform Regions," Institute of British Geographers-
Transactions, 47:	73-98.
Discussion of clustering methods in general and a presentation of
new 'dense space method' which first identifies 'natural' clusters
and joins those that intersect. The clustering criterion is based
on distance squared. Three classification levels are suggested:
nuclear, basic and complete. Empirical data is clustered as examples.
43﻿140.	Poland, Eleanor, and Paul A. Lembeke, Delineation of Hospital
Service Districts, U.S.P.H.S. Grant No. W-14, 895 pp.; 1967.
A 1962 study "addressed to the problem of developing standard
method of planning the proper size, location and scope of hospitals
for a larger region, such as one or more States." Methods are
described in full detail for outlining mutually exclusive hospital
service districts and for certain useful indices for those
districts such as:	patient population ratios; the size and number
of general hospitals; the effective physician strength on the
hospital staffs; the distribution of medical specialties; and the
diagnostic conditions being treated in the hospitals of the
district and in other hospitals.
Data were collected in 1956 and 1957 through field trips to 354
hospitals and 1,036 nursing homes. Information was also obtained
through a one day census at each hospital, and several censuses
in the cases of smaller hospitals, until data on 48,000 hospitalized
patients were gathered. Also a representative sample of households
of the region was visited (3,684) and data obtained on 13,925
persons. Detailed data relating to Kansas (one of two States
studied) is given in the appendix with tables of data from the
tate's 130 hospital service districts to illustrate methods or
problems.
The statewide regional study was estimated to be less expensive
than a series of community studies and to be more productive for
planning purposes.
141.	Pyle, Gerald F. and Lauer, Bruce M., "Comparing Spatial Configura-
tions:	Hospital Service Areas and Disease Rates," Economic
Geography, 41:	50-68, 1975.
Much has been written about the spatial variability of health
conditions within metropolitan areas. The two most commonly used
surrogate measures are analyses of hospital service areas and
medical geographies using public health data. While each method
has distinct advantages, little has been accomplished with regard
to merging these techniques in attempts to understand more about
urban health conditions given the lack of private physician-generated
information.
This study is an attempt to develop such a methodology. As this
study unfolds, the basic issue addressed becomes:	Is it possible
to understand more about urban health conditions by simultaneously
analyzing inpatient travel and reported disease rates than can be
learned by independent investigation of these topics?
44﻿142.	Rand, G.K., "Methodological Choices in Depot Location Studies,"
Operations Research Quarterly, 27(1, ii), April 1976, 241-249.
Following a brief introduction indicating that the literature
on depot location is extensive, seven choices that face the
analyst are discussed. When choosing the objective the choice
is between cost minimization or return-on-assets criteria.
The potential locations can be anywhere (an infinite set) or
restricted to a feasible set. The search procedure employed may
be an optimizing one or a heuristic. The planning horizon needs
to be determined. The location of current depots may or may not
be built into the formulation of the model and the model may or
may not include depot capacities. Finally, the modelling of
local delivery costs is discussed.
143.	Ray, D. M. and B. J. L. Berry (1966) "Multivariate Socioeconomic
Regionalization:	A Pilot Study in Central Canaada," in Regional
Statistical Studies, T. Tymes, S. Ostry (eds.) pp. 1-48. Toronto:
University of Toronto Press.
Suggests a clustering procedure based on creating groups represented
by weighted centroids of points which are nearest - using squared
distance.
144.	Reisman, Arnold, and Mary Lou Kiley, eds. Health Care Delivery
Planning. New York:	Gordon and Breach, 1973, 383 pp.
A collection of papers is presented which apply operations research
and systems analysis techniques to planning for personal health
care service delivery. It is intended for health planners,
administrators, nurses, physicians, social workers, and paramedical
personnel responsible for health care delivery. The main purpose
of the book is to develop an understanding of the tools of
operations research/management science and the practical applic-
ability of the results. In each paper, a computer or a mathematical
model is developed, and specific applications of the model to real
life problems are made. Among the models presented are:	a
measurement model for planning and budgeting; utility measurement
in public health decisionmaking; biomedical research relevance
assessment; estimation of a potential hemodialysis population; and
optimal design of prepaid dental plans. Others include:	a study
of health care in western Pennsylvania; a computerized management
information system for human blood utilization; and transportation
systems for maternal and infant care project patients. Other
models are concerned with anesthesiology manpower planning, and
with developing a basis for appropriate training of anesthesia
personnel. The development of a community health service system
simulation model is provided, as well as a model of communicable
disease spread in a rural population. The two final chapters
present a planning model for delivery of health services and a
simulation of an ambulatory health care delivery system as
illustrated by a pediatric clinic. Each model has an accompanying
bibliography.
45﻿145.	ReVelle, Charles, Constantine Toregas, and Louis Falkson,
"Applications of the Location Set-Covering Problem" Geographical
Analysis, 8(1), January 1976, pp. 65-76.
The location set-covering problem is extended to apply to three
new situations. (1) The demands are assumed to occur continuously
along arcs of a network. (2) A mobile unit departs from one of
the locations to be chosen and picks up the demand, providing
service at a still more distant point. (3) New demands and sites
occur over time.
146.	ReVelle, Charles, David Bigman, David Schilling, Jared Cohon,
and Richard Church, Facility Location:	A Review of Context-free
and EMS Models, Health Services Research, 12(2), Summer 1977,
129-146.
EMS location models are those formulated to address specific
problems of emergency medical services systems; context-free
location models are those developed without reference to particular
applications. The literature on these two types of public facility
location models is reviewed, and the development of the maximal
covering model from several earlier context-free models is described,
with emphasis on problem statements and articulation of service
objectives. An application of the maximal covering model to fire
truck location points up the ability of this model to handle
multiple objectives; its ability to compare alternative solutions
gives it great utility for planning and evaluating EMS system of
a wide range of complexity. Potential applications of the maximal
covering model are discussed regarding EMS problems involving
multiple time standards and service objectives, location of special
equipment, and siting of fixed facilities.
147.	ReVelle, Charles, David Marks, and Jon C. Liebman, "An Analysis
of Private and Public Sector Location Models." Management Science,
16(1), July, 1970, 692-707.
Recent models of location are drawn together and compared as to
structure, criteria, and constraints. Private sector models are
distinguished as those in which the total cost of transport and
facilities is isolated as the objective to be minimized. The
solution techniques of six such models are discussed. Public
sector models are characterized by a criterion function involving
a surrogate for social utility and by a constraint on investment
in facilities or on the number of facilities. Five models with
this format are discussed and compared. The two types of problem,
location in the private sector and location in the public sector,
are seen to have the same conceptual foundation, but formats which
differ of necessity due to our inability to relate social utility
to dollar value.
46﻿148.	ReVelle, Charles, and Ralph W. Swain, "Central Facilities Location,"
Geographical Analysis, 2(1), January 1970, pp. 30-42.
