DOCUMENT RESUME ED 095 779 AUTHOR Alpert, Joel J; Charney ... · DOCUMENT RESUME ED 095 779 HE 005 852 AUTHOR Alpert, Joel J; Charney, Evan TITLE The Education of Physicians for Primary

DOCUMENT RESUME

ED 095 779 HE 005 852

AUTHOR Alpert, Joel J; Charney, EvanTITLE The Education of Physicians for Primary Care.INSTITUTION Health Resources Administration (DHEW/PHS), Bethesda,

Md. Bureau of Health Services Research.REPORT NO DHEW-HRA-74-3113PUB DATE 73NOTE 69p.

EDRS PRICE MF-$0.75 HC-$3.15 PLUS POSTAGEDESCRIPTORS Educational Programs; *Graduate Study; *Higher

Education; *Medical Education; *Physicians; ProgramDescriptions; *Undergraduate Study

ABSTRACTThis monograph reviews educational programs designed

to prepare physicians to practice primary care, with particularattention to historical influences that underlie the development ofthese programs and recommends specific action to improve the qualityof that preparation. Emphasis is placed on the problems of primarycare, history of education for primary care, undergraduate programs,graduate education programs, and the elements of a successfulprogram. (MJM)

#1'(c. \cc' 0iv

-k

4.1?- NI%

CY QisS Q0

,k\ \ \

,;?,,Q.,

2Q- c,V" ,k CT\ ,ev Q'it:k

igc. _IA ttygs it,4 061 tt* ,N

C.V** c0.- 20 CP' A C' 4;0 gx'

0 \ 9 Q, \ 0 Pik' . \ 0' \ tS

Itt,tk fiL,Q. vs) $,t,- els vg, cy. .tiot%

e ,a ON ...k C Q'qs \Qk 0,-k k 9 ., \ A

0.- \ opr \ C ot,-C1/4PQ 0' (0,

CiVk 0 90 C,V y Co gikc QC\ y Co A \ q

0 \ 2 A 1;9'0, \ tg 00k 0' #

OV ti* c,t9. CV oi1/4# (es (7 -A ("6. fe'

QC\ Co A ,

t'k l? t. \

\ Ok OVSk ' 0' 0' 0*. v'(' VS\ Ot'' PS'. CP'

RO's 90 cp c,ti- 90 clik (0 A C-0 QV, c A Ci ,ik.k

t, \ v,0 \ c,Q. \ ,pRs .1,1PS 0.. ,,A, 40\ /?' ,0'

O.' 4.4001k i1/4# cpSkA c"',101. 20 Cit" A Cir- 90 , git'ick CI

4 \C,2et\O%tk,S,;(7#01"9"C'tC""?it'4;.\lestt°.°1:k

' A r \ ' F /f'r <C,# V ';''' 1

4:, vg - .. ik -tt'- \tt, i' v. # v.

# P9'gr- C-7. C' 99' 9 AC)

PQ' C- 99' \ 9 A A \ V\ sq' Iii

Pik 4 \ 0' O's OV''?' 0' e oy- oy ,g. os. Q.0- ot' c,

4 c, Osgyp, cy ,,c, ce. AC' 0 \9. <,\ 9 OA,

.q:k 9--k \ Q .,. \ 0 V'St io (2.-* 4.01/4 01;.. si`-'' C\- 41:;6. r.

ell,"10.1.4' P-9' NV c,- PQ' GP 0 e 0 -kc, ie lq,,.. c. Co cect\ 9 ,k i -..k Q \ ...k it, \ i., Q.

A 0 4., \ ySt' P-C\ . g 4 t.1 0 # %..9 ' #t

01' ov k vg., ry.54 00 (e ObSk c". oel q4 C,' ,k Cr 01/44 \ 9

le A C' 40 \ g , \ 2 0 P-5 Q, \ 0; VSk 41;Sk Q's' C\N- e?

iki, Its.`?' #P- (41,, vg, osV'' ot vg- cp 90' 90 c,P-GP 90 14

de, C, CO 4' .c) AC, .....1 \ 9 , 1 its v,Sk i, \ C. \ _k

90 ,,,k \ YF.

i 0.' , 1;C' NtV # 9ti Nles 'I'V*. Ok VS\ $'V 04*

PQ' C7. 0 99GP A C° 99'

es *0 , 99' y ei eLAC' QC\ 9 chA N0 6 \ q \ 4kA _9s t, \ 0.4

(c, \ 9,,, AsOk ttV- 0; D-0' #V Q..0Y cP. C,t'Sk GOr (e9 0..' 91`

9 v..- _,:zs- 90 c,t' C.,- 29- ., 9 y ,i,-k \ Q \ - ick vskA \

V t, \ vk0 4,v.C' cx,- \ 0' ..)S"rit41";',- PQ', CifoOt:?:2C't CAA ("6.\

citi, _RS e y CI AScivikk Co%

e-k ,I,Vg 0,q 0 \ C'Vk

\ ; °1 # Vg it SC'VScy C, Qc \ \ 2 -k 0 us uipAR,ur NT OF HEALTH.FUUCATION&Y.ELFAIPE

eit-k C" 441A

iik \ Q. -C\ V** NAtiuNALINSTITUTEOF /;°. AV

o_tt''(-,Pg_, Cl'ie#, et,..1 C'tL bk./CATION 00 CP.

9.,-\ k -- \ ,, , _T. _,, c-- ..:., !AL. ci t PRO ,,,,k fia,

6 doir,, kr ,,\ " 't!)1.A %? A, NF. ( i VF D r NOM i., &tar id?'

P12,1?.. ilsle ISS' Ots 4?\ ..; p; g, .,,":61:;. JAN /A ON OciGIN Alkr i1Z4V GI

9 -A cA. Q. P-0 ci?' Ac, , \ Q.A :. ... Po.....(, ,.Ei. Ow OPIWONS or 9, '' "r .1. ' .41. ' 1'..a.u. V RFPuF 444A/4%

1$ \ .401. `APP 09 Ot 2 t:.. .-.' . .. '... ..A..- ';'.(7''" (3 P'9;01.;IiRkWA

V. gsW Qik- y C' A \ 9 : 1 C. \ -(4- NtAt'9,

49Pvgc- -\ tiN°''929,-#Vgse'r-CC.kc' rfs. 9°s\,,,, , q,;,. v,. . ,F. cpS..k- CO? IP \

, 9%4.'4, Y cts'k V.9' c* 0' 0*. ea, C7 ,..tC \9C\_ \I? bli-k 4,0k si`i' r .tS' 01' q0,1/ .4T- 1._ ( 0 , Q 0. ,v ,(4, _Q,...- J. os cor- a .., q

t.. C,' 0(.,V*? 29.` v c,,-1 j vS,' A, s0.. ` st-A (7 N.- ,:zzl \ vgfc-Ats

t., 0, vgltt%:?-0 c,V*NIA C) \ C'Cis\ \ 9 %,9*.k t,.. R,i, ,..p,i09011,k ft Cie, \ qN'o,t,

GP' A C' r- \ /''k %S. 0' Vg # ikS C"*. A \ 9 \ VSk 406. 0.44k`' V4s..,k .ri., t, \ ck, _AV'' 414 (7 CA . it?' \ 9 0 ;.9' Q. 1/4 tt' 9 Cr 1 C' 9

O qs v i`r CR' .'k \ i, v- 4 Is` C. gq. q A 9. t, \ Q.00 CP6' y C" q ) 9 O. # v..9' ik 0 Co -1 \ t, \ 01' 04 9 cot.

9 -4 i, AtIk 0 CP' A CI q 6 \ 0 __63-14 g.t, IS" 004 4;o0 a,K4 \cAO At r .18' . 06. 00 (4' 41.-k a" olk. kik' .4P..or: CP' _4 Ci % Ili -0. \ 11.1F .1Pr..Qt* Pt

JOEL J. ALPERT M.D.and

EVAN CHARNEY M.D.Autumn 1973

U.S. DEPARTMENT OFHEALTH, EDUCATION,AND WELFARE

Public Health ServiceHealth Resources Administration

Bureau of Heatth Services ResearchDHEW Pub. No. (HRA)74-3113

Table of ContentsPage

Acknowledgement iii

INTRODUCTION

I. What is Primary Medicine 1

References 5

II. Problems of Primary Care Education 7

How Should Primary Care be Practiced? 7

Relative Importance of Education 8

Assumptions of the Monograph 11

References

III. History of Education for Primary Care 13

Selection for Medical School 17

The Medical School Experience 18

References 20

IV. Undergraduate Programs

The Physician's Office (Preceptorships)

The Patient's Home 25

The Hospital's Outpatient Department 27

Health Advisors 28

Comprehensive Care 29

Family Care 32

Model Practices 34

Formal Course Work 35

Refeiences 36

V. Graduate Education Programs 41

Pediatric and Internal Medicine Programs

Family Medicine Programs 43

Postgraduate Education 45

Primary Care Graduate Education in Great Britian 46

References 48

iii

VI. The Elements of a Successful Program 5l

The Students 53

The Faculty 56

The Patients 57

Curriculum Time 59

The Setting 61

Conclusion 64

References 65

iv

Acknowledgement

Education for primary care has been a majorconcern for both of us. Although formally trainedas pediatricians, it became apparent early in ourcareers that care for our young patients involvedtheir families and that most of these patient careneeds occurred outside of the hospital. These viewsnow seem so obvious that many of our readers willperhaps find our statement naive. Yet, the numberof colleagues who have shared this view with us inthe past have been very few, and concern forprimary care in the medical school and teachinghospital can at times be a very lonely undertaking.

Robert Haggerty stimulated us early and pro-vided invaluable guidance and counsel. It was hissuggestion that we take advantage of the oppor-tunity afforded by our coincidental study leave inGreat Britian to write this monograph. We havealso benefited from the critical comments ofProfessor Margot Jefferys, whose Social MedicineUnit at Bedford College, University of London,was our academic home during this year.

We are especially indebted to the University ofRochester, the John and Mary R. Markle Founda-tion, the 'Harvard Medical School, Boston Univer-sity School of Medicine and the Bureau of HealthServices Research, formerly the National Centerfor Health Services Research and Development, forproviding us with the time and necessary financialsupport for our sabbatical year. Specific supportfor writing this monograph was provided by theNational Center for Health Services Research andDevelopment (Purchase Order 4±123692) and theBureau of Health Manpower Education (ContractItNIH-72-4030), U.S. Department of Health, Edu-cation, and Welfare. Mrs. Jean Carmody of the Na-tional Center was especially helpful to us.

We hope that this monograph will prove timelyand assist those who share our interest in educationfor primary care.

V

Joel J. AlpertBoston, Massachusetts

Evan CharneyRochester, New York

Autumn 1973

Introduction

This monograph has two main purposes. Thefirst is to review critically those educational pro-grams designed to prepare physicians to practiceprimary care, with particular attention to the his-torical influences that underlie the development ofthese programs. The second is to recommendspecific action to improve the quality of thatpreparation.

Our central thesis is that there are two inter-related and serious problems in our present educa-tional structurenot enough physicians enterprimary care practice and those who do are notadequately prepared for the job. These dual defectsare a result of factors both within and outside ofthe medical education process, and an under-standing of their nature an.: historical developmentmust logically precede any recommendations forchange.

At the outset, we need to define the term"primary care" in order to have clearly in mindthat aspect of medical practice to be analyzed. Fol-lowing this, we outline some problems in primarycare practice that affect education. Finally, wenote several problems within the educationalsystem itself that compromise the effective teach-ing of primary care. In subsequent sections of themonograph, we discuss the history of primary careeducation and describe in some detail primary careprograms at the medical school, residency, andcontinuing education levels. In e final section, weset forth a number of factors that should be con-sidered in the planning of a primary care educationprogram and make specific recommendations fortheir implementation.

vii

I. What IsPrimaryMedicine?

The terms "primary medicine" and "primarycare" have gained wide acceptance in the pastdecade. particularly in the United States. Likemost new terms, their meaning has evolved andbeen reshaped with each succeeding author's use. Itis important to begin with a definition of primarymedicine, at least, to distinguish the term from itsoccasional fellow-travellers: comprehensive medi-cine, social medicine, preventive medicine, com-munity medicine, personal medicine, ambulatorymedicine, and family medicine. These terms over-lap not only with primary medicine, but also witheach other, reflecting areas of common interest aswell as a certain vagueness of definition. The factthat academic medical departments often have oneor several of these as titles has legitimized their use,but not always clarified their meaning. Before de-partments and learned societies adopt completelythe newer term "primary care", a definition is inorder.

To begin, there seems to be wide agreement thatprimary medicine is within the personal healthsystem rather than the public health system and,therefore, is focused on the healtl needs of indi-viduals and families (White, 1967: Hansen, 1970).These individuals live in communities and mayshare certain common characteristics with others intheir vicinity, but one starts with the individual orfamily as the reference point and then expands orelaborates. We do not start with a community as"patient." Parenthetically. it should be acknowl-edged that a public health system frequently en-compasses the responsibility for assuring that apersonal health system exists and thrives.As a con-trasting example, Deustschle and Eberson (1968)define community medicine as "a discipline .. . forstudying and solving in-depth community healthproblems. This includes an organized communityeffort in environmental health related specifically,to fundamental causes and social consequences of

all the more prevalent diseases ... the diagnosisand therapy of community and social pathology,not individual pathology, must be our majorconcern."

In a similar vein, social medicine is defined byMcKeown and Lowe (1966) as comprising twoparts, which are epidemiology and the study of themedical needs of society. It is not seen as a clinicalor laboratory discipline. According to these latterauthors, social medicine and community medicineare, first of all, scholarly rather than consultingdisciplines; and, secondly, they are focused on thecommunity or large group as the unit of care ratherthan on the individual or family.

The term "comprehensive medicine" should bedistinguished from primary medicine; because forno other reason, many programs reviewed in thismonograph are titled as such or have an implieddefinition of the term underlying their organiza-tion. The problem with comprehensive medicine isits all inclusiveness. To quote Lee (1961), compre-hensive medicine is "an attempt to apply all avail-able knowledge be it pathology, psychology, orsociology- to the maintenance of health and thediagnosis, therapy, and rehabilitation of the sick ordisabled patient." In Weinerman's terms, compre-hensive medicine is " ... the organized provisionof health services to family groups, including a fullspectrum of service from prevention throughrehabilitation, continuity of care for the individual,emphasis upon the social and personal aspects ofdisease and its management, use of the health teamconcept with personal physician responsibility andcoordination of the diverse elements of modernscientific practice" (Snoke, 1965). Sanazaro andBates (1968), in an exhaustive review of teachingprograms labelled "comprehensive medicine", useda critical incident study to attempt further defini-tion of the term and still had difficulty to in sepa-rating it from "good medicine." Indeed by the

1

definition listed, it refers to an ideal of breadth anddepth that is consonant with all good medical care.

We suggest that the term "comprehensive medi-cine" be retired nova, after two decades of yeomanservice, for two reasons. It is insufficiently restric-tive to define a subcategory of medicine: and it isdivisive, albeit unconsciously. To assume that com-prehensive medicine is he sole prerogative of theprimary physician is to insult the good consultantby assuming that he ignores a breadth of factors inhis practice. Moreover, it lets the less competentone off the hook and per :ts him to pursue histunnel vision to the last hydroxyl bond without asideways glance.

There are _three main anchoring points to ourdefinition of. primary medicine. We shall base thisdefinition on the contributions of White (1967).Hansen (1970), Maraca (1971). Pellegrino (1968).the Millis Commission (1966), and the AmericanAcademy of Family Practice (1969).

1. Primary medicine is first-contact medicine. Inits "first-contact" function, however, primarymedicine is separable from secondary and tertiarymedicine, which are based on referral rather thaninitial contact (Hansen, 1970). As suggested byWhite (1967), primary medicine is the "care thepatient receives when he first approaches thehealth-service system or formally participates inthe process of medical care." We would give thisaspect a more active connotation. There is increas-ing awareness that the decision to seek out andcontinue with medical care is not a straightforwardprocess and is influenced by a host of individualand social factors. Persons in greatest need may notseek care appropriately or follow advice ade-quately. Primary medicine is very much concernedwith such factors, which act at the interface be-tween the patient and the provider. It i:; orientedto outreach and followup as well as to helping thepatient define the conditions under which entry toprofessional services and continuation in care areappropriate. In this sense, primary medicine is in-cluded in that portion of preventive medicine thatcan be practiced at the family level and that worksthrough the patient-primary provider relationship.Here the overlap with community medicine is ob-vious. Fur example. if there is need for a group ofpatients to know about sickle cell anemia, to bescreened for the presence of sickle trait, and to becounselled accordingly, shall this be the respon-sibility of the primary care practices within thecommunity or shall it be done through publichealth auspices? Shall such a program involveschools, churches, and the public media? Ideally,both public and personal health groups--as well as

the communityought to be concerned; and theinitiative may come from any one group. The "cor-rect" approach will vary with kcal conditionsand is, therefore, proper subject matter for bothprimary and community medicine.

2. Primary medicine assumes longitudinalresponsibility for the patient regardless of thepresence or ab.sence of disc ase. I n Magraw's (1971)term, primary physicians "hold the contract" forproviding personal health service. Implicit here isthe idea of an ongoing responsibility, which maybe relinquished in part at times, but not terminatedunless the patient agrees. Specifically, it is notiimited to the course of a single episode or illness.

Emergency room medicine rarely has this on-going aspect. Although it is first contact, it is nottotal primary medicine. Care of patients withchronic disease tests the definition from anotheraspect: many chronic conditions are lifelong, andoften consultant or secondary level physicians as-sume longitudinal responsibility for the patient'scare. The issue here is whether the physician seesthe limits of his responsibility, the "terms of thecontract," as defined by the disease or by the pa-tient. The consultant practices complete primarycare only to the degree that he is willing to assumeresponsibility for all three aspects of the definition.

Some observers point out that continuity of careby one provider may be a mixed blessing. Last(1967), for example, suggests that "it may becomealmost axiomatic to think of continuity of care asa desirable, if not essential, feature of adeqqatepatient care. There is little supporting evidence,and it can be argued that continuity of care isagainst the best interests of the patient. Familiaritybreeds contempt; continuity breeds uncritical ac-ceptance of established diagnosis." His is an ex-treme position, but one that must be borne inmind. On the other hand, continuity has been as-sociated with increased patient compliance(Charlie y. 1967) and lowe.ed medical costs(Heagerty, 1970); there is no evidence that con-tinuity is in fact detrimental to patient care.

3. Primary medicine serves as the "ilitCgrat-lOiliSt" for the patient. When other healthresources arc involved, the primary care physicianretains the coordinating role. Moreover, the pri-mary care physician or team is interested in man-aging to the limit of its capability the physical.psychological, and social aspects of patient care.This c,ncept is undoubtedly the hardest one todefine with precision. It often irritates the hospitalconsultant who is wary that by implication he isbeing considered less perceptive or even less

compassionate. After all. the consultant argues. the"good hematologist" is certainly as concerned withsocial and psychological factors affecting his pa-tient with leukemia as is the family physician. Thekey distinction here is broadness of responsibilityrather than broadness of vision. The primary physician is inclusive in his attitude toward his patient'sproblems. caring for as many of them as possible.and, where referral is indicated, retaining his longi-tudinal responsibility as the' integrationist. Thesecondary level physician tends to be exclusive.concentrating his skills as much as possible andreferring patients the moment their problems straytoo far from his more limited focus of concern. Putanother way, the primary care doctor spends mostof his time thinking about the patient and the im-pact of various forces on his health or illness over aperiod of time. The secondary or tertiary leveldoctor spends most of his time thinking about a

disease state or a technical skill and how variouspatients fit into or alter that field of interest over aperiod of time. For one. the illness is the episode:for the other, the patient is the episode.

What are the limits of this broad integrationistrole? Here there is decidedly less agreement.When, for example, does the management of a pa-tient with urinary tract infection become properstudy for a urologist, and when does behavior dis-turb.ance merit a psychiatrist? We tend to makethese decisions pragmatically based on the skill ofan individual practitioner or the availability of anindividual consultant--rather than on commonlyagreed criteria.

Some would suggest th primary care endswhen the patient is hospitahLed, as is usually thecase in Great Britian. "Ambulatory medicine"could then be said to equal primary medicine.However, certain inconsistencies spring to mind.Many medical problems that require consultantmanagement arc largely dealt with on an out-patient basis, with only episodic hospitalization:for example, chronic leukemia, collagen disorders,congenital heart disease. Conversely, the decisionto hospitalize a patient is, at times, based onpsychological or social factors in managementrather than solely on medical complexity. Forthese reasons. we think that site of care alonehome, office, hospital is an insufficiently discrimi-nating indicator on which to base a definition ofprimary care: and, indeed, this fact has proved tobe a major problem for programs that attempted toteach primary care. (See Section IV.)

How then can we describe the "vertical" and"horizontal" limits 'McWhinnCV, 1967' of thisintegrationist function: that is, how far into the

medical complexity of the condition and acrosswhat range of "nondisease" factors ought primarymedicine extend? We suggest the following opera-tional criteria. The primary care physician orteam's responsibility ends or is temporarily sus-pended when any of the following situationsoccur:(a) the patient is not satisfied with the diagnostic

or management plan and wishes consultation:or,

(b) the team itself does not feel competent tomanage the problem alone or does not possessthe necessary technical skill to do so:

(c) external review reveals limits in diagnostic ormanagement ability. In this case, the issue is tobe resolved by referral of such cases in thefuture, developing diagnostic capability at thesite of care, or upgrading the management skillof the primary health team.

We include the last criterion advisedly. Theopinions of the patient and physician alone may beinsufficient to privide the kind of care now possi-ble in contemporary medicine. Studies such asthose by Peterson (1956) and ('lute (1963) suggestthat primary care is not often of good quality.Although those studies have been criticized forapplying hospital standards to primary care, noother criteria are now available. In fact. this thirdcriterion of external review may necessitate thedevelopment of such standards over some period oftime (Richardson. 1972). The mutual scrutiny ofpractice that is relatively common in hospitalmedicine is conspicuously absent in primary care,especially in solo practice. In University hospitals,Mumford (1970) has observed relay learning inwhich physicians communicate information aboutpatient management with good deal of mutualcriticism and interaction. We believe that this kindof critical communication is central to continuedprofessional growth and, by extension, to im-proving the quality of practice. Can this "relaylearning" be achieved in primary care? Techniquesof medical audit in this area are only now beingdeveloped and tentatively tested: but they holdpromise of a major development for primary careand could provide a logical basis for continuingeducation as well. At any rate, the concept of ex-ternal review appears to be gaining impetus inprimary care as third-party payers demand increas-ing scrutiny and value for money spent.

It should be noted that by our definition ofprimary care "family medicines' is properly a sub-set of primary medicine. As defined by the Ameri-can Academy of Family Practice (1969), all offamily medicine is subsumed under the three

3

criteria listed: first contact, longitudinal responsi-bility, and a broad integrationist role. However. bythe same token, most pediatric and internal medi-cine practitioners arc primary physicians as well(Young, 1964); and practicing as a family physi-cian is no guarantee that the doctor will care for all

family members (Brown, 1 9 7 1). We believe, likeMcKeown (1965), that it is unnecessarily restrictiveto insist that primary care can only be practisedwhen all family members are cared for by the samephysician, although our own bias is that this wouldbe preferable.

References

SECTION I

American Academy of Family Practice: Education forFamily Practice. Kansas City, Missouri. August. 1969.

Brown, J.W., Robertson, L.S.. Kosa, J., Alpert. J.J.: AStudy of General Practice in Massachusetts. J.A.M.A..216: 301-306. 1971.

Charney, E., Bynum. R.. Eldridge, D., FrankMeWhinrey, J., NcNabb, N., Scheiner. A., Sumpter, E. andlker, H.: How Well Do Patients Take Oral Penicillin? ACollaborative Study in Private Practice. Pediat. 40: 188,1967.

Clute, K.: The General Practitioner. Univ. of TorontoPress. 1963.

Deutschlc. K.W. and Eberson, F.: Community MedicineComes of Age, J.Med.Ed. 43: 1229, 1968.

Hansen, M.F.: An Educational Program for PrimaryCare: Definitions and Hypotheses. J. Med. Ed. 45: 1001.1970.

Heagcrty, M.C., Robertson. L.S., Kosa, J.. Alpert.J.J.: Some Comprehensive Costs in Comprehensive VersusFragmented Pediatric Care. Pediat. 46: 596, 1970.

Last. J.M.: Quality of General Practice. Med. J. Aus-tralia. 54: 780, 1967.

Lee. P.V.: The Dynamics of Medical Education TheirPortent for Medical Care. J.Med.Ed. Part II, December,1961.

Magraw, R.: Medical Education and Health ServicesImplication for Family Medicine. New England J. Med 285:1407. 1971.

McKeown, T. and Lowe, C.R.: An Introduction toSocial Medicine. Blackwell Scientific Publication, Oxfordand Edinburgh, 1966.

McKeown, T.: Medione in Modern Society. Allen andUnwin, London 1965.

McWhinney, I.R.: A Primary Physician in a Comprehen-sive Health Service. Lancet 1: 91. 1967.

Millis. J., Chairman. Report of the Citizens Commissionon Graduate Medical Education. AMA Chicago, 1966.

Mumford, E.: Interns: From Students to Physicians.Commonwealth Fund, Harvard U. Press, Cambridge, 1970.

Pelligrino, E.D.: Planning for Comprehensive and ;on-tinuing Care of Patients Through Education. J.Med.Ed.43,: 751, 1968.

Peterson, 0., Andrews, L., Spain, R.S. and Greenberg,B.: An Analytical Study of North Carolina, GeneralPractice. J.Med.Ed.. Part 2, 1-165. 1956.

Richardson, F.M.: Peer Review of Medical Care. MedicalCare 10: 29, 1972.

Sanazaro, P. and Bates. B.: A Joint Study of TeachingProgi.tms in Comprehensive Medicine. J.Med.Ed. 43: 777.1968.

Snokc, P.S. and Weinerman, E.R.: Comprehensive CarePrograms in University Medical Centers. J.Med.Ed.40: 625. 1965.

White, K.L.: PriMary Medical Care for Families. NewEng.J.Med. 277: 847, 1967.

Young, L.E.: Education and Roles of Personal Physi-cians in Medical Practice. J.A.M.A. 187: 928-33. 1964.

II. ProblemsOfPrimary CareEducation

There are now fewer primary care practitionersavailable to our population than at any rime in thiscentury, and the ratio is continuing to fall (Sever-inghaus. 19651. Grouping non-Federal generalpractitioners. internists, and pediatricians in pri-mary practice as .`family doctors", there were 94family doctors per 100,000 civilian population in1931; 60 in 1957. and 54 by 1970 (White, 1964:Millis, 1966: Kobach, 1971). Schonfeld (1972)estimates that, based on currently recommendeUstandards of care, an adequate number would be133 per 100,000 population. Some would considerthis number excessive, however, in addition tobeing unattainable for all practical purposes. Incontrast, Great Britain had between 40 and 50general practitioners per 100.000 population in1971 with reasonably good availability of care forthe population. Of course, it is not easy to com-pare the two countries.

In the United States, the availability of a pri-mary care physician varies enormously with geo-graphic location: and. hence. a single figure for theentire nation conveys an unrealistic sense of theproblem in a local area. Moreover, internists andpediatricians unlike the British generalist, assume aportion of secondary medical care for their pa-tients: and this must be taken into account in anycomparison. Of equal importance, such factors asthe use of allied health professionals, changes inpractice organization, and changing populationgrowth rates will greatly influence the primarymanpower requirements in the immediate futureand comp and any simple statemept of ideal num-bers required. Nevertheless. it can be stated thatthe number of primary care providers is now de-creasing and will probably continue to do so asaging general practitioners retire. The lack of avail-ability of primary physicians is a common com-plaint among the public. and two major publicreviews of our medical care system identity thegrowing shortage of primary care physicians as the

leading problem confronting our medical caresystem (Millis, 1966; Carnegie, 1970). It would befair to state that this is a serious problem, but wecaution that a careful analysis of the exact primarymanpower need is a complex question in itself andbeyond the scope of this monograph.

The solution to this problem, whatever its truemagnitude. is more involved than merely insistingtat medical schools must do a better job oforienting students to careers in primary care. Theissue is complicated a number of factors thatimpinge on the medical educational system. First,there is a good deal of current uncertainty abouthow primary care should be practiced, and thisconfuses the educator's task. Second, it wouldappear that factors outside education have at leastequal influence on the quality of practice; andthird, the experience of some other industrializedcountries, such as Great Britain, is that manipula-tion of employment opportunity and pay may besufficient to insure a reasonable supply of primarycare practitioners. quite apart from the influenceof the educational system itself. Let us considereach of these in turn.

How Should Primary Care be Practiced

Should primary care be practiced be by familydoctors , internists, and pediatricians, or hysomeone other than the physician alto-gether? Should practice be solo or in groups: and,if in a group, in what size organization end withwhat structure? is there one practice model that isideal for urban and rural practice alike and equallyapplicable to the needs of the affluent and thepoor? What is the appropriate role of the primarycare physician in the social. psychological, andpolitical realm?

An educator would be hard pressed to answerthese questions. for it is extremely difficult to

5

organize coherent educational programs whenthere is little agreement on how those trainedshould practice. The problem is accentuated: be-cause there is not one but many patterns of pri-mary care practice at present. In various countriesor regions of our own country, primary carepractitioners may be generalists or specialists, maypractice solo or in groups, may be involved inteams with allied health professionals. and, in someinstances, may not be physicians at all (Side].1968; Fry, 1969). Furthermore, the present pat-terns continue to fluctuate. Friedson (1971) re-minds us that the publicly acknowledged profes-sional role of the practicing physician is less than100 years old and has not in tact been handeddown in its present for:: from Hippocrates. Inshort. there is no one universally accepted modelof ,:ufficient venerability as to make change un-thinkable. There are several separable themes in thediscussion of the physician's role in primary care:

Should the doctor be involved at all? Can pri-mary care be divided into sufficiently dist:Tete andrepetitive tasks that can be learned by less exten-sively trained professionals and nonprofessionals,so that the physician can retreat to a centralizedhospita! and await the "triaged case"? Garfield'sproposals (1970) are oriented in this direction, andthe use of isolated feldshers or nurse practitionersin such roles already exists in sonic locales (Sidel,1968). .We would side here with Magraw (1971)and Jeffervs (BMA. 1968) who feel that there isand continues to be an important professional rolefor the physician in primary care. oreover, White1 967 ) has suggested that the primary care rela-

tionship is the fundamental basis for the contractbetween the protession and society and that to brogate this role would require a major reordering ofthat relationship.

0.16 physician is inroleed primar care, whatkinds of professional associates are required and inwhat kind of organizational team structure? Dueto the pressures generated by a heavy patient lead,the primary physician clearly requires assistance. Afundamental question that must first be resolved iswhether or not he should separate out and delegatespecific tasks to subordinate. or share sonic of thedecision making with allied personnel in a copro-fessional team model (Bate:;, 1970). Indeed, sonicdiscussions of primary care already assume that ateam, rather than a physician alone, is involved(Hansen, 1970).

What kind of practice setting is best for the pri-mary care team' Here citie,,tion,, of site are in-

volved, as well as the interrelationship of primarycare to the patient on one hand and to consultant

6

or referral medicine, on the other. Such questionsas the following are posed: flow can we achievethe efficiencies of a large organization without los-ing the personal and human qualities supportedlycharacteristic of a small one? Will a "technocracy"of primary care develop, as has occurred in thehos.iital, so that the often tentative and ill-defined

I

needs of the patient are lost in what could becomea computerized ulti phasic medical cen-

ter? Si.ould primary care systems be hospitalbased, or should a new system be organized withinthe community but away from the hospital(Somers, 1971)? How much comumer involve-ment and control is optimal for primary care: andwhat, if anything, should be reserved to the "pro-fessional" domain?

In the absence of needed research findings. thescarcity of data on any one of these patterns makesit difficult to answer these questions with any de-gree of finality. However. educational :irograms,because they involve practitioners of the future,will influence and be influenced by these consider-ations. As Haggerty (1969) has suggested. part ofthe university's function in prinary medical care isto conduct research that will help resolve theseproblems. For the present, medical educators willat least need to be cognizant of these issues as theyplan programs and stress the evaluative aspects andresearch component as programs develop.

Relative Importance of EducationHow important is the medical education process

itself in producing the kind of primary care prac-titioner we need? Educators often fail to appre-ciate how other influences both before and aftermedic-al school and residency may have a majorimpact on how medicine is practiced. Friedson(1970) states the case forcefully for the influenceof the ultimate work setting on practice. He citesextensive evichmcc that deficiencies in medicalschool experience do not explain some importantdeficiencies of professional performance ''half sowell as does the organization of the immediatework environment." He notes studies that foundthat the .same individual hospital physicians be-haved differently when their supervision varied. Healso sites the findings of Peterson (1956) and (lute(1963) who found little relation between variationin professional education and the technical per-formance of general practitioners many Years aftergraduation. Studies of case workers and lawyers(Carlin. 19661 also suggested little relationship be-tween education and quality of practice. Finally,

1Gray 966) in a longitudinal study of medical

students found that equally "cynical" medicalschool graduates later differed in cy nicism accord-ing co the type of practice in which they wereengaged. To Friedson, these studies emphasized theimportance of the social setting in which the pro-fessional worked rather than his education.

Funkenstein (1971) suggests that the medicalstudent reacts to factors outside of medical schoolrather than to the influences of curriculum, teach-ing, or research. He observed that changes in careerorientation

fourthsimultaneously in first,

second, and tourth year students; an event thatcoincided with apparent changes in society thatplaced more or less value on certain career choices.For example, his data suggests that, in the late1960's, interest family medicine became a publicconcern and was reflected by increased interest infamily medicine in all three medical school classesat the same time.

