‘ - , THE TEACHNG’OFEJDULT Wffgiif'j :xj-jij.;‘j;.j_ AMBULATDRY PATEEN CARE . 1- f " MU. S. MEDtCAL SCHOOLS: _. _ i;_: 4. ‘ CHARACTER SECS OF. PROGRAMS , f v Dissertationforthe Degree'ofPh.D. _ . y - . ~ MICHIGAN STATE UNIVERSITY, . _ . T - ' . ‘ "DONALD MERLEGRAGG . / .nnwr '.".I.,. 0 “My”. ‘, . , tum f Ky».— .. ,..‘.."”'r.r...z ., , .M...I.,-r _,..,.. . .v- , ’ a W\-,.‘Iv¢r’t>l -..,r_-1171"’l 4r.- .r. 03w}— __,.., ‘ .1...” ,t, '..,.. ,...,-..¢ ‘ «i LIBRA R 1/ Michigan State 7 University a This is to certify‘that the ‘3’ , s ' thesis I" THE TEACHING OF ADULT AMBULATORY PATIENT h CARE IN U . S . MEDICAL ‘ SCHOOLS: CHARACTERISTICS OF PROGRAMS presented by Donald Merle Gragg has been accepted towards fulfillment of the requirements for Ph.D. ‘ degree in Administration and Higher Education Major profe MAM27/77; ., / J / .lfl-s 0-7639 1 @4933 © 1974 DONALD MERLE GRAGG ALL RIGHTS RESERVED ABSTRACT THE TEACHING OF ADULT AMBULATORY PATIENT CARE IN U.S. MEDICAL SCHOOLS: CHARACTERISTICS OF PROGRAMS BY Donald Merle Gragg Approximately half of the student's time in medical school is devoted to clinical instruction, learning through supervised experiences with patients. The current rapid increase in the number of medical students without an accompanying increase in the number of university hospital beds suggests that new or modified approaches to clinical instruction are going to be needed. One way that this impending shortage of hospitalized patients might be alleviated would be to increase the utilization of ambula— tory patients for clinical instruction. This seems par— ticularly appropriate since the vast majority of clinical instruction in 0.8. medical schools involves experiences with hospitalized patients, but the greater proportion of medical practice involves the care of ambulatory patients. As a first step in stimulating research in this area, a descriptive and eXploratory study of the teaching of adult ambulatory patient care in 0.8. medical schools Donald Merle Gragg was conducted. The information produced is expected to be valuable to medical schools in planning and conducting ambulatory care teaching programs, and to researchers by suggesting tOpics for further study. The research was intended to produce information with which to answer the following questions: 1. What are the major problems which faculty and students perceive in current adult ambulatory care teaching? 2. Which characteristics of current adult ambulatory care teaching programs are associated with the success or failure of the program? 3. Does the degree of success of programs vary with differing program goals? 4. Are preceptorship programs more successful than clinic programs for some types of goals? A model was proposed for use in the study of clinical instruction. Patient, Student, and Faculty Variables are largely beyond the control of program designers. Program Goals are determined by the curricu— lum. Process Variables, i.e. operations variables, are controlled by the program director within the constraints imposed by the previously listed groups of variables. The Outcome Variables are dependent upon Program Goals and Patient, Student, Faculty, and Process Variables. Donald Merle Gragg The available literature on the teaching of ambulatory patient care provided a number of case studies, but proved to be inadequate to guide the design of the present study. Methodology A three-phase methodology was used in which the results of the earlier phases were used to assist in the design of the later phases. The first phase of the study consisted of a series of problem identification sessions with medical school faculty and students. The next phase of the study involved interviews at three medical schools with administrators, faculty, nurses, and students repre- senting ten adult ambulatory care teaching programs. This phase resulted in a series of case studies and assisted in the identification of appropriate variables and hypotheses for use in the questionnaire phase. The first component of the questionnaire phase was a survey of the ninety—six established United States medical schools which identified sixty—two clinic programs and twenty-six preceptorship programs appropriate for further study. These programs were selected because they dealt with medical students working with adult ambulatory patients in the fields of general medicine or family :medicine. Questionnaires were designed for completion by faculty supervisors of clinic programs, clinic nurses, Donald Merle Gragg participating students in selected programs, and pre— ceptorship coordinators. Responses were coded and consolidated for computer analysis. Each program was assigned to a series of groups based on (a) faculty and (b) student ratings of the program's success, and (c) on faculty indication of the program's goals. The data analysis was of three types: the description of all programs and subsets of programs on relevant variables; the determination of the degree of association between variables; and the testing of seven— teen hypotheses generated during the first two phases of the study. Problem Identification Phase A problem identification process to identify areas of difficulty in adult ambulatory care teaching was carried out primarily to assist in the design of Phases Two and Three of this study. One faculty group and one student group at each of two medical schools participated in the problem identification process. One hundred sixteen problem statements were produced. Problems regarding concern for the continuity of patient care and continuity of the student—patient relationship received the highest priorities. Problems included among the Process Variable group of the model presented received the most attention from both students and faculty. Students demonstrated Donald Merle Gragg more concern than faculty with the Patient Variables, while faculty showed more concern with the Student Variables. Case Study Phase Descriptions of nine clinic programs and one pre— ceptorship program at three medical schools were presented. All of the clinic programs used a similar method of oper— ation. An ambulatory patient's medical history, physical examination, and perhaps laboratory studies were performed by a student. The student conferred with a faculty member and the patient's diagnosis and recommended management were discussed and decided. The patient was then advised of the findings and recommendations. In one clinic program, in addition to the process described above, a significant proportion of the student's time was devoted to the study of the pathophysiology of selected disease processes. All of the programs studied demonstrated the fact that good ambulatory care teaching is eXpensive. Questionnaire Survey Phase Adequate data for analysis were obtained on fifty of the sixty-two clinic programs surveyed and twenty-three of the twenty—six preceptorship programs. Questionnaire results were analyzed and a profile of clinic programs was presented characterizing the pro- grams on each of the variables studied. A comparison of Donald Merle Gragg clinic and preceptorship programs revealed that: (a) pre— ceptorships were more commonly elective experiences; (b) they had a higher priority on two goals, student involvement in a health care delivery system other than a university medical center and experience to assist the student in making career choice decisions; (c) they had a lower priority on goals relating to knowledge of the pathophysiology of diseases and the natural history of chronic diseases; and (d) there was no difference in the faculty ratings of success of the two types of programs. A comparison of the programs with and without student questionnaire data revealed that those programs selected for study with student questionnaires (a) had a smaller percentage of full—time faculty teaching in the clinic, (b) involved more third-year students, (c) were larger, (d) were more commonly required, and (e) were rated lower by faculty on the educational value of the programs. There was no meaningful difference between the faculty and student responses to questions on the descrip— tive, or independent, variables. On the outcome variables, however, there was no significant relationship between faculty and student responses. Students' ratings of program strengths and weaknesses confirmed the earlier :finding that clinic and/or curriculum organization was a rnajor problem area. Donald Merle Gragg The relationships between variables were studied (a) by comparing the characteristics of groups of programs with different goals and different success ratings, (b) by correlations between variables, and (c) by multiple regression analyses using the success indices as depen- dent variables. A number of program characteristics were found to be significantly related to program goals. The ratings of the willingness of faculty to participate in the programs bore the strongest relationship to the faculty ratings of program success, accounting for over 20 percent of the variance of the Faculty Success Index for all clinic programs and programs without student questionnaire data. Student ratings of success were positively correlated with high ranking of the goal, to demonstrate the pathophysiology of disease at a given point in time, and negatively correlated with the number of students in the program. Three of the seventeen research hypotheses were accepted: (1) Students planning careers in the primary care specialties rated clinic programs as being of greater educational value than did other students. The Faculty Success Index was positively correlated with both (2) faculty willingness to participate in the program and (3) the percentage of participating faculty who were full- time faculty. THE TEACHING OF ADULT AMBULATORY PATIENT CARE IN U.S. MEDICAL SCHOOLS: CHARACTERISTICS OF PROGRAMS BY Donald Merle Gragg A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Administration and Higher Education 1973 Q\ U 6% Copyright by DONALD MERLE GRAGG 1973 DED ICAT I ON TO My Wife, Charlotte, and our three children, Kathleen, Brian, and Gary. They gave the encouragement and support needed to complete this study. ii ACKNOWLEDGMENTS I wish to express my sincere appreciation to my Committee Chairman, Walter F. Johnson, Ph.D., and my Dissertation Co—Director, Jack L. Maatsch, Ph.D., who gave excellent advice and were always available. My other committee members, Richard L. Featherstone, Ph.D., Hilliard Jason, M.D., Ed.D., and Ronald W. Richards, Ph.D., also provided much assistance. The hundreds of medical school faculty, staff, and students, who contributed to the study, deserve much credit for its success. L. Thompson Bowles, M.D., Ph.D., of the Association of American Medical Colleges, and C. H. William Ruhe, M.D. of the American Medical Associ— ation provided advice and assistance in the design of the study. The faculty, fellows, and staff of the Office of Medical Education Research and Development, Michigan State University, were extremely helpful. Susan Thrash of the Office of Research Consultation, College of Edu— cation, Michigan State University, made valuable con— tributions to the analyses used in the study. The large secretarial workload was very capably handled by Shirley Ballentine and Evelyn Baker. The study was supported in part by National Institutes of Health Contract No. 72—4276. It was conducted while the author was on duty under instruction in the 0.8. Navy Medical Corps. The opinions expressed are those of the author and are not to be construed as those of the Department of the Navy. iv ACKNOWL LIST OF LIST OF Chapter 1. TABLE OF CONTENTS EDGMENTS . . . . . . . . TABLES . . . . . . . . FIGURES . . . . . . . . THE PROBLEM . . . . . . . Educational Question. . . Administrative Question. . Need for Research on Ambulatory Teaching . . . . . . The Purposes of This Study . Research Questions . . . A Model for the Study of Clinical ing Programs . . Assumptions. . . . . . Interdependence, Independent, Interactions of Variables Control of Variables. . . Current Study . . . . . Overview of the Dissertation. 0 Care 10 10 Chapter 2. REVIEW OF THE LITERATURE . . An Historical Perspective . The Comprehensive Care Movement The Cornell Program . . The Colorado Experiment . Reviews of Comprehensive Care Programs . . . . . Teaching Fate of the Comprehensive Care Movement. Aspects of Ambulatory Care Teaching. Patient Variables . . . Faculty Variables . . . Student Variables . . . Process Variables . . . Duration of Program . . Program Goals . . . . Outcome Variables . . . Summary . . . . . . . 3. METHODOLOGY . . . . . . Overview of Methodology. . Restrictions on the Scope of Problem Identification Phase Case Study Phase . . . . National Survey Phase . . Program Identification Survey the Study. Selection of Variables and Questionnaire Design . . . . . . vi Page 12 12 14 15 l7 18 20 21 21 27 3O 31 32 37 38 39 42 42 43 44 46 49 49 54 Chapter Page Pilot Test of Questionnaires . . . . 55 Main Questionnaire Survey . . . . . 59 Coding and Consolidation of Questionnaire Data . . . . . . . . . . . 60 Data Analysis . . . . . . . . . 65 Summary. . . . . . . . . . . . 70 4. RESULTS OF PROBLEM IDENTIFICATION PROCESS (PHASE ONE) . . . . . . . . . . 72 Composition of the Groups. . . . . . 72 Lists of Problems . . . . . . . . 73 Categorical Grouping of Problems . . . 74 Analysis of Problems by Variable Groups . 79 Summary. . . . . . . . . . . . 79 5. THE TEACHING OF ADULT AMBULATORY PATIENT CARE IN THREE MEDICAL SCHOOLS 83 Adult Ambulatory Care Teaching at Uni- versity A . . . . . . . . . . 83 The "Special" Clinic. . . . . . . 86 Medical Clinic. . . . . . . . . 90 Family Medicine Program. . . . . . 91 Adolescent Clinic. . . . . . . . 93 Other Programs. . . . . . . . . 95 Adult Ambulatory Care Teaching at Uni— versity B . . . . . . . . . . 96 "Special" Clinic at "University Hospital". 98 "Special" Clinic at "County Hospital" . . 