Who
Provides
Health
Care
to Children
and
Adolescents
in the United
States?
Barbara
Starfield,
MD, Robert
A. Hoekelman,
MD,
Marie McCormick,
MD, Paul Benson,
BA, Robert
C. Mendenhall,
MS
Christy
Moynihan,
PhD,
and Stephen
Radecki,
PhD
From The Johns Hopkins University School of Hygiene and Public Health, Baltimore; The
University of Rochester School of Medicine and Dentistry, Rochester, New York; The
University of Pennsylvania School of Medicine, Philadelphia; and The University of Southern California School of Medicine, Los Angeles
ABSTRACT. Face-to-face visits by children and adoles-cents in office-based practice in the mid-1970s were stud-ied. Pediatricians, family physicians, and general practi-tioners accounted for 35%, 6%, and 30%, respectively, of
all child visits. Although 40% to 45% of preventive and medical encounters were with pediatricians, only 12% of
visits for minor surgery, 20% of visits for psychosocial problems, and 9% of visits for combined medical-surgical reasons were to pediatricians. Only in very young children did pediatricians provide a substantial proportion of care for each of the types of visits. For some common
diag-noses (acne, refractive error) most care was provided by
specialists other than primary care specialists, but less than 16% of all preventive care visits (including routine eye examinations) was provided by specialists other than primary care physicians. A substantial proportion of the prenatal care and management of minor trauma was
provided by family physicians and general practitioners. Although the limitations of the study (including an av-erage response rate of 55%, exclusion of certain
special-ties and institutional physicians, sampling at different
times of the year, lack of control for area of location of practice, and lack of information about response rates of different types of physicians within each specialty) pre-dude definitive conclusions, the findings raise important questions for future study. Pediatrics 1984;74:991-997;
ambulatory care, child health care, pediatric practice, pe-diatric specialties.
More than 75% of the health care visits made by patients less than 19 years of age in the United
States
are
to physicians working in office-basedpractice.’239 Of these visits, more than one fourth
Received for publication June 30, 1983; accepted Feb 23, 1984. Reprint requesta to (B.S.) Division of Health Policy, The Johns Hopkins University School of Hygiene and Public Health, 615 N Wolfe St. Baltimore, MD 21205.
PEDIATRICS (ISSN 0031 4005). Copyright © 1984 by the American Academy of Pediatrics.
are to specialists other than generalists or pediatri-cians.’247 The National Ambulatory Care Survey provides some information about the nature of care that is delivered. One publication is devoted to care provided to children and youth.2 In this publication, in which data are presented in four age categories
less than 2 years, 2 to 5 years, 6 to 14 years, 15 to
MATERIALS
AND
METHODS
From 1975 through 1977, surveys were conducted to obtain information about the practices of physi-cians in 24 specialties. Several reports about the
survey method and data concerning individual
spe-cialties have been published,79 and there is also a publication’0 that summarizes the major
character-istics of practices and patients for each of the
specialties in the study.
The sampling frame for the study was the Amer-ican Medical Association’s master file of physi-cians. Categorization of each specialist was based upon the categorization in this file, which is based upon physicians’ designations of their own special-ties rather than on critei’ia such as Board certifi-cation or eligibility. The specialties included in the study were allergy, cardiology, dermatology, emer-gency medicine, endocrinology, family practice, gas-troenterology, general practice, general surgery, he-matology, infectious disease, internal medicine, medical oncology, nephrology, neurology, neurosur-gery, obstetrics-gynecology, ophthalmology, or-thopedic surgery, otorhinolaryngology, pediatrics, psychiatry, pulmonary disease, and rheumatology.