The lesson to be drawn by medical educatorsfrom luch data is that there will be value in inte-grating primary care education and primary carepractice. As we elaborate in later section, consult-ant medicine is more closely linked to medical edu-cation than primary medicine. By and large, con-sultants practice in hospital settings that arc quiteintimately related to where education is going on,whereas primary care practitioners operate in soloor in groups quite isolated from the usual educa-tional milieu. Insofar as the setting is influential inaffecting the nature and quality of practice, itwould be valuable both for student and practi-tioner to be in closer contact with each otherthroughout training and practice.

There is some evidence that manipulating thecircumstances of practice alone may be sufficientto provide adequate numbers of primary care prac-titioners quite apart from what takes place in medi-cal school and in the teaching hospital. The ex-perience in Great Britain is instructional. Onentrance to medical school, somewhat fewerBritish students than American students cite gen-era; practice as their goal 16 percent compared to22 percent (Harris, 1969: Pavia, 1971. By grad-uation. the picture has reversed itself. Whileless than 10 percent of American seniors areheaded for general practice; by then, 28.6 percentof British students are so oriented; and even morewill eventually end up in this field (Harris. 1969;Calahan, 1957). What accounts for this shift? Hastheir relative educational experience been thedetermining factor? Our observation is that theBritish medical student's education is at least ashospital- and specialty-oriented as his Americancounterparts. if not more so. British schools arc

all, in fact, hospital-based programs. While almostevery medical school provides sonic experience ingeneral practice for the undergraduate, this lastsusually only one or two weeks out of five years.Inasmuch as all hospital-based physicians are, bydefinition, specialists, there is little opportunity forthe student to observe primary medical practice.The movement toward formal education in generalPractice is just beginning to take hold in GreatBritain; and, thus far, it is concentrated at the post-graduate (residency) rather than at the medicalschool level.

All of this suggests that if education has had anyinfluence on the student's choice of a primary carecareer b Britain it would have to be a negative one;namely, dissatisfaction with what he sees of hospi-tal medicine. A more likely explanation is thatth :re are only limited numbers oc specialtyconsultant posts available; and, therefore, there is astrong incentive for those who see little change ofachieving consultant status to opt for generalpractice. In short, most medical students mustenter general practice, like it or not; and bygraduation this has become increasingly evident tothem: Recent increases in the pay of general practi-tioners relative to hospital-based doctors has been afurther influence, external to the educational sys-tem, that has served to improve recruitment to gen-eral practice.

The lesson here is that factors of work environ-ment, remuneration, and relative employment op-portunity are powerful determinants of the pri-mary care manpower supply, quite apart fromwhat goes on within medical education. In fact,internal change in the education milieu alone, inour view, will be quite inadequate to redress thepresent imbalance.

Nevertheless, as medical educators, we arecommitted to develop the best educational systemwe can. If, as is abundantly evident, good edu-cation will not guarantee good practice, inadequateeducation is even less likely to do so. Our con-tention is that, in addition to the external factorswe have listed, there have been problems withinmedical education that have resulted in inadequatepreparation for primary care practice. Theseproblems have a common underlying feature;namely, that preparation for primary care practicehas not been a specific goal of most current medi-cal education programs and has not been thespecific responsibility of any one group. Jason(1970) states the case even more stronglycallingthe mismatch between student education andphysician career educational malpractice. Hereminds educators that over 90 percent of medical

7

students do end up in practice careers, mostly asspecialists: and only one medical school in thecountry has as many as 13 percent of its graduatesentering full academic careers. We will consider thispoint in further detail in the monograph, but it isworth summarizing, in this initial section.

This criticism of failure to prepare for primarycare practice applies to all levels of the currenteducational proces.;: student selection, medicalschool, internship and residency. and continuingeducation. For example, selection procedures formedical school still rely heavily on the demon-strated scientific ability of the applicant; eventhough there is little evidence that scientific abilityis predictive of success in medical school and evenless evidence that it correlates with success inpractice. The problem is that useful criteria forselecting future primary practitioners do not nowexist. Indeed, the development of such criteria hasnot been an important consideration in the post-World War II period.

Within the medical schools, there have been in-adequate efforts to orient students towards careersin primary care. The failure to educate for primarypractice represents a significart historical change,as there has been a gradual shift in the overall pur-pose of undergraduate medical education, whichhas worked to the detriment of primary medicine.The change has been stated quite explicitly. In1954, an edi- 3rial in the Journal of the AmericanMedical Association stated:

previous \Tars, Many MecliCal SCIWOIS Statedas the majOr objective of their undeNraduateteachiv propums the preparation of studentsfor the .eneral practice of medicine. Currently.hotrerer. most medical faculties em brace as themajor objecti . the provision of a 'solidframework. of fionhunental principles (Ipplicable

X111 areas of medicine- / 954In other words. alter World War II, the prepara-

tion of the undifferentiated physician" becamethe goal of medical schools, acknowledging thegrowing responsibility of residency training tocomplete his education for practice. Our percep-tion is that even this has changed giving way I

the current goal of preparing a "variably differen-tiated" graduate. 1 he student has now been ori-ented toward a field of practice that is ratherfirmly established by the time of his graduationfrom medkai school. The curriculum revisions ofthe past decade have tended to further emphasizethis early career choice for the student. A largelyelective fourth year of medical school plus astraight intermhip compel the student to make hiscareer choice well before graduation from medical

8

school. in addition, the integration of the intern-ship into the residency further pushes the studenttoward an early career decision as did the replace-ment of the rotating by the straight internship. In-asmuch as primary medicine is rarely presented as aclinical rotation in medical school, while an arrayof specialities are seen as real possibilities, a careerin primary medicine has neither the attraction ofthe carrot nor the force of the stick at precisely thetime when the student must make his choice. inshort, not only do four years in medical school nolonger, in themselves, constitute the necessary andsufficient training for primary care practice; butthat field also has been relegated to an unseenoption, downgraded by its omission.

In addition, a particular problem of under-graduate programs that is most closely related toprimary medicine is that they often confuse thisterm with "comprehensive" medicine. What theseprograms have emphasized are the social andpsychological aspects of medicine, central in them-selves to all good medical practice, but notequivalent to primary care. In addition, their set-tings (usually in the hospital), their faculty (usuallyfull-time hospital specialists), and their patientpopulation (usually the poor or the university com-munity) are sufficently atypical to represent un-fairly either the usual or the ideal in primary care.

At the graduate level as well, there is a loss, or atleast a blurring, or identity of primary care educa-tion. The residency originally developed as spe-cialty training, intended as a body of knowledgeand skill added onto that of the generalist. The factthat now most internists and pediatricians in manycases. often tend to be generalists-increasingly soas older general practitioners retire means thatthere has been considerable lack of "tit" betweeneducation and practice in these fields. The averagegraduate of a university program in medicine orpediatrics is superbly prepared for practice as achief resident- a career that does not exist.

Finally, the particular importance of continuingeducation for the generalist is only now beingappreciated. Although there are many courses andprograms for the practitioner, most have failed todemonstrate their efficacy. Primary practice is

especially isolated from continuing educationstimulation in contrast to that given to most spe-cialty practice. Furthermore, most continuing edu-cation is subject to the same specialist orientationas is the rest of medical education. This fact hasresulted in programs less attuned to the needs andviews of the learner than they ought to be.

We will return to these themes and amplify themas various programs of primary care education are

reviewed. Our main concern is that education hasbeen allowed, indeed subsidized, to educate thespecialist while nut being required to the samedegree to attend to the needs of the primary ,:arepractitioner. In the absence of outside restrictionson specialist practice, the result has been starvationof the

ofcare sector. The implications of

sonic ot the other factors we have examined is thatphysician education for primary care cannot beconceived and implemented without regard to theway medicine is practiced in our society. Medicaleducators will need to be involved in the politicalprocesses of the profession and society, which to-gether dictate how medicine will be practiced.Ideally, the education system and the practicesystem have something to offer each other-relevancy for the former and ongoing professionalgrowth and development for the latter.

Assumptions of the MonographAs we have outlined, there remain important un-

resolved issues in physician education for primarycare. Nevertheless, as with medical practice, thosein medical education will have to make decisionsnow for training based on incomplete data. At leastfor the purposes of this monograph we will makethe following assumptions:

1. The demand for primary care providers willbe increasingly articulated by the public and ex-pressed in legislative mandate.

In Friedson's terms (1970), society will reexertits control over the profession, at least in this area,and withdraw some of the autonomy it has grantedthe profession to train its own members in un-restricted fashion. As Pellegrino states (1971),"There is (now) serious discontinuity between theinterests and goals of medical faculties and the

interests and goals of the society that supports theschools. In any such counter position the educa-tional establishment, despite its strength andexpertise, cannot long prevail."

2. Civersity involvement will increase in pri-mary care education at all levels, from medicalschool through practice.

The university has been responsible for directingundergraduate medical education for the past fiftyyears. Its rcsponsbility for residency and continu-ing education have been less well defined. It is

likely that the university's role as coordinator andpossibly director of graduate and continuing edu-cation will be acknowledged and supported. Thisextended role will most likely continue to beshared with those who represent the public and thepracticing medical profession.

3. Mere be a significant professional rolefor the physician in primary care.

However. the changing nature of illness inmodern society will alter that role, requiring moreemphasis on management than on cure -"healthcare" rather than "medical cure" in Millis' terms(1971). It is unclear at present whether the pri-mary physician will be a family doctor, a generalinternist plus a general pediatrician, or both. Al-though assistance for the doctor is required, it isuncertain whether a "hierarchical team" or a "co-professional" will emerge: also unknown are otherskills that will be represented on the health team.Similarly, the optimal size and setting for primarypractice groups is as yet undetermined. Nonethe-less, the physician will likely retain, as we believehe should, the key responsibility in primary care.

9

References

SECTION II

Bates, B.: Doctor and Nurse: Changing Roles and Rela-tions N. Eng. J.Med. 28,?: 129, 1970.

British Medical Association: Primary Medical Care, May,1968.

Calahan, 1)., Collett, E.P., Hamer, N.: Career Interestsand Expectations of U.S. Medical Students. J.Med.Ed.32: 557. 1957.

Carlin, J.: Lawyers Ethics: A Survey of the New YurkCity Bar. New York, Russell Sage Foundation, 1966.

Carnegie Commission on Higher Education: Higher Edu-cation and the Nation's Health. McGraw-Hill, New York,1970.

Clute, K.: The General Practitioner. Univ. of TorontoPress. 1963.

Freidson, E.: Profession of Medicine. New York, Dodd,Mead & Co., 1970.

Fry. J.: Medicine in Three Societies. Aylesbury Bucks,M.T.P.. Chiltern House, 1969.

Funkenstein, D.H.: Medical Students, Medical Schooland Society During Three Years. In Psychosocial Aspects ofMedical Training. Coombs. R.. Vincent. C. 'Edit.). Spring-tiel d. Ill.. Charles C. Thuas. 1971.

Garfield. S.: The Delivery of Medical Care. ScientificAmerican, 222: 15. 1970.

Gray. R.: The Effect of Medical Specialization on Physi-cians' Attitudes. J. Health and Human Behavior 11: 128,1966.

Haggerty. R. J.: The University and Primary MedicalCare. New Eng. J. Med. 281: 416, 1969.

Hansen. M.F.: An Educational Program for PrimaryCare: Definitions and Hypotheses. J. Med. Ed. 45: 1001,1970.

Harris. C.M.: General Practice Teaching Undergraduatesin British Medical Schools. Royal College of General Prac-titioners. Reports tro General Practice. 1969.

Jason, J.: The Relevance of Medical Education toMedical Practice J.A.M.A 212: 2092, 1970.

Magraw. R.: Medical Education and Health ServicesImplication for Family Medicine. New England J.Med. 285:1407. 1971.

Millis, J., Chairman. Report of the Citizens Commissionon Craduate Medical Education. AMA Chicago, 1966.

Millis. J: A Rational Public Policy for Medical Educationand its Financing. National Fund for Medical Education,New York, 1971.

Paiva. R.E.: Intellectual, Personality and EnvironmentalFactors in Career Specialty Preferences. J.Med.Ed.46: 281, 1971.

Pellegrino E . D.: Research in Medical Education: TheViews of a Friendly Philistine. J.Med.Ed. 46: r 50, 1971.

Peterson, 0., Andrews, L., Spain, R.S. and Greenberg,B.: An Analytical Study of North Carolina, GeneralPractice. J.Med.Ed., Part 2, 1-165, 1956.

Roback, G.A.: Distribution of Physicians, 1970,American Medical Association, Chicago, 1971.

Schonfeld, H.K., Heston, J.F., Falk, I.S.: Number ofPhysicians Required for Primary medical Care. New Eng.J.Med. 286,: 571. 1972.

Severinghaus, A.E.: Distribution of Graduates ofMedical Schools in the United States and Canada Accordingto Specialities 1900-1964. J.Med.Ed. 40: 721, 1965.

Side', V.W.: Feldshers and "Feldsherism": The Roleand Training of the Feldsher in the U.S.S.R. N.E.J.M.278: 934, 1968.

Somers, A.R.: Health Care in Transition: Directions forthe Future - Hospital Research and Educational Trust,Chicago, 1971.

White. K. L. : General Practice in the United States,J.Med.Ed. 39: 333. 1964.

White, K.L.: Primary Medical Care for Families. NewEng.J.Med. 277: 847, 1967.

HistoryOfEducationForPrimary Care

Medical education in the United States, which isnow considered by much of the world to offer astandard of excellence, has been characterized byperiodic upheaval and reform. The names ofMogan, Osier, Flexner, and Millis stand out asassociated with major efforts to redirect medicaleducation in this country; and they have specialsignificance for primary care education. In the18th Century, Morgan advocated high-quality edu-cation and pleaded for a model similar to the full-time Europea-, university system. Osier, the greatclinician, brought teaching to the bedside of thepatient. Flexner, the educator, whose name is as-sociated with the immense upheaval in medicaleducation early in the 20th century, exposed theproprietary schools and established the universityas the major force in undergraduate medical educa-tion. The name of Millis is associated with currentattempts to alter undergraduate and graduate medi-cal education in the direction of primary care andto expand university responsibility to includegraduate education.

Medical education in the United States waschallenged by Morgan, who had been influencedgreatly by his European c.ducation (Moll, 1968,Hall, 1896). In 1765, he wrote his celebrated Dis-course upon the Institution of Medical Schools inAmerica (Morgan, reprinted 1937). He recognizedthe medical school as an effective social organiza-tion for learning and saw it as preparing medicalstudents for practice. He anticipated specializationand argued persuasively, although unsuccessfully,for the move away from the apprentice system tothe full-time system as developed in Europeanuniversities. While Morgan became the first Ameri-

can Professor of Medicine at the University ofPennsylvania, his discourse was largely ignored.Medical schools in the United States developed asproprietary schools, and education was largely

accomplished by the apprenticeship. Morgan's pleaabout university affiliation was lost.

In an attempt to provide physicians for a rapidlydeveloping and expanding country, America thenentered what has been characterized as a dark ageof medical education (Robinson, 1935). Althoughthe low quality of medical education was Ameri-ca's educational scandal, proprietary schools didproduce large numbers of primary care physiciansfor a country that needed physicians for itsexpanding frontier.

At the end of the 19th Century, Osier expressedthe view that medicine must bring its educationalhouse into order. Like Morgan, Osier emphasizedthe necessary relationship between the universityand the medical school. Deploring the criminallaxity in standards of medical schools, he accusedmedical colleges in the United States of being un-responsive both to the public and to the profession(Osier, 1905). True, the proprietary school grad-uated large numbers of physicians, but the qualityof their education was decidedly poor. Osier em-phasized the value and benefits of teachingstudents by the bedside of the patient and stressedthe importance of developing clinical skills andresponsibility. In many ways, Osier prepared theway for Abraham Flexner, whose report in 1910(Flexner. 1910) documented the vast extent of themedical education scandal.

Our interest is not with the detailed content ofthis report but with its results, which dramaticallyimproved American medical education. Proprietaryschools began closing overnight, and the remainingmedical schools established ties with universities asurged by Morgan, a century before (Evans, 1965).However, association with a university, prior toFlexner, was no guarantee of quality., and this re-port was even critical of many university medical

11

schools. Flexner demanded that the university setand control standards for medical education.

Although advances and reforms were needed,some felt that Flexner's recommendations weretoo rigid and confining. MacKenzie (1918) calledFlexner "doctrinaire" and advised against the full-time academic model for faculty, which had alsobeen suggested for the United Kingdom (Newman,1918, 1923; Flcxner, 1912). The model of medicaleducation in the United Kingdom was based in ateaching hospital usually without university con-trol ( Jefferys, 1969). MacKenzie also cautionedabout putting medical education completely in thehands of full-time teachers. He pointed out that, ina hospital, students did not see illness with non-specific physical signs. These observations couldonly be made in the community where medicinewas practiced and not in a hospital where studentswere than being educated.

Medical education in the United States is nowtotally changed. The full-time university model asdeveloped at the Johns Hopkins Hospital becamethe established educational pattern (Evans, 1965).Education for practice took place almost entirelywithin the university and its related teaching hospi-tals. Subsequent results led to today's prob-lems: i.e., the complexity, the fragmentation, andthe inflexibility of standards for graduate medicaleducation. As a consequence of the Flexner report,general practitioners could no longer be full-timefaculty; and the stage was inevitably set, at least inthe United States, for the decline in both thequantity and quality of general practice and pri-mary care (Haggerty, 1963).

Voices were raised with the complaint thatmedical schools were not educating physicians forpractice. Inadequate attention was being paid tothe art of medicine (Rappleye. 1932: Curran.1948), principally by failure to attend to socialfactors in illness and to consider the patient as aperson (Peabody. 1927; Dublin, 1947: Means,1946; Propst, 1939: Robinson. 1935: Reynolds.1939. Thornton, 1937: Rice, 1939; Cannon, 1946:Colwell. 1946). -

Some efforts were made to alter the perceivedsituation. Five years before the opening of theJohns Hopkins Medical School, Billings advocatedsending students into patients' homes (Curran,1948). In the 1890's, Osier and Welch had assignedthird-year students to investigate the home condi-tions of patients with tuberculosis. A social servicedepartment was established at the MassachusettsGeneral Hospital in 1912, where Edsall andCannon developed social clinics to view the patientin his social setting. In the 1920's at the Boston

12

City Hospital, Minot and Cannon also established asocial clinic. Their intent was to emphasize theimportance of social factors in illness for thestudent in both medical school and hospital(Minot, 1925; Cannon, 1934). Similar programsdeveloped at other medical schools (Harvey, 1946;LaSaine, 1940), and most of these had as statedobjectives that the student was to study the social,medical, personal, and sanitary backgrounds of hispatient. One result of these efforts was the incor-poration of the social and family history as part ofa traditional medical history.

Such programs were often located in the out-patient department due to the prevalent view thatthe outpatient department was the place to learnabout medical practice. Although their organizerswere generally enthusiastic, the efforts gained toolittle academic support, either financially or profes-sionally. These efforts were not subject to criticalevaluation; and, when evaluated, the evaluationusually consisted of case examples of beneficialoutcome; surveys of graduates were also under-taken (Cohen, 1941; Melaney, 1939; Cockerill,1941).

Robinson summed up these programs phil-osophically by stating that their purpose was toconsider "the patient as a person" (Robinson,1935, 1939). His program in the Eastern District ofBaltimore became a significant extension of thesocial clinic, when lie suggested that a hospitalmight have responsibility for a specific community.Within a decade, concern for the patient was alsoextended to include not only the patient, but alsohis family (Richardson, 1945) with the sl.ggestionthat the family rather than the individual should bethe logical basis for medical care.

During the 1930's, the full-time specialty systemso dominated medical education that these pro-grams either remained merely philosophical with-out implementation or, when implemented, re-mained outside of the mainstream of medicaleducation. New departments were established andtheir faculties were increasingly composed ofspecialists (Stevens. 1971). The resulting frag-mentation of medical care was considered lessimportant than the goal of achieving scientificexcellence.

In some quarters there was continued interestindeed, concern- in the relationship between medi-cal education and practice. The privately supportedCommittee on Medical Costs (1932), which hadbeen established by a concerned group of laymenand physicians. recommended the training ofphysicians in the teaching of health and the pre-vention of disease as well as restriction of entry

into the specialties. The Committee also reco-mended that teams of health professionals orga-nized around the hospitals provide complete thera-peutic and preventive services whether in thehome. office, or hospital setting. The personalrelationship between the physician and patient wasseen as important: and the Committee. recom-mended community medical centers wid gen-eralists. group health insurance, and group practice.In fact, this report recommended that 80 percentof all medical graduates should function as gen-eralists. This early report and its recommendationswere overshadowed by the depression and, unlikethe Flexner report. had no substantial impact onmedical practice .or education (R ichinond, 1969).

It was in the fields of psychiatry and preventivemedicine that further developments of some rele-vance to primary care education occurred. Forpsychiatry, primary medicine was equated with thepsychological and psychiatric aspects of practiceand psychoanalytic principles (Group. 1962). Forpreventive medicine, primary care largely meantconsidering the social, economic, and environ-mental factors that produced and influenced illness( .eathers. 1939). Programs in these departmentswere not so much attempts to educate. large num-bers of physicians for primary practice, as to equipphysicians with psychiatric or preventive medicineskills.

In the late thirties and early forties, well-definedprograms for preventive medicine teaching couldbe round in about (me-third of the medical schindsCurran. 1948!. However, only 11 (14 percent) ofthe medical schools and 13 teaching hospitals wereconsidered to have medical social departments thatcontributed to adequate teaching of medical stu-dents (Bartlett. 1939). Additional programs wereinitiated due to Luncern for the patient after hospi-talization ( Jensen.. 1944). Followup became animportant aspect of both medic. I and surgical care.

Conferences on the role of psychiatry in medicaleducation were held in 1933. 1942, and 1952(F.baugh. 1933: Ebaugh, 1942: Whitehorn, 19521.These reports noted the contribution that psychi-atry could make to the general practice of medi-cine through teaching interviewing skills, theunderstanding of psychosomatic disorders. and theobvious importance of dealing with psychologicalissues surrounding organic disease (Lid/. 1956).Psychiatry was perceived as a major component offamily. comprehensive. and. by our definition,primary care (Lid/. 1970).

The inclusion of behavioral science in the medi-cal curriculum was also suggested in the 1930's andthe 1940's. Warbasse (1932) pleaded for a course

in the medical school that would combat "themonastic seclusion .of the medical student," whomhe saw spending four yeas becoming a medicaltechnician. President Angell of Yale spoke of thestudy of medical sociology as important enough tocreate one or inure new chairs of medical sociologyin medical schools, a suggestion that was notaccomplished (Angell, 1933). In the late 1940's, aseries of lectures in medical sociology were spon-sored by the Department of Medicine at theHarvard Medical School (Sociology. 1946): thiscourse expanded to become the major focus forteaching preventive medicine in that school.

The continued ferment in medical educationover the failure to teach social and psychologicalaspects of practice. as well as the continued needto educate for practice, was noted in the nextgeneration of conferences in the early 1950's.These were the conferences on psychiatry andmedical education (Whitehorn, 1952) and theconference on the Teaching of Preventive Medicine.in Medical Schools (Clark. 1953). There were alsotwo conferences on world medical CCILICatI011 atwhich considerable attention was given to educa-tion for general practice (Proceedings 1954, 1961).

The teaching of social and psychological skills insocial medicine and psychiatry came together inthe 1950's in the concept of comprehensive. medi-cine (Matarrazzo. 1955). To Matarrazzo, the ad-Vent of comprehensive medicine was the dawn of anew approach in niedical education and ulitatelvin the practice of medicine. Yet, these efforts alsoremained outside the mainstream of medicaleducation.

Following the Second World War, research inmedicine accelerated to an unprecedented degree.Financial support to medical schools through theNational Institutes of Health helped bring into fullfruit it ion t he Flexner research model. Fundsearmarked for research subsidized education: be-cause. the faculty, growing in size, was largely sup-ported by research grants. These grants emphasizedspecialization. because research was highly techni-cal and specialized. In addition, graduates inmedical education also received post-doctoral fel-lowships that promoted this trend toward spe-cialization. The enlarged full-tie staff resulted in adecline in the influence and importance of thepart-time medical faculty.

Berry (1953) initiated a series of annual teachinginstitutes that were concerned with educationalreforms and evaluation (( gee. 1958: Comroe, 1961:Wolf, 1962). fik reported six experiments primarilyin teaching, integrating curriculum, and learningcomprehensive care skills that were. supported by

13

private foundations, particularly the Common-wealth Fund and the Kellogg, Rockefeller, andMilbank Foundations. Most of the foundation sup-port was for the comprehensive care programs thathad been developed in order to provide a morehumanistic base for the physician. Reviews of theseand related experiments appeared (Lee, 1962;Snoke and Weincrman, 1965; Sanazaro, 1968), asdid particularly detailed reports of studies at Col-orado (Hammond, 1959), Cornell (Reader.. 1967),Western Reserve (Kennell, 1961), and Harvard(Haggerty, 1962; Stokes, 1963).

What had happened to general practice in theperiod since Flexncr? Efforts had been made inthe 1920's to reintroduce preceptorships for medi-cal students (Kerr, 1926; Bardeen, 1928) in generalpractice, but these attempts were isolated and didlittle in the long run to attract students to the fieldor to reverse the decline in numbers of generalpractitioners. Although a few schools in this periodmade use of preceptors, these efforts were in theshadow of the teaching hospital. The bitter crit-icism of the proprietary schools extended to allforms of apprentice education, including precep-torships. Most of the schools that did use preceptorprograms were in rural States, where the apparentcommitment to produce a general practitioner whowould provide primary care did not disappear asrapidly as in the urban schools.

In 1941, a resolution urging creation of a boardof general practice was rejected by the House ofDelegates of the American Medical Association onthe grounds that passing a State or National boardexamination automatically certified a physician todo general practice. In 1946, a section on generalpractice in the American Medical Association wasestablished; and, in 1947, the American Academyof General Practice was founded and becameactively engaged in attempts to increase recruit-ment fur general practice. In 1959, the AmericanMedical Association defined the terms "familyphysician" and "family practice": and a series ofprograms were established that offered for the firsttime specific residency training for generalpractice, but they remained largely unfilled(Stevens, 1971). Financial barriers to specializationwere removed with support through research fel-lowships and residency programs; and the generalpractitioner or "L.M.D." became an unattrativemodel to medical students. The academic com-munity largely ignored the need for primary carephysicians.

Some general practitioners and academicians sawa future for general practice in the "new" dis-cipline of family medicine (Rardin, 1961). ThisI 4

discipline was defined as the continuing and com-prehensive care of the individual patient and hisfamily regardless of age. Some placed emphasis onthe psychosocial skills of the physician in additionto the usual pediatric, medical, psychiatric, andobstetric skills. Programs were again initiated atcertain medical schools that were consistent withthe model of a family physician (Kennell, 1961;Haggerty, 1962).

In the mid-1960's, a series of reports called for anational commitment to the education of personal,primary, or family physicians (Millis, 1966; Coggcs-hall, 1965; Willard, 1966). Millis, who was thenpresident of Case Western Reserve University,chaired an American Medical Associationcommittee of laymen, educators, and physiciansthat produced the Millis Commission Report. Thisreport, in one sense, was an expansion of the muchearlier Flexncr Report; it called for extension ofuniversity control to include graduate (residency)training. This was accompanied by a call for a na-tional commitment to produce primary care physi-cians who would, for most patients, represent thecommon point of entry into a reorganized, re-vitalized, and rational system of health caredelivery. Following these reports, the specialty ofFamily Medicine was approved in 1969 with thesupport and the leadership of the AmericanAcademy of General Practice and a small numberof academic physicians. The Specialty Boards ofMedicine and Pediatrics also responded by develop-ing a dual certification in Family Medicine. Byspending four years in the hospitaltwo in ped-iatrics and two in medicine, a physician would beprepared to function as a family physician. How-ever, pediatrics and internal medicine had theirown problems in establishing and accepting theiridentity as primary care disciplines. Thus, it wasdifficult to see how the combined program wouldsucceed in producing a family physician. This movecould also be interpreted as one whose intent wasto frustrate the developing Board of FamilyPractice.

Opposition to the specialty of family medicinewas still noted in testimony against governmentfunding for family physician programs. Despite thisopposition, the legislation, passed in 1971,provided the first funds for financing programs infamily medicine. Despite this opposition, the legis-lation, passed in 1971, provided the first funds forfinancing programs in family medicine. Medicalschools and teaching hospitals began to developtraining programs for primary care physicians aswell as nonphysician personnel. Following theestablishment of the Board., of Family Medicine,

there was a growth of postgraduate programs andestablishment of new departments of family medi-cine (Magraw, 1971). 13y 1972, there were over100 recognized residency programs.

In the 1960's, public outcries were made overthe disadvantaged population that had not bene-fited from what had been called and was thenbelieved to be the "finest medical care in theworld." The "rediscovery" of poverty and thespiralling costs of medical care forced the Nationto reexamine its medical priorities. The immediateresults of this reexamination were a slowing of theresearch effort and the involvement of the FederalGovernment in many large-scale service programs.With the establishment of the National Center forHealth Services Research and Development, re-search on health services delivery became a moreimportant activity of Government.

Medical school departments of preventive medi-cine, obstetrics, and pediatrics became involved inprimary care as funds became available in neighbor-hood health centers, children and youth programs,and maternal and infant health projects. Medicalschools and teaching hospitals were being con-fronted in the 1960's with a population who wereusing their facilities for primary care but who, infact, were receiving fragmented health care. Pa-tients made increased use, mainly for primary careneeds, of the outpatient and emergency facilities tosuch a degree that many investigators began tostudy the problems of institutions providing pri-mary care. Service programs outside the hospitaland medical schools were seen as a new and legiti-mate activity.

Public awareness of poverty and the admittedmanpower crisis accelerated students' demands foropportunities to provide service. At Harvard Medi-cal School, for example, in the two-year period,1968-1970, there was an increase from 3 to 26percent in first-year students who expressed adesire to be family physicians. Many medical stu-dents began to question the research-orientedmodel of the medical schools and identified theirdesire for service with the developing discipline offamily medicine.

New medical schools were established inresponse to an acknowledge manpower crisis, bothin the inadequate number of unequal distributionof physicians. The crisis was especially noted inobtaining primary medical care. General practi-tioners were not being replaced by younger practi-tioners. To illustrate, the ghettos of large cities had.almost no general practitioners to meet the pri-mary care needs of the disadvantaged population,and rural areas had none.

There was no evidence that the new medicalschools would actually produce primary care prac-titioners. The admission by medical educators thatthere was a need to educate for primary care was,of course, a necessary first step; but the attitude inmost medical centers toward general practitionerscontinued to be a condescending one. Clearly, thebattle to produce practitioners was unlikely tosucceed without a major change in the climate ofthe medical school and the fundamental goals notonly of the medical student, but also of the medi-cal faculty.

Recognition of the need for primary carephysicians, through the efforts of Millis and others,and the shift in career interest in medical studentsto family physician and related careers now re-quired the development of substantive educationalprograms. These programs needed to be developedin the very settings that had proved so hostile inthe past; namely, the teaching hospital and themedical schools. Efforts to develop educationalprograms outside the medical centers had alwaysbeen perceived away from the mainstream of medi-cal education; and, even when successful, such pro-grams had done little to change the basic commit-ment of the medical center. Medical students hadcome to medical school as potential general practi-tioners and had been graduated as embryo special-ists. Would it ever be possible to produce programsin this setting that would alter the failures of thepast? Because considerable study had been madeof medical education itself, we will briefly examinenwilic;t1 students, faculty, and their medical schoolexperience. We can also examine specific programsthat can provide us with suitable guidelines for cur-rent attempts to develop new programs in primarycare.

Selection for Medical SchoolIt is possible that the majority of students ac-

cepted in medical school were the ones least likelyto pursue primary care careers? Acceptance tomedical school has always been vet v competitive;and medical school admission committees havebeen faced, since the end of World War I I, with theprospect of selecting classes of 100 to 200 studentsfrom an applicant pool numbering in the thou-sands-- only 50 percent of whom were ultimatelyaccepted.

In general, accepted candidates were more likelyto have majored in the sciences, to have had highscores on their medical college admission tests, andto have had high grades in college. These studentsgenerally attended private rather than public uni-versities and overrepresented the middle and

15

upper-middle classes (J. Med. Ed. 1969; Gee,1958: Rodzinski, 1965).

Yet, the majority of these students, at least onentering medical school, indicated a desire topursue primary care careers (Hagerty, 1963). Asseniors, however, their career choices were de-finitely directed toward the specialities. Their com-mitment to a primary care career had not beensufficiently strong to resist the pressures of special-ization that existed within the medical educationsetting.

Perhaps potential primary ire physicians whocould have resisted the pressure: to specialize werediscouraged from applvingto medical school. Forexample, would students who were less scien-tifically oriented have pursued primary carecareers? In the 1950's and 1960's. there were at-tempts to encourage students who had a majorinterest in the nonscientific fields to apply to medi-cal school. However, medical school cataloguesstated that those who applied with a minimum ofscientific courses would be expected to excel inthese courses (Dove, 1970). The nonscientist couldhave been easily discouraged by such an approach.