"The problem of central facilities location consists of designating
m on n communities (m < n) as centers in such a way that the
average distance or time travelled per person is a minimum." The
procedure described here uses linear programming.
149.	Rhode Island Dept, of Health, Proposed Health Services Planning
Subareas for Rhode Island. Health Planning and Resources Develop-
ment, Technical Report No. 7, March 1977, (unpublished).
Paper defines primary, secondary, and tertiary health services
for planning purposes. Normative criteria for designating health
service areas are proposed, which relate to the three levels of
health services. Health service planning subareas (HSPs) are
made to correspond with existing boundaries of towns and Human
Service Districts, and HSPs reflect to some extent current
utilization patterns. Using the specified criteria, three sets
of planning subareas are defined, one corresponding to each level
of health services.
150.	Rider, L. L., "A Parametric Model for the Allocation of Fire
Companies," Rand Corp. April 1975, R-1615-NYC/HUD 19 pp.
Describes a method for the allocation of fire companies to regions
of a city. The method avoids the difficulty of predetermining a
specific allocation criterion by allowing the decisionmaker to
generate allocations for a complete range of criteria through use
of a single trade-off parameter. Among the criteria incorporated
by the model are the minimization of city-wide travel time, the
equalization of average regional travel times, and the equalization
of fire companies in New York City shows that, for one value of
the trade-off parameter, the model produces results that correspond
closely to the Fire Department's allocation policy.
151.	Rojeski, P. and Charles ReVelle (1970) "Central Facilities Location
under an Investment Constraint," Geographical Analysis, 2:	343-60.
The L-A problem is in discrete space and facilities have different
fixed costs and linear processing costs. There is an investment
constraint. The problem is formulated as a linear program and it
is shown how by varying the resource constraint slightly an integer
solution will result. An example is given. It is mentioned that
other integer solutions may not be found by the resource variation
method but the method's main advantage is the resource constraint
which should be useful for public sector decisionmakers.
47﻿152.	Ross, G. Terry, and Richard M. Soland, "Modeling Facility Location
Problems as Generalized Assignment Problems," Management Science,
24(3), November 1977, 345-357.
A variety of well-known facility location and location-allocation
models are shown to be equivalent to, and therefore solvable as,
generalized assignment problems (GAP's). (The GAP is a 0-1
programming model in which it is desired to minimize the cost of
assigning n tasks to a subset of m agents. Each task must be
assigned to one agent, but each agent is limited only by the amount
of a resource, e.g., time, available to him and the fact that the
amount of resource required by a task depends on both the task
and the agent performing it.) The facility location models
considered are divided into public and private sector models. In
the public sector, both p-median and capacity constrained p-median
problems are treated. (In the p-median problem exactly p on n
sites must be selected to provide service to all n. Each site has
an associated weight, e.g., its population, and it is desired to
minimize the respective service sites. The capacity constrained
p-median problem differs only in that there is an upper limit on
the sum of the weights of the sites served by each service site.)
In the private sector we consider both capacitated and uncapaci-
tated warehouse location problems in which each customer's demands
must be satisfied by a single warehouse. In addition, we show how
certain types of constraints limiting the site and capacity
combinations allowed can be incorporated into these models through
their treatment as GAP's. An existing algorithm for the GAP is
modified to take advantage of the special structure of these
facility location problems, and computational results are reported.
153.	Rushton, G. (1971) Map Transormations of Point Patterns:	Central
Place Patterns in Areas of Variable Population Density, paper
presented at the 18th Annual Regional Science Association Meeting
(Ann Arbor, Michigan).
A very interesting paper. A continuous and variable density
demand is represented by a trend surface equation which may be
complex. Then a prespecified number of sources are fitted to the
demand surface so that the capacities of the service centres (in
geography - central places) are roughly equal. A multidimensional
scaling algorithm is used to fit the source points. It is shown
how the resulting trade areas are in the shape of distorted hexagons.
The algorithms can locate several hundred sources in a good local
minimum configuration in very reasonable time. A printout of the
algorithm is given.
154.	Saaty, Thomas L. "Optimum Positions for m Airports," Naval Research
Logistics Quarterly, 19(1), March 1972, 101-109.
There is a set of n factories, which are located throughout the
United States. The wares produced by these factories are delivered
48﻿to local airports, where they are flown to one of two overseas
airports (Manila or Frankfurt), called distribution centers. The
centers distribute the wares to a number of neighboring areas.
Each center has a required tonnage to be delivered by air cargo.
The problem is to find the optimum location for a single airport in
order to minimize the total transportation costs; it is also
desired to find the optimum position for two airports, three
airports, ..., m airports by use of the minimum cost criterion.
The real situation requires the determination of subsets of one,
two, three-up to m airports, which yield minimum shipping costs
among a total of m airports already in existence. Many papers have
been written on this type of problem under various titles including
"warehouse" and "location allocation" problem. A bibliography is
included which may help the reader extend his knowledge of the
subject. Instead of a direct search for an exact numerical solution
to this nonlinear problem, we attempt to characterize the solution
under various hypotheses relating to the geometry in order to bring
out, whenever possible, some if its more interesting properties.
This characterization provides useful insight to the decisionmaker
in developing a strategy as to which existing airports to keep
operating, as to those whose potential capacity should be increased
and as to which ones to shut down.
154A.Jan Scheurwater, "Facility Location - A Review," draft paper
prepared for the 16th European Regional Science Conference,
August 24-27, 1976, Copenhagen, Denmark.
Recent developments in facility location research are reviewed
and a typology is presented for facility location problems which
distinguishes between consumer and producer oriented problems,
continuous and discrete space, static and dynamic problems, and the
type and number of facilities involved. The mathematical proper-
ties of facility location problems are discussed, and consumer
oriented facility location problems in continuous and discrete
space are reviewed in terms of possible solution techniques.
155. Schneider, Jerry B., "Measuring the Locational Efficiency of the
Urban Hospital," Health Services Research, 2:	54-69, 1967.
A measure of locational efficiency, the locations imbalance
vector (LIVOR) is developed and applied in the analysis of
nineteen hospitals which provide medical, surgical, and obstetrical
services in the Cincinnati metropolitan area. With only a few
exceptions, the locational efficiency of the individual hospitals
presently serving the Cincinnati area is found to be quite low.
Some possible policy implications of the study's measure of
locational efficiency are illustrated and its relation to a larger
question, that of explaining the pattern of occupancy pressure in
a metropolitan area, is also briefly discussed.
49﻿156.	__________ "A New Approach to Areawide Planning of Metropolitan
Hospital Systems," Hospitals, 42(8):	79-83, April 16, 1968.
The hospital service area concept is inadequate for planning
future capacity and spatial distribution of hospital facilities
in metropolitan areas, the author contends. He proposes an
alternative approach that deals with capacity estimation and
locational distribution as separate problems. He describes and
develops a "sector-gram" technique, based on methods of areal
statistics, to describe the spatial pattern of patients'
residences and hospital capacity in a consistent and standardized
manner. Despite its present limitations, the author suggests
that this technique offers the hospital planner an operational
method for planning an efficient locational pattern of hospital
facilities in metropolitan areas.