100 Adult Ambulatory Care Teaching at Uni— versity C . . . . . . . . . . . 103 Chapter The "Faculty Clinic" . . . . . . General Medicine Clinic . . . . . Family Medicine Clinic . . . . . Community Medicine Program . . . . Cost of Adult Ambulatory Care Teaching. Summary . . . . . . . . . . . 6. RESULTS OF THE QUESTIONNAIRE SURVEY . . Questionnaire ReSponses. . . . . . Missing Data Elements. . . . . . Profile of Clinic Programs. . . . . Patient Variables Faculty Variables . . . . . . . Student Variables . . . . . . . Process Variables Program Goals . . . . . . . . Outcome Variables . . . . . . . Comparison of Preceptorship and Clinic Programs . . . . . . . . . . Program Goals . . . . . . . . Outcome Variables . . . . . . . Discussion . . . . . . . . . Comparison of Clinic Programs With (SQ) and Without (NSQ) Student Question- naire Data . . . . . . . Faculty Variables . . . . . . . Student Variables . . . . . . . viii Page 104 106 108 109 110 111 114 114 115 116 116 118 118 120 124 124 124 126 126 128 130 131 131 Chapter Process Variables . . . . . . . . Program Goals . . . . . . . . . Outcome Variables . . . . . . . . Discussion . . . . . . . . . . Comparison of Student and Faculty Responses. . . . . . . . . . . Discussion . . . . . . . . . . Clinic Programs' Strengths and Weaknesses as Reported by Students . . . . . . Discussion . . . . . . . . . . Relationships Between Variables-—C1inic Programs . . . . . . . . . . . Relationships Between Program Goals and Other Variables . . . . . . . . Relationships Between Outcome Variables and Other Variables. . . . . . . Discussion . . . . . . . . . . Relationships Between Variables--Precep- torship Programs . . . . . . . . Goal Group by Educational Group Matrix Correlations Between Rankings of Indi— vidual Goals and Outcome Measures . . Tests of Hypotheses . . . . . . . . Discussion . . . . . . . . . . Summary . . . . . . . . . . . . 7. SUMMARY AND CONCLUSIONS . . . . . . . Summary . . . . . . . . . . . . ix Page 132 132 133 133 134 134 135 137 138 139 143 149 150 151 151 152 152 153 156 156 Chapter Methodology . . . . . . Problem Identification Phase Case Study Phase . . . . Questionnaire Survey Phase . Discussion . . . . . . . Use of the Study . . . . Research Questions. . . . Usefulness of the Model . . Conclusions. . . . . . . Alternative Models of Clinical with Ambulatory Patients. . Observation Model . . . . Family Study Model. . . . History—Taking and Physical Examination. Early Experience with Patient Manage- ment. . . . . . . . Management of Acute Minor Illnesses . . Management of Chronic Diseases. . . . Advanced Patient Management. Summary . . . . . . . Implications for Further Research . . . Further Study of "Faculty Willingness" Study of Appropriate Experiences for All Students . . . . . Page 158 159 160 161 163 163 164 166 167 169 169 169 170 171 171 172 172 172 173 Page APPENDICES Appendix A. Checklists Used in Interviews, Case Study Phase . . . . . . . . . . . . . 174 B. Survey Instruments . . . . . . . . . 177 C. Supplementary Data on Methodology . . . . 194 D. Identified Problems . . . . . . . . . 201 E. Supplementary Results of the Questionnaire Survey Phase. . . . . . . . . . . 212 SELECTED BIBLIOGRAPHY. . . . . . . . . . . 241 xi LIST OF TABLES Responses to the Program Identification Survey . Numbers of Programs Identified and Qualified for Further Study . . . . . . . . . . Frequency of Incomplete Data on Qualified Programs 0 O O O O O O O O O O O 0 List of Variables for Study of Clinic Programs . Group Designation of Goals. . . . . . . . Research Hypotheses and Methods of statistical Testing . . . . . . . . . . . . . Composition of Problem Identification Groups. . Major Problems Listed by Faculty at School One . Major Problems Listed by Students at School One. Major Problems Listed by Faculty at School Two . Major Problems Listed by Students at School Two. Problem Categories by Rank and Total Points for All Groups 0 O O O O O O O O O O 0 Number of Problems, Total Points, and Average Points per Problem for the Problems in Dif— ferent Variable Groups . . . . . . . . Number of Problems, Total Points, and Average Points per Problem for Different Variables by Student and Faculty Groups . . . . . . Comparative Data on Clinic Programs. . . . . Location of Ambulatory Care Teaching for Uni— versity A Senior Students, 1972-73 . . . . xii Page 51 52 53 56 64 68 73 75 76 77 78 80 81 81 84 87 Table 6.1 6.2 6.3 6.4 6.10 6.11 6.12 6.13 6.14 6.15 6.16 Questionnaire Response Rates by Source . . . Profile of Clinic Programs on Patient variables 0 O O O O O O C O O O 0 Profile of Clinic Programs on Faculty variables 0 O O O O O O O O O O 0 Career Choice Preferences of Student Respondents . . . . . . . . . . . Student Times Per Patient Visit . . . . . Primary Purpose of the Clinics. . . . . . Subjects of Conferences . . . . . . . . Average Ranking of Clinic Program Goals in Order of Mean Ranks (N = 47). . . . . . Faculty Ratings of Clinic Programs on Outcome Variables (N = 50) . . . . . . . . . Ranking of Program Goals for Preceptorship and Clinic Programs . . . . . . . . . . Faculty Ratings of Preceptorship and Clinic Programs on Outcome Variables . . . . . Chi-square Test of Numbers of Programs in Student Success Group by Faculty Success Group Matrix . . . . . . . . . . . Numbers of Clinic Programs in Each "Strength- Weakness Category" Based on Student Responses O O O O O O O O O O O 0 Student Comments on Clinic Programs' Strengths and Weaknesses . . . . . . . Correlations Between the Ranking of Goal Statements and Clinic Program Character- iSticS O O O O O O O O O O O O 0 Correlations Between Faculty Ratings on Out— come Variables and Independent Variables for Three Program Groups . . . . . . . xiii Page 115 117 119 120 121 122 122 125 126 127 129 135 136 137 142 146 Table 6.17 Results of Analyses of the Predictors of Success of Adult Ambulatory Care Teaching Programs . . . . . . . . . . . . Coding Procedures. . . . . . . . . . Frequency Distribution of Student and Faculty Indications of When Students write New Patients' Records. . . . . . Frequency Distribution of Faculty and Nurse Indications of Primary Purpose of the Clinic 0 O O O O O O O O O O 0 Correlations Among Student and Faculty Ratings of Outcome Variables for Clinic Programs With Student Data (N = 21) . . . Index Values and Number of Programs in Each Outcome Group. . . . . . . . . Problems Listed by Faculty at School One . . Problems Listed by Students at School One. . Problems Listed by Faculty at School Two . . Problems Listed by Students at School Two. . Student Response Rates for Individual Pro— grams . O O O O O O O O O O 0 0 Criteria for the Selection of Patients for Teaching Medical Students . . . . . . Additional Goals Listed by Clinic Program Directors and Assigned Ranks . . . . . Additional Goals Listed by Preceptorship Program Coordinators and Assigned Ranks. . Comparison of Clinic Programs With (SQ) and Without (NSQ) Student Questionnaire Data (t teStS) o o o o o o o o o o o 0 Comparison of Clinic Programs With (SQ) and Without (NSQ) Student Questionnaire Data (Chi square tests). . . . . . . . . xiv Page 147 194 198 199 199 200 201 205 207 210 212 213 214 216 218 220 Table Page E.7 Key to Variable Names for Intercorrelation Matrices . . . . . . . . . . . . 221 E.8 Intercorrelation Matrix - All Clinic Pro- grams . . . . . . . . . . . . . 223 E.9 Intercorrelation Matrix - Clinic Programs Without Student Questionnaire Data . . . 226 E.10 Intercorrelation Matrix — Clinic Programs With Student Questionnaire Data . . . . 227 E.1l Variables Significantly Related to Goal Groups (t tests) (Part A) . . . . . . 229 E.12 Variables Significantly Related to Goal Groups (Chi—square test) (Part B). . . . 232 E.13 Chi-square Analyses of Numbers of Clinic Programs in Goal Groups by Faculty Success Groups Matrices . . . . . . . 233 E.14 Comparison of High and Low Faculty Success Groups (FSG) on Selected Variables (t tests) . . . . . . . . . . . 234 E.15 Stepwise Multiple Regression Analysis of All Clinic Programs on the Rating of Faculty Willingness to Participate (N = 44) . . . . . . . . . . . . 235 E.16 Chi-square Analysis of Numbers of Preceptor— ship Programs in the Goal Groups by Edu- cational Groups Matrix . . . . . . . 236 E.17 Tests of Hypotheses . . . . . . . . . 237 XV LIST OF FIGURES Figure Page 1.1 A Model for the Study of Clinical Teaching Programs . . . . . . . . . . . . 8 C.1 Mean and Standard Deviation of Variables by Source-Clinic Programs With Student Data (Part A) O O O O O O O O O O l O 197 C.2 Mean and Standard Deviation of Variables by Source—Clinic Programs with Student Data (Part B) C O O O C O O O O O O O 198 Chapter 1 THE PROBLEM The education of medical students consists of the acquisition of the essential knowledge, skills, and atti— tudes for the practice of medicine. While some of this education occurs through classroom~type activities, approximately half of the student's time in medical school is devoted to learning through supervised exper- iences with patients. For at least the past half century, the vast majority of these patient care experiences have been with hospitalized patients on the teaching wards of university, or university—affiliated, hospitals. This situation raises two questions, one educational and the other administrative. Educational Question Is it educationally sound for medical students to have little, if any, experience in dealing with non- hospitalized, or ambulatory,1 patients? While the 1The term, "ambulatory patients," is used through— out this report to indicate patients who are not confined or admitted to a hospital. student certainly needs to have experience with acutely ill, hospitalized patients under controlled conditions, it is ambulatory patients who provide the opportunity for the student to have experiences that much more closely parallel typical medical practice. The student should learn to care for patients between the crises that lead to hospitalization. With ambulatory patients the student can follow a patient with a chronic disease thereby gaining a "disease—over—time" perspective that is difficult to achieve with hospitalized patients. Ambulatory patients also present a different spectrum of disorders than are found on a teaching ward. Minor illnesses, smoldering chronic diseases, and psy— chosomatic and mild emotional disorders constitute the bulk of medical practice, but seldom result in hospitali— zation. Medical advances over the past three decades have made it possible for many problems previously requiring hospitalization, to be managed on an out—patient2 basis. A continuation of this trend would accentuate the current disparity between the nature of the typical medical practice and the study and management of hospitalized patients. 2The term, "out—patient," refers to patients who are not confined or admitted to a hospital, and is used synonymously with ambulatory patient. If the educational principle that the content and setting of learning should closely parallel the content and setting of practice is accepted, it must be concluded that ambulatory patients should play a major role in medical education. Administrative Question Will there continue to be adequate numbers of hospitalized patients accessible to students to continue the past pattern of clinical teaching? The number of students entering United States medical schools annually increased over 25 percent between 1968 and 1972.3 There is reason to believe that the number of entering students may nearly double between 1972 and 1985.4 Unless current teaching hospitals will be able to handle more students, and there is no available evidence for this, the number or size of teaching hospitals will have to increase, clinical teaching will have to be carried out elsewhere, or the amount of clinical teaching per student will have to be diminished. The problem of adequate numbers of patients in teaching hospitals is aggravated by three additional 3"U.s. Medical School Enrollments, 1968—1969 Through 1972—1973," Journal of Medical Education, 48:293- 97, March, 1973. 4William G. Anlyan, "1985," Journal of Medical Education, 46:917—26, November, 1971. factors: (1) Funds for the construction of new hospitals are being curtailed;5 (2) Expanding medical insurance programs are reducing the size of the medically indigent population, which has been the traditional source of patients used in clinical teaching; and (3) The number of hospital admissions has stabilized as the length of patient stay declines, resulting in fewer hospitalized patients.6 It seems unlikely that traditional teaching hospitals will be able to maintain their nearly exclusive role in the clinical instruction of medical students. If traditional teaching hospitals will be unable to provide an adequate amount of clinical instruction, what are the alternatives? Clinical instruction could be provided by (a) community hospitals; (b) by educational strategies requiring fewer patients, such as simulation techniques; or (c) by greater utilization of ambulatory patients. All three of these alternatives have, in fact, been used to some extent. 5"NIH Funds To Be Less in Fiscal 1974 Than in Fiscal 1972," Bulletin of the Association of American Medical Colleges, VIII, 2 (February, 1973), 1. 6Anne R. Somers, Health Care in Transition: Directions for the Future (Chicago: Hospital Research and Educational Trust, 1971), p. 28. Need for Research on Ambulatory Care Teaching Since ambulatory care teaching7 is expected to assume an increasingly important role in medical education, it is essential to have knowledge of the strategies and methods of this type of instruction in order to design and operate an optimum educational program. Although there is considerable literature8 on ambulatory care teaching, no systematic research has been carried out to guide the planning of this aspect of medical education. The Purposes of This Study The purposes of this study are: (1) To provide information on selected aspects of current adult ambulatory care teaching to assist educational planners in decision— making; and (2) To identify fruitful areas for further research on adult ambulatory care teaching. The study is primarily exploratory and descriptive in nature. The research is intended to produce infor— mation with which to answer the following questions. 