Response rates varied by specialty but ranged from 34% to 82% with a mean of 55%. (Response rates for the primary care specialties were 51% for pediatrics, 44% for family practice, and 36% for general practice. Completion of log diaries among respondents ranged from 86% to 88% in the three specialties in which this was examined (pediatrics, family practice, orthopedics.) In Table 1 are shown the number of encounters in each age group for each type of physician and the weights applied to calculate the estimates for the entire poulation of encounters by correcting for the different sampling proportions of the different type of physicians. (Survey data reweighted to obtain estimates for the population of encounters are shown in Tables 2 to 4.) The survey consisted of a questionnaire and a log diary that each participating physician com-pleted for a preassigned three-day recording period. In addition, a practice summary listed the physi-cians’ encounters for a seven-day period. The log diary obtained information about the age and sex of each patient, as well as diagnoses, diagnostic and therapeutic procedures, and disposition. The data in this paper are derived from this log diary.
This study involved the creation of a data file
TABLE 1.
All Encounte
Number of Office-Based Encounters Sampled and Weigh rs by Patients’ Age and Type of Physician*
ts for Estim ates of
Patients’ Age (yr)
Type of Physician
.
. . . Family General
Pediatrician . . . .
Physician Practitioner Other
N W N W N W N W
1 1-4 5-9 10-14 15-19
4,998 2.5 652 1.6 314 10.2 8,073 2.5 1,092 1.6 584 11.0 5,332 2.5 1,027 1.6 679 11.6
2,921 2.6 1,141 1.7 838 12.3 1,214 2.5 2,023 1.6 1,401 14.3
495 2,502
3,960
4,870 10,894
2.6 2.1
1.9
2.0 2.1
Total 22,538 2.5 5,935 1.6 3,816 12.6 22,721 2.1
* Abbreviations used are: N, number of patient encounters used in analyses, ie, not
annualized and unweighted; W, weight for annual estimates of all encounters by the population, in 1,000s.
TABLE 2. D
bulatory Patie
istribution of Encounters by Physician Specialty and nt Visits, United States, 1977*
Patients’ Age:
Am-Patients’ Age (yr)
Total
Encounters . . .
(1 000s) Pediatrician ,
% of Encounters with
.
Family General Physician Practitioner
Other Physician Specialist
Total
<1 1-4 5-9 10-14 15-19
18,097 69.5 33,191 60.0 30,290 43.4 29,446 25.5 49,351 6.1
5.6 17.7 5.2 19.4 5.3 26.0 6.4 35.0 6.5 40.6
7.1 15.5 25.3 33.0 46.8
99.9 100.1 100.0 99.9 100.0
Total 160,375 35.0 5.9 29.9 29.2 100.0
* These population estimates are the estimated number of annual encounters (in 1,000s)
TABLE 3. Percent of All Encou
1977, Children of All Ages*
nters by Speci alty and Type of Encount er, Unit ed States,
Type of No. in
Encounter Sample
Type of Physician
. . .
Pediatrician Family. General Physician Practitioner
Other Total
Preventive 11,146
Medical 30,744
Minor surgery 4,838
Majorsurgery 2,804
Medical and 1,047
surgical
Obstetrics 1,433
Psychosocial 962 Environmental/ 99
economic
44.2
40.0
12.4
1.1 8.8
0.4
19.5 10.4
4.2 35.8
6.7 30.3
6.7 31.7
1.0 3.3
6.4 12.7
5.9 23.7
4.2 16.5 15.6 30.7
15.8
23.0
49.1
94.7 72.1
69.9
59.8 43.3
100.0
100.0
99.9
100.1 100.0
99.9
100.0 100.0
* Percentages are population estimates and therefore are corrected for underenumeration
and missing data. Row totals do not always add to 100% due to rounding.
TABLE 4. Diagnoses and Who M tory Care of Children*
akes Them: Selected Principal Diagnoses in
Ambula-% of Total Attributable to Different Physician Type
No. in Pediatrician Family General Other Sample Physician Practitioner Specialist
Most frequent diagnoses
Medical/special examination 5,494 46.0 4.0 42.8 7.1
Otitis media 2,222 63.3 3.5 13.6 19.5
Upper respiratory tract 1,745 35.5 14.5 42.2 8.0
infection
Pharyngitis 1,396 59.0 8.5 26.4 6.1
Tonsillitis 406 42.4 8.7 35.2 13.7
Diseases of sebaceous glands 1,834 1.0 1.4 8.1 89.6
Pneumonia 1,459 98.8 1.2 .. . ...