What about students who were underrepresentedin medical school such as blacks, women, and thepoor? Were they the students who would haveresisted the pressure to specialize? While medicalschools could point to the fact that in the fiftyyears from 1900 to 1951 the medical student poolhad become more, rather than less, representativeof the population at large (Adams, 1953). therecould be little denial that blacks, women, and thepoor were underrepresented. For example, before1910. the financial burden of a medical educationwas not great; and a poor individual could considerbecoming a physician (Stevens, 1971). The expenseassociated with university education and prolongedspecialty training meant that the expense of amedical education was now a significant barrier tothe poor; in addition, blacks and women were dis-couraged from applying, the former by economicand racial discrimination and the latter by definingmedicine generally and many of its more attractivespecialties as careers fur men.

Suggestions were made that one way to getphysicians for the ghetto was to recruit from thissetting. For example, the black, poor students wereexpected to return to their disadvantaged com-munities as primary care practitioners. While lessscientifically oriented, these students presumablywould be more humanistic and more representativeof their race. However. there is no evidence thatthese students either possessed such characteristicsor greater promise than those who had not suffered

16

deprivation or, if they did, would become primarycare physicians if exposed to the same medical ex-perience as their colleagues and predecessors.

The Medical School ExperienceHow do medical education and the medical

school setting actually influence the medical stu-dent? To what degree does medical educationshape the attitudes and career choices of the stu-dent and to what degree does the student react tothe society outside of the medical school? Thesequestions are fundamental if, as suggested earlier,we arc to believe that medical educational reformcan influence the medical undergraduate and hischoice of career.

In the 1950's, Becker and colleagues at theUniversity of Kansas described a "secret" societyof medical students, which was formed in responseto the academic shock experienced by the begin-ning medical student (Becker, 1961). This shockwas caused by the overwhelming amount of mate-rial presented to the students to master. The stu-dent's ability to memorize was emphasized ratherthan his desire to help patients. The student there-fore, did not perceive himself as belonging to theprofession of medicine. The faculty viewed the stu-dents not as professionals but as students on proba-tion and, hence, the title "Boys in White." Thestudents, through their various subcultures such asthe fraternity or cadaver groups, oriented theirbehavior to this new reality. They structured theiractivities so that they conformed to the facultyculture. Many of Becker's findings were confirmedby Miller (1962) at Buffalo.

On the other hand, at Columbia, Merton (1957)found that the faculty and students formed a muchdifferent relationship in response to the same cul-tural shock. He characterized this different rela-tionship by the use of the term "student physi-cian". Rather than develop a secret society, thestudents began from the first year a process thatallied themselves with the faculty. Merton notedthat students at schools like Cornell, Pennsylvania,and Western Reserve felt that the faculty took amuch more personal interest in them than atKansas and Buffalo; and he also observed that thisinterest was an important expression of the col-league relationship.

The colleague relationship developed in spite ofthe fact that initially the goals of the Columbiastudent and faculty were not identical. The facultysaw the production of scientific physicians as theirchief educational task: while the students, initiallyat least, identified themselves strongly with careers

in general practice. Gradually but surely, thestudent adopted the faculty goals. They aspired tobe either academicians or practicing specialistsrather than practicing generalists.

Bloom's (1971) study at State University ofNew York at Brooklyn suggested a third kind ofstudent culture, intermediate between fly! previoustwo. He found that the faculty was interested inresearch and that they believed that teaching wasan imposition on their time. The student viewedhis medical school experience as a necessary four-year initiation' that be had to endure to become aphysician. The students wanted to practice medi-cine as specialists, maintain relationships with themedical schools as teachers but not as researchers.and were interested in patients as opposed toimpersonal technical problem solving.

Whatfull

we say about the faculty ?Thefaculty, full or part-time. basic science or clinical.represent a wide range of thought and experience:and generalizations about them must be madecautiously.

Students first come in contact with the pre-clinical basic science faculty. It is this facultythat presents students with the vast amount ofmaterial that results in the previous describedcultural shock. It is the basic science faculty that ismost likely to consider students in a probationarystatus rather than colleagues (Becker. 1961): it isthis faculty that is least likely to provide the stu-dent with role models (Hughes. 1959).

On the other hand, the clinical faculty also hasits conflicts with students. Most of the full-timeclinical faculty members arc anxious to do re-search, and they view teaching as interferring withtheir major research activity (Bloom, 1971). Manybelieve their students are too "practically"oriented. Bloom found a different relationship be-

tween the part-time and full-time clinical facultyand students. The part-time faculty members weremore likely to share common goals with studentsthan were full-time faculty. Also, the full-timefaculty wished to train academically, orientedphysicians and influenced students away from theiroriginal goal of becoming generalists. Coker ( 1960)presents evidence that faculty can influence thecareer choice of students.

Mendel 1965 observed that the attitude ofstudents towards patients tended to parallel theattitude of the faculty towards students. Thus, ifthe faculty were awhoritarian and punitive, thestudents were likely to display similar relationshipswith their patients. ottheil (1969) found thatwhen the student during his third year perceivedhis environment as warm. reinforcing, and hu-

manistic. he managed his patients in a similar wayas a fourth-year student. Thus, if students werehostile to their faculty, one would be concernedabout the general attitudes that students adoptedtoward their patients.

Dowling (1964) however, reported that studentsfound their full-time clinical teachers to be moresupportive and to deal more humanely with pa-tients in the hospital than did the part-time clinicalfaculty. This fact suggests that the student whoidentifies himself with patient care could findsomewhere in the immense medical school facultya role model who would reinforce in the student'smind the kind of physician the student wishes tobe.

Despite these findings. the medical school mayhave relatively little influence on its students; itmay be. as noted in a previous section. that stu-dents, faculty, and the medical school are moreinfluenced by society at large. (See page 17.)

Becker (1958) and Eron (1955) noted thatstudents arrive in medical school humanisticallyoriented; but, by the end of their medical schoolexperience, they are more cynical. This is ex-plained not by the alleged dehumanizing process ofmedical education as has been suggested. but bythe fact that it is a temporary adaptation on the

part of the student to the pressures imposed onhim by the medical school. As the medical schoolexperience ends. the student's original idealism re-turns. By this time, however. it is unlikely that thestudent will express his idealism through thepursuit of a primary care career.

In summary. medical education is the acquisi-tion not only of skills and knowledge. but also ofthe values, interests, and attitudes of the profes-sion. The climate of the medical school and theacceptance of a student either as a boy or a col-league should have an enormous impact on the at-titudes and career choices of medical students.

In the following sections. we will considerundergraduate. graduate. and continuing educationprograms in family medicine. internal medicine andpediatrics that were directed at primary care. Thisdiscussion must have the limitation of any litera-ture review. Many programs no longer exist in theform described. Most of the articles are descriptiverather than evaluative: many describe proposedprograms that are not currently operational. Never-theless. what has been reported to date providesvaluable background for interpreting current edu-cational efforts. As we shall show. information isavailable that contains almost all of the ingredientsnecessary for development of primary care pro-grams for the decades ahead.

17

References

S LCTION I I I

Adams. S.: Trends in Occupational Origins of Phvsi- Colwell. A. H.: Social and Environmental Factors inclans. Amer. Soc. Review, 18: 404-409. 1953. Medicine, J.A.Am.Med.Coll.. 21: 160, 1946.

Autfell. lames R.: The University Today. New York. 'Medical Care for the American People: The Final ReportNew York University Press, New York. 1933. of the Committee on the Costs of Medical Care, Chicago,

Bache. 1LE.M.: The Teaching of Preventive Medicine in 1932.Europe and in the United States, Milbank Memorial Fund Conroe. J.H.: Research and Medical Education. ReportQuarterly. 42 No. 4. Part 1. 22-35. 1964. on Ninth Teaching Institute of the Association of America

Birc:en. C. R.: Extramural Clinical Teaching: The Medical Colleges, Colorado Springs. Col.. 1961.Preceptor System at Wisconsin, Wisconsin Med. J. 27: 141, oonilf!., R. H. and Stein, L.P.: Medical-Student Society1928. and Culture in Coombs. Phychosocial Aspects of Medical

Bartlett. N.M.: The Participation of Medical Social Training, Springfield. Charles C. Thomas. 1971.Workers in the Teaching of Medical Students, Chicago. Coombs, R.H. and Clark. Vincent: Psychosocial AspectsAmerican Association of Medical Social Workers. 1939. of Medical Training, Charles C. Thomas, Springfield. 1971.

Itec ker.11.S. Au d Geer. B.: The Fate of Idealism in Medi- Cooper. J.A.D.. New Directions in Programs of Medicalal S, pools. Amer. Soc. Rev. 23: 50-56. 1958. Education, J. Med. Ed. 33 Supplement 189-216, 1958.

Becker, H.S., Geer. B.. Hughes. E.C., Strauss, A,: Boys Curran. J.A. and Cokerill. E.: Widening Horizons inin White: Student Culture in Medical School, Chicago, Uni- Medical Education: A Study of the Teaching of Social andversitv ot Chicago, 1961. Environmental Factors in Medicine. New York. The Com-

Br. Ismail. WV.: SIlit: Aspects of the Clinical Teaching monwealth Fund, 1948.of the Social and Environmental Factors in Medicine. J.Ass. Dove. D.B.: Minority Enrollment in U.S. MedicalArtier. `,led. Coll., 22: 149, 1947. Schools 1969-70 Compared to 1968-69. J. Med. Ed.

Berrv. . G.: Medical Education in Transition. J. Med. Ed.. 45: 174-81, 1970.2S: 17-42. March, 1953. Dowling. H.F.and Cotsonas, N.J.: Attitudes Toward Pa-

tti:loin. S.W.: The Sociology of Medical Education, Mil: tients of Full Time and Practicing Instructors as Revealed inbank Metnoi ial Fund Quarterly 43: 143-184, 1965. Student Questionnaires, New Eng. JM. 271: 716-18. 1964,

Bloom. S.W.: The Medical School as a Social System, Dublin, T.D.:The Basis for Teaching Social and Environ-\What, k Memorial Fund Quarterly 49. Part 2. April 1971. mental Factors in Medicine in the Undergraduate Medical

1i, we rs. 12. Status Quo Editorial) J. Med. Ed. Curriculum. J.A.Amer.Med.Coll., 22: 142, May, 1947.: !

1 1 i- 3.1. 19(.2. Duff. R.S. and Hollingshead. A.B.: Sickness and Society,cdillion. 1.11.: Sonic Clinical Aspects of Social Medicine. New York: Harper and Row, 1968.

\, w 1'n..!. IM. '34: 211.1046. Ehaugh. F.G. and Noble. R.A.: Psychiatry in MedicalCannon. 1.M.: Social Case Teaching of Medical Students, Education, New York National Committee for Mental

J .A. Anier. Med. Coll.. 9, 141-43, 1934. Hygiene. 1933.(..pl.,% it . I).: Faculty Evaluation of the Cornell Com- Ebaugh. F.G. and Rymer, C.A.: Psychiatry in Medical

pi-Ohnave Care and Teaching Program in Reader and Goss. Education, New York, The Commonwealth Fund, 1942.p. 3i5 -351. Eron, L.D.. Effect of Medical Education on Medical Stu-

Caughey. 1.1 .. Jr.: Education t the Physician Io: Le for His dents, Attitudes. J.Med.Ed. 30: 559-566. 1955.it,dc. in Comprehensive Health Service. J.Med.lid., Evans. L,: The Crisis in Medical Education, Ann Arbor.?.1 381-4. 19.59. University of Michigan Press. 1965.

(.1It L. KA i.: Plot wive Medicine in Medical Schools, J. Flexner A.: Medical Education in the United States and\led. :.d.. 28: 1 123, ( k tuber. P.11 t 2. 19; 3. Dept. of Canada, A Report to the Carnegie Foundation for thet :, Jura do Springs Conference, Nov. 1932. Advancement of Teaching. Bulletin No. 4, New York,

(.hire. K. F.: The cwileral Practitioner, Toronto, Univer- 1910.its of Toronto Pres,. 1963. Flexner, A.: Medical Education in Europe. A Report to

Co, 'Neon. E. The Pti.bparation of the Medical Student in the Carnegie Foundation for the Advancement of Teaching,the Re, o:fnition of the S::ial Component of Disease, Hosp. Carnegie Foundation, New York, 1912.hog., 22: 3(,1 .1941. Flexner. A.: Medical Education. A Comparative Study,

coxeshall. 1..T.: Planning for Medical Progress Through New York: The Macmillan Company. 1925.1..H. rf o.,:. Evanston. 111., Amer. Ass. Amer. Med. Coll., 107 Friedson: The Hospital in Modern Society, New York.I' 191,;.

Free Press of (Aencoe. 1963.Lohen. 1.. and Derow. 11.A.: Training of Interns in the

5, . i,Il Aspects of Illness. I AMA. /17: 1817. 1941.Coke!. R.. Jr.. Back. K.W.. Donnelly. T.G.. Miller. N..

Strickland. I..H.: Patterns of Influence: Medical SchoolFa, oh, N,1,11,1,,is and th, Values and Specialty Interests ofMedical Students, I. Med. E...?5: 518.27. 1960.

I S

Friedson. E.: The Profession of Medicine. New York,Dodd. Mead & Co.. 1970.

Funkenstcin, D.H.: Implications of the Rapid SocialChanges in Universities and Medical Schools for the Educa-tion of Future Physicians. J. Med. Ed.. 43: 433-54. 1968.

Funkenstein. D.H.: Current Medical Stood Admis-sions: The Problems and Proposal. J Med. Ed..45: 497.509. 1970.

Garfield. S.: The Delivery of Medical Care ScientificAmerican, 222: 15. 1970.

Gee. H.H. and Glaser. R.J. (eds): The Ecology of theMedical Student. E:anston, III.. AAMC, 1958.

Gottheil. E.. Hassenfeld, L. Gronkiewicz. E.: StudentsPerceptions of Medical School and Their Attitudes TowardsPatients. Brit. J. Med. Ed.. 3: 355-58. 1969.

Gough. H.C.: The Recruitment and Selection of Stu-dents in Coombs. Pyschosocial Aspects of Medical Training.Springfield: Charles C. Thomas. 1971.

Group for the Advancement of Psychiatry. Committeeon Medical Education: the Preclinical Teaching of Psychi-atry. Report No. 54. New York. Group for Advancement ofPsychiatry. Vol. 4. 1962.

Haggerty. R.J.: Family Medicine: A Teaching Programfor Medical Students and Pediatric House Officers..Med. Ed.. 37: 531-580. 1962.

Hagerty. R.J.: Etiology of Decline in General Practice.J AMA 185: 179-181. 1963.

Hall. W.S.: Medical Education in America Past. Present.and Future, JAMA. 27: 1265. 1896.

Hammond. K.R. and Kern. F.. Jr.: Teaching Compre-ht.nsive Medical Care, a Psychological Study of a Change inMedical Education (The Commonwealth Fund) Cambridge.Harvard. University Press. 1959.

Hargrove. E.A.. Han. G.C.. Fleming. W.L.: Multidis-ciplinary Teaching of Human Ecology in the First Year ofMedicine. J.Med.Ed.. 32: 697. 1957.

Harvey. S.C.: Social and Environmental Factors in theProvision of Medical Care. Bull. Med. Soc. County of Kings.N.Y.. 25: 182. 1946.

Hughes. E.C.: Stress and Strain in Professional Educa-tion. Harvard Educational Review. 29: 319-329, 1959.

Hunt. J.H.: General Practice in the World Today and itsAcademic Needs. Lancet 2: 29-32. 1964.

Huntington. M.J. and Merton. R.: The Development of aProfessional Self Image in the Student Physician. Cam-bridge. Harvard University Press. 1957.

Huntley. R.R.: Epidemiology of Family Practice. J AMA185: 175-159. 1963.

Hutchins. E.B.: The AAMC Longitudinal Study: Im-plication for Medical Education. J.Med.Ed. 39: 265-277.1964.

Jefferys. P.M.: The Teaching of Community Medicine inthe Undergraduate Curriculum. Brit. J.Med.Ed. 3: 176-84.1969.

Jenson. F.. Weiskotten, N.G.. Thomas, M.A.: MedicalCare of the Discharged Hospital Patient. New York. TheCommonwealth Fund. 1944.

Undergraduate Origins of Medical Students. J.Med.Ed..44: 712-13. 1969.

Kane. R.L. and Kane. R.A.: Physicians Attitudes ofOmnipotence in a University Hospital. J.Med.Ed.. 44: 684.1969.

K e n dal . P. I. .: Medical Specialization in Coombs.Psvcholorical Aspects of Medical Training. Springfield.Charles ( Thomas. 1971.

Kennel . J.H.. Chichering. I).. Soroker. E.: Experiencewith Medical School Family Study. J. Med. Ed. .?6:

1649 171 n. 1961.

Kerr, W.J .: A Unique and Promising Experiment inMedical Education. California and West. Med., 24: 465.1926.

Korman. M. and Stubblefield. R.L.: Medical SchoolEvaluation and Internship Performance. J.Med.F.11.,46: 670-673. 1971.

LaSaine, T.A.: Teaching the Social Component of Medi-cal Care at Meharry Medical College. J. Natl. M.A.,32: 248, 1940.

Leathers. W.S.: Development of the Clinical Concept inTeaching Preventive Medicine. J.A.Am.M.Coll., 14: 21.1939.

Lee. P.V.: Medical Schools and the Changing Times.Evanston. Ill.. the Associa%ion of American MedicalColleges. 1962.

Lewis. C.E. ano Resnik. B.A.: Relative Orientations ofStudents of Medicine and Nursing to Ambulatory PatientCare. J.Med.Ed. 41: 162-166. 1966.

Lidz, T.: An Outline for . Curriculum for TeachingPsychiatry in Medical Schools. J.Med.Ed. 31: 115-22,1956.

Lidz. T. and Edelson. M.: Training Toni um) w'sPsychiatrist: The Crisis in Curriculum. New Haven andLondon. Yale University Press. 1970.

Lief. H.I.: Personality Characteristics of Medical Stu-dents in Coombs. Psychosocial Aspects of Medical Training.Springfield. Charles C. Thomas. 1971.

Magra %v. K: Medical Education and Health Serv-ices: Implications for Family Medicine. New Eng. JM 285:1407. 1971.

Mackenzie. J.: The Aim of Medical Education,Edin.Med.J. 20: 31. 1918.

Matarazzo, J.D.: Comprehensive Medicine: A New Erain Medical Education. Human Organization. 14: 4-9. 1955.

McKeown, T.: Medicine in Modern Society. London,George Allen and Unwin Ltd.. 1965.

Means, J.H.: The Clinical Teaching of Social Medicine.Bull. John Hopkins Hosp. 78: 96. 1946.

Melaney. H.E. and Mortara. F.: Medical Social FacultyStudies in Medical Education, J.Med. Ed.. 27: 19. 1952.

Melaney, H.E.: Environmental Case Studies by MedicalStudents. J.A.Ain.M.Coll. /4: 72. 1939.

Mendel, Wm. and Green. J.A., On Becoming a Physician.J.Med.Ed. 40: 266-72, 1965.

Menke. W.G.: Medical Identity: Change and Conflict inProfessional Roles. J.Med.Ed.. 46: 58-63. 1966.

Merton. R.. Reader. J.. Kendell. P. (Eds.): The StudentPhysician, Cambridge, Harvard Univ. Press. 1957.

Milles, G.E. Ed : Teaching and Learning in MedicalSchool. Cambridge. Harvard University Press. 1962.

Miller, H.: Fifty 'Years After Flexner. Lancet. 2:647-654. 1966.

Minot. G.K.: The Physician. Student. and Medical SocialWorker. Boston M and Surg.J. 193: 1090. 1925.

Minot. G.K.: Medical Social Aspects in Practice. Arch.I nt. Med-54 : I. 1934.

Minot. G.K.: Thoughts Concerning the Teaching of Medi-cal Social Conditions. J.A.Ain.Med.Coll., V.9: 147-149,1934.

Moll. W.: History of American Medical Education.lirit.J.Med.Fd. 2: 173 81. 1968.

19

Morgan, J.: A Discourse upon the Institution of Medical Robinson, G.C.: The Study of the Patient as a Whole asSchools in America. Philadelphia, 1765, Baltimore, Johns Training for Medical Practice, J.A.Am.M.Coll, 14: 65,Hopkins University, 1937. 1939.

Nelson, B.W., Bird. R.A., Rodgers, J .M.: Educational Robinson, G.C.: The Patient as a Person, New York,Pathway Analysis for the Study of Minority Representation Commonwealth Fund. 1939.in Medical School. J.NIed.lid. -In: 74 5-749,1971. Rodzi ns ki, E.: Social Class of Medical Students.

Newman, G.: Recent Advances in Medical Education in J.A.M.A.. 113: 95-8. 1965.London. England. HMSO. 1923. Rossi, A.M. and Neuman, G.G.: Evaluation of Faculty

Newman. G.: Sonic Notes on Medical Education in Attitudes Toward Patient Orientation in Medical EducationLondon, England. HMS0,1918. and Practice. J.Med.Ed., 42: 849-54. 1967.

Osier. W.: The Natural Method of Teaching the Subject Sanazaro, P.J. and Bates, B.: A joint Study of Teachingof Medkine JAMA 36: 1673. 1901. Programs in Comprehensive Medicine. J.Med.Ed. 777-89,

Oster, W.: Acquanimitas. Philadelphia, P. Blakistoris Son 1960.& Co.. 1905. Snoke, P.S. and Weinerman, E.R.: Comprehensive Care

Peabody F.W.. The Care of the Patient, JAMA Programs in University Medical Centers. J.Med.Ed.,40: 625-657. 1965.88: 877-82. 1927.

Peterson, 0.1- Andrews, L.P., Spain. R.S., Greenberg. Medical Sociology lectures sponsored by Dept. of Preven-IW.: An Analytic Study of North Carolina General tine Medicine, Harvard Medical School, Reprinted formPractice. 1953.54, J.Med.Ed. 31: 1-165, Part 2, 1956. New Eng. J. Med., Jan 3-Feb. ?8, 1946.

Stevens, R: American Medicine and the Public Interest.New Haven. Yak University Press, 1971.

Stokes, J., Blodgett. F.M., Rutstein, D.: An Experimentin the Teaching of Family Medicine, J.Med.Ed., 38: 539,1963.

Susser, M.: Teaching Social Medicine in the UnitedStates. Milbank Mem. Fund Quarterly 44: Nu. 4 Pt. 1, 389,1966.

Thorton, J. and Knauth, M.S.: The Social Component inMedical Care: A Study of One Hundred Cases from thePresbyterian Hospital in the City of New York. ColumbiaUniversity, 1937.

Warbasse, J.P.: Medical Sociology in the Medical SchoolCurriculum, JAMA, 9V: 779. 1932.

Webster, T.G.: The Behavioral Sciences in Medical Edu-cation and Practice in Coombs, Psychosocial Aspects ofMedical Training, Springfield, Charles C. Thomas, 1971.

Whitehorn, J.C.: Orienting Students Toward the WholePatient. JAMA 538, 1957.

Whitehorn, J.C., Jacobsen. C., Levine, M., Lippand, V.W.(Eds.), Psychiatry and Medical Education. Washington,D.C.: American Psychiatric Association. 1952.

WHO Tech. Rep. Series, The Use of Health ServicesFacilities, 355: 5-36, 1967.

WHO Tech. Rep. Series, Undergraduate Medical Educa-tion, Geneva, 69: 66, 1953.

WHO Chron. The Family Doctor, / 7: 4 58 61, 196.

Willard. W.: Meeting the Challenge of Fatnily Practice.Chicago. III. Am. Med. Assoc., 1966.

Willard, W.R .: The Primary Physician. J.Med.Ed.44: 119, 1969.

Wolf, S. and Darley, W.: Medical Education and PracticeReport of the Tenth Teaching Institute of the Associationof American Medical Colleges, Colorado Springs. Col.,1962.

l'iucoffs, M.C.: Certain Trends in Undergraduate MedicalEducation, Annals of Internal Medicine (Phil) 41: 1 250-54.1954.

Popper, H. 'Ed): Trends in New Medical Schools. Gruneand Stratton, New York and London, 1967.

Preiss, J.J. and Long, E.C.,: Student Attitudes TowardEarly Career Commitment, Brit. J.Mcd.Educ. 4: 9-12.1970.

Price, P.: Performance Measures of Physicians Salt LakeCity: University of Utah.Press, 1963.

Proceedings Firs, World Conference on Medical Educa-tion. London. Oxford University Press. 1954.

Proceedings of the Second World Conference on MedicalEducation. World Health Organization. New York, 1961.

Propst, D.W.: The Patient is the Unit of Practice, Spring-field, III., Charles C. Thomas. 1939.

Rappeleye. W.C.: Medical Education. Financial Reportof the Commission of Medical Education, New York 1932.

Rardin, T.E.: Medical Students, Family Doctors, andFamily Practice. J AMA. 176: 479. 1961.

Reader. and Gess. M.E.W. (Ed): ComprehensiveMedical Care and Teaching, Ithaca, New York, CornellUniversity Press, 1967.

Reynolds. G.P.: The Teaching of the Medicosocial As-pects of Cases. N. Eng. J. Med., 220: 1, 1939.

Rice, E. and Hiscock, I.V.: Teaching the Social Com.',talent in Medicine, Yale J. Biol. & Med.. 11: 645, 1939.

Richardson. 11.B.: Patients Have Families, New York.Commonwealth Fund. 1945.

Richmond, J.: Currents in American Medicine, Cam-bridge. Harvard University Press. 1969.

R obinson. (;.C.: Proper Attention to the Role ofEmotional and Social Factors in Illness as New Step inPublic Health. Milbank Mem. Fund Quart. 23: 20, 1935.

IV. UndergraduateProgram

he closest medical education has come to aTplanned program for primary care was in the firstdecade of this century when undergraduate educa-tion plus a rotating internship constituted suffi-cient preparation for general practice. The develop-ment of medical specialties changed this situation.Moreover, many programs we are including as pri-mary care programs were not really directed atpreparation for practice. They were attempts toimprove the hospital care of patients. such asintroducing followup to hospital care (Curran.1945: Jenson, 1944) or attempts to reverse thefragmentation of care in the outpatient depart-ment. The programs focused largely on the psycho-logical and social aspects of medicine, only expos-ing students to limited aspects of family care(Kennell, 1961).

Unfortunately. there appears to be no broad the-oretical framework that provides us with a basis forpresenting and analyzing undergraduate primarycare programs. There were a limited number ofplaces where primary care could be taught. andmost of the programs appear to have been shapedlargely liy the site of the program rather than byany major philosophical goal. For example. pro-grams in outpatient departments in different hos-pitals had much more in common with each otherthan did programs within one institution located indifferent sites. Representative sites were the physi-cian's office !which included preceptor programs),the patient's home in home care programs. occa-sionally the hospital's wards, and, most recently.mcklel practices and health centers (Sheps. 1 953:

Faulkner, 1953:- Fiarrell. 1968: Carnegie. 1970).Even when multiple sites were ..ised, as in the com-prehensive care programs. one site usually domi-nated ind provided the program wHi its distinctivecharacteristics.

It is difficult to find any program that had asa specifically stated goal the teaching of primary

care. as we have defined it. However, the primarycare programs did seek to close the gap that existedbetween the social and the technical in medicaleducation. Generally, the programs sought tofoster what could best be called positive attitudestowards patients, families, and colleagues. Areas ofprogram concern included the physician-patientrelationship, teamwork, the appropriate use of con-sultants, and record keeping. Internal medicine,pediatrics, surgery. obstetrics, and psychiatry werethe clinical disciplines in which these skills andattitudes could be taught. Psychosomatic problemsand chronic illness provided clinical experienceswhere these skills were most needed and could beapplied. The faculty included full-time andpart-time staff, usually from the specialties; onlyrarely, did general practitioners participate as

faculty.

The Physician's Office (Preceptorships)One of the immediate results if the Flexner

report was the disappearance of the proprietaryschool. Because the apprentice system was closelyidentified with the proprietary school and precep-torships are a form of apprenticeship, this form ofeducation was .j sickly abandoned by most medicalschools. The early method whereby a student al):prenticed himself to the physician for a period ofthree years. then attended lectures for one year,and then returned for another period to his precep-tor was seen as part of the proprietary schoolmodel. This period of seven years often reduced tothree years in the United States, particularly on thefrontier actually exposed the student to thedoctor-patient relationship before be began hisinedicil lectures. Exposure to what we consider thecentral experience of medicine was a benefitpoorly appreciated, h)st in the switch to theFlexner system.

The limiting of the preceptorship also occurredin the United Kingdom. Here the apprenticing of

the student to the practitionei remained a part ofthe educational traditi'n; but was of varying con-tent and quality. The oldest preceptorship programwas at Edinburgh. In 1776. the University began anassociation with the Edinburgh Royal Dispensary,which served the poor of the city and was staffed1)% general practitioners. When the National HealthService came into operation. the dispensary wasconverted into two national health general prac-tices and the University established departmentof general practice (Scott. 1950, 1956, 1960,1967).

Preceptor schemes intended to introduce stu-dents to general practice were started at St. Mary's(Barber. 1952) in 1935 and Sheffield in 1951(Hubson, 19521. In 1953, national surveys werecompleted by the College of General Practitionersand the British Medical Students' AssociationMesh p. 1 953: Mac( :lean , 1961: British Medical

Journal. 1953. (n the 23 schools studied, threehad compulsory schemes. Generally, preceptor-ships were characteriied by their site, such ashealth center or practices. attachment or residen-tial schemes. and da% visits. Other than the healthcenter schemes. which were few in number, thedifference between the others appeared to be thelength of attachment. with most students spendinga day in the day visit, a period of one or two weeksin the attachment schemes, and usually a month inthe residential scheme. The length of preceptorshipwas important as several evaluations I Dean. 1971)suggested that the longer the student spent in theattachment the more positive he telt toward hisexperience.

At a national Meeting of the British Medical Stu-dents' Association in 1965, it was unanimouslyagreed that iNnerai practice schemes '1f at least twoweeks' duration should be compulsory for all stu-dents at all schools. Strong support was also givenby students for the setting Up of departments ofgeneral practice.

The Charing Cioss precepturship is a typicalcompulsory (Me. Fifth year students spend a

period of one to two weeks with a selected practi-t ion e r Arnold. 9(!4,. The preceptorship. atAberdeen. which originated as a voluntary effortbut is now compulsory. was similar. with studentsinitially spending one week with a practitioner.Students in Ay. spend one day with the academicdcpirt mem of general practice. diree days with thepractitioner. and final day discussing and p,esellt-ing their finding. thy conference organiteu bythe de part ment Richardson. 1966'1.

Pearson's study '1968: in 1967 noted that allbut one of the 26 1;titish si pools had preceptor

)

programs. One-third of the preceptor schemes werecompulsory and two-thirds were voluntary. Attach-ment was usually for two weeks and took place inthe fifth or sixth. year. In one-third, the studentswere briefed before the attachment. Only oneschool held a seminar. Reports from students wererarely called for and rarely discussed. The investiga-tors concluded that medical schools were casual intheir approach to the problem and that the existingschemes were "amateur, haphazard, and providedlittle or no feedback either to the medical schoolor to the student, and rarely to the practitioner."

By 1969 (Harris, 1969) all schools had someCorm of attachment and half of the 28 schools hadcompulsory schemes. About half reported sonicliaison between the preceptors and the school andrequired reports from their students; about halfeven paid their practitioners an honorarium. A pat-tern can be seen of increasing interest in preceptor-ships and increasing student demand for suchexperience.

Must of the studies evaluating the preceptorshipsin the United Kingdom have been based on opin-ion. Brotherston (1959), in his survey of Edin-burgh graduates. noted that students had learnedabout unfamiliar common illnesses (common in thecommunity but uncommon in the hospital) andthe management of illness outside of the hospital;they had also acquired new attitudes towards gen-eral practice. which were almost always positive.

Other countries have made sonic. use of the pre-ceptorship where general practice is part of theestablished medical care system. These include Hol-land, Yugoslavia. and Israel. Mertens 1966) in Hol-land described a clerkship, which was an electiveone-month attachment to a general practice. Thecourse. which was subsequently lengthened to twomonths, was chosen by 75 percent of the students:45 percent had elected to become or would begeneral practitioners.

Vuletic (1964) in Yugoslavia described a two-week attachment. Students were attached to prac-tices in pairs and participated in seminars at themedical school where 'cases seen in practice werediscussed. An ambitious study is now under way totest the effectiveness of this course, which willdepend on a continuous analysis of the quality ofwork rrformed by the general practitioner. gradu-ates, and faculty.

Prvwes (1961) described a four-week programfor final-year students at the Hadassah MedicalSchool in Jerusalem.This course was the responsi-bility of the Department of Medical Education.Again, the evaluation of the program depended onquestionnaires administered before and after the

course and dealt mainly with attitudes. Studentsdescribed the course as particularly valuable forrural practices.

In the United States. till first preceptor schemesafter the Flexner report made their appearances inCalifornia ( Kerr. 1925) and Wisconsin Bardeen.1928). Kerr described his effort as "a unique andpromising experiment in which students spenttwo weeks with a general practitioner caring forprivate patients: he emphasized the importance ofcareful selection of the practitioner faculty.Bardeen established a program at Wisconsin simul-taneously with the development of the new. four-year medical school. Generally. the preceptorswere selected from among those considered thebest practitioners and teachers (Parkin. 1959). Theevaluation of the Wisconsin program (Bowers.1957. 1960) showed that 13 percent of the gradu-ates became general practitioners. Other programsfollowed; primarily in rural states such as Nebraska(Lee. 1966). Kansas (Wescoe. 1956). VermontWolf. 19571 and South Dakota (Slaughter. 1949).Using time United Kingdom classification. most ofthese were residential schemes.