7The term, "ambulatory care teaching," is used to mean the teaching of clinical medicine to medical students using ambulatory patients. 8This literature is reviewed in Chapter 2. Research Questions 1. What are the major problems which faculty and students perceive in current adult ambulatory care teaching? 2. Which characteristics of current adult ambulatory care teaching programs are associated with the success or failure of the programs? 3. Does the degree of success of programs vary with differing program goals? 4. Are preceptorship programs9 more successful than clinic programslo for some types of goals? A Model for the Study of Clinical Teaching Programs A detailed examination of a complex system requires an organizing set of principles or a model. A review of the literature on ambulatory care teaching failed to disclose existing principles or a model adequate to guide 9"Preceptorship programs" refers to programs in which medical students are involved in ambulatory care out- side the usual teaching hospital setting under the super— vision of physicians who are primarily practitioners rather than medical school faculty. These programs are sponsored or approved by the school but take place in the physician's office, or other practice site. The students are usually assigned to physicians singly rather than in groups. lo"Clinic programs" refers to programs in which groups of students are involved in ambulatory care in an outvpatient clinic which is part of a university, or university—affiliated, medical center under the supervision of university faculty, either salaried or voluntary. this study. For this reason the model described below was developed. The model is presented in diagrammatic form in Figure 1.1. Although the model was designed to assist in the study of ambulatory care teaching, it could be applied in the study of any clinical teaching program. Assumptions The essential assumptions of the model are that (a) program characteristics can be separated into groups of characteristics, or variables, called Patient Variables, Student Variables, Faculty Variables, and Process Variables; (b) Program Goals influence some of these variables; and (c) the success of the program, as measured by the Outcome Variables, is determined by the program characteristics in the Patient, Student, Faculty, and Process Variable groups, and the Program Goals. Interdependence, Independent, and Interactions of Variables The variables within each group, Patient, Student, Faculty, and Process, have considerable overlap or inter— dependence. For example, the academic rank of a faculty member may influence his willingness to teach in an ambulatory care setting. It is wise, therefore, to think in terms of profiles or clusters of characteristics within each variable group rather than dealing with each variable separately. Although a variable or the profile of variables in one group may influence the characteristics INPUT PROCESS OUTPUT PROGRAM GOALS l I ‘ \ I \ \ l \ I \ VI PATIENT VARIABLES PROCESS VARIABLES STUDENT VARIABLES / / / / / / ———> SYSTEM FLOW FACULTY - - - ~> VARIABLES INTERACTIONS OTHER THAN SYSTEM FLOW Figure 1.1 A Model for the Study of Clinical Teaching Programs of one or more variables in another group, this influence is by program design, not by the inherent nature of the variables involved. Thus, these variable groups can be considered independent of one another for research pur— poses, while recognizing that interactions, or designed interrelationships, are to be expected. For example, the characteristics of the students in a given program do not inherently change the nature of the faculty in the program, but the characteristics of the students may well influence the selection of faculty members (a program design decision). These interdependent, independent, and interaction relationships must be recognized and reckoned with in the use of this model. Some predicted inter— actions are represented by arrows in Figure 1.1. Control of Variables The different variable groups are under differing degrees of control by the program director. The student and faculty characteristics are usually determined by the overall curriculum, resources, and other factors not under the control of the program director. The program goals are usually determined largely by the curriculum, but with some input by the program director. The patient cnarac— teristics are decided by a combination of the available patient population, which is beyond the control of the program director, and the patient selection and/or screening processes which are established by the program 10 director. The characteristics of the process are specified by the program director within the constraints of the faculty, student, and patient characteristics and the program goals. The program goals and the faculty, student, patient, and process variables, determine the success of the program (outcome variables). These differ— ing amounts of control over the variables of the different groups are of concern in determining which variable groups are most important to study for purposes of assisting in the decision—making of educational planners. Current Study In the current study selected variables within each variable group are examined in order to test the usefulness of the model. The interdependence, independence, and interactions of the selected variables are also explored. Overview of the Dissertation Literature on ambulatory patient care teaching is reviewed in Chapter 2. This review consists of an his— torical perspective followed by a more detailed review organized around the variable groupings used in the pre— viously described model. In Chapter 3 the methodology of the study is described in some detail. The study was carried out in three phases. The first phase consisted of group sessions with medical students and faculty to 11 identify problems associated with adult ambulatory care teaching. Brief case studies of the adult ambulatory care teaching programs at three medical schools constituted the second phase of the study. The third phase was a series of national questionnaire surveys of adult ambulatory care teaching programs. Chapters 4 through 6 consist of the results of Phases One through Three, respectively. Chapter 2 REVIEW OF THE LITERATURE The literature on ambulatory care teaching is reviewed in this chapter with two separate styles. First, a historical perspective is taken in sketching the develop— ment of out—patient clinics and the emergence and activi- ties of the Comprehensive Care Movement. In the second part of this chapter bits of information from various sources are organized around the components of the model presented in the previous chapter. There are, therefore, sections on Patient Variables, Faculty Variables, Student Variables, Process Variables, Program Goals, and Outcome Variables. The purpose of this chapter is to provide the reader with adequate background information to understand the nature and significance of the current study. An Historical Perspective Out-patient clinics first appeared in the United States during the latter part of the eighteenth century—— the New York Dispensary in 1771, the Philadelphia Dis— pensary in 1786, and the Boston Dispensary in 1796. The 12 13 development of these clinics "stemmed from a sense of obligation on the part of society to provide medical services for all persons regardless of their ability to pay."1 During the next century dispensaries and the hospitals with which they became affiliated were dedi— cated to the sick poor. With Lister's discovery of the principles of antisepsis, and the subsequent development of improved surgical techniques, hospitals became centers for the treatment of the sick of all classes. The out- patient clinics, however, remained as a source of ambula— tory care primarily for the indigent populace. The degree to which these pre—twentieth century clinics were used to train physicians is not clear. Medical education was predominately an apprenticeship until the mid—eighteenth century when an academic trend developed. This trend led to the Flexner Report2 which "gave the final impetus for centering all medical education in hospitals and universities, and charity wards became 3 the sole teaching centers." The role of out—patient lMarvin B. Sussman et al., The Walking Patient: A Study in Outpatient Care (Cleveland: The Press of Western Reserve University, 1967), p. 2. 2Abraham Flexner, Medical Education in the United States and Canada (New York: The Carnegie Foundation for the Advancement of Teaching, 1910). 3John Schulman, Jr., "The Role of Hospital Ambulatory Service in Medical Education," Journal of Medical Education, 46:246, March, 1971. 14 clinics in medical education during the early twentieth century is quite clear. These clinics served as sources of patients who would be admitted to the wards if they were interesting teaching cases.4 Out—patient departments had second—class status. Care in these departments was fragmented due to the pre— sence of many specialty and subspecialty clinics with narrow interests and an orientation to crisis or episodic care. Society developed an increasing concern for adequate health care for all people. It is not surprising that there emerged a movement dedicated to promoting and teach- ing comprehensive medical care. The Comprehensive Care Movement Comprehensive care "implies the mobilization of all appropriate available resources for the care of the patient," including a primary concern for the entire patient, the application of preventative measures, and the early detection of disease.5 Ambulatory, or out- patient, medicine is not necessarily comprehensive, but a major portion of comprehensive care takes place in the 4Sussman et a1., op. cit., p. 4; Charles H. Good- rich, Margaret C. Olendzki, and Annemarie F. Crocetti, "Hospital-based Comprehensive Care: Is It a Failure?" Medical Care, 10:366, July-August, 1972; Schulman, op. cit., p. 246. 5Peter V. Lee, Medical Schools and the Changing Times (Evanston, Ill.: Association of American Medical Colleges, 1962), p. 29. 15 ambulatory care setting. It is for this reason that the Comprehensive Care Movement focused on ambulatory care teaching. As a result, during the past two decades the majority of the literature on ambulatory care teaching emerged from this Movement. The comprehensive ambulatory care teaching literature includes two books, several reviews of the literature, and numerous articles describ— ing individual teaching programs. The books and review articles are summarized below. Descriptions of individual programs are incorporated in a later section of this chapter. The Cornell Program In 1952 with support from the Commonwealth Fund, the New York Hospital-Cornell Medical Center began the Comprehensive Care and Teaching Program (CC & TP). Reader and G055 provide an extensive report on the first five years of operation of this program.6 In the CC & TP, as it was called, senior medical students devoted twenty—two and a half weeks to comprehensive ambulatory medicine. During this time the students attended Medical, Pediatrics, Psychiatry, and elective specialty Clinics, as well as various conferences, seminars, and lectures. The Medical and Pediatric Clinics were reorganized to provide 6George G. Reader and Mary E. W. Goss (eds.), Comprehensive Medical Care and Teaching (Ithaca, N.Y.: Cornell University Press, 1967). 16 continuity of the faculty—student-patient relationship. Consultants were available in the clinics to avoid the fragmentation resulting from frequent referrals to specialty clinics. Students were expected to assume primary responsibility for the diagnosis and treatment of their patients. "On the whole, students eagerly accepted . . . (this) responsibility."7 Since the CC & TP was considered an experimental program, arrangements were made for extensive research activities. These aspects of the Program were carried out by the Bureau of Applied Social Research at Columbia University, and had a sociological focus which centered on the effect of the program on the attitudes and values of the medical students. The results of these studies 8 make up a major portion of Reader and Goss' book, and 9 led to a major publication in medical sociology. The finding of this intensive research and evaluation effort are well summarized by Reader: . . the Cornell Program had the desired effect on students' attitudes and standards as compared with the effect of other fourthsyear courses. It resulted in students' becoming more patient-oriented and had moderate success in reversing the trend from first through fourth year, common in medical schools, toward 7 8 Ibid., p. 59. Ibid. 9Robert K. Merton, George Reader, and Patricia Kendall (eds.), The Student Physician: Introductory Studies in the Sociology of Medical Education (Cambridge: Harvard University Press, 1957). 17 an increasing preference for patients with physical illness. At the same time, professional detachment toward patients was enhanced, not at the expense of viewing patients as disease entities rather than as persons, but representing an increase in professional maturity. As a result of the Program, students also tended to attach greater importance to social and emotional problems of patients than those not exposed to it. Their standards in regard to quality of medical care and the limitations of the physician's role were improved by the CC & TP experience; they became more discriminating and more realistic. 