Eczema/dermatitis 890 26.5 6.7 47.5 19.3
Refractive error 1,328 .. . . . . . 100.0
Diarrheal disease 790 62.2 4.1 33.7 ...
Prenatal care 1,245 .. . 11.6 35.1 53.3
Hay fever 884 13.0 3.8 14.1 69.1
Other viral diseases 897 25.7 6.8 20.8 46.7
Asthma 700 44.4 2.0 4.7 48.9
Medical/surgical aftercare 380 .. . 11.6 35.1 53.3
Other selected diagnoses
Sprained ankle 240 5.0 10.6 42.6 41.8
Other ill defined sprains 29 11.2 . . . 88.8 ...
Transient situational disorders 106 .. . . . . .. . 100.0
Behavior problems 126 63.5 . . . . . . 36.6
Neuroses 234 6.4 6.2 14.7 72.8
* Percentages are population estimates and therefore are corrected for underenumeration
and missing data. Row percents do not always total 100; the difference is due to rounding.
Ellipses indicate that less than 1% of encounters are associated with the diagnosis.
from each of the 24 specialty files. This file consists of all visits by patients ranging in age from 0 to 19
years according to type of physician; there were
more than 55,000 such visits. Analysis consisted of comparisons among pediatricians, general practi-tioners, family physicians, and all other specialists combined. For adolescents of ages 15 to 19 years,
visits to internists and obstetrician-gynecologists were also studied separately.
The data concern only ambulatory care visits to
physicians primarily in noninstitutional (office-based) practice. Information was obtained for in-patient encounters, but was excluded from this
study because hospitalizations are of relatively low
requested to record ambulatory encounters wher-ever they occurred (offices, clinics, or elsewhere).
RESULTS
Pediatricians accounted for about 35% of ambu-latory encounters with children and adolescents aged 0 to 19 years that were made to office-based physicians (Table 2). Family physicians and general practitioners accounted for about 36% of office
visits and other specialists accounted for about 29%. Only in the preschool period did pediatricians provide more than 50% of the outpatient care of children. Conversely, pediatricians provided little of the care to teenagers: 25% of the visits by 10 to
14 year olds but only about 6% of the visits by 15 to 19 year olds were to pediatricians.
In general, Board-certified pediatricians ac-counted for about 75% of visits to pediatricians, or about 25% of all visits. Even during infancy, a scant majority (51%) of encounters were with Board-certified pediatricians.
The proportion of all ambulatory visits of differ-ent types by children to office-based physicians in the United States according to the type of physician
is shown in Table 3. For all ages combined, pedia-tricians were involved in 40% to 45% of preventive and of medical encounters, about 12% of all visits for minor surgery, about 20% of visits for psycho-social problems, and about 9% of visits with a combined medical/surgical focus. As might be ex-pected, specialists other than primary care physi-cians accounted for most of the major surgery and for substantial proportions of combined medical-surgical visits and obstetric visits. They also pro-vided almost 50% of the minor surgery and ac-counted for about 60% of visits made primarily for psychosocial reasons. These patterns differed some-what in the age groups. For children aged 1 to 4 years and for those less than 1 year, pediatricians accounted for a much greater proportion of care in all categories: 66% to 75% of the preventive and medical visits, 25% to 40% of the minor surgery,
2% to 10% of the major surgery, 16% to 40% of the combined medical/surgical visits, and 33% to 90% of the psychosocial visits, respectively. For those children aged 5 to 9 years, the distributions of encounter types by physician type were similar to those for all ages combined except for a higher proportion (44%) of psychosocial visits to pediatri-cians. For children and adolescents aged 10 to 14 and 15 to 19 years, pediatricians accounted for an increasingly lower percent of visits for prevention; conversely, general practitioners accounted for
pro-gressively greater proportions of such visits. For adolescents aged 15 to 19 years, pediatricians were involved in less than 10% of all types of visits. In
this age group, 85% to 90% of visits of all types were to general practitioners or other specialists (including internists and obstetricians). In contrast to the other types of physicians, family practition-ers accounted for a relatively stable proportion of all types of encounters across the different age groups.