Perhaps one of the best described programs inthe United States is the one at Kansas (Wescoe.1956: Rising. 1962). To quote the investigators.this course of tour and one-half weeks in thefourth year. "received the unqualified endorsementof students. faculty, and preceptors." The purposeof the compulsory program was to give the studentrural practice experience. Practices were selected intowns of a size no larger than 2.500.

In addition to these attachment schemes a nun)ber of medical schools in the United States and theUnited Kingdom offered elective periods enablingstudents to spend one to two-month periods in res-idence with general practitioners. In the UnitedStates. especially in rural practices. students selln-ingIV are permitted more patient responsibility. Inthe United Kingdom. some resistance to studentsassuming responsibility has come from the generalpractitioner who is concerned about his own relytionship to and responsiAity for his patient. Arecent study in the United Kingdom. however. indicated that only one out of twenty patients ob-

jected to having a medical student present or evenprovide services in the office of the general practi-tioner (Richardson, 1970),

The selection of faculty for these programsalso been of some concern. Parkin (19i9.. usingcriteria similar to Peterson's study of general prac-titioners. found that there were no unqualifiedpractitioners among the Wist onsii, preceptors and

the stlidents were C spuscd to good medicalpractice.

.41111111114.11y

LIMO ITCCIIIIV pl'eCeptUES111111 11,11V not been amajor considermion, at least in terms of currk ultuntime in the I limited States. Preceptorships have beenused widely in other countries. Both students(Rosenberg, 19591 and the general practitionerfaculty t Kindsc hi. 19:19) had high regard for theseexperiences. Often preceptor programs are poorlyorganised. although this mav be the results of cutriculum restrictions as much as by any inherentproblem with the method. Full -tinge medicalfaculty have nut made any major effort to i.inprovedie preceptorship experience. thereby often re-

vealing their bias toward the hospital as not onlythe principal. but the 4,nlv site for medical studenteducation. Precepturships. however, bring studentsin contact with physicians who are practicing varying degrees of primary care. The experience is par-ticularly valuable. because. at least in the officesetting. the patients represent a cross section ofsociety rather than solely the disadvantaged. as willbe subsequently noted. who are most often seen inhospital programs. This seems especially so in therural as opposed to the urban setting.

The essence of the preceptorship is to watch askilled and experienced clinician practice medicineand. in some programs. to participate in the care

providing process. Watching an experiencedprofessional Call be valuable in anv number of set-.tings. and preceptorships shothd not be thought ofonly in terms of the practicing physician's office.! example. preceptorships could be developed inthe hospital. where too often the student is givenresponsibility without the opportunity to observeor be observed by an experienced teacher. Although this experience appears to have a useful

place in medical education. it has not \ beenfully developed as an educational es pericnce.Perhaps the lack of interest in the United Statescan he explained by the absence of a defined general practitionel service (Sweet. 1951 inaantull asthe preceptorship fur the most part has been associated with general practice.

The Patient's HomeHome care proK,.anis, like the preceptorship.

have a history that antedates the Flexner era andhas been particularly developed in he UnitedStates. Most often these programs originated in thehospital and developed out of a concern to initiatefAlowup for previously hospitalized patients. Onlyrarely were they developed to serve unhospitalizedpatients. Initially, il.e Were a service for indigentfamilies with no general practitioner who lived near

large institutions. The two oldest programs in theUnited States were started at Boston University(Bakst, 1950, 1957, 1959) and Tufts University(Olef, 1939; Gibson, 1965).

The Boston University program. has been asso-ciated with the Home Medical Service that wasestablished in 1776; Boston University began itsp 'irticipation in the program in 1874. The programprovided fourth-year medical students with the op-portunity to make home visits to families or pa-tients who would call the service for emergencyillness care. The student, who acted independently,would check with an available preceptor only if hefelt there was a need. The supervision was providedby residents and faculty internist, psychiatrist,and director. Most services were provided with theexception of obstetrics. The patients were indigentand lived near the hospital. The program was theresponsibility of the Department of Medicine, but,since 1949, has been the responsibility of the De-partment of Preventive and Community Medicine.The course was described by the students as one ofthe most popular at the medic, school; but noother evaluation was undertaken. it did not pro-vide for of care as the students couldonly see those patients who could be accom-modated by the number of students and theirsupervising faculty participating on any given day.An elective opportunity was also offered in thehome care program in which third-year studentscould follow one family for nine months. This partof the program. apparently short - lived, was notfurther described. A record system for the patientswas maintained separate from the hospital recordsystem.

Alt!,ough the Tufts program is almost as oldthe Boston Dispensary. where the program wasbased. originated in 1796. student teaching did notbegin until 1929.. The program. administered bythe Department of Medicity:. also involved indigentpatients. Students were assigned in their fourthyear for a one-month rotation. The students madehouse calls with district physicians and residents ofthe Boston Dispensary, One of the very few con-trolled studies to examine the effectiveness ofhouse calls was carried out in this program byGibson and Kramer (1965). They found that thehome as a site of care resulted in additional med-ical diagnoses being made compared with the officesetting. A similar program developed at Georgia(Svdenstricker. 1939) with indigents for patientsand hospital residents for faculty.

Home visits to selected outpatients were made ina number of medical schools (Hiscock. 1939;Robinson. 1939: Wciskotten. 1944: Neiderman.

1958). Usually organized by departments of pre-ventive medicine, these programs made use of thecase study method, which involved the study of anindividual case in great detail. Occasionally. a homevisit was made to a hospitalized or previously hos-pitalized patient (Baily, 1937).

At the Medical College of Virginia (Holmes,1953), a program was developed in coordinationwith the Richmond Health Department. Fourth-year students made home visits during a three-weekclerkship in medicine. The faculty were primarilyresidents in pediatrics and medicine. At Syracuse,students made home visits to hospitalized patientsand followup visits one year later (Weiskotten,1944). The purpose of the program was to makestudents apnreciate the role of time, which oftensolved problems that, in the short run. appeared tohave no ready solution. Unfortunately, there is noquantitative or qualitative information as to howsuccessful this followup program was.

A home care program at Johns Hopkins wasinitiated after a study showed adverse social condi-tions existed for 65 percent of an unselected series

-of patients admitted to the hospital (Robinson,1939). These conditions contributed directly tothe need for admission, and the program was de-signed to prevent future hospitalization by focus-ing professional attention on adverse problems inthe home.

Shrand (1966) described a home care service incentral London where children, who would other-wise have been hospitalized, were cared for in theirhomes by a team from the teaching hospital. Therewas little undergraduate teaching in this course, al-though the potential for this was considered excel-lent by the investigator. Students could makehome visits and occasionally were accompanied bythe registrar assigned to the program.

Some programs combined features of preceptor-ship and home care. At Vanderbilt (Melancy. 1934.1949), volunteer fourth-year students were as-signed to a practicing physician preceptor whotook the student on house calls and introducedhim to patient care in private practice. There werecase studies of hospitalized patients whose homeswere investigated either by the student himself orby the medical social worker. At Tennessee(Packer, 1954). the family of a patient examinedby the third-year student in the outpatient depart-ment became his total responsibility; and, thus, thestudent hei...ame the family physician. Generalpractitioners. who constituted the staff of the gen-eral practice clinic. served as preceptors for boththe intramural and extramural aspects of theprogram.

Summary

In general. the home care programs dealt withepisodes of illness rather than continuity of care.The faculty was hospital based and too often stillin training and inexperienced with respect to careoutside of the hospital. In alnr.)st all of the pro-grams, the patients were indigent and often elderly.The programs may have demonstrated to studentsthe value of going into the patients' home: but,again, there was no measurable impact on thecareers of physicians and no evidence that studentsbenefited from the experience. This experiencemight also have produced a negative impact; inso-far as many physicians now practice as if homecare, or more specifically, the need for a house call,is totally unimportant. As a site of health care, thehome is obviously important both for episodes ofillness and tor continuing care. Like the preceptor-ship, it belongs in any substantial program of pri-mary care an integral part of an overall programand not merely an isolated activity.

The Hospital's Outpatient DepartmentThe outpatient department is the part of the

hospital said to closely resemble a primary caresetting. It was developed as a consultative orsecondary care clinic also serving an emergencycare function. It usually provided only that part ofprimary care considered first contact. Moreover,the outpatient department. especially in large

urban hospitals, had become the place whereincreasing numbers of indigent families soughtcare, especially as general practitioners becameincreasingly unavailable.

Programs in primary care in the outpatient de-partment :vac among the easiest to establish due tothe availability of patients and the existence ofoutpatient departments in most hospitals. Becauselarge numbers of ambulatory patients were in-digent. meeting their needs did not put the institu-tion directly in Competition with private practi-tioners. The hospital was also assured that patientswould occupy its inpatient beds. Fragmentation ofservices was recognized by the staff in the out-patient department: consequently, many of theearly programs they developed attempted toprovide better services to patients whose primaryneeds were lost in specialty oriented clinics.

For these reasons the outpatient department wasa logical place to begin programs identified withgeneral or integrated care. Integration generallymeant reorganizing the outpatient department toput emphasis on total patient needs rather than ona disease. as had been the case in the specialty

clinics. However, these programs were essentiallysmall demonstration efforts and did little to changeactual delivers' of hospital ambulatory services. Theoutpatient department at Vanderbilt was organizedwith this goal in mind in the 1930's (Burwell,1935): this was followed by integrated programs atSyracuse (Weiskotten, 1944) and at Cornell (Barr,1953). After the Second World War, outpatient in-tegrated programs were described at a number ofschool 5: Washington University of St. Louis(Shank, 1956), Pittsburg (Gregg, 1956), NorthCarolina (White. 1957; Fleming, 1956), Buffalo(Bunnell, 1951), Yale (Solnit, 1954), Duke( Bogd o n o ff, 1963), Cornell (Reader, 1956).Colorado (Kern, 1956) and Northwestern (Snyder,1959).

The program at North Carolina evolved from areorganized outpatient department, which was partof the expansion of the medicv I school from a two-to a four-year school (Fleming, 1956; White,1959). As a general medical clinic, 't combinedmany of the specialty clinics with hospital special-ists as the faculty. Fourth-year students spent ap-proximately half of that year working in this clinic.The satin was especially fertile for studying refer-ral patterns to the outpatient department (White,1959). But there is no indication of its success orfailure in educating practicing physicians for theState, even though the school had as a stated goalthe education of practicing physicians.

In pediatric teaching, the outpatient departmentassumed considerable importance. At Yale (Solnit,1954), fourth-year students, during a six-weekclerkship, gained coordinated pediatric-psychiatricexperience that stressed interviewing skills. Thisteaching was also part of a program for interns andresidents.

A general medical clinic, in which six clinics hadbeen combined, was also developed for teaching atBuffalo. The course, described as successful. re-ceived no further evaluation (Bunnell, 1957).

ofearly innovation was the reorganization

of the Vanderbilt Clinic at Columbia Presbyterianto allow students to see patients of private referralphysicians (Cadmus. 1948). This was one of thefew attempts to have private patients participate inthe outpatient setting in student teaching, but itinvolved consultative services and not primary carepatients.

At Oklahoma, the reorganization of the out-patient department was accomplished as a demon-stration of liaison between the University MedicalCenter and the referring physician with a second-ary goal of increasing senior faculty participation

25

in the ambulatory area Colinore. 19541. There wasno reported evaluation to indicate whether or notsenmr faculty did become involved in the out-patient as opposed to inpatient care.

In summary. almost all of the outpatient depart-ment _programs attempting to improve teachingwere:faced initially with the need to reorganize theoutpAtient department. They did not have an ex-pressed goal of producing primary care practi-tioners. The student, resident, or undergraduatewas attracted to the outpatient department, be-cause he could assume almost complete responsi-bility for a new patient: whereas on the inpatientservice, he always shared responsibility for the pa-tient. While working in the outpatient department.the student might recognize the seriously ill pa-tient: but the experience did not help him to man-age the patient without significant organic disease.Moreover. the student might not be providing high-quality care. In auditing charts of students.Bauniont (1967) found that of 250 records audited39 percent tailed to meet the criteria of qualityused in that study. The most obvious omission wasinadequate communication with the referringphysician. Also. little attention was accorded thepenalty

ofby the families and patients who. in-

stead of a phsician. saw a less-experienced studentin the outpatient department.

outpatient programs began to deliver more pri-mal-% care in the post-World War II period. Thiswas particularly true in

ofemergency clinics,

because fewer sources of primary care existed inthe community. The situation created major prob-lems. inasmuch as the outpatient department. as asetting. shares the overwhelming constraints of ahospital: namely. its ,n-ganizational relationshipsand high costs.

Fur esample. Bogdonoff (1963) noted theadministrative hurdles that faced any outpatientdepartment to reorganize to achieve a more com-prehensive program. These included emphasis onspec ialties., the hospital bureaucracy., the de-emphasis ut the total patient. and the difficulty incoordinating services. He concluded that, even M.the best of circumstances. care of patients in theoutpatient department was a difficult task and thatmi\ing educational. research, and service goals onlycomplicated the situation.

Any program in an outpatient department suf-fered due to its inability to alter in a major way itsrelationship with the hospital. Neither did changesin the organization of the outpatient departmentfacilitate the treatment of Common disorders.26

because the relationship between physician and pa-tient was disease-and crisisoriented. discontinuous.and noncomprehensive. The faculty remainedprimarily specialists and subspecialists, and the in-fluence of specialty organization geared to findingdisease was overwhelming. Despite the expressedinterest in the outpatient department. it was theless experienced physician who continued to workin the outpatient department and the senior staffwho taught on the wards. A totally differentpicture ex isted in Great Britain where theoutpatient unit remained a secondary and consulta-tive clinic and was staffed by a consultant or hiswell-trained registrar.

Perhaps most important. the outpatient depart-ment is part of a two-class health care systemusually providing care to patients whose morbidityand mortality are among the highest in our society.Locating educational programs in this settingmeans that students would deal almost exclusivelywith a disadvantaged population as they did inhome care programs.

Thus, an important criticism of outpatientdepartment programs remains its inappropriatenessfor primary medical care. Outpatient departmentswere properly designed for the referral patient.either for special consultation or for a genuineemergency. but not for that important part of pri-mary care that involves health maintenance, healtheducation, and treatment of the family as a unit.

Health AdvisorsOne attempt to introduce students to medical

care outside of the hospital while developing arelationship with patients was to have the studentact as a health advisor rather than as a physician

to patients and families in a number of sites, suchas in the patient's home. the outpatient. department.and the hospital. Most. but not all, of these pro-grams were offered in the preclinical years. AtLong Island College of Medicine (now State Uni-versity of New York at Brooklyn) shortly beforeWorld War II. cases were assigned to medical stu-dent., in the third year and discussed one year later(Curran. 1945). There was no followup as to theeffectiveness of this program.

At Cornell. a plan .was developed for third-yearmedical students to follow a family for two yearsthrough frequent house visits. This program wascalled the Family Health Advisor Project and wasfirst offered to third-year students who followedtheir assigned families for 15 months. World War IIinterrupted the project. but it was reestablishedafter the war 1953).

A similar program at Western Reserve was anearly precursor of the extensive curriculum revisionthat occurred in that institution after the war(Kennel'. 1961). An entering student was intro-duced as a health advisor to a family often thefamily was disadvantaged and black; and the wifepregnant. Through periodic health and home visits,the student was to follow the family through hisfour years of medical school.

At Vanderbilt (Quinn, 1960). the student, actingas a health advisor during his first two Years atmedical school, served primarily as a liaison be-tween his assigned families and their so.irce of hos-pital care. A smiilar program was developed atLouisville as an elective (Miller. 1961).

Perhaps the best reported example of the FamilyHealth Advisor system was that developed atPennyslyania in 1949 (Appel. 1953: Hubbard,1952: Hubbard. 1954; Mc Mitchell, 1952). In hisfirst year of school, the student became the healthadvisor to a family. Families representing all eco-nomic strata were especially selected for the pro-gram by the program social worker. The studentvisited his families regularly and discussed his ex-periences at preventive medicine seminars. He wassupported by an interdepartmental staff consistingof two clinicians, one psychiatrist, and a socialworker. Although described as successful in 1952and expanded. the program was not further re-ported on or evaluated.

Parmalee (1960) developed a similar program at.UCLA in a well-baby clinic. In the first year ofmedical .school, students made home visits to achosen family: in the second year. a second familywas added. In the third year, the program includeda monthly seminar as well as visits to an additionalfamily in which the mother was pregnant. The pro-gram occupied a total of 116 hours through thefour years of medical school, documenting justhow little curriculum time was devoted to familycare teaching.

A more contemporary attempt at family advisorprograms has been accomplished by students theselves in health advocacy programs (Rogati, 1971.McGarvey, 1968. Students reached out to a dis-athan taged family and advise to thein how to getmedical care. They screen populations to work toestablish care facilities. They also have as theirgoal the establishment of courses in their own med-ical school. Often the students are overwhelmed bythe complex needs of the families and the com-munities. But, on the plus side, student initiativecare can have very tangible results, such as theestablishment of clinics and neighborhood healthcenters ( Johnson. 1969),.

SuptimaryThe family advisor programs were generally

described as successful in the preclinical years,most probably as a result of the patient contactthey offered students. However, there was no ex-tensively plumed evaluation of these programs. Inthe clinical years, the dominant hospital cultureagain took over: and these programs suffered bycomparison with the drama of ward medicine(Kennel'. 1961). Where most successful, the pro-gram provided students some degree of responsi-bility, which was subsequently enlarged upon inthe comprehensive and family care programs.

Comprehensive Care

Following upon the Family Health Advisor pro-grams, the efforts in psychiatry, preventive medi-cine, medicine and pediatrics came together in pro-grams of comprehensive care. Generally theseprograms, although differing in title, sliared apatient-oriented rather than a disease-orientedapproach. These efforts. unlike the earlier pro-grams, were largely experimental and substantiallyfinanced by private foundations, particularly bythe Commonwealth, Rockefeller, and KelloggFoundations. An important innovation of theseprograms was the introduction of the behavioralscientist, such as the medical sociologist, to clinicalmedicine departments (Weiner, 1961),

Representative programs were developed atT:.inple (Steiger. 1956. 1957. 1960). Colorado(Hammond, 1959) and Cornell (Reader, 1953.1956. 1959, 1964). Additional programs wereundertaken at a number of schools on a smallerscale (Peterson, 1959: Weinstein, 1956; Johnson,1959). All of these schools had undertaken earliereducational efforts to create a climate lor thesechanges. All had strong leadership and, in eachschool. sympathetic and understanding administra-tive support (Magraw, 1971).

The program at Temple began in 1952 as aweekly conference for senior medical students dur-ing a one-month clerkship in the general medicalclinic under the collaborative direction of an in-ternist and a psychiatrist (Steiger. 1957: Neibuhr,1960. Teaching was given in 16 hours of confer-ence during the clerkship. Lecturers. seminars. andclinics were added for first-, second-, and third-yearstudents with a six-week clerkship in the medicalclinic for fourth-year students. Third-year studentsalso followed a patient with a chronic disease forone year. Later this was replaced by student super-vision of the health care of a family. Although

27

there were no fundamental changes in the curricu-lum, the program gave fourth-year students inten-sive outpatient clinic experience. The primaryobjective of Temple was stated to be the educationof broadly oriented generalists (Lee, 1962). How-ever, there is no record of success or failure inachieving this goal.

In 1953, the University of Colorado initiated aGeneral Medical Clinic for teaching comprehensivemedical care at the Denver General Hospital. Com-prehensive care in this program meant that thephysician assumed total responsibility for his pa-tient's health care. The investigators felt that thesuccess of comprehensive care was related primar-ily to an attitude rather than a skill. The purposeof the General Medical Clinic was to provide anopportunity for the medical student to learn atleast as much about fundamental medical skills asin the classically organized medical clinics: to pro-vide him with additional knowledge, particularly inthe areas of sociology and psychology: and to pro-vide a setting in which the attitudes leading to thepractice of comprehensive medical care would bedeveloped and maintained.

The curriculum, of the General Medical Clinicwas covered in a six-month block period in thefourth medical school year. The student spent fiveone-half days a week for eighteen weeks and twoone-half days a week for six weeks in the clinic.The remainder of his time was spent outside of theGeneral Medical Clinic in medical, pediatric, andobstetrical services. In addition, the student in theGeneral Medical Clinic was assigned to traditionalspecialty at Colorado General Hospital fortwo one-half days a week. The other two one-halfdays a week provided a modified clinical experi-ence on the medical wards at Denver General Hos-pital. Here, the student followed his clinic patientsthrough their hospitalization. The program pro-vided: continuity of student-patient relations for aslong as six months. supervision by a team, en-hanced sense of responsibility for patients, and fre-quently opportunity to deal with family gro.,.ps aspatients.

A family and home care program gave each stu-dent an opportunity to follow a family during histraining in the General Medical Clinic. He also ac-quired similar experience with pediatric and obstet-rical patients, attended comprehensive care confer-en...es. and participated in a preceptor program. Healso attended weekly seminars and participated insix conferences in psychosomatic medicine.

Comprehensive care conferences were held everyother week: and, at this time. the fundamentalphilosophy of the General Medical Clinic was

discussed. Towards the end of the experiment,these conferences were curtailed, and there was anassociated lack of interest by students and facultyin the family and home care program. There wasalso a decrease in emphasis on the preceptorprogram.

The Colorado program was an experiment with aclassical research design. Each of three classes(1954-1956) was divided into control and experi-mental groups. the former assigned to ColoradoGeneral Hospital and the latter to Denver GeneralHospital. The overall effect of the program wasthat it mitigated increasingly negative attitudes to-ward comprehensive care.1 observed previously,without impairing the acquisition of traditionalmedical knowledge and skill.

From data on hand, it is apparent that the Gen-eral Medical Clinic was more successful in achievingits goals in the first half of the senior year than inthe second. There were several explanations of-(erred for this observation. First, the student con-sidered the learning of traditional organic medi-cine much more important than learning com-prehensive care. Second, he believed that his Gen-eral Medical Clinic program presented inadequateopportunities for learning traditional medicine.Third. anxiety associated with the learning of tradi-tional medicine increased as graduation and intern-ship approached. Although the students believedthat the General Medical Clinic hampered theirlearning of medical knowled and skill, the datashowed that it did not. For all of the statedreasons, the investigators concluded that the fourthYear of medical school was too late in the curricu-lum to introduce comprehensive care, because thestudents had already largely adopted the dominantculture of hospital medicine. The research conclu-sion was that both the students and faculty be-lieved that comprehensive care was an attitude andnot a skill. The students resented being taughtcomprehensive care by faculty who they believedhad no special skills in this area.

Despite their commitment to the comprehensivecare program, the faculty was almost obsessed intrying to establish an organic diagnosis. They werefrustrated and hostile toward patients for whom nosuch organic diagnosis could be made or for thosewho seemingly were not sufficiently motivated toget well. In the comprehensive care setting thesefeelings, while conspicuous and temporary, did in-terfere with teaching. It was concluded that al-though faculty did not need advanced training inpreventive medicine and public health, it did re-quire basic concepts in psychosomatic medicine.They needed to know sociological principles in

addit;,n to being well prepared as clinicians. Part-time specialists did .not work as effectively in theclinic as full -time faculty. Also, due to the.exces-sive emphasis on teaching,. the part-time facultyfelt that its time was not efficiently utilised. It wasalso concluded that a program devoted to teaching,comprehensive care would be most ettective it padents from a wide variety of social backgroundswere included rather than the totallydisadvantaged.

Despite the widespread support of the Univer-sity and the administration. whenever day -to-dayproblems anise that put the program in conflictwith traditional specialty and inpatient services.decisions usually favored the more traditional view.Also. a fundamental conflict existed in the DenverGeneral Hospital where the administrative goal wasto see patients. rather than to devote the time toteaching that the General Medical Clinic effort re-quired. It is not surpirising. therefore, that flit..General Medical Clinic was elimMated in 1961(~nuke. 19651.

Another important experiment was carried outat Cornell beginning in 1 952 ;Reader. 1964!. Thecomprehensive care and teaching program was alogical development of the programs of familyhealth advisor. home care, and pediatric outpatientdepartment initiated at Cornell before and duringWorld War II ( Barr. 19461. The goal of the programin terms of patient care was to provide continuityto ambulatory patients. For students, the goal wasto learn about comprehensive care. Using a before-and-after design. the research goal was to measurechanges in attitude and values as well as the abilityof the medical students to use psycho)gical andsociohigical methods.

Fur six months in the fourth year of medicalschool, students participated in the continuity careprogram. This involved the general medical clinic.the pediatric clinic., the psychiatric outpatient department. the home care. and family care pro-grams. At any one time, one-halt of the class parti-cipated in the continuity program. Precepting wasdone by the appropriate specialist. A number ofinnovative teaching techniques Were developedUsing tare recordings. one-ay-screen interviewing.and small-group seminars..

Criteria tc.ir selecting families in the program in-cluded the following characteristics: a member hadan illness that required continuing medical super-vision; ioung children: location close enough formembers to receive home care: and freedom fromoverwhelming. complec, social problems.

A high percentage of students. particularly thosewhose families had no member with organic

disease. were dissatisfied in the role of familyphysician: because it offered tl.em a meaninglessexperience. Students satisfied in the role of thefamily physician were those engaged in traditionalmedical activity: i.e., with patients with organic ill-ness. The students were also satisfied if the familiesaccepted the students as their physicians. This is animportant point: because the data indicated that, ifthe student telt involved as the family physician, itwas likely that his attitude would change. l'he fam-ily care program was discontinued in 1959, in part.because families could not be found who met thestated criteria and. probably. because the studentsexpressed ex,essive frustration in dealing withwell families. The home care program, whichhad more disease Content ;tioltkin. 1960). wasfound more satisfying to the students: it continuedas part of the program.

A major conclusion was that the comprehensivec. are chnic students needed .to work for at least atour-month period and preferably for six monthsto experience any sense of continuity. Like thestudents at Colorado, the Cornell students wereconcerned that they were not learning, the facts ofmedicine. even though the research findings indi-cated otherwise. For example. there was no changein national board scores in those classes that partic-ipated in the program compared with the twopreceding classes that did not participate in thisexperiment.

The administrative structure, educational organi-/anon, and physical siie of the medical centerpresented formidable obstacles to comprehensivecare Magra. 1 97 1 1. For example. when com-prehensive care or related programs are estab-lished. it is very difficult to find appropriate physi-cal quarters for them: because they must competewith established programs for finite resources. Thehospital was not only physically incompatible, butit also presented an intellectually hostile or at leastunaccepting environment. After four years ofoperation. 35 percent of the faculty at Cometl didnot know that the program existed Caplovitz..1967 . )ver one third of those who did know ofthe program had reservations about it. In fact. onlyone of five of the faculty believed that the medicalschool should educate More general physicians.Faculty in the specialties of psychiatry, publichealth. preventive medicine. and medicine weremost tavorof the program. Only -10 percent ofthe full proic,ors endorsed the program. andprofessors and residents generally shared a IlL"ptiVeview of the program. Caploviti pointed out that ifcomprehensive care programs were perceived asproducers of general physicians and if this was

equated by the faculty with general practice. thenopposition to the program would increase evenmore. In 1967. the program was reduced from theoriginal 22-week schedule to 15 weeks and theprogram ceased entirely in the late 1960's.

SummaryThese were ambitious programs with elements of

success and failure. The succeeded as experimentsin medical education but failed, because the majority of the faculty never viewed them as mute thanexperiments, well insulated from the main work ofhospital medicine.

Additional valuable findings emerged from thesestudies. Given the climate of the medical schooland the teaching hospital, there was a minimum oftime needed for the programs to accomplish evenminimal goals: and this period appeared usually tobe six months. The fourth year of medical schoolwas not the best time to introduce the programs:because by that time, senior students had adoptedthe dominant hospital culture and preferred pa-tients with organic disease.

Both Colorado and Cornell had family care pro-grams as part of their overall comprehensive careprograms; and at both institutions problems wereexperienced with the family care programs due toadministrative conflicts and student. frustration.Because family .care was only a part of the compre-hensive care effort, the family care aspects werereadily expendable. There .were, however, a groupof programs that made family care their majorthrust: and it is these programs that we will nowexamine.

Family CareFamily Care Programs were located sometimes

outside of In)spitals, generally in a special setting.The majority. however. even th()LIgh Colin flitted tomeeting needs outside of the hospital, were forceddue to financial and physical needs to locate withinthe hospital complex. usually in the outpatientdepartment.

Representative programs were establish i atVermont (Haynes, 19611', Louisville (Eller. .9.57).Harvard (Haggerty. 1962: Stokes, 1963), Yale(Beloit*. 1967, 1968), and Western Reser\ e Kenlicit, 1961).

A Family Care Program was started in Vermontin 1959 (Haynes. 19601. Faculty consisted of gen-eral practitioners. public health nurses, and socialworkers. Third -and fourth-year students providedcare as family physicians to at least two indigentfamilies for two years. Seminars were

oftwice

weekly. Directed by the Department of PreventiveMedicine., the program had an officc facilityseparate from the hospital.

30

Two programs were launched at Harvard; one, atthe Massachusetts General Hospital; the other, atthe Children's Hospital Medical Center. The pro-gram at the

for

General Hospital wasestablished tor a five-year period beginning in1955.. Each student was assigned one or two fam-ilies for whom he provided care during his thirdyear. An attempt was made to follow patients inthe fourth year, but the experience was unsatisfac-tory and consequently terminated due to curricu-lum conflicts. The curriculum was not changed.and students cared for selected families in theirfree time. Only one event took preeedenc, overmedical school scheduled activities and that was theonset of labor in pregnant patients. because thestudent was expected to be present and to partici-pate in the delivery. Students actually rendered asmall amount of service, which averaged 20 percentof the total services offered to the participatingfamilies. Faculty were hospital -based internists andpediatricians, and the program office was locatedin the hospital outpatient department.

From 1957 to 1959, third -year students wereassigned at random to the program. Approximately20 students troll' each of the three classes wereselected, with the remainder of each class as a con-trol. Students were compared with their costa ison such indicators as grades. class standing, sla-clonal Board scores, attitudes as measured byquestionnaires. and history taking. At the end ofthree Years, it was not possible by these measuresto demonstrate any differences between experi-mental and control students. Consequently, theprogram at the Massachusetts General Hospital wasterminated in 1960.

The program at the Children's Hospital MedicalCenter was started in 1956 with objectives similarto the Massachusetts General Hospital program. Itwas a more complex organization. because care fora family required the services of three hospitalsThe Peter Bent Brigham Hospital. Children's Hos-pital Medical Center, and Boston Hospital forWomen as well as the usual team of specialtypreceptors (Haggerty. 1962). Finding faculty withappropriate skills to teach family care was a majorproblem at that time, because there were no familypractitioners on the faculty. The program had itsown building. separate from, but in close proxi-mity to the associated teaching hospitals.

The program gave third-year medical students anopportunity to work as family physicians ;n a teamrelationship with a nurse and social worker. Stu-dents were on call at all times through an answer-ing service backed up by preceptors primarilypediatricians and internists plus consultantpsychiatrists and obstetricians. In addition, they

attended a weekly seminar. Due to several curricu-lum changes at Harvard, the course was movedfrom the third to the hmrth year and, most re-cently, is being offered in both the third

forfourth years. Students participated originally for annnimuin of nine months and currently participatefor 'twenty months. Family ph% skim's joined thefaculty in 1967 (Alpert, 197().

Students were called to deliver necessary servicesduring other scheduled class activities: and, even-.tuallv. a program was developed that had many ofthe characteristics of a group practice. 13Y the late1960's students were involved in 90 percent of themedical services to their assigned families. Some 30students or 25 percent of each class participated inthe later years.

In an attempt to offer medical students experi.ence with families of a broabr social and eco-nomic spectrum. families from Boston area under-graduate colleges were included. Families withsmall children and pregnant women were alsoselected on the basis of observations at Cornell andColorado that suggested students needed some ac-tual practice to get the experience of being a fam-ily physician. Like the Cornell program, it becameincreasingly difficult to recruit families who metthe program's requirements; and, thus, screeningwas eventually abandoned.

In 1971. a survey of all students who had par-ticipated in the Children's and Massachusetts Gen-eral programs, together with their classmate con-trols, was completed. Because in the period. 1957to 1960. the students had been assigned at random.this presented an opportunity to measure the pos-sible long-range impact of the program. No majordifferences were noted, although 24 percent of thecontrol students had taken surgical training. com-pared with 18 percent of the family health stu-dents. Otherwise, there was no apparent differencein present practice patterns. Only one student inthe total sample identified himself being in fam-ily practice. Students who had been volunteers inthe Children's programs for 1961-1965 were morelikely to pursue pediatric careers than were theirclassmates, and the 1966 to 1970 cohort containeda small number of students who were consideringcareers in family medicine. Thus in recent years,the program has given evidence of attracting stu-dents interested in primary care careers.

Following an extensive review of family andcomprehensive care programs. Beloff and Weiner-man (1967) established a Family Care Program atYale, which contained many features of thepreceding efforts. Third -year medical studentsworked as physician members of the health team.

and the program was based in the outpatient de-partment. An early .descriptive analysis noted thedevelopment of a family record system and thegeneral popularity of the course with the par-ticipating students. The program was terminated in1971 despite its reported successes due to minimalinstitutional support.

At Western Reserve, a Family Care Program wasestablished that offered students experience in alltour years of the curriculum (Kennel!, 1961). Theprogram was part of an extensive revision of thetotal curriculum (Wearn. 1956; Caughey, 1956;Caughey, 1959; Adams. 1958; Ham, 1962). Thestudent began as a health advisor in his first yearand was a student physician in his clinical year.Initially compulsory for all four years, the FamilyCare Program became an elective for the last twoyears of medical school. It had the advantages of acontinuity clinic and an integrated outpatientdepartment program. In the clinical year, thepsychiatry and pediatrics departments providedconsiderable input into the student's training.(Adams. 1958).