0 The Colorado Experiment A year after the beginning of the Comprehensive Care and Teaching Program at Cornell, the University of Colorado established an experimental comprehensive care teaching program at the Denver General Hospital.11 This program was centered around a specially organized General Medical Clinic (GMC). The GMC served a limited number of patients, both adults and children. The principles of comprehensive care that were taught were similar to those of the Cornell program, but at Colorado there was more emphasis on family care. A randomly selected half of each senior class attended the GMC program approximately half of their time for twenty-four weeks. In addition to the supervised patient care experiences, the students participated in 10Reader and Goss, op. cit., p. 287. 11Kenneth R. Hammond and Fred Kern, Jr., Teachin Comprehensive Medical Care (Cambridge: Harvard UniverSity Press, 1959). 18 special conferences. The control group of students spent an equal amount of time in varied experiences at a dif- ferent hospital. The research activities associated with the Colorado program focused on psychological studies, and 12 The are reported in detail by Hammond and Kern. experimental group acquired at least as much traditional medical knowledge and skill as the control group, but there was only a slight increase in their understanding of psychological and sociological principles. During the senior year the control group students developed increas- ingly negative attitudes toward the aspects of comprehen— sive care, while the attitudes of the experimental group students showed little change. Reviews of Comprehensive Care Teaching Programs As part of a series of case reports on experi- mentation in medical education, Lee includes four com- prehensive care teaching programs——Cornell, Colorado, Temple, and North Carolina.13 He points out the contrasts among the different programs——Temple emphasized psycho— somatic medicine and North Carolina stressed preventive lzIbid. l3Lee, op. cit., pp. 29-59. 19 medicine, whereas Cornell and Colorado carried out research in the sociology and psychology of medical education. Snoke and Weinerman reviewed twenty—three compre- hensive care programs at nineteen medical schools. Of their conclusions, the following are particularly relevant: 1. Medical education in the current era is incom- plete without attention to the principles and practice of comprehensive patient care. 2. The comprehensive care unit should be modest in size, with controlled patient load and low pre— ceptor—student ratio. 3. The patient group should be selected to represent a cross-section of the community. 4. The student should have direct responsibility in the care of patients assigned to him.14 In an attempt to determine the effectiveness of teaching programs in comprehensive medicine, Sanazaro and Bates conducted a study comparing the performance of stu- dents in schools with and without explicit comprehensive care programs.15 They failed to demonstrate significant differences in the relevant behaviors of the two groups of students, but the great majority of the students in both groups were judged to perform at a satisfactory level. The authors acknowledge severe short—comings in their methodology which limit the usefulness of their findings. l4Parnie S. Snoke and E. Richard Weinerman, "Com- prehensive Care Programs in University Medical Centers," The Journal of Medical Education, 40:625—57, July, 1965. lsPaul J. Sanazaro and Barbara Bates, "A Joint Study of Teaching Programs in Comprehensive Medicine," The Journal of Medical Education, 43:777—89, July, 1968. 20 Fate of the Comprehensive Care Movement The Cornell Comprehensive Care and Teaching Program continued in operation with minor changes for over fourteen years, but then declined until currently only a small vestige remains.16 The Colorado Program was modified and incorporated into the regular curriculum after the five- year experimental study. In 1961 the program was dis- continued.l7 Snoke and Weinerman noted that six of the twenty-one programs they reviewed had been discontinued by 1964.18 Goodrich, Olendzki, and Crocetti claim that the first generation comprehensive care programs were successful in their area of teaching and research.19 They report, however, that second—generation projects are failing to replace the traditional out—patient departments. Over the past few years the number of reports on comprehensive ambulatory care teaching programs has diminished. It appears that many of the principles of comprehensive care have been incorporated into traditional educational programs, and that the major concern of the 16 Personal correspondence. l7Snoke and Weinerman, op. cit., pp. 632—33. lBIbid., pp. 628—31. 19Goodrich, Olendzki, and Crocetti, op. cit. r______________i 21 Comprehensive Care Movement has shifted from medical edu- cation programs to the health care delivery system. The Movement produced much information on various aspects of ambulatory care teaching that is useful in studying medical education. Aspects of Ambulatory Care Teaching The literature that resulted from the Comprehensive Care Movement and related ambulatory care teaching programs is predominantly in the form of case studies or compilations of case studies. In this section the information from these reports will be reorganized to fit the model for the study of ambulatory care teaching described in Chapter 1. Patient Variables The patient is a critical element in the ambulatory care teaching program, however, as Hammond and Kern point out, patients who do not fit the program "are likely to be barriers, rather than aids, to education."20 Snoke and Weinerman concluded that "ideally, the patient group should be selected to represent a cross—section of the 21 community." Students are concerned if the patients do not present a concentration of specific diseases, but 20 . Hammond and Kern, op. C1t., p. 141. 21Snoke and Weinerman, op. cit., p. 648. 22 faculty may worry more about the patients' lack of motivation to get well and communications problems." The high percentage of chronically, but not seriously, ill patients with insoluble psychological problems and hopeless social situations, together with the high missed appointment rate, definitely detracted from the effective— "22 ness of the . . . (Colorado) program. Students complain of "the frustration of being unable to achieve any 23 Although observable treatment gain in many cases." indigent, chronically ill patients present difficulties in the educational program, they are more satisfied than acutely ill patients with the care they receive in the out—patient clinic.24 Finding suitable patients for inclusion in an ambulatory care teaching program is not easy. The Cornell Comprehensive Care and Teaching Program had to revise patient selection criteria in order to obtain an adequate 25 number of suitable patients. A major concern in the selection of patients for a teaching program is the 22Hammond and Kern, op. cit., p. 159. 23David E. Reed, "Twelve Years' Experience with a Comprehensive Ambulatory Care Program," Journal of Medical Education, 45:1043, December, 1970. 24Sussman et a1., op. cit., p. 85. 25Merton, Reader, and Kendall, op. cit., p. 250. 23 acceptance by the patient of a medical student as his "physician." In a study of clinic patients Wasserman, et al., found "that 78% (of patients) accepted without question the involvement of medical students in the team; and 22% were fully aware that the students were not graduate physicians, yet welcomed their interest."26 Concern has been expressed for the fact that expanded medical insurance programs are reducing the number of "medically indigent" patients, and that there may be inadequate numbers of such patients to carry out tra~ ditional ambulatory care teaching programs. Reed reports the results of a study of private patients in physicians' offices. Of these patients, 25 percent desired contact with medical students, 65 percent did not care, and 10 percent felt that medical student involvement in their care would be unacceptable.27 This report suggests that private patients may be available for the teaching of medical students. In addition to this general view of the role and problems of the patients in ambulatory care teaching, the literature provides information on some of the specific aspects of the patient role. 26Edward Wasserman et a1., "Medical student Involvement in Comprehensive Health Care," Journal of American Medical Association, 215:2098, March 29, 1971. 27Reed, op. cit., p. 1046. 24 Socio—economic status. Sussman discusses the impact of the differences between the values of the indigent patient and the middle—class practitioner or medical student.28 Snoke and Weinerman conclude that the dependence of ambulatory care teaching programs on the low'income clinic population has been a definite dis— advantage. "The problems presented by such patients are discouraging and difficult, particularly for inexperienced medical students, and have been the source of much of the 29 negative student attitude." Sussman concurs that "medical students have not been seeing the type of patients likely to be encountered in their later "30 Hammond and Kern also argue that the practice. patient population should represent a variety of social and cultural types.31 One study also showed "that more preventive and health education work was possible with moderate-to—high income families than with lower-income 32 families." There seems to be little question that the 28Sussman et a1., op. cit., p. l. 29Snoke and Weinerman, op. cit., p. 646. 30Sussman et a1., op. cit., p. 191. 31Hammond and Kern, op. cit., p. 25. 32Snoke and Weinerman, op. cit., p. 639. 25 typical clinic patient population consisting of the lower socio—economic classes has presented an education problem. Psychological and psychosomatic problems. It has been noted that physicians tend to resent patients with emotional problems because of the frustration produced by being unable to affect a cure.33 In spite of this, the literature suggests that this may not be as great a problem in ambulatory care teaching as the patients' socio—economic status. In the Cornell Program less than 5 percent of the diagnoses established by students were in the cate— gories reflecting predominantly emotional problems.34 Reed observed that students complained of "the predominance "35 but that in of 'functional' (psychosomatic) illness, recent years students have demonstrated more acceptance of this type of problem. He attributes this increased acceptance to changing social values.36 A report of the program at Washington University indicates that one—third of the patients presented with early, poorly manifest, or functional illness. This initially produced considerable 33Anne R. Somers, Health Care in Transition: Directions for the Future (Chicago: Hospital Research and Educational Trust, 1971), p. 10. 34Reader and Goss, op. cit., p. 61. 35Reed, op. cit., p. 1043. 361bid., p. 1044. 26 student dissatisfaction, but this was partially overcome by teaching the students to understand and cope with these problems.37 It appears that psychological and psychoso— matic illnesses present difficulties in ambulatory care teaching, but that these difficulties can be at least partially overcome by an appropriate educational program. Missed appointments. Studies in two clinics revealed that approximately 20 percent of patients failed to keep their appointments.38 This frequency of missed appointments was one of the students’ top three complaints about the teaching program at one institution.39 Sussman found that this irregular attendance by the patients not only produced severe scheduling problems with a waste of student and faculty time, but also lowered staff morale.40 None of the authors proposes a solution to this problem, but it is intimately related to the patients' socio~ economic and cultural backgrounds and the type and severity of their medical problems. 37Robert E. Shank, "Three Years' Experience in the Coordinated Outpatient Program at Washington Uni— versity," Journal of Medical Education, 31:283—93, May, 1956. 38Hammond and Kern, op. cit., p. 53; Sussman et a1., op. cit., p. 9. 39Reed, op. cit., p. 1043. 0Sussman et a1., op. cit., p. 10. 27 The literature clearly indicates that patient variables play a major role in the design and success of ambulatory care teaching programs. Another major component of such programs is the faculty. Faculty Variables The importance of the faculty is well documented in the literature. Reader and G055 state "that the teach— ing of comprehensive medicine requires a large number of dedicated and skilled faculty members."41 Hammond and Kern comment on the importance of the faculty as follows: Staffing a program of this kind is a major problem. Ideally, almost all the staff should be permanently associated with the program on a full—time basis; they should be skillful teachers, willing to devote long hours to working with students; and they should be highly competent in general medicine as well as the medical phases of psychology and sociology, and in their own medical specialty. Such people are not easy to find.42 The problem of acquiring an appropriate faculty is quite complex. Sussman et al. found that nearly half of the physicians working in teaching clinics were dis— 43 satisfied with their roles. Hammond and Kern noted that faculty tend to arrive late at teaching clinics, 41Reader and Goss, op. cit., p. 291. 42Hammond and Kern, op. cit., p. 159. 