157.	Schneider, Jerry B., and John G. Symons, Jr., "Regional Health
Facility System Planning:	An Access Opportunity Model," Regional
Science Research Institute:	Discussion Paper No. 48, July 1971.
The purpose of this report is to describe a model designed to
assist a comprehensive health planning agency evaluate and plan
the growth of the health facility system in a large metropolitan
region. The general objective that the model is designed to
serve is that of improving the level of opportunity that consumers
have with regard to gaining access to the health care system.
In this sense, the philosophy of the approach is that the CHPA
should seek to encourage changes in the location and/or size of
health care facilities within a region such that the access
opportunity of consumers is improved as much as possible over
time.
The major concept upon which the model is based is that of
access opportunity. This term is meant to refer to the ease with
which the health care services in the region can be reached and
obtained by any resident of the region.
158.	Schultz, George P. (1969b) Facility Patterns for a Regional
Health Care System, Discussion Paper No. 34, Regional Science
Research Institute, University of Pennsylvania, Philadelphia.
The same basic conceptual framework as in Schultz's earlier work
is given except that now there is a hierarchy. The important of
planning and accessibility are stressed. Central place theory
serves as the conceptual basis of the model developed and the
theory is reviewed. The hierarchy of health care facilities and
external economies are stressed but specialization in different
kinds of health care is assumed away in a three level hierarchy.
After discussing all the realistic concerns, a great many
simplifying assumptions are made (e.g., uniform continuous demand)
and the optimal service area for each level is determined which
50﻿maximized per capita net benefit. As it is assumed that each
higher order facility contains the services of lower order
facilities, the well-established geometrical relations of central
place theory with its hexagonal lattices is invoked to supply
the optimal locations and nesting for service centers.
157A.Schneider, Jerry B., "Measuring, Evaluating and Redesigning
Hospital-Physician-Patient Spatial Relationships," Inquiry 5(2):
24-42.
This paper examines the spatial distribution of physicians'
offices and residences in the context of patient residential
pattern and the locational pattern of the general short-term
hospital system in Cincinnati. Two statistics, the center and the
measure of dispersion about the center, are presented for each of
four spatial distributions:	patients, by medical service and
residential location; hospital admissions, by medical service and
hospital location; physicians' offices, by medical specialty; and
physicians' residences, by medical specialty. The use of a
"sectorgram" is suggested as a more precise technique for measuring
the size, shape and directional orientation of a spatial pattern,
and its application in redesigning the system of obstetric services
in the Cincinnati metropolitan area is demonstrated.
159.	__________ "The Logic of Health Care Facility Planning," Socio-
Economic Planning Science, 4(3), September 1970, 383-393.
This study presents a logical model for the planning of health
care facilities. The objective is an optimal facility pattern,
that is, one which maximizes net social benefit for the population
of a metropolitan region. Several orders of services are defined
and a successively inclusive facility hierarchy assumed for the
provision of these services. A method is stated for determining
the optimal scale and service area for each order. Net social
benefit is computed from a set of basic functions which describe
the demand for services, the unit cost of services, travel cost
and benefits derived. The geometry of location and a region-wide
view of the system are discussed.
160.	Schuster, Stewart A., "Optimal Contiguous Area Location in
Geographical Parcel Configurations, Computers and Operations
Research, 2(2), September 1975, 101-108.
A definition of optimal area, path, and point sites is proposed
in terms of contiguous geographic parcels. This definition is
formalized using a linear graph model of a parcel configuration.
The graph model is used to develop integer programming formula-
tions of site location problems. The paper deals primarily with
the area problem showing that the problem of locating an area can
subsume the problems of path and point location.
51﻿161.
162.
163.
Scott, Allen J. (1969a) "Combinatorial Programming and the
Planning of Urban and Regional Systems," Environment and Planning,
1:	125-42.
The importance of combinatorial aspects of planning problems is
stressed. Some of the recent developments in integer programming
including branch and bound and backtrack programming methods are
reviewed. He suggests that the methods might be combined for
large problems so the former is used until storage is gone and
then the backtracking scheme would take over. The importance of
heuristics in solving these problems is noted and the algorithms
are classified generally into those that progressively alter an
initial feasible solution and those that start with any solution
and proceed to improve it by expansion or contraction. Illustra-
tions are given. Problems of direct relevant to spatial planning
are summarized under a network heading, grouping and partitioning
and miscellaneous problems. Under the second heading the discussion
centers on the L-A problem on a plane, on a graph; taxonomic
description and regionalization, and optimal districting and
apportionment. Examples are given of each case.
___________ (1969b) "On the Optimal Partitioning of Spatially
Distributed Point Sets," in Studies in Regional Science, A. J.
Scott (ed.) London:	Pion Ltd.
Using an example of two groups from a set of equally weighted
points, an exact solution procedure is developed for the continuous
space L-A problem. The centroid is used for the locate subroutine
for computational convenience. The method used is backtrack
programming. At each level of the tree a new destination point
is considered and either added to one group or the other. As the
method would be good for only small problems, an efficient but
rather complex heuristic is devised based on Scott. This heuristic
traces a locally optimal path through a completely connected
network representing the solution space through a completely
connected network representing the solution space of the problem.
In the course of constructing the partition, a shuffling reallo-
cation routine is entered at appropriate intervals. This
heuristic was tested against the locate-allocate heuristic and
found to be superior. In fact, it was 93% effective on more
than 1700 problems. Solution times were not reported.
__________ "Location-Allocation Systems:	A Review," Geographical
Analysis, 2(2), April 1970, 95-119.
A review of continuous space, discrete space and specifically
network oriented L-A models. Selections from the literature are
used as examples of the variety of problems (and applications)
that are possible. Mathematical formulations are given for a
52﻿variety of problems and some solution procedures are described
briefly. The relationship between a wide variety of problems
characterized by the classification scheme at the end of this
bibliography is shown. Such problems as the political districting
problem, the problem of taxonomy, and Hotelling type problems are
all shown to be part of, or related to, the general L-A problem.
An extensive but selective bibliography is provided.
164.	__________ "Dynamic Location-Allocation Systems:	Some Basic
Planning Strategies." Environment and Planning, 3(1), (1971), 73-
82.
The general location-allocation problem is defined. Then various
generalizations of the problem are indicated. The dynamic extension
of the location-allocation problem is shown to be of especial
interest and significance. Two major approaches to the formulation
and analysis of this dynamic problem are discussed. The first
approach makes no attempt to anticipate future events and thus
leads over the long-run to suboptimal solutions. The second approach
guarantees full optimality over the range of definition of the
problem. Some numerical examples are presented.