From our view, the Western Reserve curriculumwas especially significant; because students saw pa-tients early in their careers, medical education waspatient oriented and involved patients with chronicdiseases, and students had close and continuingcontact with their preceptors. In addition, the pro-gram carefully integrated seminars and clinicalwork.

The Major goal of the Western Reserve experi-ence was to develop a curriculum that stressed in-terdepartmental teaching. Achieving this goal didnot change faculty attitudes toward primary care.The one published survey of faculty at WesternReserve suggests that in its beginnings the facultysupported change due to the belief that changemeant better education (Horowitz. 1960). Of allthe curriculum changes. the Family Care Programand continuity clinic were least well accepted bythe faculty. The faculty. although initially commit-ted to the important goal of education reform,never defined their curriculum goals in terms ofsubsequent career choice of their graduates. Be-cause there never was a goal for recruiting studentsto primary care, Western Reserve graduates verylikely pursued careers no different from careers atother medical schools, despite this very importantand major education experiment.

New schools have recently turned their attentionto the family care model in the teaching of familymedicine (Harrell, 1968; Walker, 1966). (Also seegraduate education, page 143.) As a major activity,it is too soon to evaluate the impact of these

31

programs. The development of an undergraduategeneralist track has also been suggested (Pellegrino,1966), but it has vet to be generally implemented.Summary

In general, family care programs like comprehen-sive care programs have been part of the medicalschool curriculum not as a result of a commitmentto primary care, but because the programs wereexperiments in medical education. The majority ofthe faculty was unwilling to see the programs asrepresenting any major commitment on the part ofthe medicalschool. Often the participating studentwould find himself in a position of conflict be-tween his family care activity and his scheduled,usual medical course. Where the students volun-teered. this conflict was not a major factor: butwhere students were assigned usually for experi-mental purposes. there were major antagonisms andresistance.

There also were students whose experience leftthem impressed with .the negative aspects of

ofcare, a view reinforced by the majority of thefaculty. Certainly the programs at least in theirearly years, did nut influence students to selectprimary care careers. But. the programs did teachcontinuity and focused on first contact outside ofthe hospital. Moreover, the student family physi-cian was placed in the position of providing healthas well as illness care for a family.

Only within recent years and in a few programshave family physicians and other primary care pro-viders been part of the faculty. Consequently. mostprograms did not provide students with a bonafide model of a .primary care practitioner. More-over. providing family care required longitudinalrather than block experience: and even in an elec-tive curriculum, the student would continue in aposition of conflict. However, these programs didemphasize the importance of the family as the unitof health care and came the closest of any pro-grams in emphasizing the important contributionthat can be made by the primary care physician.

Model PracticesChanges presently taking place in our health care

system offer a number of newer sites where educa-tional models for primary care can be developed.For example, one of the requirements of the newlycreated Board of Family Medicine calls for a modelpractice if there is to be an accredited residencyprogram. once again, these model practices arebeing located most often in the hospital outpatientdepartment, although sonic have developed in aspecial facility outside of the hospital.

Some of the newer settings are group practicesin hospitals. in u Itispecialty group practices.

community hospitals, and, increasingly, some formof neighborhood health center (Kark, 1957). Somemay be associated with prepaid insurance planscurrently developed at a number of medicalschools (Ebert. 1967). One of the issues to be re-solved in developing programs in neighborhoodhealth centers is the need for community approvalof student participation, either as undergraduatesor residents. In general, most community programsin the United States resist student physicians dueto reactions to past impersonal experiences inteaching hospitals that many feel represented asignificant degree of exploitation. In some com-munities the students themselves have organized aprogram (Record, 1969: Waserman, 1971), butthese efforts are generally limited to the studentsorganizing the clinic rather than ulitmately givingcare.

To this date, very few of the health center set-tings have been available for education for primarycare. However, sonic centers are accepting residentphysicians who are deemed suitable by the respon-sible community board. Conversely, most privategroup practices resist having students due to thevery real expense involved in their education. Inthis setting particularly, plans to give studentsgraded responsibilities as in hospitals have not beenfully developed.

Perhaps it was Edinburgh (Scott. 1967) thatpioneered in the establishment of a model acade-mic practice. Here students spent block periodsobserving physicians in a

forpractice but

did not themselves care for patients. Additionalunckrgraduate university programs in family medi-cin: are now underway at Oklahoma ( Lienke,1970), Rochester (Haller. 1969). Hershey (Harrell,1968) and Miami (Carmichael, 1965). Lienke( 1970) has described a group of family physiciansworking in a university medical center. A grouppractice of family physicians has been developed atHershey (Harrell, 1968). Prepaid group practices atHarvard, Yale, and Johns Hopkins may eventuallybe used for education but, initially at least, theoperational emphasis has been on organization andfinances. Early experiences show that even modelpractices develop many problems, as (I ands forservice continually comprk4me the professionaltime available for education. Obtaining experi-enced faculty of these programs has been amajor problem.

Sum wary

Through its model practice, a department of pri-mary care, family medicine, or general practicemay be the proper vehicle for coordination of pri-

mare care education in a medical school. (Medalie.1969). However. unless a medical school acceptsthe goal of producing primary care practitioners,these departments are likely to be overwhelmed bythe stronger and snore traditional departments.

Formal Course WorkThe classroom is vet another site for the teach-

ing of primary care The course could be located inthe basic science department of a medical school, aconference room in the hospital outpatient depart-ment. or on an inpatient ward. But each of thesesettings call fur didactic, rather than practical orapplied instruction.

Numerous atter pts have been made to in-troduce formal classroom teaching to primary med-icine. This teaching was particularly easy in theprechnical %Tars because it did nut requires actualpatient responsibility on the part of the student.Courses were often taught by departments ofpsychiatry and preventive medicine (Aldrich. 1953:Come IL 1957: Engel, 1957: Fox, 1951: Cute,1953: Greenhill, 1950: Simian, 1967). However,as previously stated, the psychiatric efforts were,primarily. developmentally or psychoanalytically(Saslow. 1948) rather than socially oriented: andthe preventive medicine courses were socially orcommunity and not clinically oriented (Antolloysky. 1966: Wegman. 1969).

Many of the seminars were held on the inpatientservice. One of th. earliest programs was that givenat the Beth Israel Hospital in Boston (Derow.1933: Cohen. 1935: Cohen, 1941). Beginning, in1929 weekly medical social-work rounds includedhospitalized patients as well as outpatients. Stu-dents participated in their third year during theirmedical rotation in the outpatient department andin the fourth year as ward clinical clerks. Althoughearly graduates participated in the programs withenthusiasms. the program social worker stated thatshe was never fully utilized. Many of the programswere Libelled as multidisciplinary efforts 'Bakst.195': Bates, 1965.-. At Niashington University ofSt. Louis, an interdepartmental program i;;volvingmedicine and public health was developed. Students studied cases during medical clerkship andreported on these cases at seminars. which wereheld during preventive medicine. Home visits werealso included in the program *Shank. 1956..

At the Medical College of Virginia. .1 first-yearcourse was offered covering patients' physical en-vironment. interviewing. evolution, genetics,growth and development. history of medicine.physical examination. epidenniulugy. and behavior.The course, which included a general practitionerand other clinicians on its supervising committee.occupied 300 hours over a 42-week period in thefirst year Arington, 1964

In the 1970's, teaching of social and environ-mental factors in medicine were extensions of basicclinical instruction, including ward round teachinv,,home visits, case conferences, clinical discussion,and expanded history taking. There were alsoformally organized case -study projects sponsoredby one or more departments. A typical plan wouldinclude a home visit on a hospitalized patient,study by a team, and presentation by the studentat a seminar (Griffith. 1971). In general, thesemethods proved popular when the seminar leader,lecturers, and social workers were able, dynamic,and sensitive to the needs of the involved students.

Some efforts have been made to evaluate theshort-run results of these programs. In a clinicalCourse offered to first-year students, the percent-age of those who believed psychosocial factorscould be a cause of illness increased from 45 per-cent to 72 percent (Bruhn. 1969). Lewis (1965)studied a random sample of first-year students whoparticipated in a series of seminars dealing withhome care. The participating students did better onthe study measures than did their controls. Inanother study of a short-term, comprehensive careseminar course for fourth-year students, there wasactually little difference between experimental stu-dents and their controls: what little differencesthere were suggested a less positive view takenby the experimental students towaitl clinical hos-pital medicine. (Shaffer. 1965). These studies didnot follow students long enough to assess the pos-sible long-term impact. Engel (1971) suggests thatefforts directed at first -year students might beharmful to long-term professional growth, inas-much as these students have not been prepared forearly patient contact. Moreover, in the clinicalears, the interested student was exposed to a

house staff who found these "social efforts in-terfering in the ''real work'' of the ward. Prechnicalstudents. On the other hand, not involved in thework of the ward, would participate eagerly inthese courses: because, to them, it represented realcontact the closest that they had withpatients. However, these courses could not com-pete with the drama of the hospitalized patient andhis organic disease. When organic disease wassuccessfully treated, the results were dramatic andalmost always influenced the patient's episode ofillness. Broader issues in the primary care sphererequire time for their alteration, or resolution, aswell as for coping me, hanisms to change. The stu-dent and resident did nut see this. What they didsec was that these complex social and behavioralproblems could not be resolved at case confer-ences. The values of the dominant culture predonli-nated to the extent that the studemit attenipted toavoid rather than to participate in these conferences.

33

In our judgment. attempts to incroduce dis-cussion of social and Lundy facturs in illnesstailed when they were made spec i.tl and separateexercises.

SummaryConflict between the goals ut the primal.% care

program and the goals of hospital medicine apparet.itly is a repetitive theme. The theme expressesitself in a number of ways. The prechnical studentwho has participated with some enthusiasm in theprechnical primary care program arrives on theward where he collies face to face with hospitalmedicine. The clinical student find, himself indisagreement with his intern when his primary care

or "whole patient" responsibility conflicts withward duties that are primarily disease oriented.Conflict is also seen when the faculty member on award joins the resident in belittling the referringphysician. Not only are there no faculty members.who present the student with satisfactory primaryare models. but also there are very few residents

who otter models cut the resident in training. Thus.the student identities with his intern and resident.and his primary care interest diminishes by theoverwhelming demands of hospital medicine. Toavoid these conflicts. the student needs models notonly of the practicing primary care physician. butalso of residents who are preparing for a career inprimary care.

References

SECTION

Adams. W.R.: The psychiatrist in an Ambulatory Clerkship for Comprehensive Medical Cire in .1 New Curriculum.J. Med. Ed..?.?: 211 20. 1938.

Aldrich. C.K.: Psychiatric Teaching on an InpatientService. J.Med.Ed.. 2S. 36. 1953.

Alpert. J.J.: Continuing Education tor Family Practice.Postgrad. Med. 43; 237-239. 1969.

Alpert. J.J.. Kosa. J.. Haggett. 12.J.. Robertson. I...Heagarty. M.(:.: Attutudes and Satisfactions of LowIncome Families Receiving Comprehensive Pediatric Care.Amer. J. Pub. Health 60: 3. 499 506. 1970.

Antonov,ky. A.: Teaching .t Social Medicine Orientationto Medical Students. J.Med.Ed.. 4/. 870 76. 1%6.

Appel. K.G.: Putting the Family Bak k in Medical Utication. N E JM..249 397 99. 1953.

Arnold. M.: The Charing, Cross Hospital General Practitintier Student Attachment Scheme. Practitioner. /9.1.:783 88. 1964.

Arington. (;.E.. Jr.: Man and His Environment. A NevCourse of Studv at the Medical College of Virginia. J.Med.Ed.. 39: 704 II. I %4.

Bah. A.A. and Weiskotten. H.G.: The Training of theStudent in What is Involved in Adequate Medical Care.JAMA /09: 2136 37. 1937.

Bakst. H. A Domiciliary Care Program as an !megrated Facility in the Medical Curriculum. J.Ass.Amer.Med.Cull.. 25: 406-16. 1959.

Bakst. H.J. and Malamud. W.: Combined Teaching inHuman Ecology. J.Med.Ed. 32: 603 09.1937.

Bakst. H.J.: The Case for Emphasis on Soda( Forces andEnvironment. J.Med.Ed. Supplement 34: 6 9.1959.

Barber. G.: Education in General Practice. An Experinient at St. Marv's. Brit. Med. J. 2: 490-1. 1952.

Bardeen. C.R.: Extramural (finical Teaching: The Preceptor System at Wisconsin. Wise .M.J.. 111.192.8.

34

Barnett. G.D.: The Introductory Course in ClinicalMedicine. J.A.Amr.Med.Coll.. 225. 1938.

Barr. D.P.: The Next Ten Years in Medicine. New YorkMed. 2: 14. 1946.

Barr. I).: The Teaching of Preventive Medicine. J.Med.Ed.. 2N: 49 56. 1953.

Bates. B.,: Comprehensive Medicine: A Conference Ap-proach with Inpatient Emphasis. J.Med.Ed.. 4(): 778-84.196,5.

Baumont. (i.. Feigal. Magraw. R.. Defoe. E.C.. Carey.J.k;.: Medical Auditing in a Comprehensive Clinic Program.J.Med.Ed.. 42: 359-67 .1967.

Bean. W.: The Rise and Fall of the Correlated ClinicCourse at Iowa. J.Mcd.Ed...?/: 97-100. 1956.

Belot!. J.S. and Weinerman. E.R.: Yale Studies in Fami-ly Health Care. I. Planning and Pilot Test of a New Pro-gram. J AMA. /v9: 383-89. 1967.

J.S. and Snoke. P.: Yale Studies in FamilyHealth Care. II Organisation of Comprehensive Family CareProgram. JAMA. 204: 355 60. 1968.

J.S.. Kemper. M.. Weinerman. E.R.: Medical Stu-dent Response to a Program fur Teaching ComprehensiveCare. J.Med.Ed.. 45: 1047.1059. 1970.

Berl e. B.B.: The Family Health Advisor Plan in MedicalTeaching. J .Med. Ed.. 2S: 0-45. 1953.

Bogdonoff. MD.. Elwell. S.W.. Ref Iris. J.: Medical Outpatient Department Teaching. J.Med.Ed.. 35: 88589.1963.

Bowers. J.Z. and Parkin. R.C.: The Wisconsin PreceptorProgram. A 3(1 Year Experiment in Medical Education.J.Med.Ed.. 32: 610. 1957.

Bower.. J.Z. and Page. I I.A.: Study of Preceptor Prograins. JAMA. / 1921. 1960.

British Medical Journal Editorial : A Sight of GeneralPtae dec. 2 I79 80. 1951

Brotherston. i; Family 1 1...eaca.ng Programas Evaluated by it% Filmier Student,. Lancet, 2. 202 205.1959.

Bruhn. J.G. and Natignton. I.: Ettect, of a Course inInternal _Medicine on the Attitudes of First Year Students.J.Med.Ed.. 44: 51)4 09, 1969.

Bunnell, 1.1..: Notes on Developing a General MedicalClinic. J.Med.Ed.. 32: i39. 1957.

Burwell. C. Sidney and Younian,. J.I1.: Method, andobjeitices in the Organi/ation of the Outpatient Departmeat (It a 'reaching Hospital. J.A.Amer.Med.Coll_V65-69. 1935,

Cadmus. R.A. and Lamb. A. k.: The Group Clinic. AnExperiment with the Group Practice Principle in a LargeMetropolitan Teaching Dispensary. 1.Assoc.Amr.Med,Coll.. 23: 121 24. 1948.

Cplovitt. D.: FaitiltY Evaluation of the Cornell Com-prehensive Care and Teaching Program in Reader and Goss.p. 335 355.

artni.hael. I .P.: A Program of Instruction in FamilyMedicine. J.Med.E.(1.. 40. 371) 74. 1965.

Carn...gie Comission on Higher Education. Higher Education and Nation's Health. McGraw Hill. Oct., 1970.

Casewell. M.W : Report on the Existence and Value ofElective Period, in British ktedical Schools. British MedicalStudents Association. 1963.

Castle. C.H.: Department of Community and FamilyMedicine. A New Program at the University of Utah. RockyMt. Med.J. OS: 3-8, 1971.

Caughey. Jr.: Clinical Teaching During Four Years,J.Med.Ed...?/: 530 34. 1956.

Caughey. J.1... Education of the Physician for hisRole in Comprehensive Health Service. J.Mcd.Ed., 34:381-. 1959.

Clyne. M.B.: Undergraduate Teaching in PsychologicalMedicine as Applicable to General Practice. J.Med.Ed.,$.': 961-69. 1 96 i.

Cohen. E. and Derow. H.A.: Teaching Medical StudentsOhjeitii es for Care of Patients and Social Aspects of Illness.Arch. hit. Med..c1): 351. 1935.

Cohen. E. and Derow. 11.A.: Training of Interns illSocial Aspect, of Illness. J AMA /11: 1817. 1941.

Cohnore. J.P. and Wolf. S.: A New Design for Serviceand 'Teaching in the OPD. J AMA 156: 830 33. 1954.

Conwell, D.C.. Pv.. C.. Lyons. B.. Gaventa. W.: Introdu( don of the Concept of Comprehensive Medicine intothe Curriculum. J.Med.Ed.. 32. 210 14. 1957.

.;:tirran. J.A.: Provision of Better Medic d Care. MilbankMent. Fund Quart. 2.1:7 1945,

Drlev. W. and Somers. A.R.: Medicine. Money andManpower. New England.J. Med., 276: 1291 (16. 1967.

Dean. T.M.: Medical Students' Response to Undergradual; Instruction in General Practice. Brit. J. Med. Ed_5.169-171. 1971.

Derow. H.A. and Cohen. E.: The 'raining of Interns inthe Social Aspects of Medicine. New Eng.J.Med.. 200: 827.1913.

Dents( Ide. K.W.: The Community. Medicine Clerkship,Mill ink Mem. Fund Quart. 47: 411 .447. 1969.

Deutschle. K.W. and Fulmer, H.S.: Community Medleine: A -New- Department at the University of KentuckyCollege of Medicine. J.Med.F.,(1., 37: 434 45. :962.

Dorsey, Knapp. P.W., Frank, Education forContinuity ut Care. J.Med.Ed. 20 21) 28, 1954.

G.G.: A Report 2 Years Following the Creation ofa Chair of General Practice, J.S.C.Med.Ass. 57. 436-38,1961.

Ebert. R.: The Role of the Medical School in Planningthe Health Care System. J.Med,E(1. .1.1. 481-88, 1967.

C.H. and Adams. B.: The Family Health Prop amof the Univetsity of Louisville School of Medicine. TheHanover Forum 1 1 19. Winter. 1956-57.

Elliot. G.A.: The Training of Student; in General Prac-tice. South African Med.J.. 29: 134-7. 1955,

Engel. GA._ Green. W.L.. Jr.: Reiclisman. F.. Schmidt!,A.. Ashenburg, N.: A Graduate and Undergraduate Teach-ing Program on the Psychological Aspects of Medicine,J.Med.Ed.. 32: 859-71. 1957.

Evans. L.J.: The Forces That Shape the Medical School,J.Med.Ed.. 40: 272.283, 1965.

Falkenstein. D.H.: The Learning and Personal Develop-ment of Medical Students and the Recent Changes in Uni-versities and Medical Schools. J.Med.Ed.. 883-897. 1971.

Great Britain. The Field of Work of the Family Doctor.65 pp London. Her Majesty's Stationary Office. 1963.

Faulkner. J.M.: Symposium on Extramural Facilities inMedical Education. J.Med.Ed. 28: 9-8. 1953.

Fisher. T.M. and Stevenson. I.: The Use of Home Visitsin Teaching Psychological Attitudes to Medical Students,J.Med.Ed.. 30: 203. 1955.

Fleming, W.L.: Teaching of Family Physicians' Ap-proach by a Department of Preventive Medicine. JAM&/6/. 711-13. 1956.

Fox. H.M.: Teaching Integrated Medicine, J.Med.Ed. 24:421-29. 1951.

Fre iden. R.B.. Barsky. A.J.. Levine. II.. Williams,S.: Medical Education and Physician Behavior, PreparingPhysicians in New Roles. J.Med.Ed. 47: 163-167. 1972.

Freidson. E.: The Hospital in Modern Society. NewYork. Free Press of Glencoe. 1963.

Garfield. S.: The Delivery of Medical Care. ScientificAmerican. 222: 15. 1970

Gee. H.H. and Cowles. J.T.: The Appraisal of Applicantsto Medical Schools. Evanston. Ill.. AAMC, 1957.

General Practice: The Teaching of General Practitioners.Report of the Undergraduate Ed. Comm. of the Founda-tion of the College of General Practitioners. BMJ 2: 36.38,1953.

Gneraf Practice: Preparation of the Physician: Papersprepared fur a European Conference on the Training of theDoctor fur his Work in the Community. Edinburgh. 1961,WHO Public Health Papers ::20. Geneva. 1963.

Gih,on. C.D. and Kramer. B.M., Site of Care in MedicalPractice. Mt. dical Care 3: 14. 1965.

Gray. R.M., Moody. P.M.. Newman, 1,V.R.E.: An Analycis of Physicians' Attitudes of Cynicism and Humanitarianism Before and After Entering Medical Practice. J.Med.Ed..411: 760, 1965.

Gteenhill. M.H.. Fitipatrick. W.N.. Berldinger, K.W.: Recent Developtonts in the Teaching of Comprehensive Medi-cine. North Catalina Medical Journal / /: 615.619. 1950.

Gregg. L.A. and Moses. C.: OPD Program for SeniorMedi( al Students. J.Med.Ed. 3/ : 757-63. 1956.

35

Griffith. J.A.: Indepen,:ent Learning in Community Pe-diatrics. J 46: I000 -92. 1971.

The Organization of Group Practice. a Report of a Sub-committee of the Standing Medical Advisory Committee.London. HMSO. 1971.

Cute. S.B.. Matarazzo. Saslow. G.A.: Formulationof Principles of Comprehensive Medicine with Special Ref-erence to Learning Theory. J. Clin. Psycho'. 9: 127-36.1953.

Haggerty. R.J.: Family Medicine: A Teaching Programfor Medical Students and Pediatric House Officers. J.Med.Ed.. 37: 531-580. 1%2.

Ham. T.H.: Medical Education at Western ReserveUniversity. Progress Report for 16 Years. 1946-62. NewEng. J.Med. 26 7: 868-916. 1962.

Hammond. K.R. and Kern. F.: Teachinl ComprehensiveMedical Care. A Psychological Study of a Change in MedicalEducation The Commormealth Fund. Cambridge.Harvard. 1959.

Harrell. I .T.: .Education of the Family Physician. J AMA.203: 495. 1968.

Harris. C.M.: General Practice Teaching of Under-graduates in British Medical Schools. Reports from GeneralPractice National Council of the Royal College of GeneralPractice. April 1969.

Haynes. M.A.: An Approach to the Teaching of FacultyCirc../ AMA. 173: 1340.41. 1960.

Haller. J.W. and v E.S.: A University Family Medi-cine Program. J.Med,Ed.44: 225. 1969.

Heller. R.F. and Heller. C.A.: Studentships in GeneralPractice. Lancet. 1745-46. 1963.

Herzberg. F.. Inkley. S.. Adams. W.R.: Sonic Effects onthe Clinical Faculty of a Critical Incident Study of thePerformance of Students. J.Med.Ed.. 35: 666-674. 1960.

Heslop. R.W.: A Survey of the General Practice Schemesin British Medical Schools. British Medical Students Associ-ation. 1953.

Hiscock. I.V.: Clinical-Public Health Case Studies byMedical Students with Special Reference to Home Visits.J.A.AniMed.0011. 14: 79. 1939.

Hobson. W.: Training of Medical Students in GeneralPractice in Sheffield. BMJ 2: 491-492. 1952.

Holmes. EM.. Nelson. K.. Harper. C.L.: The RichmondHome. Medi,:al Cate Program. Amer. J. Pub. Hlth..596-602. 1953.

Horowitz. M.J.. and Herzberg. P.H.: Faculty AttitudesToward Experiences with Two Revise' Programs of MedicalEducation. J.Med. Ed.. 35: 277-82. 1960.

Hospital Council of Greater New York. Organized HomeMedical Care in New York City: A Study of Ninteen Pro-grams. Cambridge. Harvard University Press. 1956.

Hubbard. J.P.: Mitchell. J. Nick.. Pool. M.L.. Rogers.A.M.: The Family in the Training of Medical Students.J.Med.Ed. 27: 10. 1952.

Hubbard. J.P : Integrating Preventive and Social Medi-cine in the Medical Curriculum. New Eng. J. Mcd. 25:513-519. 1954.

Johnson. J.G.: The Family Care Program. Health News(N.Y. State Dept. of Health .i"6: 4-12. March. 1959.

uhnson. P.C.. Barnhill. B.. Tranquada. R.: Evaluationof a Student Health Project. J.Mcd.Ed. 44. 945 50. 1969.

36

Kark. S.L. and Stewart. G.W.: Health Education andNeighborhood Family Practice. Health Educ. J. /5: 131.1957.

Kendall. P.: Evaluating an Experimental Program in Med-ical Educatio.. in Innovation in Education, Miles, M.B.(Ed.). New York Teachers College, Columbia University,p. 343-60. 1964.

Kennel'. J.H.: Experience with a Medical School FamilyStudy, J.Med.Ed. 36: 1649-1716. 1961.

Kern, F.J and Hammond. K.R.: Research in TeachingComprehensive Medicine. J.Med.Ed. 31: 544-552. 1956.

Kerr, W.J.: A Unique and Promising Experiment in Med-ical Education. California and West Mcd. 24: 465. 1926.

Kindschi, L.G.: The Role of the Preceptor in the U. ofWisconsin Preceptor Program. J.Med.Ed., 34: 649-53. 1959.

Knowles, J.H.: Views of Education and Medical Care.Cambridge, Harvard, 1968.

Korach, A.: Extramural Teaching of Preventive Medicineand Public Health, M.Bull. University of Cincinnati, 9: 1

Oct.. 1942.Kraus. A.S.. Botterel, E.H.. Einarson. D.W.. Thompson.

D.W. Initial Career Plans and Subsequent Family Practice,J.Med.Ed.. 826-830.

Lee, P.V.: Medical Schools and the Changing Times,Evanston. III. The AAMC, 1962.

Lee. L.W. and Smith. F.: Preceptorship Program1966: Sixteen Years at the University of Nebraska, Neb.St.Med.J.. 51: 359-6t, 1966.

Lewis. C.E.: Home Care Revisited. An Experiment inMedical and Nursing Education. J.Med.Ed. 40: 84-91. 1965.

Lienkc. R.J.: The Family Practice Model in HealthEducation. J AMA. 2/2. 2097-101, 1970.

Lynch. H.T.: Teaching Comprehensive Medicine in theHome Setting. Nebraska Med. J. 54: 516-8. /969.

MacClean. I.M.: Report on General Practice AttachmentSchemes in British Medical Schools, British Medical Stu-dents Association. 1961.

Magraw, R.: Medical Education and Health Services:Implications for Family Medicine, New Eng. J. Med. 285:1047. 1971.

Mathewson. H.O.: Two Experimental General PracticesU.K. and U.S.A.. Arch. Envir. Health (Chic.), 14:

809-820, 1967.McCandles. F.D.. and Weinstein. M.: Relation of Stu-

dent Attitude Changes to Teaching Techniques A Prelim-inary Study. J.Med.Ed. 31: 47. 1956.

McGarvey. M.R.. Mullan, F.. Sharfstein. S.S.: A Study inMedical Action The Student Health Organization, NewEng. J.Med. 279: 74-80. 1968.

McMitchell. J.: The Laboratory and the Community inUndergraduate Training of a Physician. Yale J. Biol. Med..21: 252-56. 1952.

Medalie, J.H., Devries. A.. Shachor. S.: Family Medicine..t Tel Aviv. Lancet 1: 979-81. 1969.

Melaney. H.E.: Preventive Medical Services for the Fam-ily. Milbank Mein. Fund Quart. 27: 251-259. 1949.

Melaney. H.E.: Environmental Case Studies in theTeaching of Preventive Medicine. South M.J., 27: 167.1934.

Mertens. A.: Experience with a Cle4ship in FamilyPractice. J .Med. Ed. 41: 36267. 1966.

Meyer, R.G.: Medical Education and Medical PracticeDemonstrations, J .Med. Ed. 3S: 596-602, 1963.

Millet, G.E.: Bedside 'I caching for First Year Students,J.Med.Ed. 29: 2h-32, 1954.

Miller. L.B. and Erwin, E.F.: Attitude Changes in Medi-cal Students During a Comprehensive Care Program. J.Med.Ed. 36: 412.28, 1 961.

Miller. T.G.: The Care of the Hospital Outpatient. NewEno. J .Med. 224: 1043. 1941.

Moore. J.W.: Preceptor Method of Teaching. J.A.Am.M.Coll. 21: 323, 1946.

Neibuhr, H., Steiger, W.A.. Hoffman. F.H.: Comprehensive Medicine and Medical Attitudes, J.Med. Ed., 35:1154-1162, 1960.

Niederman, J.C. and Paul, J.R.: The Home Visit as aTeaching Exercise: A Clinical Concept, J,Med.Ed.. 33:29-35. 1958.

Oki, I. and Pratt, J.H.: The Teaching of DomiciliaryMedicine at Tufts College Medical School. J.A.Am.Med.Coll. 14: 93, 1939.

Packer, H.: Family Care Programs in Medical Education.Southern M. J. 47: 689, 1954.

Parkin, R.C. and Peterson, D.L.: The Role of the Precep-tor:drip in Medical Education, J.Med.Ed. 34: 64.1-48. 1959.

Partnalce, A.H., Jr., Swengel, E.. Adams, J.M.: The Fam-ily in Medical Education. Ped.. 14: 387-95, 1 954.

Parmalce, A.: First Year Curriculum: The Family Medi-cine Course at U.C.L.A, J.Med.Ed.: 35: 675-83. 1960.

Pearson, R. J.C., Eimerl, T.S., Byrne, P.S.: Medical Stu-dents and General Practice, Lancet. /: 81-83, 1968.

Pellegrino, E.D.: Generalist Function in Medicine,JAMA. /98: 541-45, 1966.

Peterson. E.S.: Evolution of the Comprehensive CareConcept in the Medical School Clinics, Quart. Bull. North-western University Med. School 33: 352-57, 1 959.

Pollach. S. and Manning. P.R.: An Experience in Teaelting the Doctor-Patient Relationship to First-Year M.:dicJStudents. J.Med.Ed. 42: 770-74. 1967.

Puncher. H.G. and Richmond, J.B.: The Teaching ofSocial Medicine in the Clinical Clerkship: J. A. Amer.Med.Coll. 24: 97 99, 1949.

Prywcs, M., Sarell, M., Yodfat, Y.: Evaluative Study of aPreceptorship Program in General Practice for Senior Med-ical Students. Brit. J.Med.Ed., 1: 265-74. 1961.

Querido, A.: The Efficiency of Medical Care. H.E. Sten-fert Kroese N.V.. Leiden.

Quinn. R.W.. Zcidberg. 1..D., Christie. A.: A TeachingPr(q!ratti Based on Comprehensive Care of the Family,J.Med.Ed.. 35: 142-47. 1960.

Rakel. R.E.: Undergraduate Education in Ftcnily Medi-cine. J. Iowa Med. Soc.. 61: 231-5. 1971.

Reader. G.G.: Organization ant. Development of a Com-prehensive Care Program. Am. J. Pub. Health. 44: 760-65.1954.

Reader. G.G.: Comprehensive Medical Care. J.Med.Ed.28: 34-40. 195.3.

Reader. G.G.: Some of the Problems of Teaching Consprehensive Medicine, J.Med.Ed. 31: 544, 1956.

Reader, G.G.: Aims and Methods of Clinical Teaching atCornell Medical College A Consensus. J.Med.Ed.466-68. 1959.

Reader, G.G.: Organization and Development of a Com-prehensive Care Program. Amer. J. Pub. Health i:760-765, 1964.

Record. N.K. and McKegney, F.P.: The CommunityHealth Visitor Program: A Student-Organized Experiment,J.Med.Ed., 44: 193, 1969.

Reynolds, R.H.: Undergraduate Preparation for FamilyPractice, Proc. Royal Soc. Med.. 61: 1059-61, 1968.

Richardson, 1.M.: Student Opinion on General PracticeAttachment, Brit. Med. J. 2: 101-104, 1965.

Richardson, 1.M.: Patients and Students in General Prac-tice, J.Roy.Coll.Gen.Pract.. 20:285-287. 1970.

Rising, J .D.: The Rural Preceptorship, J. Kans. Med.Soc., 63: 81-84. 1962.

Ridey, D.A.W. and Ross, W.F.: An Introduction to Medi-cine in the Community: Doctor-Student Attachments inRhodesia, Brit.J.Med.Ed., 3: 94-100, 1 969.

Robinson. G.C.: The Patient as a Person, New York,Commonwealth Fund. 1939.

Robinson. G.C.: The Study of the Patient as a Whole asTraining for Medical Practice, J.A.Atn.Med.Coll. 14: 65.1939.

Rogatz, P.: Affecting Medical Students' Attitudes Toward Community Medicine. Bull. N.Y. Academy of Med.,47: =3. 236-248. 1971.

Rosenberg, L.: A Student's -View of Preceptorship Pro-grams. J.Med.Ed. 34: 654 656, 1959.