43Sussman et a1., op. cit., p. 194. Y—iw— 28 thus keeping students and patients waiting.44 By checking the accumulated patients quickly the faculty can increase their own efficiency at the expense of the students' and patients' satisfaction. Other problems occur with regard to faculty atti- tudes. They may demonstrate an inappropriate reluctance to accept a functional diagnosis and even demonstrate frustration with, and hostility toward, the patients.45 Also, faculty members with highly specialized backgrounds may be reluctant to teach students utilizing mundane, undifferentiated patients.46 Prestige of teaching in the outtpatient clinic. "The out-patient department has traditionally had low "47 A period status in the medical center pecking order. of teaching in the out-patient clinic has frequently been considered a necessary step prior to being permitted to teach on the in-patient wards, or even to admit patients 44Hammond and Kern, op. cit., p. 16. 45Ibid., p. 19. 46Kerr L. White and William L. Fleming, "Improving Teaching on Ambulant Patients," Journal of Medical Edu— cation, 32:30-36, January, 1957. 47E. Richard Weinerman, "Yale Studies in Ambulatory Medical Care," The New England Journal of Medicine, 272:947-54, May 6, 1965. 29 to the hospital.48 This low status image of teaching in the ambulatory care setting was modified in two programs by somewhat different strategies. In one case senior faculty members undertook full teaching responsibilities in the clinic program;49 and in the other, the personal commitments of departmental chairmen were considered to be critical to the program's success.50 Full-time faculty, voluntary faculty or house staff. Although reports cited above have tended to favor the use of full-time faculty in the teaching of ambulatory care, there appears to be some difference of opinion. Reed's study indicates that students value their exposure to practicing physicians (volunteer faculty).51 Shank reports the successful use of a combination of full-time faculty, volunteer faculty, and house staff, but indicates that careful selection is essential.52 Merton's study 48John P. Colmore and Stewart Wolf, "A New Design for Service and Teaching in the Out-Patient Department," Journal of American Medical Association, 156:830—33, October 30, 1954; White and Fleming, op. cit., p. 34. 49Colmore and Wolf, op. cit., p. 831. 50Lee, op. cit., p. 58. 51Reed, Op. cit., p. 1043. 52Shank, op. cit., p. 291. 30 suggests that volunteer (part-time) faculty may feel less involved in the teaching clinic and, therefore, have fewer ideas for improvements. He also noted that their private practices interfered with participation in clinic planning sessions.53 Reader and Goss found house staff (residents) to be a source of great help in clinic teach— ing, but stressed that the length of service was critical; "a six month assignment might be acceptable, but a year was ideal."54 Student Variables In most reported ambulatory care teaching programs students assume, under supervision, much of the physician responsibility for patient care.55 Reader and Goss noted that, on the whole, the students eagerly accepted this 56 Before students can assume this responsibility. responsibility, however, they must "have an adequate store of factual knowledge and be confident of their abilities."57 53Merton, Reader, and Kendall, op. cit., pp. 259-60. 54Reader and Goss, op. cit., p. 261. 55Sanazaro and Bates, op. cit., p. 778. 56Reader and Goss, op. cit., p. 59. 57Ibid., p. 283. ‘IFI--------------_----__--_-____—__________———________——___—“.-‘l!!=fifiliiifii'- 31 When are students adequately prepared for this responsibility, and what have been the results of schedul— ing ambulatory care experiences at different times in the student's educational program? In planning the Colorado experiment, it was decided that third—year students were inadequately prepared for the patient care responsibili- ties expected in the program. The evaluation studies carried out as part of the Colorado program showed that students participating in the clinic during the first half of the fourth year achieved the program goals to a higher degree than students participating during the second half of the year. This was shown to be due to concern with traditional medical knowledge brought about by the impending responsibilities of an internship. This dif— ference disappeared when all students were in the program and there were no control group peers to make partici— pating students feel they were lacking important exper— iences. Even so, Hammond and Kern suggest that the first half of the fourth year may be the optimal time for ambulatory care experiences for medical students.58 Process Variables There are many aspects of the operation of an ambulatory care teaching program included under the term "process variables" in the model described in Chapter 1. 58Hammond and Kern, op. cit., pp. 134—35, 158, 550. 32 This section contains a summary of information from the literature on several of these variables. Duration of Program The importance of the length of the student's participation in the ambulatory care teaching program is well stated by Reader: Responsibility for patients is the most effective way for the student to learn wisdom in patient man— agement and gain the ability to think through a clinical problem. It also motivates him to read about clinical entities he encounters so that he may better help the patient, thus stimulating more intensive study of scientific medicine. The length of the student's contacts with the same patients is believed to be the most important determinant of development of a sense of responsibility. . . .5 Reed reports that students responded favorably to being able "to follow a patient over a period of time and establish a satisfactory doctor—patient relationship."60 Reader and Goss, in summarizing their experience state that "a four month period is a minimum; six months to a year would be optimal." Several programs have had students participating in a clinic for one or two half—days per week for one or two years, thus permitting greater continuity of care than a concentrated period of a few weeks or months. The Colorado planners abandoned this design because of 59 . Merton, Reader, and Kendall, op. c1t., pp. 87-88. 60Reed, op. cit., p. 1043. 33 concern for scheduling problems and conflicting pressures for the student's time and interests.61 A program at the University of Vermont, reported by Smoke and Weinerman, used a somewhat unique arrangement to provide continuity in the student—patient relationship. A junior student and a senior student worked as a team with patients, so that the student had two years of contact with the patients, and the patients had a gradual, rather than abrupt, turn- over of student—physicians.62 This program was discon— tinued in 1964.63 Reed noted that most of the comprehensive ambula— tory care programs remaining in 1970 were either on an elective basis or consisted of a block—time assignment to the out—patient department. Although shorter, block assignments lack some of the advantages noted above, they have their own advantages. Scheduling of student time is greatly simplified. The Colorado study suggests that students are more satisfied with a shorter involvement in comprehensive ambulatory care programs.64 A longer period of involvement may also aggravate the students' 61Hammond and Kern, op. cit., p. 9. 62Snoke and Weinerman, op. cit., p. 642. 63Reed, op. cit., p. 1041. 64Hammond and Kern, op. cit., p. 160. 34 frustrations with the patients' social, psychological, and economic problems. Lee attributed the high student morale in one program, where these problems were particu— larly severe, to the relative brevity of the clinic experience.65 Although the literature provides considerable insight, it does not give definite answers to questions regarding the optimal duration of ambulatory care teach- ing programs . Time per patient, number of patients, and number of visits. The literature provides little information on these details of ambulatory care teaching clinic oper— ations. The Colorado program emphasized giving the stu— dent "ample time to investigate and manage all of the many "66 The amount of time problems his patients presented. that is required is not specified, but students would seem to typically see one new patient and one or two follow-up patients per half day as a maximum. Reports on the Cornell program indicate that, on the average, each student saw about thirty patients. The student saw each patient an average of between two and three times.67 65Lee, op. cit., p. 50. 66Hammond and Kern, op. cit., p. 6. 67Reader and Goss, op. cit., p. 53. IIIIIIIIIIIIIT___________________________________________________________:;;;Jgflfiaifi5379" 35 Patient selection. Although some case studies indicate the specific criteria, if any, used for the selection of patients for inclusion in the teaching program, no good discussion of this variable was found in the literature. One interesting strategy is worthy of mention, however. In one program, students partici— pated in a screening and emergency clinic and selected from this group of patients those that they would evaluate further and follow in another clinic.68 This student par- ticipation in patient selection would appear to have some definite advantages in achieving student interest in clinic patients. Faculty~to-student ratio. Reports of programs from the literature indicate some variety in the faculty— to-student ratio in the clinic. One author indicated that their one—to—one ratio was a strong point in the program,69 while others advocated, or accepted, ratios of one faculty member for two to four students.70 One program which had a ratio of one to one found a ratio of one to two more satisfactory. With the one—to—one ratio the faculty felt their time was used inefficiently because 68Snoke and Weinerman, op. cit., p. 639. 69Reed, op. cit., p. 1043. 7oShank, op. cit., p. 290; Colmore and Wolf, op. cit., p. 831. 36 of frequent long waits for the students to require assistance. The faculty tended, therefore, to arrive late and show a lack of interest. When the ratio of faculty-to-students was changed to one to two, the faculty were busier, more satisfied, and arrived on time.71 Conferences and seminars. Bogdonoff, et a1., provide insight into the role of conferences in ambulatory care teaching by drawing a parallel to in—patient teaching. Traditionally, the most exciting learning situation in a department of medicine has been the teaching rounds on the invpatient service, the major feature of which has been the opportunity to spend two or more hours in the midst of the day's activity in pausing to reflect upon the clinical experiences of the training physicians. . . . In the out—patient department the actual work of the day, namely the examination and care of patients, must go on during the very hours that such a midmorning pause may be taken on the in—patient service; therefore, it is not feasible for a group of physicians to take out such a mid-day time for reflection in the ambulatory care area. This has always represented a major short- coming of the out-patient teaching program. Since we considered such group conference time as an important learning experience, it was decided to schedule hours for reflection and discussion at 72 both the beginning and at the end of the working day. Reader and Goss observed that students reacted better to small discussion groups than to large meetings.73 Hammond 71 . . Personal communication. 72Morton D. Bogdonoff, Stanley W. Elwell, and Julian M. Ruffins, "Medical Out—Patient Department Teach- ing," Journal of Medical Education, 38:885—89, October, 1963. 73Reader and Goss, op. cit., p. 250. 37 and Kern noted that there was a generally unfavorable reaction by students toward conferences which were pri— marily concerned with the social and psychological aspects of the patients' problems.74 This reaction of students explains why the conferences associated with the Cornell program had "some shift over the years to a somewhat more traditional medical content."75 Program Goals The goals of comprehensive care teaching programs in general followed the definition of comprehensive care presented previously. There is little additional mention of program goals in the literature. Two sources suggest an additional goal, however. The Cornell program might be considered to have achieved another goal . . . in that it enabled some students to discover that they were not interested in patient care and, accordingly, to choose a specialty career that allowed them to avoid contact with patients or limited their responsibility as physicians to a specific organ system.76 Involvement in another program was shown to influence 80 percent of the students in their choice of elective 74Hammond and Kern, op. cit., p. 83. 75Lee, op. cit., p. 34. 76Reader and Goss, op. cit., p. 291. 38 courses in a subsequent year.77 These bits of evidence suggest that ambulatory care teaching programs can provide students with valuable information and experience for career choice decisions. Outcome Variables Except for those aspects of ambulatory care teach- ing covered in the special research studies reviewed as part of the Comprehensive Care Movement, there is little data on the outcomes of these programs. Although students in the Colorado program felt they had learned less during their time in the program, an objective assessment of their acquisition of medical knowledge showed no deficit.78 In the assessment of another program students reported their educational return for time spent in the clinic was only fair, but most students indicated that it was as good as, or better than, other clinical assignments. Students from this program were polled three to six years after graduation. The percentage of favorable reactions to the clinic experience was quite stable over time, how— ever, the percentage of mixed responses increased and unfavorable responses decreased after the students grad— uated and entered practice.79 77Wasserman et a1., Op. cit., p. 2098. 78Hammond and Kern, 0p. cit., p. 312. 79Reed, op. cit., p. 1043. 