165.	______ (1971c) Combinatorial Programming, Spatial Analysis
and Planning, London:	Methuen Ltd.
The book brings together much of the material in Scott's previous
work. The book is well-organized to point out the relationship of
L-A problems to related combinatorial problems in spatial planning.
Chapter 7 deals with L-A problems and Chapter 8 deals with dynamic
extensions. One of the few published books dealing with L-A problems
at length.
166.	Seley, J. E. and J. Wolpert (1971) "A Strategy of Ambiguity in
Locational Conflicts," in Locational Approaches to Power and
Conflict, K. Cox, D. Reynolds and S. Rokkan (eds.) Beverly Hills,
Calif.:	Sage Publications.
The facility package concept makes explicit the political strategy
of recognizing that there are those groups that must be placated
and those that can be safely ignored through one mechanism or
another.
167.	Semple, R. K., C. E. Youngman, and R. E. Zeller, "Economic Region-
alization and Information Theory:	An Ohio Example," Dept, of
Geography, Ohio State University Discussion Papers Series, 28,
1972 (26 pp.)
Develops a statistic based on information theory which appears
appropriate for problems of multivariate economic regionalization.
The State of Ohio was partitioned sequentially into 2, 3, 4, 5, 6,
and lastly 7 economic regions on the basis of sectorial income
structure of individual counties and their relative contribution to
the state as a whole.
53﻿168.	Serck-Hanssen, J. (1970) Optimal Patterns of Location, Amsterdam:
North Holland Publishing Company.
A book that surprisingly enough does not deal at length with
usual L-A problems per se. The book is written by a mathematical
economist and takes the central planning perspective. It is
divided under four headings:	Weber Models, Programming Models,
Losch models, and an attempt to determine whether there should be
only one centre. Under the first heading great stress is put on
the effects of agglomeration on location (of largely single
plants). Although no attempt will be made to summarize the work
in the book, it will be noted that a variety of realistic price,
scale economy, agglomeration, and other assumptions are incor-
porated in a rather complete mathematical development that uses
Lagrangian multipliers extensively. Of particular interest to
the L-A modeller are Chapters 6, 7, and 8. The system of
incentives for promoting optimal location patterns in a decentral-
ized setting (Chpater 7) is particularly interesting. Throughout,
however, no integer programming solution procedures are given and
the theoretical level is emphasized.
169.	Shanholtz, Mack P., "Regionalization of Health Services," Virginia
Medical Monthly, 100 (July 1973), 652-657.
Suggests a regionalization pattern with five levels of health
services, from medical outpost to regional base hospital.
170.	Shanker, R. V. (1972) Facility Location and Planning via Cluster
Analysis, paper presented at the Operations Research Society of
America Meeting (New Orleans).
An operations researcher recognizes the similarity between clustering
and facility location. A modified k-means clustering technique
allows the center to locate in continuous space (an advantage).
It can cope also with facilities which are given a priori. He
notes that variable capacity constraints may be associated with
any facility and this makes the*technique particularly useful for
problems involving marginal costing and multiple periods. The
method can also cope with different metrics and a large number of
demand and source (clusters) points.
171.	Shanker, Roy J.; Ronald E. Turner; and Andris A. Zoltners, "Sales
Territory Design:	An Integrated Approach," Management Science,
22(3), November 1975, 309-320.
A sales territory design procedure should solve the dual problems
of boundary definition and call frequency. Furthermore, it should
be possible to base the design on several workload and potential
criteria. A procedure is presented which meets this specification.
54﻿It employs recent developments in set-partitioning and includes
a computer code which makes it possible to handle design problems
of realistic proportions. The outcome of the design procedure,
termed the market matrix, is a schedule which specifies which
salesman calls on each customer, and the sales call frequencies
which maximize the total sale from all territories. A case example
illustrates how the procedure is applied.
172.	Shannon, Gary W.; Rashid L. Bashshur; and Charles A. Metzner,
The Concept of Distance as a Factor in Accessibility and
Utilization of Health Care," Medical Care Review, 26:	141-61, 1969.
This article is a comprehensive and systematic review of the
literature on the concept of distance between patient and provider
as it relates to utilization of health services that was published
prior to 1969. Although a considerable amount of research has
been published since that time, this review is still highly
recommended as an introduction to this subject.
173.	Shannon, Gary W.; and G. E. Alan Dever; Health Care Delivery:
Spatial Perspectives, New York, McGraw-Hill Book Co., 1974.
This book extends and complements the traditional concepts of
medical geography with a spatial approach to the structure and
organization of medical care delivery. Overall emphasis is
directed toward current problems in the U.S. with additional
illustrations from other countries. Specifically, the text covers
spatial perspectives, model requirements, factors determining
physician location decisions, comparison of the spatial-functional
organization of the U.S. health delivery system with other
selected nations, and suggestions for improvement in the U.S.
health system.
174.	Shannon, Gary W.; James L. Skinner; and Rashid L. Bashshur;
"Time and Distance:	The Journey for Care," International Journal
of Health Services, 237-244, Spring 1973.
Time and distance, the two common measures of travel, are examined
as they relate to the use of various types of medical service
facilities, using correlation and regression analysis. Different
conclusions about the relative accessibility of health care
facilities for selected subpopulation groups result from the use
of each measure separately. Differences in travel patterns to
hospitals and those to physicians and dentists are observed when
measured by time alone or by linear distance alone. Methodologic
and health care implications are discussed.
55﻿I
175.	Shannon, Gary W.; Carl W. Spurlock; Steven T. Gladin; and James L.
Skinner, "A Method for Evaluating the Geographic Accessibility
of Health Services," The Professional Geographer, 27(1),
February 1975, pp 30-36.
Discusses the importance of geographic accessibility as a variable
in recent models to explain the utilization of health services.
The method described here describes how well an existing distri-
bution of health facilities serves a population and measures the
impact of any proposed change in the location and/or level of
health services on the geographic accessibility of those services.
"Briefly, the methodology includes: 1) the computation of the
rectilinear distance between the residences of any selected group
of potential consumers and the nearest provider of a designated
level of service; 2) graphic display of the cumulative percentage
(Lorenz Curve) of potential consumers against the distance to any
selected level of service: 3) statistical measurement of the
difference in geographic accessibility between any pair of curves
for a designated level of service."
176.	Shuman, Larry J.; C. Patrick Hardwick; and George A. Huber;
"Location of Ambulatory Care Centers in a Metropolitan Area,"
Health Services Research, 8(2), Summer 1973, 121-138.
A mathematical model is developed for determining the best locations
for prepaid group practice of HMO ambulatory care clinics within
a metropolitan area. Several possible forms of a function for
estimating enrollment are investigated on the basis of available
data, and the model is applied to metropolitan Allegheny County
in Pennsylvania to determine clinic locations under uncertainty
in enrollment estimates, using existing ambulatory care facilities
as potential sites. Both fixed and variable costs are considered
in the model, which projects the unit cost per plan subscriber
with various combinations of constraints and enrollment functions.