The 15 most common (principal) diagnoses in child health care according to type of physician seen are shown in Table 4. Because there were many different diagnoses recorded by physicians in the survey, only those that accounted for at least
1 %
of the visits in each age group for each specialty were examined. Therefore, the data are useful only to the extent that they provide information on the source of care for the most common specificprob-lems in childhood. Data in Table 4 include all childhood ages combined; however, when the ill-nesses are the same across the ages, the patterns are similar for each of the separate age groups. In interpreting Table 4, it is important to recognize that family physicians and general practitioners were sampled during October or July and Septem-ber, respectively, whereas pediatricians were sam-pled in November. This difference may account for a substantial part of the large concentration of pneumonia present in all age groups in the practices of pediatricians. To the extent that other diagnoses are seasonal, distortions in their distribution un-doubtedly occurred as well.
The percentages of diagnoses attributable to the different types of physicians do not change sub-stantially when second or third diagnoses are in-cluded. In most instances, the differences between percentage of principal diagnoses and percentages of total diagnoses are 1% to 3%. Exceptions to this percentage difference generalization include ec-zema/dermatitis (pediatricians 33.7%, general practitioners 41.3%), prenatal care (family physi-cians 6.0%, general practitioners 19.4%, other spe-cialists 74.6%), hay fever (general practitioners 20.8%, other specialists 62.7%), asthma (general practitioners 10.5%), medical/surgical aftercare (general practitioners 21.5%, other specialists 69.9%), transient situational disorders (pediatri-cians 11.6%, other specialists 88.4%), behavior problems (pediatricians 71.5%, other specialists
28.5%), and neuroses (general practitioners 9.4%,
other specialists 78.0%). The nature of the differ-ences suggests that there might be differences in
although it appears as a second or third diagnosis. Whether the differences in order of the diagnoses are due to differences in perceived importance to patients, to differences in the interests of the
phy-sician, or to true differences in the problems of
patients seen in the different types of practices cannot be determined from the data.
Despite the limitations of the data, some patterns are clear enough to be considered definitive. As might be expected, primary physicians rarely made the diagnosis of refractive error, even as a second or third diagnosis. Most of the care for diseases of the sebaceous glands (including acne) was concen-trated in the practices of specialists other than primary physicians. Much of the prenatal care, on
the other hand, was done by family physicians and
general practitioners, although not by pediatricians.
Virtually all of the primary diagnoses of transient situational disorder and the vast majority of all diagnoses of transient situational disorder were made by other specialists, although pediatricians diagnosed more than 60% ofthe childhood behavior disorders. Family physicians and general practi-tioners made few of these diagnoses, although they accounted for at least as much of the diagnoses of childhood neuroses as did pediatricians.
Sprained ankle (International Classification of Diseases [lCD] code 845) and other or ill-defined
sprains (lCD code 848) were diagnoses made more
often in older (10 to 19 years) than in younger children. The small proportion of care provided by pediatricians for minor injuries may partly be a
result of the concentration of very young patients in their practices. However, other data (R. Menden-hall and S. Radecki, unpublished data) indicate that injuries of all types accounted for only 3% of all primary diagnoses in the practice of all pediatri-cians, including those who were in institution-based practice. Even under the unlikely assumption that none of these diagnoses were made by the physi-cians in institutional practice and that other types of physicians saw none of the many other types of injuries including lacerations, sprains, contusions, or open wounds, pediatricians still would have ac-counted for less than 25% of all care for injuries. Although differences in the time of year during which pediatricians and generalists were sampled might have exaggerated the extent to which
gener-alists (as compared with pediatricians) accounted
for visits prompted by injuries, the inference that
generalists contribute a disproportionately large
share of injury care to children is supported by other data. Unpublished data from the National
Ambulatory Medical Care Survey (NAMCS) (1975
to 1976) indicate that pediatricians accounted for
only 21% of all office-based injury care to children
less than age 18 years whereas family and general
practitioners accounted for 44%. In that survey,
injury diagnosis accounted for 3.6% of all primary
diagnoses made by pediatricians. It may be, how-ever, that pediatricians did care for a greater pro-portion of minor injuries in the younger age groups that comprise the bulk of their practices, but none of these diagnoses individually were of sufficient frequency to be evident in this analysis.