Saslow. G.: An Experiment with Comprehensive Med-icine, Psychosotn. Med., 10: 165-75, 1948.

Schwartz, D.. Ullman, A., Reader. G.G.: The Nurse,Social Worker and the Medical Student in a ComprehensiveProgram:! Nursing Outlook. 6: 39, 1958.

Scott, R.: A Teaching General Practice, Lancet 2:695 98, 1950.

Scott, R.: General Prac ice Teaching. J.Med.Ed. 31:621-33, 1956.

Scott. R.: General Practice Teaching, Brit. Med. J.5194: 293-99. 1 960.

Scott, R.: Academic Depts. of General Practice. Proc.Roy.Soc.Med., 611: 1311 , 1967.

Shaffer. J.W.. Harvey. J.C.. Reed, J.W., Schoenrich, E.H.,Thomas. A., Tubler. A.B., Teaching Comprehensive Care:An Exploratory Stink of Some Short Term Effects.J.Med.Ed.. 40: 0 IX 1965.

Shank. R.: 'Three Years Experience in the Co- omdinatedOutpatient Program at Washington University. J.Med.Ed.31:. 283 293. 1956.

Shaprio. L. Lifson. N.S.. Dublin. T.l).: An Experiment inInstruction in Health Education for Undergraduate MedicalStudents. J.Med.Ed. 2.1 229 235. 1949.

Shops,...: Medic al Schools and Hospitals. J.Med.Ed..Part IL I 169. 1965.

Shops. C.G. and riming, W.L.: Student Participationand Supervision: Symposium on Extramural Facilities inMedical Um ation. J.Med.Ed. 28: 44 8. 1953.

Shrand. H.: Th Teat hing Potential of a Home CareUnit. Lancet /: I416 7. 1066.

Silver. G.A.: Fundy Services as a Teaching Program.J.Med.Ed. 33: Imo 1058.

37

Silverton. N.H.: Student Attathment Schemes. J.Roy.Cull. General Practice 16: 480 4, 1968.

Slaughter. D.: Clinical Clerkships for Sophomore Medi-cal Students. J.Med.Ed. 24: 193199. 1949.

Smyth, C.J.: Postgraduate Training Through a GeneralPractice Residency, J.Med.Ed. 28: 20-28, 1953.

Snoke. P.S. and Weinerman, E.R.: Comprehensive CarePrograms in University Medical Centers. J.Med.Ed..625-657..1965.

Snyder, R.A.: Comprehensive Care in Medical Education,Qu art er I v Bulletin. Northwestern University MedicalSchool. 3i: 271-74, 1959.

SoInft, A.J. and Senn. M. J.E.: Teaching ComprehensivePediatrics in an Outpatient Clinic. Ped., 14: 547. 1954.

Sonkin.. L.S.: Home Care in Medical Education. J.Med.Ed.. 35: 465-510. 1960.

Spradlin. W.W.: Family Roles of the Elderly as a FocusIn Teaching Comprehensive Medicine. J.Med.Ed. 42:1045.58, 1967.

Steiger, W.A.. Hanson, A.V., Jr., Rhoads, J.M.: Experi-ences in the Teaching of Comprehensive Medicine, J.Med.Ed. 31: 241-48, 1956.

Steiger. W.A. and Hanson. A.V.: The Teaching of Med.icine at Temple University School of Medicine and Hos-pital. J.Med.Ed.. 32: 580-585. 1957.

Steiger. W.A.. Hoffman. F.H.. Hansen, A.V.: A Defini-tion of Comprehensive Medicine, J. Health and Human Be-havior, 1: 83-86. 1960.

Stokes. J.. Cliff, N.. Riche. C.. Rutstein. D.: The Effectof a Course in Family Medicine on Medical Student Skillsand Attitudes. J.Med.Ed. 38: 547 555, 1963.

Stunkard. A.J.: An Experiment in Teaching Psycho-therapy. J .Med.Ed. 32: 795. 1957.

Sturzaker, H.G. and Hoskisson: Report on Medical Edu-cation Suggestions for the Future British Medical StudentsAssociation. 1965.

Sweet. N.J. and Kerr. W. J.: Further Report on the Pre-ceptor System in Medical Education. J.Med.Ed. 26: 208-10.1951.

Sydenstricker, V.P. and Lee. F.L.. Domiciliary Medicineat the University of Georgia School of Medicine. J.Ass.Am.Med.Coll. 14: 88. 1939.

Tapp. J.W. and Densdale. K.: Medical Teaching in theCommunity. Med.Care 2: 214-218, 1964.

Ullmann, A.: The Role of the Social Worker in TeachingFourth Year Medical Students, J.Med.Ed.. 34: 239-46.1959.

38

Vuletic, A. and Jaksic, A.: Teaching of Undergraduatesby General Practitioners, Indian Med. J., 58: 114.118.1964.

Walker, J.H.: Teaching of Family and Community Medi-cine. BMJ, 2: 1129, 1966.

Waserman, E.: Medical Student Involvement in Compre-hensive Health Care. J AMA, 215: 2096-8. March, 1971.

Wearn, J.T.: The Training of the Physician, NEJM, 271:237-38, 1964.

Wegman, Sch oenbaum, S.C., Side', V.W.: Pilot Ap-plication of a Teaching Technique in Social Medicine.J .Med.Ed., 44: 1017-2344. 1969.

Weiner. H.: Behavioral Science Consensus. Their Func-tion and Relevance in Medical Education. Arch. Gen.Psychiatry. 4: 307. 1961.

Weincrman, E.R.: Yale Studies in Ambulatory MedicalCare IV. Outpatient Services in the Teaching Hospital.NEJM, 272: 947-54. 1965.

Weinerman, E.R. and Steiger, W.A.: Ambulatory Servicein the Teaching Hospital, J.Med.Ed., 39: 1020-29, 1964.

Weinstein. M.: Patients' Attitudes in a Medical Care Pro-gram, J.Med.Ed. 31: 168-171. 1956.

Weiskotten, H.G.. An Experiment in Medical Care,J .Med. Ed.. / 9: 159-168, 1944.

Wescoe, W., W.C.: Preceptors as General Educators inMedicine. J.Med.Ed. 31: 598-603. 1956.

White. K.L.: Primary Medical Care for Families, NEJM:277: 847. 1967.

White, K.L.: The Medical Schools Responsibility forTeaching Family Medicine. Med. Care I: 88-91. 1963.

White. K.L. and Flem;ng. W.L.,: Improving Teaching onAmbulant Patients, J.Med.Ed., 32: 30-36. 1957.

White. K.L.: An Outpatient Department and the Teach-ing of Preventive Medicine, Canadian M.J. 80: 506-508,1959.

Williams, R.H.: Medical Education and Patient Care,New Eng. J.Med. 271: 602-605, 1964.

Wolf, G.A.: The Preceptorship System at UVM,J .Med. Ed.. 32: 199. 1957.

Youmans. J.B and Russell. M.E.: Teaching Social As-pects of Medicine in the Medical School. J.Med.Ed.. 29:9-16. 1954.

V. GraduateEducationPrograms

nternship and residency training for primary carehas struggled with two main problems: the inade-quate overall coordination and direction that hashindered all graduate medical education, and thegraduate programs that were intended originally totrain for specialty practice rather than for generalmedicine.

Millis and others have elucidated the first ofthese problems (Millis, 1966, 1969: Kinney, 1972).Even though graduate training of the physiciannow constitutes the longest portion of his educa-tion, there has been no single professional or publicbody with overall responsibility for establishingstandards or allocating priorities for residencytraining (McKittrick, 1967). The university, whichhas accepted this responsibility at the medicalschool level, has not acknowledged its role in grad-uate education to the same degree. Moreover,"graduate education is unique among fields ofgraduate and professional education in being a re-sponsibility of institutions which have servicerather than education as their primary function...responsibility is divided among more than a thou-sand hospitals instead of among a few score univer-sities or medical schools. It is in a class by itself inthe extent to which responsibility reposes in indi-viduals rather than in faculties" (Millis, 1966). Ofparticular relevance to primary care, this serviceobligation has been rendered to hospitalized ratherthan to community-based patients.

rile second problem is more critical in regard toprimary medicine. As Stevens (1971) has indicated.the development of the residency itself was predi-cated on the need for further subspecialty educa-tion, over and above that necessary for generalistpractice. It is understandable, therefore, that medi-cally complcx hospital-based Practice has been seenas central to this education. Internal medicine andpediatric graduate programs, which evolvedprimarily to train the consultant specialist, have

been in particular conflict about their dual role.While they have the obligation to educate consult-ants, most of their graduates are engaged in pri-mary practice (Young, 1964; Bogdonoff, 1970).For example, in a discussion of internal medicineresidency training, one department chairman says,"to compete successfully in the future, if not to-day, each young intenist must have his subspe-cialty" (Meyers, 1964). Another notes, "The foun-dation for all training in internal medicine must bea period of intensive work on the wards, withdirect responsibility for patients" (Ebert, 1964).Our quarrel is not with what such pronouncementssay, but rather with what they omit and what pri-orities they reflect. As Bogdonoff (1970) states,"When a physician who is trained almost solely in asetting where Desperation Medicine makes up mostof the clinical lndeavor, the patients he sees incommunity practice turn out to have the wrongdisease."

It would seem that pediatrics and internal med-icine need to reach some rational decisions in theirrelation to family medicine. Either they ought torelinquish their role in primary care education andpracticeand also recruit proportionately fewermedical graduates to their own fieldsor else ac-knowledge their own obligation, to primary careand reassess their educational efforts with this inmind. To continue, however, to recruit the major-ity of medical students without accepting the factthat most of them should be prepared largely topractice primary care strikes us as almost unethical,given our current shortage of primary care physi-cians. In this regard, at least, the new family medi-cine residency programs have been fortunate. Start-ing with the single purpose of educating theprimary card practitioner, these residency programsDave been freer to formulate their plans withoutmultiple goals and prior hospital service obliga-tions. Even in their short history, however, their

39

educational and service requirements are unfortu-nately beginning to develop their own rigidity; andthis may be unavoidable.

The following section describes programs ingraduate and post-graduate primary care education.beginning with pediatric and internal medicine pro-grams; assesses family medicine programs: and con-cludes with comments on the situation in GreatBritain.

Pediatric and Internal Medicine ProgramsPediatric and in ternal medic;ne graduate

education have been shaped largely by the forcesalready mentioned. These gracieuate programs haveas their historic mission the training of the consult-ant physician: and the "core content" of suchtraining occurs on the acute inpatient service(Ebert, 1964: Bogdonoff, 1970: Lawson, 1969).Primary care programs need to be understoodagainst this background, because a kind of"grafted -on ". or at least peripheral, rather thancentral, quality has resulted many of them.

Mumford (1970) has alluded to the powerful im-printing effect of the internship, the year when"core content" is most strongly stressed. What con-cepts are emphasized in this period? While there issome evidence that interns come in contact with abroad spectrum of both common and uncommonmedical problems during this time (Wallace. 1971).several studies suggest that this is conveyed in an"instrumental" rather than in an "expressive" envi-ronment. As disease and procedures loom large,patient priorities and worries are of necessity lesscompelling. Payson (1961). in a time study of theinternship, observed that two straight medical in-terns averaged less than ten minutes daily witheach patient after the admission work up, and muchof that was impersonal in nature. A more recentstuds' of three West Coast intern programs revealedmore time spent with patients, although the vastmajority of this time was still concentrated in theformal history and physical examination (Gillan-ders. 1971). In a later study, Payson (1965) notedthat teaching rounds ratelv dealt with patient-physician communication skills and, indeed, ex-cluded the patient most of the time. What is con-veyed in such an atmosphere is a priority of valuesor a hierarchy of what is most important and whatis less so. As one resident recently observed. "Whenthe attending physician can't say anything intelli-gent about the ,disease, he usually talks about thesocial aspects of the case."

One cannot conclude, however, that physicianswho undergo this educational experience will be

inadequate primary care clinicians. Indeed, it ishard to see how the realities of acute hospital med-icine can be learned in any other fashion. Somefamily medicine educators have solved this prob-lem by contending that acute hospital medicine isreally a very peripheral part of their task and,therefore, needs far less emphasis during theirtraining programs.

Recognizing the neophyte doctor's urge initiallyto learn the management of the acute and seriouslyill patient, Hagerty (1969) has suggested that theacquisition of skills in long-term care, or in caseswhere continuity is important, be deferred untilthe later years of' training. In fact, at present, this isthe most common model in pediatrics and internalmedicine. In the light of Mum ford's observations.the model grants that the initial "imprinting" willbe a hospital-disease orientation and relies on thegrowing maturity of the resident and the alteredwork setting of the later years of training to restorethe balance in his clinical outlook.

It is within this context that a number of pro-grams in pediatrics and medicine, which emphasizeprimary care content material, have evolved. Theseprograms more typically segments within a largerprogram can be ordered along a spectrum ofintensity or of how much time and allegiance theyrequire of the student. At one end are those pro-grams that describe special rounds or educationalsessions on the inpatient service (Bates, 1965), inwhich social and psychological factors in the man-agement of hospitalized medical patients are em-phasized. Inasmuch as this aspect is often omittedfrom inpatient education (Payson, 1965), a sepa-rate sessiLin that involves social service, nursing,and continuing care personnel is arranged.

Somewhat more Intensive are programs wherehouse staff may follow patients over a period oftime in the outpatient clinic. These range from anoptional followup of inpatients or those considered"interesting" by the house staff to more intensive,

.required programs with assigned families or pa-tients in a separate comprehensive clinic( Bogdonoff, 1963: Miller, 1964; Wise, 1966:Haggerty. 1969). In these instances, house staffhave regular assigned sessions in which they seetheir own patients by appointment. Arrangementsfor ongoing responsibility for these patients be-tween appointments vary from completely adIvor, depending on the interest or largesse of theresident and the hospital switchboard operator, tomore (:rganized programs with separate secretaries,telephone answering services, and nurse practi-tioners to enhance communication.

At a further level of involvement, some pro-grams have established "model practices" within

the university center. which may have full-timepracticing staff to complement the residents. TheKaiser-Permanente residency training can be seenas, a version of this (Shearn. 1971). In an extensionof this model to the conttnunitV sett ing. theMontefiore Hospital Center in New York has estab-lished graduate training in medicine pediatricsthat uses a neighborhood health cente, .is the baseof its educational program (Kindig. 1969).

The hospital outpatient department itself hasbeen the site of many of these programs: and sev-eral authors suggest. as we have noted earlier. that,ambulatory patient education should be basedthere due to its similarity to medical practice(Wingert. 1966; Knowles. 1966). On the otherhand. a comment by Wedgwood (1969) reflects an-other truth about such programs: "Over half ofour (pediatric) residency (at the University ofWashington) at the present time constitutes ambu-latory pediatrics. much of it related to primaryhealth care...These programs are not popular withresidents. not because they are not given emphasisadministratively, but perhaps because of the qual-ity of instruction within the programs themselves.and because of the need for the physician in train-ing to get the acute care and the unusual off hischest."

SummaryIn summary, a kind of schizophrenia exists

about primary care education within traditional pe-diatrics and internal medicine. We mean this bothin the inaccurate lay sense of the term a splitpersonality, in this case a split allegiance to con-sultative and primary medicine identities and inthe truer definition. which is a separation ofthought and affect. often at an unconscious level.Although many departments verbalize the impor-tance of primary care programs for their trainees.their effort remains invested on the ward. Theproblem is complicated by the reality. These fieldsdo have a dual responsibility, and it has been gen-erally difficult in practice to integrate these oftenconflicting obligations.

Perhaps the most common current attitude ofpediatrics and medicine departments is reflected inthe retention of %vigil-developed secondary and ter-tiary medicine programs and the recruitment of ed-ucational faculty to develop ambulatory and pri-

ar y care under the departmental umbrella.Primary care is seen in this view as a new subspe-cialty. in a sense like hematology or endocrinology,which will be offered to the trainee as anothercareer choice. There are reasons to believe that this"let a thousand flowers bloom- philosophy willnot be successful without a more basic reordering

of department priorities. The fact is that the major-ity of the trainees will need to choose this careerprimary care in preference to all, the subspe-cialties: this is unlikely to occur in the present con-text. The final section of this monograph willconsider the issue further and suggest some pos-sible resolution of this educational dilemma.

Family Medicine ProgramsThe structure of most current family practice

graduate programs has been influenced both by thesuccess of the specialties in attracting candidatesand by the notable failure of the general practiceresidency programs established after World Was IIto do so.

As noted earlier, four years of medical schoolplus a rotating internship was considered adequatetraining for general practice until the 1950's. Thefield of general practice progressively excludedfrom influence within medical education from thetime of the Flexner report-was, therefore, not in astrong position either to appreciate the growing im-portance and attractiveness of residency training orto present an attractive postinternship program.The general practice residencies of the 1940's tothe mid-1960's were largely centered in hospitalsunaffiliated or only peripherally affiliated with uni-versities. They were chronically undersubscribedand considered less adequate educational experi-ences 'by the trainees (Gee. 1961). Furthermore,the growing popularity of the straight specialty in-ternships dud, Ile 1950's and 1960's served fur-ther to identify the student with that specialty andlessen his likelihood of entering a general practicecareer thereafter (Saunders. 1961). Finally, the at-tractiveness of entering a "specialty." coupled withthe absence of any limits on employment opportu-nity in this regard. further confirmed the relativeunattractiveness of a career in general practice. AsStevens (1971) notes. the rapid development andpopularity of residency training was specialty dom-inated. and "once again (the general practitioner)was left outside the specialty provisions. Onceagain he was identifiable by what he lacked, ratherthan what he had." The family practice programsthat have developed since the late 1960's have dif-fered in several important regards from these ear-lier programs. The new programs consider theirfield a specialty, and they hope to attract candi-dates from medical school into an integrated in-ternship and residency. Most importantly. follow-ing the Millis report. they are university based.

There were 59 approved family practice resi-dencies by mid-1971 (Gemini. 1971): over 15 bymid-1973: and more than one holf of the Nation's

41

medical schools had existing or planned familypractice programs. Most of the programs share sev-eral common teatures. They are based on the as-sumption that family medicine is a discipline dis-tinguishable from that taught in other clinicaldepartments, which therefore requires a separateunit within the university in order to develop itsbody of knowledge and to attract and traincandidates. In some schools the unit has been adivision of existing departments- typically medi-cine, preventive medicine, pediatrics, or psychiatry

or is itself a department. In general, local factorssuch as the degree of acceptance and support ofthe major clinical departments have influeoced thedecision to assume department status. The trend,however, seems to be toward a separate depart-ment, with a family physician or generalist ratherthan subspecialist as department chairman.

The field is, by definition, concerned exclusivelywith primary medicine, and most programs adhereto the definition formulated by the AmericanAcademy of Family Practice in this regard (1969).Established programs have tended to f''ollow the recommendations of the Willard Commission (1966)in developing a residency program. in %\ hich a totalof three years, including internship, is divided intotwo aspects: block rotations through subspecialtyservices, inpatient ur outpatient: and continuing in-volvement in a model family practice unit operatedby the medicine unit. Most programs emphasizethe ambulatory rather than the inpatient experi-ence as central to their purpose.

There has been an effort to attract a group offamilies representative of the general population ofpatients (Carmichael. 1965: Phillips. 1971: Smith.1971). This is in contrast to mans' existing univer-sity programs, which tend to involve either thevery poor or very specialized sub-groups that areless typical of those with whom most trainees willwork in practice. Finally, the programs have em-phasized that their faculty should be largely familypractitioners. rather than subspecialists, in order todemonstrate a role model for the student.

These programs share common problems as well.Most utilize existing specialty services for part oftheir training usually inpatient medicine and pedi-atrics; and this has produced conflict. "The familymedicine resident was not freed of his responsibilities for in-hospital care when he was assigned toambulatory care duty." observed Phillips andHoller (1971). Family medicine departments havethe alternative of operating their own inpatientservices. with the responsibility that entails, or con-tinuing to negotiate with the subspecialt:.s. We areunaware of iinv major institudon where a familymedicine department can operate its own inpatient

service: as a result of this, the departments are lessautonomous than the traditional services.

Second, as with any new discipline, there are nutyet family medicine faculty in sufficient numberswith experience in university teaching and researchto staff the rapidly expanding academic depart-ments. Skilled family practitioners have not oftenbeen teachers, and vice versa. Also, family practi-tioners have not often been researchers. What littleresearch they did was on educational methodsrather than on the merits of family medicine orhow to improve it. Insofar as most programs at-tempt to integrate and present material in a newfashion, they will likely need to develop their ownfaculty from among the new young graduates(Vuletic, 1966)..

Third, it is too early to tell whether the newfamily medicine programs will develop sufficient"legitimacy" in the minds of medical students toattract them in preference to the more establishedresidencies. In the past, more intellectual ableintellectual)

have chosen subspecialties in pre erenceto general practice programs (Monk, 1956). Thereis of course a "critical mass" phenomena that fam-ily medicine programs will need to overcomestudents who are attracted to primary care need tobe convinced that they will not be alone aftertraining and that enough family practitioners willbe produced to share the primary care burden. Inother words. graduate education for family prac-tice can only be successful if medical school experi-ence has assured the student that he is entering alegitimate and acceptable part of the profession.The past educational inadequacies of the generalpractice programs remain as something of a spectreto be overcome in the minds of students and fac-ulty. The nest decade will be critical in determiningwhether family medicine programs will overcomethese difficulties and attract a significant numberof students interested in primary care.

A final problem, commonly heard in discussionwith medical educators but less often writtenabout, concerns the field's "legitimacy' in theminds of other medical school faculty. (Sec page130.)Many academic internists, pediatricians, andother specialists do not believe that family medi-cine represents a viable form of practice or a realbody of knowledge distinct from their own. Thisattitude is not lost on the medical student. Othersare quite willing to accept the new field but "notin the university medical center." Locating the pro-gram's home in a community hospital or outside ahospital altogether Lonvevs to many a second-classstatus. It will be essential for all primary care pro-grams to educate the medical educators to the fact

that -University Hospital" does not equal "Univer-sity." In Short, programs central to student learn-ing may be located at varying sites in the commu-nity, all of importance to nicdical education.

Probably the appeal of the new family medicineresidencies has been the opportunity they affordf todesign a program with the educational goals welldefined and without the very real hospital serviceobligations that have so influenced other academicmedical departments over the past twent% years.Family medicine does not have a major researchprogram. This may he a real disadvantage. andacademic family medicine will need to develop itsown research program. Sonic programs are alreadyoverwhelmed by ambulatory service demands andare not developing the research base so essential tofuture growth.

Postgraduate EducationContinuing education for primary care practi-

tioners is an active field. judged by the number ofcourses and seminars offered. The Academy ofFamily Practice requires evidence that its members.in order to maintain accreditation, attend coursesannually the only specialty group to do so: andseveral states now require participation in continu-ing education for renewal of licensure. However,evidence of the benefit of these educational pro-grams is scanty. For example. Lewis (1970) couldnot correlate participation in postgraduateeducation by Kansas practitioners with improve-ment in health indices in their areas nor with in-creased use of certain recommended operative pro-cedures. Both Peterson (1956) and Clute (1963)found that attendance at postgraduate courses didnot correlate well with their measures of the qual-ity of a physician's medical practice. Uhl (1971)in a review

fewcontinuing education efforts noted

that. "The few studies of physician participation incontinuing education all document the fact thattraditional programs do not have a measurable of'feet on medical care in the institutional setting orin the physician's office.'' The Committee on Medi-cal Education of the New York Academy of Medi-cine reached a similar conclusion (1970). as ! aveother observers :Brown. 1970). It may be, ofcourse, that benefits of these programs do exist interms of physician satisfaction and stimulation.The value of a brief respite from a busy practice inan education milieu may be beneficial even if theresults are difficult to measure.

Must of the programs being criticized are short.several day courses. usually featuring speakersfrom subspecialty areas who report recent advances

in their own fields. Those programs devoted to up-grading specific technical skills also presentproblems. for example. McGuire (1964) notedthat those practitioners whose diagnostic arnmenhad significantly increased at the end of a briefintensive course in cardiac auscultation techniquehad regressed to their precourse level after severalmonths. Miller (1967) suggests that the problemlies in the fact that such education must be"learner based" to he effective. It must start withwhat the practitioner wishes to know, and itshould deal with problems that are common to hisexperience and about which something can bedone therapeutically.

Suggestions or descriptions of' other types of ed-ucational experience for primary care practitionershave included a one-year sabbatical program forpractitioners in a university setting (Brent, 1969),an exchange of jobs tor a one-month period be-tween academician and practitioner (Bergman,1969), and home study courses (Storey. 1971).The use of retired subspecialists (Hicks, 1972) andradio or television closed-circuit networks have alsobeen employed. In Great Britain, the British Broad-casting Corporation has an extensive series of pro-grams on public television specifically aimed at up-dating the general practitioner's fund ofinformation.

Ongoing seminars in the management of behav-ioral problems in practice have been reported bothfrom England (Balint, 1964) and the United States(Sumpter, 1968). Although success has beenclaimed for such efforts, a number of these pro-grams seem to have a defined "life span" of only afew years. The Rochester program of postgraduatebehavioral "worksliips" has continued and ex-panded over an eight-year period. with the additionof new practitioners and the withdrawal of others.These graduate programs do have the appeal ofadhering to the essentials suggested by Miller. Theirsubject matter is common, it is of perceived impor-tance to the participant, and improvement in pa-tients is often evident.

Sonic, other countries have addressed this prob-lem inure directly (Storey. 1971). Postgraduateeducation in the Soviet Union is accorded a highpriority. according to Storey. A stated goal is thatall practitioners enter teaching medical centers forthree months every three to four years. Home-study courses are provided. meanwhile, to enhancethe value of the time spent in the medical center.Although there is evidence that this policy has notbeen universally implemented (Muller. 1972), theidea that it is an accepted and stated goal is animportant step. While we lack any evidence thatsuch a program would result in changed behavior

43

of the practitioner much less in improved patientcare, it underscores a commitment to link educa-tion and practice. The opportunity for enhancedcommunication between educator and practitionerwould offer, at the very least,mutual benefit.

Our opinion is that continuing or postgraduateeducation is particularly pertinent to the primarycare field and ought to be more closely integratedwith education efforts at the medical school andgraduate level than it is at present. The idea ofplanning medical education as a single continuumhas of course been recommended by a number ofobservers (Millis. 1966: Haggerty. 1969). Perhapsthis advice has most meaning for primary care.however, as a result of the relative isolation of thepractitioner from the very influential educationalenvironment of the hospital and the university.Mum ford (1970) observed that within the univer-sity hospital two norms obtained that promotedcommunication and education: "the open mind"and "relay learning." These concepts help to stimu-late mutual observation and criticism by physi-cians, which seems central to continued learning.Most medical specialties are either closely affiliatedwith universities or hospitals in the United Statesor partake in varying degrees to mutual observationin this .cnvironment. Primary practice. conversely.

typified by the opposite a private. independ-,ent setting that tends to resist or at least not en-courage independent scrutiny. While this may havesome benefits in terms of the freedom and inde-pendence of the physician, its implications for con-tinuing education of the practitioner must be per-nicious. Particularly in light of the rapid increase inmedical knowledge. a system that fails to providethe physician with continuing, education is seri-ously deficient in its responsibility.

There is a trend toward locating continuing educa-tion programs as separate units within the universitymedical center or as responsibilities of specialtyboards or government agencies. Due to the intimatelink to basic primary-care education, we recommendthat family medicine, pediatric. and internal medi-cine units engaged in undergraduate and graduateprimary education be responsible for -at least theintegration of continuing education as well. Theeducational setting need not be the university itselfbut could include area health education centers orthe satellite programs of the university.

Primary Care Graduate Educationin Great Britain

The structure ,,f graduate level primary-care ed-ucation in Britain is ea,iier to comprehend than it isin the United States, largely because primary care

44

practice is more clearly separated from specialistpractice than it is in our own country. This divisionof responsibility separates personnel, site of careand method of remuneration.

All primary care, by our definition, is providedby general practitioners who practice largely out-side the hospital. All consultant care is provided byhospital-based specialists who have essentially noprimary care function. Remuneration for the spe-cialist is based on the consulting sessions he pro-vides for the general practitioner. This fairlyclear role division is longstanding: it antedatesthe introduction of the National Health Servicein 1948. In fact, that Act served to legitimizeand, unfortunately, in some ways to rigidify theseparateness of generalists and specialists.Changes anticipated in the reorganization of theNational Health Service in 1974 are aimed atimproving communication between the twogroups rather than altering the structure of theirroles,

As in the United States, completion of medicalschool plus one hospital year after graduation wereconsidered adequate preparation for general prac-tice until the late 1940's. At that time, a NationalTrainee General Practitioner scheme was intro-duced in which candidates spent usually one addi-tional year as apprentices with selected generalpractitioners. Although the trainee probably saw arepresentative sample of patients during this time(Richardson. 1972), there was wide dissatisfactionwith the results of the program (Whitfield. 1966).Criticism largely centered around the wide variabil-ity in the caliber of instruction provided. Further-more. the scheme never attracted more than tenpercent of those entering general practice, becausea physician could still earn far more by enteringpractice directly either as a principal or a paid as-sistant following his single hospital year: and therewas no evidence of the efficacy of spending addi-tional time in formal training.

In inure recent years a good deal of discussionand effort has gone into the design of primary-caregraduate education. The impetus for change hascome largely from the efforts of an energetic andarticulate group of general practitioners within theRoyal College of General Practitioners (1965.1967). together with suggestions of the RoyalCommission on Medical Education (1968). The lat-ter commission recommended that general practicebe recognized as a separate specialty and that allspecialties require an internship and three years ofcommon hospital-based training, followed then byseveral years of specialty training. This recommen-.dation has not been implemented to date. As of

1972. a consensus of educators suggests that allgeneral practitioners should receive three years ofspecial training atter the internship equivalentwith a long-range goal of five years of training(BMJ, 1972).

What currently exists area number of local pro-grams different regions of the country with posiOtitis fur a total of approximately I 7t) candidates:about 1.000 enter general practice annually (BMJ,1971: Lancet, 1972). Most of the programs have asimilar structure and usually last three years. Stu-dents spend one or two of their three years in hos-pitals, rotating through specialty departments ofmedicine and pediatrics and, in some places obstet-rics/gynecology, surgery, and psychiatv as well.An additional year is spent either in a family prac-tice teaching unit affiliated with a medical schoolor. more commonly, with a selected general prac-titioner. This apprentice year usually follows thetwo hospital years or is divided in some fashionbefore and afterward. Some programs include a"day release" feature in which trainees meetweekly at a postgraduate medical center with a tu-tor to evaluate their experiences, listen to formallectures, or to carry out small research projects.

In general, the new programs have been popular.with nine of ten available spaces filled. Criticismsfrom the trainees have centered around the exces-sive service demands of the hospital rotations, withinadequate attention to the educational needs of thefuture generalist or, again, around the variablequality of the general practitioner trainers them-selves. Only one in three of the latter have partic-ipated in a course in teaching. and the vast major-ity offer less than three hours a week of formalteaching to their trainees ( Lancet, 1972).

As in the United States, the impetus for develop-ing graduate education for primary care has comelargely from those within general practice andthose representing the public interest. rather thanfrom medical school and medical education repre-sentatives. The latter group. hospital-based and spe-cialty oriented, have in general resisted the estab-lishment of general practice teaching units withinmedical schools. When challenged. sonic argue thatcurrent medical education is adequate for generalpractice needs and that the newer programs are asvet unproved. At present. less than one-quarter ofBritish medical schools have general practice units.This may pose a long-term problem. If medical

school settings are considered the logical site formedical research to occur, then it Mak be difficultfor the field of general practice to develop furtherits body of knowledge unless it has this attach-ment. Alternate possibilites would be for this newfield to become affiliated with departments ofcommunity medicine within medical schools, withuniversity. departments of social and behavioral sci-ences or, most likely to develop further their ownbasic research resources within the Royal Collegeof General Practitioners. At any rate, some univer-sity or equivalent resource would seem highly de-sirable for a field that has great need to develop itseducational content.

A further interesting phenomenon is the rela-tionship in Britain between Departments of SocialMedicine and the developing field of academic gen-eral practice. In general. social medicinedepartments have adhered to an epidemiologicresearch orientation that has not included clinicalinvolvement, although there are exceptions. Whilesome general practice programs have been, by mu-tual consent, sponsored by departments of socialmedicine, most have sought a separate identity.Social medicine has historically perceived its orien-tation as being toward community or populationmedicine rather than personal medicine (McKeownand Lowe, 1966): and what little organizationalintermix has occurred has taken place, because so-,7ial medicine has provided a (temporary') base forgeneral practice within the medical school setting.

SummaryIn Summary, the picture is one of rapid growth

and change, as it is in the United States. There is apublic commitment to graduate primary care edu-cation, as reflected in official recommendationsand evolving plans and programs for such edu-cation. But a two-fold problem continues to exist.Only a small minority of practitioners currentlyelects to undergo this training prior to practice as aprincipal: at present, the number of training posi-tions is insufficient. It is unclear as vet whether ornot the problem can be resolved by the develop-ment of attractive new programs alone. Currently,the Royal College of General Practitioners has sug-gested that mandatory graduate education be re-quired after 1977, a coercive policy not yet com-pletely agreed upon even within general practice.

45

References

SECTION V

American Academy of Family Practice. Commission onEducation for Family Practice. Kansas City. Missouri. Au-gust. 1969.

Ba lint, M.: The Doctrr, His Patients and the Illness,London. Pitman Medical. 1964.

Bates, B.: Comprhensive Medicine: A Conference Ap-proach. I .Med.Ed., 411: 778. 1965.