39 Summary During the eighteenth and early nineteenth cen- turies out-patient clinics and hospitals primarily pro— vided care to indigent patients. With the advent of antisepsis and other medical advances, hospitals became centers for the care of all socio—economic classes, but out-patient clinics continued to serve mainly the poor. As medical education developed an academic base, the educational activities became centered in universities and teaching wards. In the early 1950‘s the Comprehensive Care Move— ment developed in reaction to many shortcomings of ambula— tory care clinics and their associated teaching programs. Several special clinic programs were started, some of which included considerable evaluative research. The programs at Cornell University and the University of Colorado were the most notable of these. Studies of these programs indicated that an intensive educational program could have an impact on medical student attitudes. The usual trend for medical students to become more disease—oriented and less patient—oriented during medical school was at least partially reversed in these programs. By the mid-1960's the Movement appeared to be diminishing, probably because many of its principles had been incor- porated into traditional educational programs. 40 Of the Patient Variables discussed, the most per- vasive problem appears to be for teaching clinics to obtain patients who represent a cross—section of the community population, rather than being predominantly indigent with insoluble social, economic, and psychological problems. The presence of dedicated, enthusiastic faculty strengthens an ambulatory care teaching program, but the traditional low status of the clinics makes it very diffi- cult to obtain such a faculty. Most authors agree that learning is increased when students assume a great deal of patient care responsibility, and that students are not adequately prepared for this before their final year in medical school. There is also some evidence that stu- dents are more receptive to ambulatory care teaching before they become overly concerned with the acquisition of skills required for their internships. A number of aspects of ambulatory care teaching are discussed under the heading, Process Variables. It is suggested that students should have at least four to six months experience in a comprehensive ambulatory care setting. Programs designed to give students part—time experiences over a long period have usually suffered from scheduling difficulties and poor student acceptance. The sources reviewed agree that there should be no more than four students per faculty member in the clinic. Some 41 authors believe that one faculty member for every two students is necessary, and a one-to—one ratio is desirable. The presence of conferences or seminars as part of the program is desirable, if not essential. Experience has shown, however, that it is difficult to maintain stu— dent interest in conferences on the social and psychologi— cal aspects of patients' problems. Program goals have usually been based on the defi— nition of comprehensive care. Two reports suggest that ambulatory care teaching programs can also be valuable in assisting the student in making career choice decisions. Other than the Cornell and Colorado studies of changes in student attitudes, there is little written on the outcomes of ambulatory care teaching programs. The students' perceptions of learning gains in the programs have varied from mediocre to fair. There is some evidence that students may tend to value these eXperiences somewhat more highly three to six years later than immediately following the experience. Chapter 3 METHODOLOGY Since the available literature on ambulatory care teaching in medical schools was considered inadequate to guide an exploratory, descriptive research study, a methodology was used in which the results of early phases of the study guided the design of subsequent phases. Overview of Methodology The initial phase of the study consisted of con- ducting small group problem identification sessions with faculty and students from two medical schools. This pro— duced a listing of perceived problems in adult ambulatory care teaching, and an indication of the relative importance of the different problems. The second phase of the study was a series of inter— views of administrators, faculty, nurses, and students involved in ten adult ambulatory care teaching programs at three medical schools. The lists of problems identified in the previous small group sessions were used to guide the scope of the interviews. The information obtained was used to produce case studies of the ambulatory care 42 43 teaching at these institutions and to guide the data collection and analysis in the final phase of the study. The final phase involved the collection of data with two, sequenced survey instruments. The first survey form was sent to all established United States medical schools and identified adult ambulatory care teaching programs suitable for further study. The major data collection consisted of a questionnaire survey of faculty, nurses, and students involved in selected clinic or pre- ceptorship programs for the teaching of adult ambulatory patient care to medical students. The data obtained were submitted to extensive analysis, and the hypotheses gen— erated during the case study phase were tested. Restrictions on the Scope of the Study The scope of the problem identification phase was restricted only in that participants were instructed to identify problems involving teaching medical students in an adult ambulatory (out-patient) clinic setting. It was considered necessary to further limit the sc0pe of the study for the subsequent phases. For this reason only programs dealing with general medicine or family medicine were included. Programs dealing pre— dominantly with children, surgical specialties, emergency departments, and medical subspecialties were excluded from the study. Also excluded were programs exclusively 44 involving students in the "pre-clinical" portion of their medical education, which usually constituted the first two years of medical school. Only those aspects of the ambulatory care teach— ing programs which fit the model presented in Chapter 1 were included. This eliminated such topics as how a program fit in an institution's overall philosophy of medical education and in the curriculum. Additional specific limitations of scope will be presented with the description of the main data collection survey in a subsequent section of this chapter. Problem Identification Phase Two medical schools were selected for use in the initial portion of the study on the bases of their (1) con- trasting characteristics, (2) willingness to participate, and (3) geographic proximity to the investigator. One of these schools is an old, relatively traditional insti— tution with no required adult ambulatory care teaching programs. The other school is a new, innovative insti— tution, which has an ambulatory care teaching program which is required for some students. This school was also in the process of planning an extensive ambulatory care teaching program. Separate faculty and student sessions were held at each school. At one school, the participants were (a) five faculty members from various specialties, who r , fir , . r 45 were interested in ambulatory care teaching, and (b) three senior medical students with varying ambulatory care experiences. At the other school six faculty members from the Department of Medicine and four senior students with experience in a comprehensive ambulatory care clinic participated. The nominal group problem identification process as described by Van de Ven and Delbecq was utilized.1 Participants were requested to "list the subjective and objective problems you have experienced or perceived as a medical student in an adult ambulatory (out-patient) clinic setting," or "list the subjective and objective problems you have experienced, perceived, or anticipate as a faculty member teaching in an adult ambulatory (out— patient setting." After each participant in a single group had listed the problems which he recognized, the individual lists were consolidated and each group member selected the ten problems which he felt were most important and ranked them from one to ten. After a brief discussion of these rankings each participant reconsidered his pri— ority assignments. He then distributed one hundred points 1A. H. Van de Ven and A. L. Delbecq, "The Nominal Group as a Research Instrument for Exploratory Health Studies," American Journal of Public Health, 62:337—42, March, 1972; A. L. Delbecq and A. H. Van’de Ven, "A Group Process Model for Problem Identification and Program Planning," Journal of Applied Behavioral Science, 7:466— 92, 1971. 46 among these ranked problems in proportion to his per— ception of the problems' importance.2 After each group session the total assigned points were used to rank the problems generated by that group. After all sessions were held, each problem statement was placed on a card and the cards were sorted into groups having common elements. The categories formed by this process were then ranked by the total points assigned to problems within that category. The problem statements were also sorted by their association with the Patient, Faculty, Student, or Process Variable Groups of the model presented in Chapter 1. The total points assigned to the problems in each Variable Group were determined for faculty and student participants. The results of this problem identification process and the associated analyses are presented in Chapter 4. Case Study Phase The purposes of the case study phase of this research were: (1) To apply the information obtained in the problem identification phase to the study of 2The distribution of one hundred points among the ranked problems is a modification of Van de Ven and Del- becq's procedure. They suggest having participants assign one hundred points to their top—ranked problem and values between zero and one hundred to the other nine ranked problems.3 The modification was made in an attempt to equalize the influence of group members in the final con— solidated problem priority listing. 3 Van de Ven and Delbecq, 0p. cit., pp. 339—40. 47 individual adult ambulatory care teaching programs; (2) To determine the aspects of program design and oper~ ation suitable for study by survey methods; (3) To gen— erate hypotheses to be tested in the survey phase of the study; and (4) To produce a series of brief case studies of adult ambulatory care teaching programs. In order to achieve these purposes it was necessary to identify several medical schools with a variety of adult ambulatory care teaching programs. Consultations with persons familiar with the educational programs of a number of medical schools and a review of the information con— tained in the directory of medical school curriculums published by the Association of American Medical Colleges4 provided considerable information. Based on this infor- mation, representatives of four schools were contacted regarding further details of their programs and their willingness to cooperate. Three schools with ten programs meeting the criteria for inclusion in the study agreed to participate. Their programs were believed to repre~ sent an adequate variety of program types to fulfill the purposes stated above. Two days were spent at each of the three schools studying their adult ambulatory care teaching programs and interviewing key personnel in each program. An associate 4A.A.M.C. Curriculum Directory (Washington: Association of American Medical Colleges, 1972). 48 dean for medical education or curriculum and several stu- dents were interviewed at each school. The supervising faculty members (program directors) of all programs, and additional faculty members and a clinic nurse and/or secretary for most programs, were also interviewed. The interviews were largely unstructured with the interviewee being asked to describe the operation of the programs being studied. Check-lists of topics to be covered in the interviews were utilized and direct questions were asked on those topics which were not covered in the more openwended portion of the interview.5 The check-lists consisted of those topic areas produced by the problem identification phase of this study, as well as, some descriptive measures considered to be of interest. The interviews were tape recorded with the consent of the interviewee or detailed notes were taken by the inter— viewer. The case studies produced as a result of this phase of the study are presented in Chapter 5. The hypotheses generated as a result of the interviews are listed in a subsequent section of this chapter. 