DISCUSSION
The data in this paper present a view of child and adolescent health care provided by office-based physicians in the United States in the mid-1970s. Both this survey and the NAMCS present infor-mation that describes care provided in individual encounters rather than care provided to individual children over time. For example, individuals may be seen more frequently (either by self-initiation or
by physician suggestion) in the practices of some types of specialists than other types of specialists. If this is the case for particular diagnoses or types of diagnoses, then the distribution of care may be
only an approximation of the distribution of chil-dren and youth actually exposed to care by the different specialists. Thus, the implications of the data in this survey and in NAMCS pertain to the nature of child health care visits and not necessarily to the nature of health needs experienced by chil-then and youth. A study of the distribution of care provided to people over time requires a different type of sampling than the one used in this survey and in NAMCS.
The response rate in NAMCS is relatively high (>80%), and statistical procedures for sampling and making population estimates are rigorous. Surveys
such as the one from which data in this paper are
derived are useful and important to the extent that
they contribute types of data not included in
NAMCS (such as more detailed information on
diagnostic and therapeutic procedures) or provide
insights that help in the interpretation of NAMCS data. In this survey, more visits (>50,000) were sampled than is possible in NAMCS; this made it possible to describe care provided to different age
groups within the child population. Moreover, until
recently it has not been possible to obtain data on family physicians as a separate group because the
sample of family physicians in NAMCS was too
small. Another advantage was that physicians were
instructed to include all ambulatory care provided, not only that in their offices as is the case for NAMCS.
Even though the types of information that are the subject of this paper were found to be relatively accurate in a study of the reliability of the data,”
mentioned above. Response rates of primary care practitioners were of the order of 40% to 50% so that generalizations to the entire population of primary care practitioners is hazardous. In the re-liability study,” respondents were more likely than nonrespondents to have a lower patient volume and to be Board certified; the extent to which this was the case in the survey itself or the degree to which it was the case for each of the different specialties is unknown. Moreover, the number of physicians sampled, although generally adequate for popula-tion estimates, was small compared with the Uni-verse of physicians. Therefore, estimates of rela-tively infrequent events have a wide margin of possible error when projected to the population of physicians. For example, the range of standard errors for the data on diagnoses is from about 0.1% to about 3%; the frequency of diagnoses in that range may be considerably over- or underestimated.
Another limitation results from the exclusion of some specialties from the survey. In contrast to NAMCS, which includes a sample of all office-based physicians, this survey excluded certain sub-specialties, among them child psychiatrists, pedi-atric allergists, and pediatric cardiologists. The ex-clusion of these specialists overinflates the relative proportion of children with certain types of prob-lems (such as behavior problems) who are seen by general pediatricians. On the other hand, their ex-clusion has the effect of underestimating the pro-portion of children with these problems who are seen by physicians likely to have at least some pediatric training. However, in instances for which the data are comparable with those in NAMCS, the findings are similar. For example, in the NAMCS survey,2’” 43.7% of visits in 1975 of children less than age 15 years were to pediatricians compared with 47.9% for this survey.