177.	Shuman, Larry J.; Harvey Wolfe; and R. Dixon Speas; "The Role of
Operations Research in Regional Health Planning," Operations
Research, 234-248, 1973(4), 22(2) March-April, 1974.
Operations-research workers have not met with much success in
being accepted as integral members of regional health planning
teams, owing in part to a lack of understanding by health planners
of the skills the operations researcher has to offer and in part to
the analyst’s inability to demonstrate that he can close the gap
between theoretical modeling and the implementation of his results.
This paper explores the growth of regional health planning in
the United States and highlights its important problem areas.
The literature of operations-research applications to health
56﻿planning is reviewed critically with respect to the feasibility
of models and the appropriateness of assumptions. Specific
problems with the types of studies currently in the literature
are identified and recommendations are made for improved
coordination between operations-research workers and health
planners.
178.	Smallwood, Richard D.; Edward J. Sendik; and Fred L. Offensend;
"Toward an Integrated Methodology for the Analysis of Health-Care
Systems," Operations Research, 19(6), October 1971, 1300-1322.
This paper introduces patient state and physician state of
information as two unifying concepts for the analysis of health-
care systems. The concepts are presented in the context of
four specific problem areas. As a demonstration of the utility
of these concepts, a technique for evaluating alternative facility
macroplans is described and some results from an actual application
are presented and reviewed.
179.	South Central Pennsylvania Health Planning Council, Patient
Origins in the Acute Care Hospitals and the Identification of
Medical Service Areas in South Central Pennsylvania, Harrisburg,
February 1975, 44 pp.
A patient origin survey was conducted of the acute care general
hospitals in an eight-county area. Data were obtained from 22
hospitals, representing 98 percent of the hospital beds in the
area. Admissions during the sample period totaled 12,975,
representing eight percent of the admissions for the year ending
September 1973. Patient addresses were coded geographically by
minor civil division, census tracts, county, and region. The
data then were examined to determine medical service areas. The
results indicated that county lines do not totally control patient
flow. Presence of natural barriers such as rivers or mountains
tends to direct patient movement in some cases, but accessibility
and availability of facilities and physician referral patterns
have greater bearing. The study resulted in the realignment of
10 acute care health service areas into seven medical service areas.
Maps showing acute care facilities by county and by medical service
area are included. Population and acute care bed totals are
given for each service area. Portions of this document are not
fully legible.
180.	Spence, N. A.; and P. J. Taylor (1970) "Quantitative Methods in
Regional Taxonomy," in Progress in Geography, Volume 2,
B. Christopher, R. H. Chorley, P. Haggett, D. R. Stoddard, (eds.),
London:	Edward Arnold, pp. 1-64.
A reasonably complete review of taxonomy in general, and taxonomy
as applicable to regionalization, is given. Under "Agglomerative
Procedures" methods of ordination, measures of similarity, grouping
57﻿strategies, and grouping and regionalization are discussed. Under
"Divisive Procedures" hierarchic division, non-hierarchic division,
and division and regionalization are discussed. The paper also
contains a description of two statistical tests of significance
and multiple discriminant analysis. The bibliography is extensive,
and most works are mentioned in the body.
181.	Studnicki, James, "The Geographic Fallacy:	Hospital Planning
and Spatial Behavior," Hospital Administration Summer 1975, 10-21.
Paper investigates influence of distance and travel time on where
hospitalization occurs. Research studies dealing with spatial
behavior of hospital inpatients are reviewed, and the overwhelming
conclusion from these studies is that geography is a poor means
for arriving at a useful denominator for hospital planning,
especially in metropolitan areas. The geographic planning unit
is considered useful only when there are only one or two facilities
serving a large geographic area or when areas compared are so
large assumptions about homogeneity of consumer and provider
elements can be made about them and their interactive behavior.
Distance research also indicated while physical distance is a
factor in frequency and chosen hospital utilization, its effect
is not consistent for all hospitals or all patients. Race, age,
ethnicity and socioeconomic status are variables that need to be
further researched. Implications of these findings for hospital
planning are discussed, particularly the influence of physicians
who have privileges at a number of different hospitals and can
control patterns of hospital occupancy.
182.	___________ "The Minimization of Travel Effort as a Delineating
Influence for Urban Hospital Service Areas," International
Journal of Health Services, 5(4), 1975, 679-693.
Using a study population of 16,080 live births occurring to
residents of Baltimore City in 16 hospitals in 1969, this research
measured the existing flow of these patients against the flow
"expected" in an optimal accessibility model (where each birth
would occur at the hospital with the shortest travel time to the
residence of the mother). The results of the study indicate that
there is a general pattern of distance minimizing in travel for
hospital admission with 50 percent of the births occurring to
women who travelled to one of the four closest hospitals of the
16 alternatives. However, a suprisingly large proportion (20
per cent) of the study population exhibited extreme spatial
inefficiency by traveling to those four hospitals of the 16
alternatives which were farthest from their residence. A step-
wise regression analysis identified five variables which best
explained variation in the spatial efficiency of these urban
obstetrical patients:	low hospital occupancy, high total hospital
admissions, average extra travel time potential (a measure of the
58﻿difficulty of "spatial choices" facing patients depending upon
the location of their residence with respect to the alternative
hospitals), race, and the importance of the hospital's obstet-
rical service (a ratio of births to total admissions).
183. Sultz, Harry A.; Frank J. Rens; and Donald A. Brothers. CHIP
System Based on Computer Graphics. State University of New York
at Buffalo. Dept, of Social and Preventive Medicine. 1971, 12 p.
System thinking and its application to health planning is discussed
in this report on the Community Health Information Profile (CHIP)
of western New York. Based on a combination of computer technology,
geographic theory, and graphics, the CHIP system utilizes health-
related data collected and retained in a number of agencies and
institutions. By using computer-drawn maps, the health planner can
locate health facilities within population concentrations of
greatest utilization, and identify population groups at risk.
Precise geographic identification is particularly important for
the analysis of data about small areas such as locally designated
special areas, hospital service areas, air quality control regions,
mental health catchment areas, and areawide health planning regions.
Basic demographic data, vital health statistics, facility and
manpower resources, and information on the various types of
community service programs is easily obtainable, and these combined
with clinical information, can be of great help to the health
planner. A technique developed by the U.S. Census Bureau called
Dual Independent Map Encoding System (DIME) for geographically
relating data from different sources is described. Figures
illustrating computer-drawn maps, mentioned in the text, are not
included. Portions of this document are not fully legible.
184.	Swain, Ralph W., "A Parametric Decomposition Approach for the
Solution of Uncapacitated Location Problems", Management Science,
21(2), October 1974, 189-198.