5Copies of the interview check—lists are included in Appendix A. 49 National Survey Phase This phase of the study involved three rounds of data collection. A survey form was sent to all estab— lished United States medical schools to identify ambulatory care teaching programs suitable for further study. Detailed questionnaires were pilot tested by sending them to the program directors, clinic nurses, students, and a preceptorship coordinator involved in the programs studied in the case study phase. The final, revised questionnaires were then used to collect data on the programs selected from the responses to the program identification survey. Program Identification Survey A survey instrument was designed to obtain a list— ing of adult ambulatory care teaching programs, both clinics and preceptorships, in the fields of general med— icine, internal medicine and its subspecialties, primary care, and family, community, and comprehensive medicine. At the time the instrument was designed it was not possible to estimate the number or types of programs which would be identified, therefore, information was requested to permit programs to be categorized to permit stratified random sampling, if appropriate, for subsequent data collection. For this reason, information was requested on (a) the medical school year (first, second, third, or fourth) of the participating students, (b) the number of hours per week and number of weeks that the students 50 participated in this ambulatory care teaching program, and (c) whether the program was required or elective. The percentage of the students in each class participating in each program was requested to determine if the program was of adequate size to warrant further study and to determine the feasibility of obtaining questionnaire data from students. The name and address of the faculty supervisor for each program and the nurse supervisor for each clinic program was required for the subsequent questionnaire mailings. A mailing list of all senior students was requested to permit random sampling of stu— dents for the questionnaire survey. A copy of the program identification survey instrument is in Appendix B. The Director of Curriculum of the Association of American Medical Colleges was consulted regarding the overall study. He endorsed the project and recommended that the Group on Medical Education Representative from each school be contacted for the program identification survey information. The survey form, a brief description of the research study, and a covering letter were mailed to the Group on Medical Education Representative of each of the ninety-six United States medical schools which would be granting M.D. degrees in 1973. A follow—up mailing was sent to those individuals who had not responded within one month of the initial mailing. ““i 51 1 Results of the program identification survey. The types of responses to the survey are presented in Table 3.1. Fifty of the ninety-six schools returned the completed forms and an additional seven schools provided useful information without completing the survey forms. Thirty— eight schools provided a mailing list of the students in their senior class. Table 3.1 Responses to the Program Identification Survey Number of Schools Percentage 1. Completed Forms Returned 50 52 2. Partial Data, but Forms Not Completed 7 7 3. Indicated Intent to Cooperate, but No Information Provided 4 4 ‘4. Refused to Participate 5 5 5. No Response 30 31 Total 96 99a aTotal percentage less than 100 due to rounding eezrror. The numbers and types of programs identified are listed in Table 3.2. A total of 276 programs were jvéieentified, which represented an average of 4.8 programs E3631: school. Sixty—two clinic programs and 26 preceptor— Eslldip programs qualified for further study. The reasons 15(3): programs to be judged not qualified for further S31:11dy because of specialty were that (a) They were 52 concerned with a surgical or other specialty field not included in the study, (b) they dealt exclusively with a medical subspecialty,6 or (c) they dealt exclusively with students in the "pre—clinical" phase of their education. Programs were disqualified for size if (a) they were elec- tive and less than 25 percent of the class participated in the program, (b) the general program was required but less than 10 percent of the class fulfilled the requirement in that specific clinic, or (c) the students spent less than three hours per week on a regular basis in the program. Table 3.2 Numbers of Programs Identified and Qualified for Further Study Percentage of gumber Of Total Programs rograms Identified 1.. Clinic Programs Qualified 62 22 2 - Preceptorship Programs Qualified 26 9 £3 - Not Qualified, Specialty 156 57 4:- Not Qualified, Size 32 12 Total Programs Identified 276 \_ _ 6After reviewing the results of the program iden— 1::1-f5ication survey it was decided that efforts to further $31911dy ambulatory care teaching in the medical subspecial— -t;3L13 students participating in individual programs. ~—_— 53 p Table 3.3 lists the frequency with which data requested in the survey form were not completed for those programs which qualified for further study. The only information which was commonly missing was the name and address of a nurse qualified to respond to a questionnaire about clinic programs. Table 3.3 Frequency of Incomplete Data on Qualified Programs Number of Percentage of Programs With Programs With Data Missing Data Missing 1. Medical School Year of Participating Students 2 2 2. Program Required or Elective 0 0 3. Percentage of Class Participating 3 3 ‘4. Student Time in the Program 3 3 5. Name and Address of Faculty Supervisor 0 0 6 . Name and Address of Clinic Nurse 27 44a aOf clinic programs. Based on the data from the survey, twenty-three <3:L:inic programs were selected for study by questionnaires t1<> students, as well as to faculty and nurses. The cri— t1'E=—‘1:‘ia for selecting these programs were: (1) a senior (3:1—613s mailing list was available; and (2) at least 50 per- czfiallt of the class participated in the program, or those S31211dents participating in the program were identified. huh—A 54 Selection of Variables and Qpestionnaire Design Those aspects of program design which were involved in the problems identified in the first phase of the study, plus additional program characteristics recognized as being significant in the case study phase, were used as a list of variables from which those suitable for further study by a questionnaire survey were selected. The criteria used for the selection of clinic program variables were: (1) the variable should be categorical or quantifiable; (2) the variable should be of importance in describing the nature of the program, or have a hypothe— sized relationship to program success; (3) all of the ‘Variable Groups from the model in Chapter 1 should be :represented; and (4) the number of variables selected should be such as to permit the student and nurse (guestionnaires to be limited to one page and the faculty caizestionnaire to about three pages. After the variables were selected, each variable vve1s assigned to one or more of the questionnaire forms, iféicrulty, student, or nurse, on the basis of expected com- £>€3tzence to respond to the question. Since student Sltleestionnaires were not used in all programs, variables 6153signed to it were also assigned to the faculty and/or I‘lllrse questionnaire whenever this was feasible. Some “V'Einriables were assigned to both the faculty and nurse 55 questionnaires to serve as a reliability check on reSponses. The variables selected and the questionnaires to which they were assigned are listed in Table 3.4. Because of the great variety in the specific activities of individual instructors and students in pre- ceptorship programs the variables selected for the study of these programs were restricted to (a) whether the program was required, selective, or elective, (b) program goals, and (c) two outcome measures, educational value and goal achievement. Pilot Test of Questionnaires Questionnaires were constructed based on the variables assigned to each of the questionnaire types, .faculty, nurse, student, and preceptorship. For each of 'the nine clinic programs studied in the case study phase, 51 faculty questionnaire was sent to the program director, at nurse questionnaire was sent to the clinic nurse, and sst:udent questionnaires to all, or a random sample of, sstzudents depending on the size of the program. A pre— Ceptorship questionnaire was sent to the coordinator of illiee preceptorship program included in the case study EDIléise. Participants were requested to critique the Silleestionnaire in addition to responding to the questions. Since the investigator was familiar with all of -t:k1<3 programs involved in the pilot testing process, it was EDC>Ssible to determine whether or not questions had been mi S interpreted . 56 Table 3.4 List of Variables for Study of Clinic Programs Patient Variables 1. 2. 3. Percentage of patients medically indigent. (N) Percentage of patients over age 60. (N) Percentage of patients with primarily psychological or psychosomatic problems. (F) Percentage of patients with problems not previously diagnosed. (F) Percentage of patients failing to keep appoint- ments. (N) Percentage of patients cancelling appointments and failing to schedule new appointments. (N) Selection or screening of patients assigned to students. (N) Patients are adults only or adults and children. (F)a Faculty Variables QWNH one. Willingness of faculty to teach in the clinic. (F) Are faculty full-time, part-time, or volunteer. (F) Seniority of faculty. (F) Percentage of teaching done by house staff. (F) Student Variables Percentage of students who are in their fourth year. (F) Student's career choice. (S) Required, selected, or elected program. (F) Number of students in program per year. (F) Process Variables l. 2. 3. Student time per new patient. (F, N, S) Student time per follow—up patient. (F, N, S) Student-faculty interaction time per patient. (F, S) Student time per new patient spent in writing the medical record. (F, S) Record written during or after clinic hours. (F, S) Number of patients seen per student during program. (N, S) Percenfiage of patients seen three or more times. (N. S) 57 Table 3.4 (continued) 11. 12. 13. 14. 15. 16. 17. Primary purpose of clinic (teaching students or patient care). (F, N) Presence of teaching conferences. (F) Types of teaching conferences. (F) Student time in clinic (hours per week, number of weeks). (F) Students’ degree of patient care responsibility. (F, S) Faculty—to-student ratio in the clinic. (F) Percentage of clinic patients seen by students. (N) Do faculty see patients without students during teaching clinic. (N) Percentage of patients seen by faculty without students. (N) Percentage of patients on whom a complete medical workup is performed. (F, S)c E. Program Goals (F) F. Program's Strong and weak Areas (S)a G. Outcome Variables 1. Education value to students of the program. (F, S) 2. Student enjoyment of experience. (F, S) 3. Degree of goal achievement. (F) aVariable added after pilot test of questionnaires. bVariable modified after pilot test of question- riaiires. cVariable deleted after pilot test of question— Iléiires. Note: F = Faculty Questionnaire; N = Nurse Questionnaire; S = Student Questionnaire 58 Responses to, and comments on, the questionnaires resulted in the following changes: (1) A question was added to the faculty questionnaire concerning whether the clinic dealt with only adults or adults and children; (2) As a result of students' remarks, a question was added to elicit students' impressions of the program's strong and weak areas and comments on these areas; (3) The question on the percentage of patients receiving a "com- plete medical workup" was deleted because of apparent ambiguity and the lack of importance of the question; and (4) The style of questions regarding the number of patients seen per student, and the average number of times each ‘patient was seen, were modified to facilitate responses. The response rate of students on the pilot survey xMas 67 percent. This response rate was used in determin- inmg the number of students per program to be sampled in tflae subsequent, main questionnaire survey. Because the changes to the questionnaires between 1:11e pilot and main surveys were considered to be minor, t:Ileprograms studied by the pilot questionnaires were not JTGBssurveyed with the final form of the questionnaires.7 _ 7C0pies of the pilot and final sets of question- 1iat-‘Lres are contained in Appendix B. 59 ? Main Questionnaire Survey For each of the fifty-three clinic programs selected for further study and not involved in the questionnaire pilot test, a faculty questionnaire was sent to the faculty supervisor previously identified and a clinic nurse, if a nurse was identified in the program identification survey. If no nurse was identified, the nurse questionnaire was also sent to the faculty super— visor with the request that it be passed on to the appro— priate nurse or other clinic staff member. Student questionnaires were sent to a random sample of final year students at the institutions involved with the fourteen programs selected for the collection of data from students and not involved in the questionnaire pilot test. For gprograms in which 90 percent or more of the school's :students participated, twenty—five students received gluestionnaires. If 50 to 90 percent of the students ‘Emarticipated in the program, forty students were sur— xz‘eyed.8 Faculty and nurses, who had not responded within tlkrree weeks of the first mailing, were sent follow-up letters and additional questionnaires. Three faculty ‘nnfalnbers who supervised programs in which students had \ 8The sample sizes were chosen in an attempt to eIlsure at least ten usable student responses per program, 3h>Eised on an estimated response rate of 50 percent for ‘tikusse students who had participated in the program. 