Moreover, the exclusion of institution-based phy-sicians from analysis may distort the extent to which physicians with at least some pediatric train-ing provide care for children. A greater proportion of pediatricians (44.9%) were estimated to be in institution-based and in “other” practice locations than was the case for family physicians (29.2%) or general practitioners (12.3%),b0 a fact that suggests that inclusion of institution-based physicians might contribute more child visits to pediatricians than to generalists. However, three in five (59%) of the pediatrician respondents in these categories were in “other” (rather than institutional) locations and hence not in sites where they would be likely to be providing direct care to children. Furthermore less than one fifth of all visits by children are to hospital outpatient departments and emergency rooms. Not all of these visits are with institution-based physi-cians and a certain (but unknown) proportion are
for diagnosis and treatment of unusual problems in pediatrics rather than for the common problems that are the subject of this report. Thus, although the data are not representative of all ambulatory care provided to children and may underestimate the proportion of care provided by pediatricians, the findings with regard to the specific types of information in this paper are not likely to be sub-stantially divergent from reality for the population in general.
Because this survey was conducted several years ago, it is possible that the current practice of pedi-atrics differs from that described for that time. Recent data from the NAMCS suggest, however, that there is not likely to have been much change with regard to the subjects of concern in this paper. For example, the percent of visits with the principal diagnosis of injury or poisoning in the practice of pediatricians was 4.7% in 1975 and 4.6% in 1980 to
1981.12
Despite these limitations, some of the study’s findings are so striking that the following appear
to be reasonable subjects for more definitive study:
1.
Specialists other than pediatricians, family physicians, and general practitioners appear to pro-vide little primary care to children, at least with regard to care that is considered preventive in na-ture.2. Pediatricians, as compared with family
phy-sicians and general practitioners, appear to provide a relatively small percentage of the minor surgery or care for the predominant injuries in childhood.
3. Pediatricians provide most of the care to
in-fants and more than 50% of the care to preschool-aged children. As more than 50% of the care pro-vided to infants (as compared with approximately 20% in other child age groups) is preventive, 44% of all preventive care to children is provided by pediatricians.
4. Pediatricians appear to account for a relatively
small proportion of the “psychosocial” visits by children.
Expected small surpluses of pediatricians, as
pre-dicted by the Graduate Medical Education National Advisory Commission, provide pediatric educators the opportunity to initiate studies that will provide more definitive information than was possible with
this survey, and to assess the appropriateness of
pediatric training in meeting the needs of children. The American Academy of Pediatrics’ Task Force on Pediatric Education recognized that acute med-ical illnesses are declining in relative frequency.
Because psychosocial problems are assuming
greater importance, the Task Force on Pediatric
Education’3 recommended greater emphasis on
find-ings do suggest that pediatric practice is largely oriented toward preventive care and care that is traditionally “medical” in orientation. Inasmuch as the survey could not determine the extent to which the practices of pediatrician-generalists focus on unusual or more severe problems, subsequent stud-ies might determine whether pediatricians do see a greater variety of more uncommon or more serious illnesses, might determine whether these are seen primarily by pediatric generalists or by pediatri-cians with subspecialty training, and might deter-mine if it is because pediatrician-generalists are more likely to make unusual diagnoses, because parents selectively seek their care when a condition
is perceived as unusual, or because children are
referred to these pediatricians by other types of
physicians. Subsequent studies should also examine
care provided for injuries, for it may be that pedi-atric residency training programs should better pre-pare practitioners to deal more adequately with the minor surgical aspects of at least some of these problems. Answers to all of these questions will require studies of a design different from the ones
used in this survey and in the NAMCS, as the
frequencies of the individual events are too small to be detected reliably by the procedures used for sampling. In addition, subsequent studies should include area of practice (region and urban-rural location) as a control variable, as different special-ties do distribute differently by these characteris-tics.’#{176}
Deficiencies in child health care as of the early
1980s, as documented in part by Budetti et al,’4
provide additional impetus for a careful evaluation of the role and training of the pediatric practitioner. These deficiencies are most pronounced in central cities and rural areas and among the poor, blacks, and adolescents. These are the groups in the pop-ulation with the greatest unmet need, particularly with regard to psychosocial problems and
inju-ries.’’7 Pediatricians are much less likely than
family practitioners to provide care in rural areas,’8 perhaps, at least in part, because training prepares them less adequately to practice where there are not a variety of types of specialists to care for the nonmedical and nonpreventive care needs of chil-dren. To the extent that this is the case, the findings of this national survey of child health care may
suggest areas requiring attention by medical
aca-demia.