Bergman, A.B. and Skinner. A.L.: Professor PractitionerExchange Program. Ped. Clin. NA. 16: 815, 1969.

Bogdonoff, M.D.: A Change in the Training Model forthe Practicing Internist. Arch. Int. Med.. 126: 694. 1970.

Bogdonoff. M.D.. dwell, S.W., Ruffin, J.M.: MedicalOPD Teaching. J.Med.Ed. 38. 885-89. 1963.

Bogdonoff, M.D.: On a Contradiction Between Intentand Enactment. Arch. Int. Med. 126: 1070, 1970.

Brent. R.L. and Morse. H.B.: Not in Our Own Image:Educating the Pediatrician for Practice. Ped. din. NA.16: 793. 1969.

British Medical Journal. Vocational Training in GeneralPractice, 2: 176. 1972.

British Medical Journal. Vocational Training in GeneralPractice III. Wessex 3: 41. 1971.

Brown, C.R.. Jr. and Uhl. H.S.M.: Mandatory Continu-ing Education: Sense or Nonsense? J.A.M.A. '13: 1660.1970.

Carmichael. L.P.: A Program of Instruction in FamilyMedicine. New Eng. J.Med.Ed, 40: 370. 1965.

Clute, K.: The General Practitioner. Univ. of TorontoPress, 1963.

Ebert, Richard V.: The Residency I New Eng.J.Med.,271: 547.59, 1964.

Gee, H.H. and Schumacher. C. F.: The Internship.J.Med.Ed. 36: Part 2. 34-60. 1961.

Geyntan. J.P.: Family Medicine as an Academic Disci-pline. J.Med.Ed., 46 815, 1971.

Gillandem W. and Heiman. M.: Time Study Compari-sons of Three Intern Programs, J.Med.Ed. 46: 142-149.1971.

Haggerty, R.J.: The University ar:d Primary MedicalCare. Nev: Eng. J.Med. 281: 416. 1969.

Hicks. N.: Doctors Briefed by Ex-Professors. N.Y.Times. April 24, 1972.

Kindig. D.: Social-Medical House Officerships,New Eng.J.Med., 1078: 1969.

Kinney. T.1).: Chairman-AAMC-Ad Hoc Committee onGraduate Medical Education J.Mcd.Ed. 47: 77. 1972.

Knowles. J.: Ambulatory Patient Care. J.Med.Ed,41: 722, 1966.

Lancet: Training for General Practice: The Trainee'sView /: 943. 1972.

Lawson, R.B.: Pediatrics: The Relation of Training; toMultiple Tracks of the Future. J.Med.Ed. 44: 791. 1969.

Lewis. C.F. and Hassancin. R.S.: Continuing Medical Ed-ucation. An Epidemiologic Evaluation. .M.. 282: 254.1970.

46

McGuire. C.. Hurley. R.E.. Babbott, 17.. Butterworth.J.S.: Auscultatory Skill: Gain and Retention After Inten-sive Instruction J.Med. Ed. 39: 120. 1964.

McKeown. T.. and Lowe. U.K.: An Introduction to So-cial Medicine. Blackwell Scientific Publications, Oxford andEdinburgh. 1966.

McKittrick, L.S.: Rational Responses to Graduate Edu-cation of Physicians. J AMA, 2W : 112. 1967.

Mechanic. I).: General Practice in England and Wales:Results From a Survey of :a National Sample of GeneralPractitioners. Medical Care 6' 245. 1 %8.

Meyers. J.D.: A Consideration of Residency Training.New Eng. J.Med. 271: 573. 1964.

Miller. G.E.: Continuing Education for What? J.Med.Ed.42: 320, 1967.

Miller. W.R.: A System to Provide and Teach Compre-hensive Medical Care. J.A.M.A. 189: 1-5. 1964.

Millis. J.S.: Chairman -- Report of the citizens Commis-sion on Graduate Medical Education. AMA. Chicago. 1966.

Millis. J.S.: Objectives of Residency Training, CanadianMed. Assoc. J. 100: 599. 1969.

Monk, M.ft and Terris. M.: Factors in Studeat Choice ofGeneral or Specialty Practice, New Eng. J. Med. 25.5:1135. 1956.

Mumford, E.: Interns: From Students to Physicians.Commonwealth Fund. Harvard Univ. Press, Cambridge.1970.

Muller, J.E., Abdellali, F.G.. Billings. F.T.. Hess. A.E..Petit. D., Egeberg, R.O.: Soviet Medical System: Aspectsof Revelerice for American Medicine, New Fag. J. Med.286: 693, 1972.

New York Academy of Medicine, Committee on Med.Ed.: Continuing Education f the Practicing Physician,Bull.. N.Y. Acad. Med. 46: 893, 1970.

Payson. H.E.. Gaenslen. E.C., Jr.. Stargardtcr, F.1..:Time Study of an Internship on a University Medical Serv-ice. N. E. J .Med. 264: 439-43, 1961.

Payson. H.E., and Barchas. J.D.: A Time Study of Medi-cal Teaching Rounds, New Eng, J.Med. 27.1 1468,

P?;erson. 0.. Andrews, L.D.. Spain, R.S.. Greenberg.li.G.: An Analytical Study of North Carolina General Prac-tice 1953-54, J.Med. Ed. 31: Part 2. 1956.

Phillips, T.J. and Holler. J.W.: A University Family Med-ical Program. J.Med.Ed. 46:: 821.1971.

Richardson, I.M.. and Howie. J.G.R.: A Study ofTrainee General Practitioners. British J.Med.Ed, 6: 2g.1972.

Royal College of General Practitioners Report on Im-plementation of Vocational Training. Published by Councilof R.C.G.P.. August. 1967.

Royal College of General Practitioners. Special Voca-tional Training for General Practice. Reports from GeneralPractice Nos. 1-6. 1965-1967.

Royal C(mrmission on Medical Education Report.London, H.M.S.O.. 1968.

Rutstein, D.: Physicians for America: Two Medical Cur-ricula, J.M.E..?6: Part 2. 129. 1961.

Saunders. R.H. and the Committee on Internships, Resi-dencies and Graduate Medical Education of the Assoc. ofAmerican Medical College, The University Hospital Intern-ship in 1960. J.Med.Ed. 36: 561, 1961.

Shearn, M.A.: Professional Education in a Service Sys-tem in the Kaiser-Permanente Medical Care Program.Somers, A.R.. Edit. Commonwealth Fund, New York,1971.

Smith. R.: Family Medicine in the Academic and Com-munity Settings. Proc. Roy. Soc. Med. 64: 523, 1971.

Stevens, R.: American Medicine and the Public lnterst,Yale Univ. Press.. Ncw Haven, 1971.

Storey, P.R.: Continuing Medical Education in the So-viet Union. New Eng. Med. 285: 437. 1971.

Sumpter. E.A.. and Friedman, S.B.: Workshop DealingWith Emotional Problems: One Method of Preventing theDissatisfied Pediatrician Syndrome" Clin. Pediat. 7: 149,1968.

Uhl. H.S.M.: Continuing Medical Education, New Eng.J Me d. 50. 1971.

Vuletic, A.: Undergraduate Teaching by General Practi-tioners, WHO Med. Ed. Bull. No. 2, 26-34, 1966.

Wallace, P.D., and Silber, D.L.: Analysis of a StraightPediatric Internship, J. Pediat. 79: 110, 1971.

Wedgwood, R.J.: Quoted in Crook. W.G. Criticismand Dissent Ped. Clin. N.A.. 16: 997, 1969.

Whitfield, M.J.: Training for General Practice: Result ofa Survey into the General Practice Training Scheme, Brit.Med. J. 1: 663. 1966.

Willard, W.R. (Chairman): Meeting the Challenge ofFamily Practice. Report of the Ad Hoc Committee on Ed-ucation for Family Practice of the Council on Medical Edu-cation. Chicago: Amer. Med. Assoc., 1966.

Wingert, W.A.: The Utilization of the Intern in a Pedi-atric Outpatient Setting, J.Med.Ed. 41: 756-65, 1966.

Wise, H.B., Spear, P.W., Silver, G.A.: A Program in Com-munity Medicine for the Medical Resident, J.Med.Ed.41: 1071-76.1966.

Young, L.E.: EducaKm and Roles of Personal Physi-cians in Medical Practice. J.A.M.A. 187: 928-33, 1964.

4X/8

VI. The ElementsOf ASuccessfulProgram

In light of the historical trends we have outlinedand the programs we have reviewed, what can werecommend so that more physicians will be trainedwho are prepared to practice competent primarycare? (1f equal importance, the education in-sure that most will adapt to. and some lead in. theevolution of primary care practice during their owncareers?

We feel confident in the validity of our propos-als as they concern educational programs as a resultof our own personal experiences. But why limitourselves to this level of discussion? Let us see howthe history of medical education has been in-fluenced by actions and events occurring at muchbroader levels -within the entire medical school.the practicing profession, the public funding agen-cies. and the climate and priorities of the time.

We can begin arbitrarily with the Flexner report.It led to a decision to link medical school and uni-versity in the 1910-to-1920 period and fostered thesubsequent growth and organization of the medicalspecialties. Consider these extrinsic factors affect-ing medical education: the limitation on the num-ber of medical student positions during the 1940'sand 1950's, the enormons Federal investment inbiomedical research in the 1950's and 1960's. thegrowth of consumerism and the pressure to admit-minority" students in the 1970's. These are allexamples of actions largely emanating from forcesoutside medical education itself through. signifi-cantly, interacting with persons within the educa-tional system which have had as much impact onshaping the kind of doctor who now enters prac-tice as any set of curriculum changes. special pro-grams, ur charismatic instructors.

If another example of the influence of publicpolicy on medical education is needed. we shouldconsider the cutbacks in federal funding of bio-medical research during the period. 1969 to 1970.Faculty and schools whose income was largely

derived from this source-in increasing .annualincrements -were rudely awakened to the fact thatthis was not "natural law", but reflected Politicalskills and realities as much as the importance andworth of the research itself. So. it will be with thepresent "natural law". that medical schools turnout large numbers of practitioners and do it imme-diately.

Medical educators will. therefore. need to bemore actively involved in the political processesthat influence medical care organization in generaland medical education in particular, if their in-fluence is to count. Although we do not possessthe knowledge for elaborating on the tactics of thispoint. we are quite convinced of the correctness ofthe strategy. This involvement can be at the localand State levels by regional planning and by devel-oping communication with elected representatives.At a national level, spokesmen for medicaleducation the Association of American MedicalColleges and the professional societies alreadyhave contact with Congressmen directly or throughregistered lobbyists. These efforts should be under-stood. supported by the membership. and ex-panded if possible. Recent events clearly indicatethat government support of medical education ischanging. It is the responsibility of medical educa-tors to attempt to influence so that the inevitablestringy attached to this funding are not tangled, asso often happens. in all irrational fashion.

At the medical school level, a major task of allconcerned individuals is to determine exactlywhere the responsibility for primary care educationshould lie. Just as medical education and medicalresearch must compete for finite resources at thepublic level. so primary care must compete for fi-nite educational resources at the medical schoollevel. Part of the failure to accord primary careeducation a high enough priority in that it has notbeen the Main coMillithient of any one department

49

Within the sc hool. When the time collies to selectthe medical students, to divide up the curriculum,to select the resident staff, and outline theirprogram obligations. no one speaks loudly and consistently enough for the needs of the student whowill enter primary care. There .ire exceptions tothis. of course: as with -comprehensive meth, me-,most faculty members think about the problemsonic of the time. The major clinical departmentsparticularly internal medicine and pediatrics havejust enough interest in this issue, so that they areoften unwilling to relinquish the responsibility to anew department of family inedicint but trotenough to develop effective education effortsthemselves. The Pellegrino Committee (1 968)spoke to the issue as follows:

Inevitably this vexing question will arise:which department should teach the generalistfunction? . . . In some instances a departmentof general practice might well be contemplated:in others, the department of medicine, pediatricsor community medicine might take the lead. Aninterdisciplinary program calling on all depart-.went, but totally dependent on nu single one ofthem might be the optimal solution.Our feeling is that public pressure ma help to

force the issue. Outside funds earmarked for pri-mary care education may serve as the necessarystimulus for the medical school to define where itsprimary care commitments lie. Without intendingto equivocate, however, there is danger in imposingtoo precipitous or too rigid a solution. The -currect" primary care practice model still remains un-resolved, although there are probably several satis-factory ones. It would be a mistake to insist on asingle template at this tiine, fur the situation ateach medical school varies enonnoway with inter-ested and capable people located in various clinicaldepartments and in the Dean's Office. Moreover,medical schools have the responsibility t," evaluatethose programs developed in primary care to pro-vide needed data about the pathways travelled

With dies:. cautions in mind, we suggest steps tobe tal..en at two levels, First. at the National level,those professional organisations that representmedical educators should begin a inure active dis-cussion of the problem, both internally and withrepresentatives of other generalist and specialtygroups. This need is particularly apparent in inter-nal medicine and pediatrics: their obligations toprimary care education and practice and, specifi-cally, their relationship to family medicine shouldbe defined more precisely. Do they wish largely torelinquish their role in primary care to familymedicine the must radical and least likely

511

solution. though possibly the most rational one?Given the great sire and diversity of practice in ourcountry. it is more likely that several coexistingpatterns will emerge. Indeed, some variation maybe desirable, inasmuch as present evidence for theclear superiority of one pattern or the other is lack-ing. Nevertheless, national education and practicegroups must spell out their positions, with the MIpliations of those positions for student educationwell elaborated. For example, if internal medicinewishes to retain its current de facto primary careobligation, how will it help insure that the majorityof its trainees du enter primary practice and, just asimportant, are well prepared fur their career-. Ifinternal medicine opts to defer to family medicinein this matter, how will it limit its recruitment ofinedicl students to the minority required for consultant work?

Along with this "vertical'' debate, a similar "hor-i/untal- discussion should take place within eachmedical school. Here, two pertinent decisions areto be made. Which department or interdisciplinandivision shall ''hold the primary care contract '',and what are the obligations of the other clinicaldepartMentS to primary care' All medical schooldepartments ought to have a stake in any programthat accounts for the majority of its graduatesassuming the physician retains his role in primarycare. Moreover, each department should state itspolicy and program at the undergraduate, r.si-dewy, and continuing education levels for the ad-vice and discussion of the medical school as awhole. As for (Hier departmental obligations, wehave in mind those specialties with important butpresently ill-defined roles in primary care, such aspsychiatry, obstetrics/gynecology. and communitymedicine. For example, in our reaiiew of liaisonprograms between psychiatry and medicine, wenote that psychiatrists tardy have accepted thechallenge of developing a body of knowledge andtechnique appropriate to primary care. Most often,they selected items from general psychiatric theoryand practice and adapted them for consultant pur-poses, rather than developing ideas front the view-point of a primary care participant. Moreover,tr "'els that have been developed in these programsaL none appropriate to a hospital inpatient or out-patient setting, rather than to primary care.

The establishment of a special clinic within ahealth center is not What we have in mind c ither.Although there is a place IAA such clinics, they donot address the issue of major concern. The ques-tion is not what the usefulness is in a health centerof an adolescent or an orthopedic clinic. but ratherwhat the implications of orthopedic or adolescent

medicine are for the organization and practice ofprimary care. The latter is a much broader andmore difficult charge that must be accepted by spe-cialty divisions within schools of medicine if newknowledge is to develop. Much of the problemstems from the facts that we do not. at present.know the answers and that the specialties havebeen more concerned with elaborating their owndiscrete areas rather than attending to the needs ofthe generalist. Once again, this points out the needfor research and evaluation. In addition, the veryreal service and education obligations of the spe-cialties have rarely left sufficient time or physicalenergy. much less the intellectual energy. for theinvestment required in developing primary careprograms. We see this as a job for the primary andthe consultant departments to undertake together.There are persons within the specialties who wouldfind such questions challenging. ant! they should beencouraged to develop their ideas within the pri-mary care educational setting. It will be necessaryto secure sources of funding directly for this taskwithout attaching it so tightly to service demandsthat, again, a makeshift model is constructed.

In addition to developing and defining its rela-tionship with other departments within the medi-cal school, the unit responsible for primary careneeds to establish tics with those involved in otherhealth science fields. particularly in nursing andsocial service education. At present the relativerules of the physician, nurse, and social worker inprimary care is in flux: but communication amongthese disciplines is essential as new programs de-velop. Although physicians often prefer to devisemodel programs without the advice and participa-tion of these allied health professionals. the !Milo-dons of such an approach become evident whenattempts are made to expand the program beyondthe local level. The fact of the matter is that pri-mary medicine does have ill-defined borders withconsultant specialties. on the one hand. and withallied health professions. on the other. Althoughthis makes for organizational complexity. it wouldbe better to recognize and legitimize these relation-ships overtly rather than to develop programs inisolation.

With these broad charges in mind, let us turnnow to specific elements within educational pro-grams for primary care. Even with a supportivepublic climate, allocated funds, and a committedmedical school administration and fl:.ulty, thequality of any program will be influenced by anumber of internal factors. Hansen and Reeb(1970) have outlined a very complete curriculumfor primary care education. However, in translating

their material into a viable progranr for student andhouse staff education, attention to the componentsof the system is of equal importance. As we havenoted earlier. many of the programs reviewedherein have shared identifiable and common inter-nal problems that have compromised their educ4-tiunal effectiveness; and, so, attention to these ele-ments may be 'useful.

Specifically. these elements or ingredients of anyprogram are the students, the faculty. the patients.the curriculum structure, and the setting. They areall interconnected, each affecting the other to formthe "learning environment." There is value, how-ever. in considering them separately. in turn, whilerecognizing, that they form an int:gral patternwithin the larger setting of medical education.

The StudentsWhatever the other characteristics of the educa-

tional program are. they must all funnel into andbe processed by the student. the "final commonpathway" and a most important ingredient in edu-cation. We have discussed student selection of amedical career and correlates of their success dur-ing training in an earlier section, but would herestress several aspects of central importance to pri-mary medicine. These are the attraction and selec-tion of suitable candidates, the concept of studentreadiness and maturity for various aspects of theprogram. and the responsibility of the student forhis own ecaication.

In general. there .has been more work towardcorrelating test performance and personality ofmedical school applicants with how they behaveduring medical school and residency than with howthey behave in practice. The reasons for this areunderstandable: although performance duringmedical education is only an in ermediate or proc-ess variable, it is the medical faculty's measure of astudent's development. Besides, the performanceof students in course work appears easier to meas-ure than how they do in practice. where agreedcriteria of adequate performance are still lacking.Unfortunately. the correlation of standardized pre-admission tests, like the Medical College AdmissionTest (MCAT) with performance in medical schoolis not good. much less with performance in resi-dency and practice. Nevertheless, in the absence ofdemonstrated validity of other measures. admis-sions committees still lean heavily on demonstratedscience skills in their applicants (Rutstein. 1961).

Student personalities are fairly well establishedand conform generally to specific patterns by thetime of admission to medical school. It is hardlylikely that all with above-average academic ability

51

and a strung natural science background will beinure productive and satisfied in careers as primarYcare physicians careers than as competent special-ists, We need to identify those personal qualitiesthat prove to be valuable assets in primary practice,and here the data is woefully inadequate. The taskmight properly begin with an attempt to definecertain desirable qualities in all those who have en-tered medical careers. Jeffeys i 19711 has outlinedseven "ideal characteristics" of a doctor as follows:

1. Above-average academic ability, in order to un-derstand the scientific basis of medicine and di-quire the diagnostic and therapeutic kill to applyit:

2. Above-average ability to sustain concentratedstudy:i. Wei-developed humanistic values, including wil-lingness to hircgo personal comfort and postponegratification in order to meet health priorities;4. Willingness to make decisions and carry respon-sibility:5. Physical energy and emotional stability;6. Interpersonal skills, including sensitivity to theneeds of others;7. Capacity to teach, especially in face-to-face clin-ical settings.

Ihere mahighly desirable fur the practice of primary medi-cine. For example. Mechanic (I 968) identified aset of attitudes and orientations that distinguishedsatisfied from dissatisfied general practitioners in(;re it Britain. Satisfied doctors ''tend to acceptinure readily than discontented doctors the per-sonal and ,ocial aspects of midicine and . . . . incontrast to dissatisfied doctors they report thatthey prefer to work th complicated byemotional factors and with patients who questiondiem and ask for inure detailed examinations.'' Onthe other hand, work from our own country sug-gests that students who now select g,ened practiceas a career share certain characteristics as a groupmost of them nut what one would consider verydesirable low

ofperformance. low scores

un measure's of "theoretical interest,- low intrinsicmotivation. and high authoritarianism. (Sanazaro,1965: Monk and Terris. 1956; Coker. 19651. Whatthis ma indicate. however. are selective featureswithin the education:d structure that propel stu-dents with these qualities away from the specialtiesand therefore toward general practice. The indliCe-Illeilt to students with above average achievementlevels to choose specialty careers by specialtyfdeult\ is obVioll'dt, a ,trong influence.

be other characteristics that are

51

It may be fairly argued that qualities desirable ina primary care titioner may DUI be easy todefine, much less IlleaNure, and that attitude andpersonality measurement are not sophisticated orrefined enough to he useful, The reality is thatsome standard. ate used already and that there isconsiderable )11 to question their appropriate-ness, Although there is evidence that the student'sclinical competence in practice derives from a com-bination of hi. personality and background withthe length and quality of his training (Lyden,1968), there is need for more research on the for-mer.

Our recommendation is that attention now bedirected to defining qualities that correlate withsatisfaction and performance in practice and to de-vising methods for measuring these qualities. Theconcept of peer review in primary., care may pro-vide an entering wedge into the definition of clini-cal competence. The process of attending to selec-tion of students for primary care is, itself, likely tobe a beneficial one. even if solid techniques arcslower to evolve. With a department or divisionwithin the mcdi, school responsible for primarycare, this would be an appropriate topic for re-search and a high priority for admissions com-mittees.

A second concept involves the students' readi-ness fur various aspects of the curriculum.Hagurty (1969) suggests a bimodal curve of activ-ity in community programs for educationalpurposes high in medical school and late resi-dency, low irr tiIthe internship year when the stu-dents deal with acute illness management. The dan-ger of an approach that omits primary care at theinternship stage is that the powerful "imprinting.'of the internship experience may be difficult toreverse (Mum ford, 1970). Again there are eve dataun which to base a judgment. It does seem logical,however. to argue that all the content areas in pri-mary care cannot be learned equally well at anygiven stage of training, Areas involving behavioralur such' aspects of care are often attractive andpertinent to the student in the late stages of hiseducation, particularly in practice (Sumpter,1968), when they have nut proven to be so earlier.Part of this effect may relate to factors of setting,curriculum, and faculty priorities to be discussed;but "mutual participation medicine may require amore mature person that active passive'' medicinedoes (Sias/. 1956.. Dealing with patient problemsthat require sharing responsibility for managementbetween therapist and patient calls for a degree ofsecurity and clinical judgment in the doctor that

needs nurture and time to evolve. Therefore, ef-forts need to be made to integrate practitionersinto educational programs. so that they may con-tinue and deepen their skills.

According to many .practitioners. a full-time.practice gives them insufficient time ter participa-ting in continuing education programs. Also. notbeing able to get coverage of their practice duringan extended absence poses an additional problem.Although a decrease in their income would un-doubtedly be a deterrent. time and coverage prob-lems are considered crucial. Here, the universitycan play an active role. Involvement of practition-ers in programs of collaborative research (Haggerty,1969) can be achieved if secretarial and researchassistant support for the practitioner is provided, arelatively modest expense that saves his time. In-volvement in longittdinal behavioral "workshops"and preceptorships should also .verve to forge linksbetween the medical center and practice that areeducational in themselves and can serve as the basisfor further sabbatical-type arrangements. Judicioususe of new allied health manpower can also betime-saving. If the addition of nurse practitionersenables pediatricians to care for the same patientpopulation with 25 percent less physician-time in-volved (Chimney. 1971). this, in effect, can free thetime of one practitioner in a four-man group prac-tice. The staff might elect to use this "bonus" todevelop rotating educational leave program. Somewould argue that such an arrangement would de-feat the main purpose of the employment of alliedhealth professionals. namely the ability of the samenumber of physicians to handle an increased pa-tient case-load. On the other hand. there may belong-term benefit in having arrangements to attractand retain more candidates in primary care. espe-cially if they are given the opportunity of periodicrelease time for study ;,.nd change of pace.

Part of the incolve for the practitioner duringhis sabbatical may by derived from involvement incertain Hospital - based ambulatory programs thatcomplement his skills in practice; e.g., working sev-eral half-days in a referral diagnostic unit or in acommunity-based consultation program in mentalretardation. cerebral palsy, or school health. How-ever. direct grants to supplement these fellowshipswill be required as well. Insofar as these programsare oriented toward improving skills appropriate toprimary care practice rather than toward wooingthe practitioner into a specialist career. they shouldbe encouraged. Again, sponsorship of thesepostgraduate programs by the primary care unit

should help insure that their focus is indeed appro-priate. We wish to emphasize that these suggestionsare meant more as a stimulus to thought and initiative rather than as a blueprint for specific actions.Ideally, the educational experience should be tai-lored to the student's level of skills and maturity.

A final consideration should be given to the re-sponsibility of the student for his own education.Milks (1969) observes that graduate medical pro-grams "seem to be training (to form or habituate)but ought to be an education (to develop, culti-vate, expand)." Central to this distinction is theassumption that a student should be responsiblefor his own learning throughout his professionalcareer, an avowed goal of all medical education. Webelieve that the best way to strengthen this as-sumption is to encourage this self-teaching pattern,while he is still in an educational setting. The rapidtrend for more elective studies within medicalschool is consistent with this goal, but this hasbeen less true of the residency period. In large part,this reflects that ambivalent position of graduatemedical education that is a shared responsibility ofthe hospital, with its heavy service obligation, andthe university, whose primary mission is education.The graduate medical studentthe residentmustbe given the opportunity to create and be responsi-ble for his own education to a greater degree--in asense. given the right to experiment. The FamilyMedicine Residency at the University of Miami is agood example of this innovation (Carmichael,1972). Presently, the only way this can be accom-plished is to limit the student's service burden tosome degree or limit the time now spent in subspe-cialty education. The utter dependence of mostuniversity hospitals on house staff for patient careconflicts with this goal. For example, how can aresident work with a migrant worker group tryingto determine its own health needs and, at the sametime, deal with the never-ending flow of patients inthe emergency department?

We do not advocate the abandonment of clinicalresponsibility by house staff. On the contrary. thisresponsibility is an essential ingredient in their edu-cation and must be retained. Rut we must be cogni-zant of an imbalance existing in many residencyprograms that is detrimental to the student. Ob-viously. he must learn his responsibility to the indi-vidual sick and needy patient. He also needs thechance to learn . his responsibility to the sick andneedy community and to define his own role inthat community as well. 'Phis learning process re-quires time and experience.

53

How can time be secured for elective programswithin the constraints of an already crowdedschedule? Several mechanisms might be consid-.:red: first, effective use of technicians and physi-cian assistants who are now part of the hospitalsetting amt., also should be maximally utilised tosave the resident's time. For example. residentsneed not perform routine laboratory tests such ascollecting blood samples and setting up intravenousinfusions, which now are increasingly carried outby technicians. Infant and adult intensive careunits. premature nurseries, burn units and otherspecialty wards now function largelysome wouldsay more effectivelywith technician and nursemanpower with the advantage of greater personnelstability than rotating house staff. As hospital spe-cialty care becomes more technologically complexrather than "intuitive," it is increasingly amenableto direction by a specialist physician with techni-cian assistance. At the primary care level, growingevidence that nurse practitioners can assume por-tions of the traditional physician role lends supportto selective apportionment of the student's time inthose areas as well. Moreover. at current housestaff salary levels. there is less financial inducementto consider the resident a source of cheap labor.

A second and probably more important ap-proach requires that specialty services be more se-lective in the experience they provide the residentheaded for primary care. While we consider it valu-able for the student to be intimately involved inthe complex care of the critically ill patient duringpart of his education. it is difficult to justify theextensive time required for such care in some inter-nal medicine lnd ,ediatric training at present. Inpart, this reflects the dual responsibility of bothdepartments for preparing both prim.lry practition-ers and consultants. But. in this combined pro-gram, the primary care trainee is shortchanged.Whereas his basic education is finished at the endof residency. most consultant specialists will havetime for the sharpening of their skills during afellowship.

Certainly most of the techniques now taught formanaging specialty disease will change within avery few years, in many cases before the studentsees another case in practice. The major justifica-tion for his participation must be in coming tounderstand the approach of the specialist, in sens-ing the potentialities and limitations of his fieldand in learning what will happcn to patients herefers. While the multiplicity of specialty areas areinherently interesting disciplines themselves, unbalance the student may benefit more from timespent in programs that direct his energy to primary

54

care problem areas. Of course. the student requiressufficient time in the specialties, su that he mayaccurately identify a patient's need for the special-ist referral and also learn how the specialist's skillsare best adapted to primary care practice.

In short, we suggest "buying time" for electiveexperiences by maximizing use of technician andallied health manpower as well as specialty train-ees. rather than automatically staffing expandedspecialty and ambulatory services with primarycare students. The value of each segment of theprogram must be justified on educational grounds.

Much of the success of the specialty aspects ofthe residency relates to the calibre of the consult-ants as teachers, quite apart from the applicabilityof their teaching to primary care. The challengefaced by primary care programs is that they needto create stimulating and challenging research andeducational projects that arc as attractive as thoseof the specialties: This leads to a consideration ofthe role of the faculty. a second ingredient in theeducational structure.

The Faculty"Do as I say, not as I du." can be as fallacious in

primary care education as it is in child rearing. Forexample. the University of Rochester offered atwo-year rotating internship between 1949 and1961. one purpose of which was to train the physi-cian for general practice. In a followup study.Romano (1964) observed that only seven percentof the trainees, in fact. ended up in general prac-tice. The program consisted solely of rotationsthrough specialty services, wit!: no general practi-tioners at all on the faculty. In fairness to what wasfelt to be a successful program, a second goal wasthe provision of a "broader base for the specialist."Outcomes such as these suggest the important in-fluence of the faculty as role models.

It may seem a truism to state that a good pro-gram requires good faculty, so let us be more spe-cific. A review of several programs in "comprehen-sive medicine" reveal faculty who are not engagedin primary practice, either never having done so orhaving ceased to do so. It would certainly seemincongruous if cardiologists or endocrinologiststaught their skills to students and house staff with-out themselves practicing their disciplines. In theoccasional instance where this situation does occur,students are quick to perceive the inconsistency. Itseems to imply that primary medical practice is aless demanding or involved field, which can be ade-quately taught by specialists or nonpractitioners.The impact of this nonverbal communication is notlost on the student. We do not mean to imply that

an administrator, a researcher, or indeed an ex-practitioner has no place in primary care educationprograms. leather, we suggest that a program withfew actual practitioners resembles that descriptionof William Jennings Bryan, when he was likened tothe River Platte: one mile wide at the mouth andone foot deep.

A common assumption in many programs is thatprimary care education requires no special faculty.that subspecialist faculty alone are competent totrain the generalist. This assumes that primary carepractice is equal to the sum of several specialists'practices. The experience of the Rochester two-year internship suggests otherwise. That is, giventhe opportunity in such settings. students will optfor specialty careers. indeed, the current scarcityof primary care practitioners being graduated fromour programs is sufficient evidence that specialtyoriented training will produce specialty orientedpractitioners in a.: pen market setting. What ismore difficult to demonstrate is that pediatriciansand internists who du end up in primary carepractice--the majority- have been shortchanged intheir education and would have been better pre-pared by primary care faculty. Our impression isthat, even if the specialist practitioner is an effec-tive teacher, the disadvantages of inappropriate pa-tiejts. curriculum, and setting within which hefunctions militate against the educational expo -ence being a sufficient one for primary care.

One problem in faculty selection involves theissue of academic rank and promotion. Should pri-mary care teachers be judged on the same basis astheir clinical and basic research colleagues--thequantity and quality of their research, participa-tion in learned societies, teaching responsibilitiesand skills? This is part of the larger issue of therelative merits of teaching versus research that con.cerns most university faculties, but it should notpresent any special or unique problems for primarymedicine.

In general. we see three kinds of faculty involvedin primary care training two of them being part-time appointments and one, full-time. Part-timefaculty are those whose major source of incomeand (tinge benefits derive from the practice ofmedicine. One group has a level of involvementtypical of most part-time faculty: they participatein vine clinical teaching or preceptorships, attendward rounds, and supervise outpatient clinics.Theseactivities have been performed in the past in returnfor staff privileges and, generally, are not salaried.

A second group of part-time faculty consists ofthose who wish to be involved in more extensiveprimary care education and so reserve ,rtion of

their time- on the order of two or three one-halldays weekly for supervision and involvement withstudents at various levels. This group should be se-lected carefully for their teaching skills and be re-imbursed for their time. They should be able tosupervise the student's development of clinicalskills in primary care, which includes interviewtechnique, diagnosis and management of the rangeof problems commonly seen in practice, rapportwith coprofessional, and the techniques of researchin practice. For these faculty, university promotionor fringe benefits properly applied are not a centralissue. They are working part-time at a job thatcomplements and enriches their practice, and theyare remunerated accordingly.

Can funds be obtained for this level of facultywork and from what source? We do not have aready answer at a time when funding medical edu-cation is a complex situation influenced by cate-gorical programs and shifting government priori-ties. If each part-time physician is paid approxi-mately 55,000 airnually for two one-half days perweekin addition to "homework" required. thenthe equivalent of one full-time position can be usedto obtain rive or six committed faculty located invarious settings in the community. The value ofthis group both as role models for the students andas advocates for the needs of primary practicewithin the faculty would be considerable.