This paper presents a new approach to determining the optimal
facility locations for uncapacitated location problems in two
stages. First, it is shown that a subset of all solutions to
the uncapacitated public facility location problem can be obtained
by considering a closely related private location problem. The
exact nature of the problem is established using the Generalized
Lagrange Multiplier results given by Everett. In the second section
of the paper a particularly efficient parametric decomposition
algorithm, which is based on the results of the first section,
is presented. Computational results are given which summarize
the current level of experience with the algorithm.
59﻿185.	____________ "Central Facilities Location with Multiple Planning
Periods." April 1976, (unpublished).
A simple model is given for location of uncapacitated central
facilities in problem settings where several development periods
exist. The structure of the model permits it to be employed in
a number of situations where dynamic behavior exists. A column
generation algorithm is given which can be employed to solve the
problem and computational results using this scheme are briefly
summarized.
186.	Swoveland, C.; R. Ugeno, I. Vertinsky and R. Vickson; "Ambulance
Location:	A Probabilistic Enumeration Approach," Management
Science, 20(4 Part II), December 1973, 686-698.
In this paper a probabilistic branch and bound procedure is
developed for the ambulance location problem. Information on
system characteristics under various ambulance assignments and
dispatch rules was provided by a digital simulation. The output
of the simulation was also used to construct the objective
function for the optimal problem. The procedure was used to
determine ambulance locations in Vancouver, Canada. Solutions
obtained from the branch and bound procedure were verified by
simulation runs.
187.	Taliaferro, J. Dale; and W. W. Remmers, "Identifying Integrated
Regions for Health Care Delivery," Health Services Reports,
99(4), April 1973, 337-343.
Before health services can be planned and made available to
consumers, the regions for their delivery have to be defined.
Any health service facility planning to develop, contract for,
offer, or evaluate health services requires some knowledge about
the population to be served.
Regions for health planning or analysis can be defined by
(a)	arbitrarily dividing the area under consideration into units,
(b)	devising a system idealized on some variable such as the
travel time to the service, or (c) observing the actual behavior
of the inhabitants of the area in seeking services.
A systematic analysis of the observed data is necessary to convert
raw data on patients’ origins into regional definitions. The
mathematical procedures of graph theory provide a way of
operationalizing a set of rules so as to define highly integrated
regions of health care utilization.
60﻿188.	Tapiero, Charles S. "Transportation-Location-Allocation Problems
over Time." Journal of Regional Science, 11(3), December 1971,
377-384.
Paper presents a solution to the dynamic transportation-location-
allocation problem when the number of destinations and sources
are fixed integers. Linear and nonlinear programming are used.
Method appears more useful for private facilities such as ware-
houses than for public facilities.
189.	Taylor, P. J. (1969) "The Location Variable in Taxonomy,"
Geographical Analysis, 1:	181-95.
An interesting general discussion initially of the popular
geographical question:	Are locations unique? An assessment is
given of Berry's regionalizing approach. Alternate theoretical
constructs are described. Taylor then presents a modification
of the Rogers and the Tanimote grouping method which includes
location as the dominant variable. It is based on nuclear
clustering where certain units are isolated because of high
typicality scores based on location (the number of other units
with a certain distance) and a discriminant (statistical) space.
Then units are clustered about a nuclei according to an objective
function and the cluster is removed. After the process is
repeated and only a few units are left, these are allocated to
the nearest (locationally) cluster. A contiguity constraint
was added in defining typicality and in the actual grouping.
The method works well when the units are regular (hexagons were
used as an example)5 but modifications are necessary for irregular
base demands. Because the approach was difficult to program,
an example is worked by hand.
190.	Terrell, P., nd L. L. Blank (1972) Mathematical Formulation of
a Police Precinct Design Problem, paper presented at the Operations
Research Society of America Meeting (New Orleans).
The problem is formulated so as to minimize the average response
distance, but with a maximum distance constraint. Other
restrictions are also considered. An algorithm is developed
which considers all designs without violating constraints. The
sensitivity of different deployment stategies is studied to
determine the best design.
191.	Teitz, M. B. (1968a) "Toward a Theory of Urban Public Facility
Location," Papers of the Regional Science Association, 21:	35-51.
Teitz points out that public facility location theory is almost
nonexistent and that when public facilities are dealt with it is
usually in an unsophisticated manner. A careful comparison of
public and private locational systems is given and it is noted
61﻿that public systems are best characterized by multiple location
problems. He classifies outputs and associated facility systems
and further differentiates them by the geometrical nature of the
system (includes L-A) and other characteristics of public
facilities including nature of the regionalization. He discusses
the implications of "cost utility" analysis for public decisions
and the trade-off between decentralization and scale economies
under limited	budgets.	The difficulty in	assuming elastic demand
is noted.	He	then sets	up a model to maximize the use of a point
pattern of facilities which have uniform capacities. Then he
analyses some of the relationships between number of facilities,
cost and budget and scale of facilities. A maximization condition
is obtained under simplifying assumptions. The paper concludes
with an important discussion of the difference between short term
solutions	and	a dynamic	locational system	under increasing demand.
In short,	the	paper has	probably provided	a great deal of stimulus
for much of the research that was forthcoming despite the fact
that it is infrequently referenced.
192.	Thoreson, J. D. nd J. M. Liittschwager (1967) "Legislative
Redistricting by Computer Simulation," Behavioral Science,
12:	237-247.
Discusses legislative districting in general with emphasis on
the criteria. Two very heuristic algorithms are proposed. The
first aggregates small population units and the second, uniform
grid cells. One district at a time is formed. This crude method
was evidently used in Iowa with some success.
193.	Toregas, Constantine, and Charles ReVelle, "An Algorithm for the
Location of Emergency Service Vehicles," ORSA Bulletin, 20, Sup.
2, Fall 1972, B-468.
This paper presents efficient solution methods for location
problems which can be formulated as set-covering problems. The
basic formulation examines a n-node network where each node
represents a point demanding service; demand is assumed identical
for all nodes. The goal is to determine the facility pattern
which minimizes the number of facilities subject to the restriction
that no demand be more than s time or distance units from at least
one service facility. The maximum service time is justified as a
measure of system efficiency. An efficient reduction algorithm is
given for diminishing the size of the constraint matrix, and it is
shown that, in many instances, an optimal solution can be found
using the algorithm alone. The basic formulation is extended to
include the location of facilities in time where now demands
occur in future periods.
62﻿194.	__________ "Optimal Location Under Time or Distance Constraints,"
Papers of the Regional Science Association, 28 (Fall 1972),
pp. 133-143.
Authors discuss the general problem of the location of public
facilities and review several approaches to it. The problem
examined in this paper is stated, "Locate the minimum number of
facilities at possible sites on the network which will ensure that
users at each point of demand will find service no more than s
distance or time units away. Of course, s could be chosen
differently for each point of demand." A method for solving this
problem is described. The problem is referred to as a "set-
covering problem," the solution makes use of "reduction techniques,"
which were first introduced in the field of electrical engineering.
195.	__________ "Binary Logic Solutions to a Class of Location Problem."