ACKNOWLEDGMENT
This work was supported by grant no. 5240 from The
Robert Wood Johnson Foundation to the Ambulatory
Pediatric Association. The views expressed in this paper are those of the authors and no official endorsement by The Robert Wood Johnson Foundation is intended or
should be inferred.
REFERENCES
1. Better Health for Our Children: A National Strategy, The Report of the Select Panelfor the Promotion of Child Health, Vollil, US Dept of Health and Human Services, publication No. 79-55071. 1981, p 239, 247
2. Ambulatory Care Utilization Patterns ofChildren arid Young Adults: National Ambulatory Medical Care Survey. United
States, January-December 1975, US Dept of Health, Edu-cation, and Welfare Series 13, No. 39, DREW publication No. (PHS) 78-1790. Vital and Health Statistics, August 1978
3. Aiken L, Lewis C, Craig J, et al: The contributions of specialists to the delivery ofprimary care: A new perspective.
N Engl J Med 1979; 300:1363-1370
4. Starfield B, Hoekelman R, McCormick M, et al: Styles of
care provided to US children. J Fam Practice, in press 1984 5. Hoekelman R, Starfield B, McCormick M, et al: A profile of
pediatric practice in the United States. Am J Dis Child
1983;137:1057-1060 6. Deleted in proof
7. Mendenhall R, Girard R, Abrahamson 5: A national study of medical and surgical specialties: I. Background, purpose, and methodology. JAMA 1978;240:848-852
8. Mendenhall R, Lloyd J, Repicky P, et al: A national study of medical and surgical specialties: II. Description of the survey instrument. JAMA 1978;140:1160-1168
9. Rosenblatt R, Cherkin D, Schneeweiss R, et al: The struc-ture and content of family practice: Current status and future trends. J Fam Pract 1982;15:681-722
10. Medical Practice in the United States, Mendenhall RC, principal investigator. Princeton, NJ, The Robert Wood Johnson Foundation, 1982
11. Perrin E, Harkins E, Marini M: Evaluation of the reliability and validity ofdata collected in the USC medical activities and manpower projects. Final Report. Seattle, WA, Battelle Human Affairs Research Group, Nov 15, 1978
12. Patterns of Ambulatory Care in Pediatrics: The National Ambulatory Medical Care Survey. United States January
1980-December 1981. US Dept of Health Education, and Welfare Series 13, No. 75 publication No. (PHS) 84-1736, Vital Health and Statistics, 1983
13. Task Force on Pediatric Education: The Future of Pediatric
Education. Evanston, IL. American Academy of Pediatrics,
1978
14. Budetti P, Frey J, McManus P: Pediatricians and family physicians: Future competition for child patients? J Fam Pract 1982;15:89-96
15. Budetti P, McManus P, Stenmark 5, et al: Child health professionals: Supply, training, and practice, in Better Health for Our Children A National Strategy: The Report of the Select Panel for the Promotion of Child Health, Vol IV,
US Dept of Health and Human Services (PHS) publication No. 79-55071, 1981
16. Starfield B, Gross E, Wood M, et al: Psychosocial and psychosomatic diagnoses in primary care of children. Pedi-atrics 1980;66:159-167
17. Mare R: Socioeconomic effects on child mortality in the United States. Am J Public Health 1982;72:539-547
18. Budetti PP, Kletke PR, Connelly JP: Current distribution
and trends in the location pattern of pediatricians, family physicians, and general practitioners between 1976 and