The third group of primary care faculty arethose with full-time appointments. They direct theeducational and research efforts as their principalwork, and they practice to the degree necessary tomaintain and develop competence and to achievetheir educational and research goals. This group isclosely identified with the general functions of themedical school and should be promoted and judgedon the same basis as their clinical department col-leagues. These are the faculty who must carry outthe needed research and evaluation in primary care.There is a good supply of faculty in the first twogroups, and there appear to be sufficient numbersof students who are attracted to these roles in pri-mary care education to permit the development ofa competent total faculty over the next severalyears. However, this will require the developmentof special programs to train the faculty, whichshould be a high priority matter for private andpublic funding agencies.

The PatientsAlthough in some ways difficult to separate

from the setting. the patients in a priniary caq.program need to have certain characteristics for theprogram to achieve its aim. For example, variations

55

in age. education. occupation. racial and ethnicbackground. as well as the living environmentrural, urban. or suburban all influence patients'medical care behavior, needs. and demands. In ad-dition, particular disease patterns and their preva-lence within the population need to he considered..There may be reason to oversample some kinds ofpatients for the program on any one of these bases.While any hundred families will provide the stu-.dent abundant experience in the management ofcommon respiratory and gastrointestinal infec-tions, they are less likely to provide experiencewith long-term management of some chronic dis-eases, such as diabetes. It the student's experiencewere otherwise limited to hospitalized patientswith ketoacidosis. he would be unlikely to learnthe primary care role with such patients. In otherwords. there is value in allocating the patient loadto achieve a distribution of cases that may not oth-erwise be achieved. Yet this expedient has gener-ally proven a difficult undertaking in the universitymedical center.

Practically speaking. it is not possible for eachstudent to work with a full spectrum of patient,disease. and setting. In general. it is easier to addpatients with selected disease characteristics, who.ire already concentrated at a university medicalcenter, than it is to provide each student with pa-tients from a range of environmental backgrounds.These cases can represent the oversampling of med-ical conditions or of well children for longitudinalgrowth and development observation that studentsshould be in contact with over 1(.11g periods oftime.

An example of how a program might operate ata graduate level would be as follows: Interns areencouraged to select patients to be cared for overthe next few ye irs from anion'', those seen duringward ur outnatic lit rotations who either have noidentified source of primary care or where arrange-ments satisfactory to patient and primary care pro-vider can be arranged. Some guidelines may be ob-tained from dental education where students haveto complete a quota of certain restrictions and pro-cedures before being considered well-rounded inhis practice. Faculty supervision is required to helphim select appropriate numbers and types of cases.If the internship has a large time commitment toacute block rotations. must trainees will not beable to tininage and learn from a large continuitypanel of patient.. There is great individual variationin the interests and capacities of interns and.hence. value in combining good faculty supervisionwith maximum responsibility by the intern. Essen-

S6

tial are adequate supportive services such as secre-tary answering service, appointment scheduling.nursing, and social work: these can make thedifference between a successful or a frustrating ex-perience for patient and student.

In addition, during the resident's subsequentyears, he works with one of the ongoing primarycare units with which the university has an affilia-tion. The variations in these units depend on thelocation of the medical center. In very large urbancenters, it may be more difficult to provide therange o setting that may be practical in a smallercity, where exurban or rural settings can be ar-ranged that are within 30 minutes driving distance.However, variation in social classparticularly.with neighborhood health center and private grouppractice affiliations-- would certainly be possible inmost cities. Two or three one-half days per weekover a year's time can be adequate for the residentto learn the style of the practice and the needs andhabits of the patient population. In other words.he mav have one single panel of continuity pa-tients, if the primary care setting is located in thehospital as well: or they may be located in twosites. Through periodic formal conferences, simpleresearch projects, and informal ex-periences of all the residents can be shared. If eachprogram has faculty members who themselves arcpracticing, then patients may be returned to theirlull tittle care after the student leaves. In practice.many will be satisfied to have another resident.especially if their right to change physicians isknown to them and respected. In addition. thepresence of allied health professionals-particularly, nurses, lends an important stabilityand ongoing continuity to thy' patient's care.

Finally, patients as "whole people" and as "con-Anners" are more of an influential factor in pri-mary medicine than is true for secondary or terti-ary care. Like it or not. the complexities of thedisease and the technology of care occupy more ofthe time and energy- of the consultant than they dothat of the primary care doctor. Primary care edu-cation should allow the trainee to shape his owndefinition of how a physician relates to the com-munity or. at least, to begin to think along theselines. This is far better done by eqerience than bylecture. For example, trio student ought to see andwork with a program's consumer group (Does ithave one?) or have the chance to become involvedin school health programs, health education, or so-cial action efforts. Students ought to have the op-portunity of working with patients or communitygroups during various phases of a health program's

development its inception and planning, the iden-tification of new service needs, ongoing health edu-cation. The essential ingredient here is that thestudent is involved at a time of experimentation orflux so that he comes to know the processes ofchange, conflict, and planning as a yarticipant(Dixon. 19651. These are the most difficult kindsof educational experiences to program; and inte-grating them into a curriculum while respecting theneeds of patient, student, faculty, and universitycan be a trying experience. Avoiding such conflictaltogether has its price as well in the productionand endorsement of the "uninvolved" physician.

Although concern has been expressed about theacceptability of medical students by patients in theprimary care sector, our experience has shownthat, with tact and honesty. patients of all eco-nomic classes accept the physician-in-training ifthey are assured that he is adequately supervisedand if their right to change physicians is respected.

In summary. primary education programs ideallyshould introduce the student to a variety of pa-tients in a variety of roles. The program shoulditself direct or be affiliated vith primary care units.with organizational and patient diversity, at differ-ent levels of development. It should function as alaboratory with case material for primary carestudy as patients with different kinds of heart

.

lesions make up the caseload of a cardiologytrainee. If primary care education is indeed the ma-jor mission of the department, then this approachis a natural one; i.e.. core training in primary carewith specialty experience selectively added and notthe reverse.

Curriculum TimeThere must be adequate time devoted to pri-

mary care education. but perhaps more importantis that this tinie be arranged appropriately withinthe larger curriculum for both medical student andgraduate: It is essential to match the "natural his-tory" of the clinical problem to be studied withthe student's time allotment. By natural history.we mean the time it takes for key elements of theproblem to become detectable or symptomatic.evolve through critical phases. and either stabilizeor be resolved in some fashion. The student needsto experience these critical phases himself. So, forexample. the natural history of an episode of pneu-monia ur otitis media lasts a few days or a fewweeks in most cases. If students only see such casesfor a few minutes in an emergency room setting.they may miss the fact that not all cases are re-solved in the same fashion and that patients theythemselves have carefully instructed in druwtaking

and symptomatic care frequently ignore all suchadvice and break return appointments, Usually thiscontinuity of care can be arrangril by allowing thisstudent to see his own patients in followup and byhaving an assignment that lasts on the order of amonth. Similarly. inasmuch as the average acutehospital stay is of approximately one week's dura-tion, rotations of a month or two on an inpatientservice usually provide the student with a goodgrasp of the course and the crises of most acutehospitalizations.

However, many important clinical content areasin primary care take a good deal longer to maketheir natural history evident, and failure to takeaccount of this can lead both to inadequate educa-tion and inadequate care. For example. Brook(1971) evaluated the followup care of 403 patientsdischarged from the Baltimore City Hospital. De-spite adequate inpatient care by university housestaff, one-third of the patients had poor subse-quent medical care, even with the use of minimalcriteria of evaluation. The fact that members of thehouse staff do not often learn what happens tochronic disease patients after discharge can lead toa distorted perspective in the trainee as well as toinadequate medical practice. Another example ofinadequate experience with natural history leadingto inappropriate practice can be seen in the advicegiven to new mothers by hospital nursery person-nel. Although most nurses are quite competent atidentifying and caring for the sick neonate, theirsuggestions to mothers at discharge about suchcommon problems as breast feeding often sufferfrom lack of further contact with the family overthe first few months of life. Similarly, one mightspeculate that liberal visiting hours for hospitalizedchildren took so long to gain acceptance. becausehospital staff in large measure were unaware of thereaction to hospitalization that is displayed formonths afterward by sonic young children.

The following table suggests how a number ofclinical topics in primary care can be divided intoshort one day to one month, intermediate -twoweek s to three months, and long-term twomonths to several years categories, based on theirnatural history. The list is meant to illustrate theconcept rather than be exhaustive. These categoriesoverlap in time, as indicated; and their separationis. to an extent. arbitrary.

Medical education cannot provide the studentwith experience in every problem he will face inpractice; but it does need to convey to him a senseof how a range of problems arises, evolves, and isresolved. so that he does not assume an opportun-istic, short-term view. Moreover, these content

57

Table 1.

Short-Term(I day to 1 month)

'Natural History'Of Primary Care Content Areas

Intermediate LongTerm(2 weeks to 3 months) (2 months to years)

1) Most medical and surgicalemergencies

2) Common Infectionsa) pharyngitisb) otitis mediac) gastroenteritisd) upper- and lower-

respiratory infections

3) Average acute hospitalization4) Minor surgical trauma

5) Relationship with the patientwhich asks why he conies andwhat needs the professionalmust meet

1) Acute or presenting phase ofsome chronic disease - /

) congenital abnormalitiesb) diabetesc ) asthmad) leukemia

2) Certain behavioral disordersa) child rearing conflictsb) school adjustment

problemsc) sonic marital conflicts

3) Recurrent abdominal pain

4) Cardiovascular disorder (acutephase) infarction, hypertension.congestive heart failure

5) Observation of the "milieuof practice." the life-style ofthe practitioner

6) Learning to work on a hierar-chically organized team

7) Observation of the "Milieu"of patient care research. Thetechniques of researchprocedures.

1) The family (patient) as thefocus, the disease as theepisode. Sociology of thefamily

2) Most chronic diseasesa) asthmab) cerebral palsye) mental retardationd) Psychosis and neurosise) diabetes

3) Growth and developmentof children

4) Working as a coprofessionalmain member

5) Design and implementationof a patient care researchproject

6) A working relation repeatedbetween professional andconsumer

/"Natural History:" The time it takes for the problem to become detectable or for symptoms to evolve through crit-ical phases and either stabilize or be resolved.

/Although this does not define the condition's entire "natural history" it does indicate the duration it usually takesfor the condition to be diagnosed and initial management pattern established.

areas cannot all be experienced in one year. Somerequire more clinical maturity and readiness in thestudent if they are to have their maximum impact.

In general, programs for both medical schooland house staff training have emphasized block,short-term experiences at the expense of longitudi-nal ones. We would point out that this has beendetrimental to subspecialist as well as to primarycare education, insofar as since the management ofone chronic disease patient over time, for example,is not the educational equivalent of managing sev-eral such patients through acute crisis episodr s.Both experiences have educational value.

Developing a curriculum with this concept inmind poses a number of practical problems. Whomanages the acute intensive care patients when thetrainee leaves the ward to see his long-term pa-tients? Who sees the long-term patient when thetrainee is detained by a crisis on the ward? Onesolution is to work in pairs or teams as the students

58

did in practice; Another is to assign nurse practi-tioners to the trainees in the same way dental as-sistants are assigned to dental students in theirtraining. The same team functions together forthree to six months. How is continuity meshedwith rotation through other services or other hospi-tals? Equally difficult for students, particularly atearlier stages of development, is the problem ofcoping with the change of pace required in movingfrom acute care, when the basic need is to extractinformation quickly, to the management of long-term problems where a different interview mannerand relationship with the patient is required. Thetrainee may find it difficult to shift his mentalgears without grinding his teeth.

During medical school and residency, the stu-dent has not beer shown that these different situa-tions may require different or more flexible tech-niques of patient workup. Indeed, except for thework of Weed (1969), medicine has been slow todevelop such tools itself. The student soon learnsthat his all-purpose, complete "New Patient

Workup" rarely fits the clinical situation. More-over, there is insufficient guidance to help him de-vise a suitable model for the more common brief-but-long-term contact he will have with patients.

Our suggestion is to construct a curriculum byfirst determining the actual content material to belearned; next, setting priorities within those con-tent areas; third, deciding how intensive and exten-sive the learning experience must be to match thesecontent areas; and fourth, specifying that stage ofstudent maturation when the material is most ap-propriate. Finally, the curriculum should be shapedto meet all these needs. It is far less rational to firstdecide what service commitments exist and thenassign students to fit those needs, as occurs inhouse staff programs. Undergraduates should beginmedical school with exposure to and contact withpatients whose needs arc in the primary cam area.Residency should include initial involvement in aprimary care setting as a beginning for a longitudi-

nal experience.

The SettingNi The setting of the primary care educationprogram -- specifically its size, organization, and re-lationship to secondary and tertiary care systemsand to the patient population -is the last of ouringredients and it crucial one for the success of theprogram. We refer here to the "style" of the settingas well as to its formal organizational and physicalaspects. Specifically. what kinds of problems areconsidered important or trivial by the staff? Howwell do physicians and allied professionals commu-nicate with each other? How isolated or integratedis the program from the problems of the commu-nity?

For example, evidence was cited earlier aboutimpact of work setting on performance of those inpractice. No less significant is this influence in edu-cational programs. Using National Board Examina-tion scores as criteria, Levit (1963) showed thatinterns in hospital program with a full comple-ment of house staff demonstrate greater gains inclinical competence after one year than those inhospitals that do not fill internship positions, re-gardless of the intern's competence on entry to the

program.In a discussion of educational programs for pri-

mary care, Hansen (1970) notes. When primarycare responsibilities or functions compete withconsultant or tertiary care responsibilities, the pri-

mary care functions are consistently underrated by

both teacher and student." Although this may re-flect qualities of the teacher and student, we be-lieve that the observation holds true largely due to

the influence of the setting. If the style, pace, loca-tion, and organization is not basically cwicernedwith primary care. and it is seen only as an un-wanted but necessary chore, then teacher andstudent will not be concerned either. It is its im-

practical to demonstrate primary care practicewithin most university hospital settings as it is toteach techniques of gall bladder surgery in a neigh-borhood health center. One of the reasons for thesuccess of the Kansas rural preceptor program(Dimond, 1954: Rising, 1962) was the chance forthe "student to participate almost totally in a'medical way of life' and identify with the precep-tor in many ways." (White, 1964). This is a clearindicator of the impact of the setting on education.

.Learning how to diagnose and manage psycho-logical problems has been a particularly vexing anddifficult problem in the education of the generalphysician. A major factor in this difficulty hasbeen the inappropriateness of the setting of theprogram. Although exemplary techniques of inter-viewing patients in the hospital (Engel, 1971) ur ofsocially oriented ward rounds (Bates, 1965) havebeen described. their success seems in large part tostem from the enthus'am or skill of an instructor.When this is missing, the programs are less success-ful; and when the original staff changes. the tech-nique is abandoned altogether.

Ward attending rounds, for example, often omitdiscussion of the social or psychological aspects ofthe case and, in fact, often ignore the patient alto-gether. 'Give us the lab results and we will du thejob" conveys the spirit. Payson (1965) observedthat regular attending physicians spent less than afifth of their time with the patient during rounds;and most of that one-fifth was spent dealing withphysical factors. He concluded that there was "lessemphasis on bedside demonstration of individualor personal aspects of medical care than most at-tending physicians realized. Rounds appeared toshow how senior physicians arrive at decisions andrelate case findings to medical theory. They didnot emphasize the physician's approach to the pa-tient and the establishment of the doctor-patientrelationship." As one attending physician statedwith candor. "I never discuss what I feel uncertainabout. I try to limit my comments to the aspectsof scientific medicine that I feel expert in."(Payson. 1965.) And at times, inappropriate deci-sions are made by physicians due to missingpsychological information (Duff and Hollingshead,1968).

Our reaction is not to point with horror at suchincidents, but to admit the basic validity of theseobservations. If we wish to teach about social and

59

psychological factors. this is probably best donewhere the setting. among other factors, is appropri-ate. In the instance of inpatient rounds. the curric-ulum time may not permit the student to L4 nraVelthe psychological factors that have led to the hos-pitali/ation nor to the sequelae after discharge. Thesetting maximi/es acute organic medicine, and thestudent responds accordingly. In a sense. for hireto ((Well on social and psychological factors may beunproductive. He would need the time and facil-ities to follow all his patients, :1 practical impossi-bility when the next case of cardiac failure ormeningitis is arriving from the emergency depart-ment. The student is most likely to learn how toelicit, to appreciate. and to utilize social andpsychological data when the curriculum permitslong-term contact with and responsibility for somepatients and when the setting prompts him to consitter such problems as pertinent. However, die_cur-riculum, including attending rounds, is not so fixedthat the hUntall aspects of care cannot be included.

Our recommendation, then, is that universityschools of medicine become involved primaryCare settings and that they conduct their educa-tional and research business in settings either pur-posely built or within existing practices adapted tomeet educational needs. We use the term "in-volved- advisedly. :)btaining a balance betweenjust enough involvement to inure that the experi-ence is educationally valuable, but not so muchthat the service burdens are overwhelming is easiersaid than done. We have been critical in the past ofsome earlier "comprehensive clinic- programs.largely because they arc unrepresentative of pri-mary care. However, investing all the effort 01 aschool into a single large health center practice inorder to achieve "reality- has its own drawbacks.It will demonstrate only one kind of practice orga-nization with one kind of patient population. andit will assume a service burden not easily or ethi-Lally terminated at a later date if the situationshould chang,e. In short, one large "model pro-gram- may absorb faculty time and allegiance todegree that may limit flexibility and prechide the!hinge for continuing experimentation and "tin-

kering- that should characterise a laboratory set-ting. As another alternative, new programs mightconsider the following approach as another.

The university sponsors a teaching practice of nomore than several hundred families large enoughto have an air of reality and small enough to ensurethat all of the practitioners ;other than the saldents, can be full-time faculty. The allied healthprofessional staff are chosen for their teaching as

60

well as their practice skills. The practice is housedin or very close to the main teaching hospital inorder to facilitate integration with inpatient andsubspecialty education. However, it is sufficientlyindependent of the hospital so that professionalroles, record systems, patient intake .,rocedures,and other matters can be changed without conflictwith existing hospital policies. Within the practiceare the overrepresented- patients suggested ear-lier. This kind of program sacrifices the reality ofpractice to a degree because heavy educational pri-orities i.e.. supervised interviewing and consulta-tion with faculty -preclude concentrating on high-volume patient flow and efficiency to the -degreerequired in practice. In general, this is the modelmost common in the new family medicine pro-grams, and it contains many elements of the earlierFamily Care Programs.

Complementing the above program are relation-ships, developed slowly and selectively over severalyears with existing or new group practices; healthcenters, or solo practitioners. The university "con-tracts.- as described earlier, for the teaching timeof some part-time faculty within these practices.Responsibility for service does not depend here onthe students to the degree that it does in the hospi-tal program. Programs are selected for affiliationnot only because they exemplify quality care, butalso because they offer diversity in setting or clien-tele. This would represent a contemporary applica-tion of the preceptorship.

During elective portions of the curriculum, med-ical students and house staff work with consumeror practice units at varying stages of program devel-opment. Commitment for service or involvement islimited to that student group's tenure for the mostpart. In some instances, more permanent affilia-tions characteristic of the first two groups of pro-grams may evolve.

We have cautioned against the problem of exces-sive service obligations incurred in the hospital as aresult of the need to find financial support for dieresident, and we would not like to see primary careprograms end up with the same conflict. Our pointis that resident staff are capable of providing high-quality primary care service under supervision inthe same way they provide service in the secondaryand tertiary care settings. I income from thesesources is available, it should be utilized. Ideally,direct funds for education are needed as well, toavoid sole dependence on this one source fo_

income.A significant Itandicap in the development of

new primary care programs is the obligation tomeet service needs that already exist in the hospital

inpatient and outpatient department. Most pro-grams will have to contend with this reality in addi-tion to fashioning new models, Existing hospitalorganization and the traditional clink system con-stitute formidable barriers to change,

We have suggested that judicious use of alliedhealth manpower, and greater selectivity in the in-volvement of trainees in specialty services are re-quired. However, the hospital ambulatory servicespose a special problem. The number of patient vis-its has increased rapidly in most urban locales overthe past decade. In addition, the responsibility forclinic management has usually fallen to those onthe faculty who are most closely allied with pri-mary care education, absorbing all their teachingand administrative energies in less than ideal set-tings. In fact, hospital outpatient and emergencydepartments increasingly provide the first contactportion of primary care to the community as thesupply of general practitioners dwindles, and train-ees are involved in a large share of that work. Avicious cycle ensues. Fewer generalists in practicemean more people using hospital ambulatory serv-ice, In response to this demand from the commu-nity, hospitals modernize and expand their facili-ties and, thereby, attract even more patients, Whathas evolved is an ad hoc pattern of medical care,facilitated in part by the availability of hospital-based trainee manpower. We consider this an inap-propriate, short-term response to a long-term need.Indeed, insofar as it endorses short-order emer-gency room care as the primary practice model, itmay have serious long-term consequences. Al-though this is one solution to the primary careproblem, it is not the only one. By investing ourtrainee manpower in the ,neratiGn of this model,we limit our option to support and develop others.Equally important, the student confronted with anunsatisfying model of care will be convinced thatprimary care or what he sees of it is the last thinghe wishes to practice.

What can be recommended to resolve the con-flict between new program needs and old programdemands? First of all, we would emphasize that theproblem is not likely to be solved unless we edu-cate more primary care practitioners, a goal towhich this monograph is devoted. In the meantime.two approaches arc suggested:

Where educational programs are already respon-sible for significant ambulatory care service, thecaseload should he analyzed into its component.parts. These usually involve some combination ofemergency medical and surgical services, short-termconsultation, long-term management of, chronicdisease, and primary care. These separate functions

each lend themselves to different organizationalstructures and staffing patterns, For example,emergency care needs rapid patient intake andprocessing facilities, specialized nurse and techni-cian manpower, easy access to surgical and medicalspecialty consultation, and relatively expensiveequipment.

Short-term consultation service requires a goodprior sorting system, so that efficient use is madeof subspecialist time. The unit stresses good work-ing Low ledge of an integration with communityhealth resources, particularly for primary care andchronic disease, so that effective and pragmatic rec-ommendations for followup care can be made. Ac-cess by the patient is indirect, through primarycare resources in the community. The pace of theunit is slower and more capable of regulation.

Long-term management of selected chronic dis-ease involves active participation of the patient inplanning his care. Efficiency and speed in patientflow arc less vital than a staff that is sensitive tothe support-and-caring aspects of medicine.

Primary care embodies aspects of all of these.but it especially stresses easy access for the patientand a staff that has the capability and skills ofoutreach and followup in the community, ratherthan the highly technical skill required for emer-gency care or the indepth knowledge of certaindiseases required in chronic illness management. Ifthe entire outpatient service is not large, several ofthese functions can, of course, be combined suc-cessfully by one well-trained and flexible staff. Stu-dent physicians who staff several of these servicessimultaneously often have a difficult time shiftingroles, especially if the support staff structure is notdesigned to meet the needs of the service required.

In short, a good deal more than the "diagnosis-prescription" function of the physician is requiredto carry out these several tasks successfully, andsome reorganization of the services based on pa-tient need would improve the quality of serviceand the efficiency of the staff.

With the other ambulatory functions separatedout, the needs of those using the service for pri-mary care should then be defined accurately. Whoare the patients? Where do they come from? Arethere actually several different populations usingthe service for somewhat different purposes-suburbanites for occasional care when their doctoris unavailable or others for all their health needs?What other resources are available? Finally, shouldan attempt be made within the hospital setting toprovide care in a setting more suitable for serviceand education? This last point requires some diffi-cult decisions. A proportion of emergency room

users may not be able to tolerate a long-term, inti-mate relationship with a health program, which iswhy they use the emergency facilities in the firstplace. They may resist being incorporated into a"model practice." Although this is an interestinggroup to study and to learn more about, they arefrustrating to physicians in training and to experi-enced physicians as well. Having a mature group offull-time faculty and allied professionals share theircare can enable students to live with and learnfrom the rejection they encounter at times. Ourrecommendation here is to aim for some kind ofpatient diversity if at all possible. Multiproblem,disorganized families should be part of any teach-ing program, because they are a part of the realityof practice or should be. They should not, how-ever, be the only group involved in the teachingprogram. Until adequate primary care resources areavailable in the community, some programs mayhave to live with two standards of care provided: amore complete service to a selected group in a lon-gitudinal teaching program and first-contact serviceon an episodic basis to others.

As a more adequate, long-term solution to theproblem, medical education personnel should par-ticipate with area-wide health planning units to en-courage and stimulate the development of ade-quate primary care education and service programsoutside of the hospital. Planning agencies should beeducated to the need for allied health manpowertraining tied into stable primary care settings. Forexample, the integration of public health resourcesnow involved in some aspects of primary care suchas well-child conferences and visiting nurse serviceswith existing or planned primary care practicesmay stretch the resources and capabilities of both.In Great Britain, the attachment of health visitorsand district nurses (public health nurses) to generalpractice groups has been accomplished in morethan one-half of physician practices with evidenceof benefit to patients and providers (Amblers,1965; McGregor, 1969).

While the problem of ambulatory service de-mands in hospitals is a growing one, we wouldhope that a combination of more rational, commu-nity-wide planning for primary care needs of thetotal population, adequate funding through a na-tional health insurance scheme, and an increasedoutput of primary care personnel in more effectiveorganizations will be sufficient to reverse the cur-rent trend toward inappropriate use of facilities.

ConclusionIf a rapid increase in the number of primary care

practitioners is the paramount objective of the62

Federal Government, we acknowledge that influ-ences outside the medical education system may beall that is required to effect such a change. A shiftin terms of financial support to education directedspecifically to this endnamely, incentives for theproduction of primary care physicians, changes inthe medical practice system, and some limitationon the availability of subspecialty careers wouldprobably have the desired effect. Our strong prefer-ence, however, is that not only more practitionershe prepared, but also that they be better educatedfor their practice. To accomplish this aim, changeswithin medical education are needed as well.

At the national level, professional societiesshould enter into discussion with family medicinerepresentatives on their relative roles and obliga-tions in primary care. The implications of any deci-sions for manpower recruitment and educationmust be spelled out.

The university should acknowledge its role incoordinating primary care education at the medicalschool and at graduate and continuing educationlevels. Such coordination should be accomplishedin conjunction with those representing the publicand the practicing professions.

The medical school must set as a priority thedevelopment of criteria for selecting students whowill be suitable candidates for careers in primarymedicine practice.

Within each medical school, a department or di-vision responsible for primary rare educationshould be identified or developed. This departmentshould have the responsibility to develop an overallprogram for primary care education at the under-graduate and graduate levels. It has a particular re-sponsibility to devise continuing education pro-grams that will link the practitioner and the educa-tional unit. Extended leave educational pr.-)gramsfor practitioners, a collaborative research effortwith physicians in practice,' and part-time facultyroles should be particularly encouraged. Researchand evaluation must, bean important activity ofthe academic medical centef in primary care.

The obligations of medical school specialty de-partments to primary care education must be fur-ther defined. These departments in the past havecorrectly considered the development of their owndisciplines as a priority. However, at present, thereis an important gap between specialty medicine'sbody of knowledge and technique and the applica-tion of this knowledge at the primary care level.The fact that most physicians in training will endup in primary care practiceshort of a major revo-lution in the way medicine is practicedunder-scores the importance of utilizing appropriate as-pects of all of medicine to primary care practice.

The university should develop sites for primarycare education under its own auspices. These willrequire a variety of contractual relations, rangingfrom ownership and direction to short-term andloose affiliations for educational purposes only. Inselecting programs to establish or to affiliate with,a diversity of patient population and practiceorganizations should be sought. These settings to-gether should be considered a part of the universitymedical effort. They should be closely integratedwith the hospital, but continue to remain inde-pendent.

In short, all components of the medical educa-tion process, which includes the medical school, itsparent university, and the teaching hospital, haveimportant work to do. At present, there is a gooddeal of uncertainty as to how we ought to providehigh-quality medical care to our Nation. On theother hand, the uncertainty provides a climate: thatfavors change and in which the education of theprimary physician can be reshaped and improveddramatically.

References

SECTION VI

Ambler, M., Anderson, J., Black, M., Draper, P., Lewis,J., Moss, W., Murrell, T.: Attachment of Local Health Au-thority Staff to General Practices: Medical Officer 295,May 1965.

Bates, B.: Comprehensive Medicine: A Conference Ap-proach J.Med.Ed. 40: 778, 1965.

Brook, R.H.. Appel, F.A., Avery, C. Orman, N..Stevenson, R.L., Effectiveness of Inpatient Follow-up Care.New Eng. J.Med. 285: 1509, 1971.

Carmichael, L., Shore. W., Frey, J.: Teaching Ambula-tory Pediatrics in Family Health Care Programs: WorkshopPresented to Ambulatory Pediatric Association, 12th An-nual Meeting. May 23. 1972.

Charney. E. and Kitzman, H.: The Child Health Nurse(Pediat:ric Nurse Practitioner) in Private Practice: A Con-trolled Trial, New Eng. J.Mcd. 285: 1353, 1971.

Coker. R.E.. Greenberg, B.G. and Kosa, J.: Authoritari-anism and Machiavellianism Among Medical Students.J.Med.Ed. 40: 1075, 1965.

Dimond, E.: The General Practitioner and the MedicalSchool. J.A.M.A. 156: 95. 1954.

Dixon, J.P.: Teaching Physicians to be Agents of SocialChange. Arch. Envir. Health 10: 713, 1965.

Duff. R.S. and Hollingshcad, A.B.:New York. Sicknessand Society, Harper and Row, 1968.

Engel. G.: Care and Feeding of the Medical Student.J.A.M.A. 215: 1135, 1971.

Haggerty. R.J.: The University and Primary MedicalCare. New Eng. Mcd. 281: 416, 1969.

Hansen, M.F.: An Educational Program for PrimaryCare. J.Med.Ed. 45: 1001. 1970.

Hansen, M.F. and Reeb. K.G.: The Outline of a Curricu-lum. J.Mcd.Ed. 45: 1007, 1970.

Jefferys, M.: Factors Determining the Choice of Medi-cine as a Career. Presented at WHO Conference on Selectionof Students for Medical Education. Bern, Switzerland,1971.

Levit, E. J., Schumacher. C.F., Hubbard, J.P.: The Effectof Characteristics of Hospitals in Relation to the Caliber ofIntents Obtained and the Competence of Interns After 1Year of Training. J.Med.Ed. 28: 909-19, 1963.

Lyden, F., Geiger, H.J., Peterson, O.L. : The Training ofGood Physicians. Harvard Univ. Press. Cambridge, Mass..1968.

McGregor. A.: Total Attachment of Community Nursesto General Practices. Brit. Med.J. 2: 291, 1969.

Mechanic. D.: General Practice in England and Wales:Results From a Survey of a National Sample of GeneralPractitioners. Medical Care 6: 245, 1968.

Millis, J.S.: Objectives of Residency Training. CanadianMed. Assoc. J. 100: 599, 1969.

Monk. M.A. and Terris, M.: Factors in Student Choiceof General or Specialty Practice, New Eng. J.Med. 255:1135, 1956.

Mumford, E.: Interns: From Students to Physicians.Commonwealth Fund, Harvard Univ. Press. Cambridge.Mass.. 1970.

Payson, H.E. and Barchas, J.D.: A Time Study of Medi-, ' Teaching Rounds. New Eng.J.Med 273: 255 1468,

5.

ellegrino, E.: Research in Medical education: Views oferiendly Phillistine. J.Med.Ed. 46: 750, 1971.Pellegrino, E. et al: Planning for Comprehensive and

Continuing Care of Patients Through Education. j.Med.Ed.43: p. 751, 1968.

Rising, J.D.: The Rural Preceptorship. J. Kansas Med.Soc. 68: 81-84, 1962.

Romano, I.: Study of a Two Year Rotating Internship.Univ. of Rochester Medical Center, 1949-61, JAMA 189,283-89, 1964.

Rutstein. D.: Physicians for Americans: Two MedicalCurricula. J.Med.Ed. 36: Part 2, p. 129, 1961.

Sanazaro, P.: Research in Medical Education: Explora-tory Analysis of a Black Box. Ann.. N.Y. Acad. Sci. 128:519, 1965.

Sumpter, E.A. and Friedman, S.B.: Workshop Dealingwith Emotional Problems: One Method of Preventing the"Dissatisfied Pediatrician Syndrome". Clin. Pediat. 7: 149,1968.

Szasz. T.S. and Hollender. M.H.: A Contribution to thePhilosophy of Medicine. Arch. Int. Med. 97: 585, 1956.

Weed, L.: Medical Records, Medical Education and Pa-tient Carc. Cleveland, The Press of Case Western Reserve,1969.

White, K.L.: General Practice in the U.S. J.Med.Ed. 39:333, 1964.

63

U.S. DEPARTMENT OFHEALTH, EDUCATION, AND WELFARE

PUBLIC HEALTH SERVICEHEALTH RESOURCES ADMINISTRATION

ROCKVILLE, MARYLAND

OFFICIAL BUSINESSPENALTY FOR PRIVATE USE, $300

DHEW Pub. No. (H RA) 74-3113

POSTAGE AND FE ES PAIDU. S. DEPARTMENT OF HEW

HEW-396

Documents

DOCUMENT RESUME ED 095 779 AUTHOR Alpert, Joel J; Charney ... · DOCUMENT RESUME ED 095 779 HE 005 852 AUTHOR Alpert, Joel J; Charney, Evan TITLE The Education of Physicians for Primary