Geographical Analysis, 5(2), April 1973, 145-155.
Facilities are to be located to serve a number of communities
represented by points on a plane. Distances or times between
communities are given. Facilities may be assigned to any
community point, but all communities must have a facility within
a specified maximum distance.
The problem is to locate the fewest number of facilities possible.
This 'location set covering problem' can be solved using a
graphical procedure, which may be implemented on a computer. The
method is illustrated. It is not infallible, but it is capable
of producing a valuable result:	the curve relating the minimum
number of facilities needed in any problem to the maximum distance
allowed. The method is applicable to regionalization when the
task is to group communities into the minimum number of regions
whilst no community is further than a certain distance from the
central community of its region.
196.	Toregas, Constantine, Ralph Swain, Charles ReVelle, and Lawrence
Bergman. "The Location of Emergency Service Facilities,"
Operations Research, 19(6), October 1971, 1363-1373.
This paper views the location of emergency facilities as a set
covering problem with equal costs in the objective. The sets
are composed of the potential facility points within a specified
time or distance of each demand point. One constraint is written
for each demand point requiring 'cover,' and linear programming
is applied to solve the covering problem, a single-cut constraint
being added as necessary to resolve fractional solutions.
63﻿197.	Transaction Systems Inc., Evaluation of Alternative Health Area
Definition Methods, Atlanta, Ga., November 1976, unpublished.
(NTIS PB-260-672).
Surveys the literature on geographic area definition for health
manpower planning purposes. Categories are:	overview studies,
administrative subdivision studies, shared needs studies,
environmental analysis studies, community/catchment studies,
consumer use studies, economics and transportation area planning
studies and manpower planning references tangentially related to
geographic area planning. Develops a "Health Care Commuting
Analysis" algorithm for definition of health manpower units of
analysis.
198.	Volz, Richard A., "Optimum Ambulance Location in Some Rural Areas,"
Transportation Science, 5(2), May 1976, 193-203.
As large medical centers become ever more capable, the existence
of well-equipped, well-staffed, and rapid emergency ambulance
service becomes increasingly important. This paper presents a
method for determining the optimum location of ambulance stations
to minimize the average response time to emergency calls. A new
point-to-point driving time model is introduced, and a computer
optimization algorithm used to determine the optimum locations.
A constraint that the average response time to any point in the
service area be less than some specified minimum is also considered.
The method is applied to Washtenaw County, Michigan.
199.	Wagner, J. L.; and L. M. Falkson, "The Optimal Nodal Location of
Public Facilities with Price-Sensitive Demand," Geographical
Analysis, 7(1), January 1975, pp. 69-83.
The models presented here are general; yet they are mathematically
equivalent to the plant location problem and are therefore amenable
to solution procedures developed for it. The models presented
here distinguish between two institutional environments that
reflect the degree of power of the consumer to choose which facility
to patronize. If consumers can be assigned arbitrarily to facilities
and can be denied service, then the environment is one of public
fiat. If consumers must be served at the facility of their choice,
then a "serve-all-comers" environment exists.
200.	Weiss, J. E.; M. R. Greenlick, and J. F. Jones; "Determinants of
Medical Care Utilization:	The Impact of Spatial Factors," Inquiry,
8:	40-57, 1971.
This study is concerned with the effect of distance between patient
and provider on the utilization of physicians' services in a
metropolitan area. The specific objectives of the study are:
64﻿(1) to examine patient travel patterns to determine the extent to
which utilization is affected by location of different physicians'
services within a given medical care system; (2) to examine
differences in travel patterns when patients have access to
otherwise equally available but spatially separate sources of
physicians' care; and (3) to examine the impact which distance
from residence to each source of care has on patient travel.
It is found that most physician visits made by the study population
are to the facility nearest to the patient's residence. The data
also suggests that when the differences in distance between
equivalent but spatially separate sources of medical care decline,
the proportion of visits made to the closest facility also declines.
201.	Wendell, Richard E.; and Arthur P. Hurter, Jr., "Location Theory,
Dominance and Convexity," Operations Research, 21(1),
January-February, 1973), 314-320.
This paper explores the nature of optimal solutions to a plant-
location problem on a plane under general distance measures. It
develops conditions that guarantee an optimal location of a
facility to lie in the convex hull of source and destination
points. The effect of restricting the solution to some pre-
determined set is explored. The development is based on a
generalization of Kuhn's characterization of a convex hull by
dominance. When a 'Manhattan' norm is employed, it is shown to
be sufficient to consider, as optimal locations, the finite
number of 'intersection points' in the convex hull.
202.	Wennberg, John; and Alan Gittelsohn, "Small Area Variations in
Health Care Delivery," Science, 182 October-December 1973,
1102-1108.
Health information about total populations is a prerequisite for
sound decisionmaking and planning in the health care field.
Experience with a population-based health data system in Vermont
reveals that there are wide variations in resource input,
utilization of services, and expenditures among neighboring
communities. Results show prima facie inequalities in the input
of resources that are associated with income transfer from areas
of lower expenditure to areas of higher expenditure. Variations
in utilization indicate that there is considerable uncertainty
about the effectiveness of different levels of aggregate, as well
as specific kinds of health services.
Informed choices in the public regulation of the health care
sector require knowledge of the relation between medical care
systems and the population groups being served, and they should
take into account the effect of regulation on equality and
65﻿effectiveness. When population-based data on small areas are
available, decisions to expand hospitals, currently based on
institutional pressures, can take into account a community's
regional ranking in regard to bed input and utilization rates.
203.	Wesolowsky, George 0.; "Location in Continuous Space."
Geographical Analysis, 5(2), 1973, 95-112.
The development of Weberian-type problems is reviewed. These
problems deal with the optimum location of new facilities on a
plane among existing populations or facilities (for instance,
the least transport cost location for a factory, given raw
material sites and a market). Subsequently, the problem of locating
a new facility among many-point populations is discussed, using
both Euclidean and rectangular distance metrics. The model is
generalized to locate several facilities simultaneously. The
problem of handling a very large number of population points is
met by dividing the study area into rectangular subareas where
each has a uniformly distributed population. A further modification,
useful in the location of essential facilities such as fire
stations, discards the objective of cost minimization and aims
simply to minimize the greatest weighted distance between population
points and facilities.
204.	___________ "Dynamic Facility Location," Management Science,
19(11), July 1973, 1241-1248.
This paper extends the static single facility location problem
to a model that permits location changes within a planning horizon
of r periods. An algorithm is developed to optimize the
sequence of locations in order to meet changes in costs, volumes,
and locations of destinations.
205.	___________ "Probabilistic Weights in the One-Dimensional Facility
Location Problem," Management Science, 24(2). October 1977,
224-229.
This note deals with the one-dimensional facility location model
in which the weights, which can represent either demand volumes
or demands and costs combined, are known only proabilistically.
The demand points themselves, or at least the "feeder" routes
to the demand points, for the facility are located on a straight
line which represents a road or some other transporation route.
It is assumed that the weights of the demand points have a multi-
variate normal distribution. The probability of the facility
being optimally located at any point on the route is derived; it
is shown that only the demand points have non-zero probabilities.
In addition, the expected value of perfect information (EVPI) is
found. In this problem, the EVPI is the expected difference in
costs between the actual best location and the optimum location
obtained by using expected weights. The EVPI, therefore, is the
maximum amount the decisionmaker should pay for information about
weights if expected values are acceptable as a decision criterion.
66﻿206.	Wesolowsky, George 0.; and William G. Truscott; "The Multiperiod
Location-Allocation Problem with Relocation of Facilities,"
Management Science, 22(1), September 1975, 57-65.
A dynamic or multiperiod location-allocation formulation is
developed from the static problem of locating G facilities among
M possible sites to serve N demand points. This dynamic model
provides a tool for analyzing tradeoffs among present values of
static distribution costs in each period and costs of relocating
facilities. The objective is to specify the plan for facility
locations and relocations and for attendant allocations of demands
which minimize these costs. Two methods of solution are presented.
First, a mixed-integer programming approach is used to solve sample
problems. From computational results reported for structurally-
similar problems, it seems that efficient general purpose codes
for this method would be capable of solving problems with at least
5 periods, 5 potential sites, and 15 demand points. The second
method, dynamic programming, is capable of increasing the size of
problems that are computationally feasible. The dynamic program-
ming, is quite attractive when the relative values of G and M
restrict the state space to a manageable size, and constraints on
the extent of location changes in each period limit the number of
alternate decisions. Possible extenions of the model and solution
procedures are discussed.
207.	Westwood, J. B.; "Location Analysis:	The Decisionmaking Process,"
International Journal of Physical Distribution 5(1), 1974-75,
22-30.
The progress of location analysis from its origins in analytical
geometry through the work of economic geographers to the current
emphasis on the systems approach to distribution is examined.
A variety equation model is discussed in which it is assumed that
the two dominant features of any location model are the search
procedure used in obtaining an optimum and the cost function used
in evaluating such a result. Location models are classified
using this concept, and a model of the decision-making process
developed to explain the types of variety reduction that takes
place and the mechanisms available within an organization to
secure such reduction are described.
207A.White, Andrew N., "Spatial Planning for Public Facilities:	The
Need to Integrate Location and Transportation Strategies,"
presentation to the 73rd Annual Meeting of the Association of
American Geographers, Salt Lake City, Utah, April 1977.
Considerable effort has been devoted to specifying models and
evaluation procedures for public facility planning, which is
currently oriented twoards structuring the service delivery system
for optimal provision. By misinterpreting the spatial interaction
67﻿basis for delivery, however, this narrow approach ignores the
process of service utilization as well as the transportation
options which might facilitate it. The aim of this paper there-
fore is to expand the framework for public facility planning by
identifying integrated spatial strategies. Such strategies include
both locational approaches which determine the spatial pattern of
service delivery and transportation approaches that seek to reduce
immobility and transportation disadvantages constraining service
users. This paper reviews these options and indicates how their
combined use in public facility planning more closely reflects
spatial realities and planning contingencies. The future use of
facility models and alternative spatial policies to improve the
effectiveness and equity of service delivery are also discussed.
208.	Wrobel, A. (1967) "Perspectives and Limitations of the Application
of Mathematical Methods in Economic Regionalization," Proceedings
of the Brno Conference on Economic Regionalization, Prague:	Czech
Academy of Science, pp. 123-27.
A discussion of the definition of a region and a summary of some
regionalizing procedures used by geographers. Stresses the
difficulties of providing initial areal units and aggregating
data to represent them. Discusses the difficulties of formulating
regionalization problems in mathematical terms. He notes that,
frequently, useful grouping criteria may be revealed only by a
grouping attempt.
209.	Yeates, M. (1963) "Hinterland Delimitation:	A Distance Minimizing
Approach," Professional Geographer, 15:	7-10.
Uses the transportation problem (where obvious assignments are
done manually) to analyze the efficiency of present school regions
in Wisconsin. The overlap between optimal regions and present
regions was found to be small. He notes that the method would be
useful in clustering and aggregation problems in general.
210.	Zimmerman, James P.; "Service Areas and Their Needs Must Be
Reassessed," Hospitals, 49(17), September 1, 1975, 46-48.
Hospitals are increasingly the focus of the health care system,
and new Federal planning legislation makes it even more important
for them to take a leadership role. If hospitals are to have
substantial responsibility for organizing community health services,
they must be able to define their service areas. For metropolitan
areas, suggests that hospitals use Sigmond's "index of relevance"
and "index of commitment," which are described. Recommends
collection of data by census tracts or other such standard areas.
Describes the use of these indices.
* U.S.G.P.O. 628-331/1302-2603
68﻿BUREAU OF HEALTH PLANNING
Health Planning Bibliography Series
Title
NTIS Accession No.
1.	Selected Bibliographic References for Community Health	
	Status Assessment	HRP-0300101
2.	Consumer Participation in Health Planning: An Annotated	HRP-0300201
	Bibliography	
3.	Services Shared by Health Care Organizations: An Annotated	HRP-0300301
	Bibliography	
4.	Women and the Health System: Selected Annotated References	HRP-0300401
5.	Selected Bibliographic References on Computerized Axial Tomography	HRP-0300501
6.	Certificate-of-Need/1122 Project Reviews: A Bibliography	HRP-0300601
7.	Community Nutrition in Preventive Health Care Services	HRP-0300701
8.	Methods for Setting Priorities in Areawide Health Care Planning:	HRP-0300801
	An Annotated Bibliography	
9.	Guidelines for Planning Health Services: An Annotated Bibliography	HRP-0300901
10.	Emergency Medical Service Systems and Health Planning:	HRP-0301001
	A Bibliography	
11.	Mental Health Planning: An Annotated Bibliography	HRP-0301101
12.	Certificate of Need Programs: A Review, Analysis, and Annotated	HRP-0301201
	Bibliography of the Research Literature	
13.	Planning for Outpatient Surgery Services: An Annotated	HRP-0301301
	Bibliography	
14.	Cost Containment: A Bibliography	HRP-0301401
15.	Hospices and Related Facilities and Services for the	HRP-0301501
	Terminally 111: Selected Bibliographic References	
16.	A Review of Planning Methods and Criteria for Neonatal Intensive	HRP-0301601
	Care Units: An Annotated Bibliography of the Clinical and Health	
	Literature	
17.	Methods to Determine Geographic/Population Boundaries for Specific	HRP-0301701
	Health Services: An Annotated Bibliography	﻿Metnoas 10 Determine ueograpmc/population Boundaries tor specific health services: An Annotated Bibliography
c
cc
ill
o
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Public Health Service
Health Resources Administration
Bureau of Health Planning﻿﻿