60 been sampled did not respond within three weeks of the second mailing. These faculty members were contacted by telephone and urged to cooperate. The response rates achieved in this survey are presented in Chapter 6. Coding and Consolidation of Questionnaire Data All responses were coded and key—punched on data cards for computer analysis. The coding procedure is described in Appendix C, Table C.1. Consolidation of clinic program data. Since data on a number of variables were obtained from more than one source per clinic program and most of the planned analyses involved the program as the unit of interest, it was necessary to consolidate the data so that there was a single measure on each variable for use in analysis. Measures of central tendency and dispension and a fre— quency distribution were determined for each type of data source (faculty, nurses, and students) on those ‘variables involved. The results of this analysis are ‘presented in Appendix C, Tables C.2 and C.3 and Figures (3.1 and C.2. The agreement among the different sources <3n.the descriptive, or independent, variables was con— Eiidered adequate to indicate that the reliability of 13esponses was satisfactory. Since faculty data were Etvailable on more programs than were nurse data and Ilurse data were available on more programs than were 61 student data, it was decided to use the data on each variable by faculty sources, if available. When faculty data on a given variable were not available nurse data were used, and if nurse data were not available student data were used. This procedure produced the maximum possible consistency of the source of data on a given independent variable across programs, while providing measures on the largest possible number of variables. An intercorrelation matrix was formed using the student and faculty ratings of the outcome variables for clinic programs with student data.9 No statistically significant correlations were found between student and faculty ratings of the outcome variables. The faculty and student ratings of both educational value and student enjoyment were, therefore, treated as separate measures. Outcome indices and groupings. To facilitate analysis, outcome measures were combined to produce out— come indices, which were used as dependent variables. Faculty ratings of educational value, student enjoyment, and goal achievement were averaged for each clinic program, ‘and called the Faculty Success Index. A Student Success Itndex was formed by averaging the students' ratings of :ries, the program was assigned to the category containing e top-ranked goal and the next highest ranked goal in either of the overlapping categories; unless the second, 64 Table 3.5 Group Designation of Goals "Continuity Goals" "Clinical Goals" "Practical Goals" ————— Goals (1) Demonstrate the continuity of the doctor-patient relationship. r________ (2) Demonstrate the natural history of chronic diseases. (3) Demonstrate the pathophysiology of disease at a given point in time. (4) Provide experience with diseases which do not commonly lead to hospitalization. (5) Provide student with experience to assist in career choice decisions. (6) Provide student with instruction and experience in abbreviated workups and seeing a volume of patients. (7) Involvement in a health care delivery system other than a university medical center. 65 tertiary set assignment was the group of goals containing no goal ranked higher than any goal in the other two groups. This categorization system would permit the for- mation of six mutually exclusive goal groups,1 however, because of the relatively small sample size it was decided to use the primary and tertiary groupings separately in -the analysis procedures. The same procedure was used to assign preceptor- sship programs to primary goal groups. Since all but two Iareceptorship programs were in the same tertiary goal «group, it was decided that this grouping would not be of \nalue for purposes of analysis. IData Analysis The analysis of the data obtained from the <1uestionnaire survey was carried out for three different SIeneral purposes: (1) To describe all programs and sub— SSets of programs on the relevant variables; (2) To deter- ‘hnine the degree of association between different variables; 61nd (3) To test the hypotheses generated during the first \ t:hird, and fourth ranked goals were all in the same cate— Eyory, in which case that category determined the primary sgoal group. A similar procedure was used to determine the 1:ertiary set assignment in borderline cases. 13The six mutually exclusive groups would be (a) primary continuity, tertiary clinical; (b) primary (zontinuity, tertiary practical; (c) primary clinical, ter- 1:iary continuity; (d) primary clinical, tertiary practical; (e) primary practical, tertiary continuity; and (f) pri- Inary practical, tertiary clinical. 66 two phases of this study. These analyses were designed with the assistance of the Office of Research Consultation of the College of Education at Michigan State University. Program profiles. The means, standard deviations, and frequency distributions of coded responses on all ‘variables were determined for (a) all programs, (b) pro- grams with student data, (0) programs without student ciata, (d) the various faculty and student success groups, and (e) each primary and tertiary goal group. Findings <:onsidered to be of general interest are included in (Shapter 6 and Appendix E.14 Correlations between variables. A series of inter— <2orrelation matrices were formed to determine the cor- :relation coefficients between the numerous variables Ioeing studied.15 Descriptive, or independent, variables, \Nhich correlated significantly with one or more outcomes Ineasures in any of the intercorrelation matrices, were :submitted to stepwise regression analyses against the zappropriate success index. 14Findings not included in Chapter 6 or Appendix E (are available on request. 15Because there was evidence that the programs ‘with student data differed significantly from those with- ‘out student data, separate intercorrelation matrices were formed for these two groups of programs in addition to the matrix for all programs. 67 a Tests of hypotheses. The seventeen research hypotheses generated during the first two phases of this study and the statistical techniques used to test the appropriate null hypotheses are presented in Table 3.6. The 95 percent confidence level (p < .05) was used in (determining the statistical significance of the test sta— tistic in all cases. Comments on data analysis procedures. Although it is considered justifiable to employ multiple analytic Inethods in an exploratory study, it must be recognized ‘that the stated level of significance may be misleading. lFor example, in this study the relationships between some ‘Jariables were analyzed by three methods, thus tripling ‘the probability of reporting a significant difference, VVhich was in fact due to random chance. The importance <>f identifying significant relationships in an exploratory sstudy excuses some reduction in the level of confidence fiLn the significance of the findings. Some of the assumptions of the statistical tests Gamployed were knowingly violated. The data on many of 1:he variables were not normally distributed. The Eissumption of independence was respected, except that 1:he rankings of different goals were treated as if they Vvere independent of one another. In some cases data Vvhich might be considered to be on a nominal scale were 68 Table 3.6 Research Hypotheses and Methods of Statistical Testing Research Hypothesis Statistical Test Used Faculty ratings of educational value more closely approximate student ratings of student enjoyment than student ratings of educational value. Students planning to enter pri— mary care fields (general prac— tice, family practice, internal medicine, or pediatrics) rate clinic programs as being of higher educational value than do students planning to enter other medical fields. Selective and elective clinic programs are rated higher by students than required pro— grams on both educational value and student enjoyment. Program success is not uni— form across goal groups for both clinic programs and preceptorship programs. Faculty ratings of the success of clinic programs (Faculty Success Index) are: a. Positively correlated with student participation in patient care decisions. b. Positively correlated with the percentage of patients whose primary problems have not been previously diagnosed. c. Negatively correlated with the percentage of patients over age 60. Test of significant difference between two correlation coef— ficients. t-test of difference between means (indepen- dent samples, one— tailed test). Analysis of variance. Chi—square analysis of goal groups and success (educational) groups for clinic and precep— torship programs. Significance of a cor- relation coefficient. Significance of a cor— relation coefficient. Significance of a cor- relation coefficient. 69 Table 3.6 (continued) Research Hypothesis Statistical Test Used Negatively correlated with the percentage of patients who are medically indigent. Negatively correlated with the number of students per faculty member in the clinic. Positively correlated with the percentage of partici— pating faculty who are full-time faculty. Positively correlated with the seniority of partici- pating faculty. Positively correlated with the willingness of faculty to teach in the clinic. Positively correlated with the percentage of students in the program who are fourth-year students. Negatively correlated with the percentage of patients with problems which are primarily psychological or psychosomatic. Negatively correlated with the percentage of patients failing to keep appoint— ments or to cancel the appointments. Positively correlated with the presence of con— ferences. Higher if the primary pur- pose of the clinic is to provide patient care or graduate education rather than to teach medical students. Significance of a cor- relation coefficient. Significance of a cor- relation coefficient. Significance of a cor- relation coefficient. Significance of a cor— relation coefficient. Significance of a cor- relation coefficient. Significance of a cor— relation coefficient. Significance of a cor- relation coefficient. Significance of a cor- relation coefficient. Significance of a cor- relation coefficient. t-test of difference between means (inde- pendent samples, one- tailed test). 70 5 treated as ordinal or better. For example, whether a program was required, selective, or elective was treated as three levels of "requiredness." Summary Since the available literature was considered insufficient to guide the design of an exploratory, descriptive study of adult ambulatory care teaching, a three—phase methodology was used in which the results of the earlier phases were used to assist in the design of the later phases. The first phase of the study consisted of a series of problem identification sessions with medical school faculty and students. The nominal group process was used in conducting these sessions. The next phase of the study involved interviews at three medical schools with administrators, faculty, nurses, and students representing ten adult ambulatory care teaching programs. The selection of aspects of the programs to study was guided by the results of the problem identification sessions. This phase resulted in a series of case studies and assisted in the identification of appropriate variables and hypotheses for use in the questionnaire phase. The first component of the questionnaire phase was a survey of the ninety—six established United States medical schools which identified sixty-two clinic programs 71 and twenty-six preceptorship programs appropriate for further study. Questionnaires were designed for com- pletion by faculty supervisors of clinic programs, clinic nurses, participating students, and preceptorship coordi- nators. These questionnaires were pilot tested on the ten programs involved in the case study phase. After minor revisions, questionnaires were sent to faculty and nurses representing the remaining programs to be studied. Questionnaires were also sent to a random sample of stu— dent participants in twenty—three of the clinic programs. Responses were coded and consolidated for computer analysis. Each program was assigned to a series of groups based on (a) faculty and (b) student ratings of the programs' success, and (c) on faculty indication of the program's goals. The data analysis was of three types: the description of all programs and subsets of programs on relevant variables; the determination of the degree of association between variables; and the testing of seven- teen hypotheses generated during the first two phases of the study. Chapter 4 RESULTS OF PROBLEM IDENTIFICATION PROCESS (PHASE ONE) During February, 1973, four nominal group problem identification sessions were held with faculty and stu- dents from two medical schools. The primary reason for including this phase in the study was to provide infor- mation for the design of Phases Two and Three. This chapter consists of the results of these problem identifi- cation sessions. Composition of the Groups Separate problem identification sessions were held with one group of faculty and one group of students at each of two medical schools. The number of participants in each group is shown in Table 4.1. Six faculty members per school and eight students per school were invited to participate in the process. The participating faculty from School One con— sisted of members of the Department of Medicine. They were all full—time faculty physicians with varying amounts and types of prior experience in ambulatory care teaching. 72 73 None of them were currently involved in an organized ambulatory care teaching program. The faculty from School Two consisted of two internists, a family phy— sician, a pediatrician, and a psychiatrist. The family physician was a volunteer faculty member and the others were full—time faculty. All of them had been involved in various ambulatory care teaching programs. Table 4.1 Composition of Problem Identification Groups Faculty Students School One 6 4a School Two 5 3 aOne of these students did not participate in the priority setting phase of the process. The participating students from School One had all spent one half day per week for over a year in a comprehensive care clinic. The students from School Two had been involved in various out—patient clinics, mostly emergency rooms and specialty clinics. Lists of Problems The output of each group session was a list of problem statements generated by group members in response to the request, "List the subjective and objective problems you have experienced or perceived as a medical student in an adult ambulatory (out-patient) clinic setting"; or 74 "List the subjective and objective problems you have eXperienced, perceived, or anticipate as a faculty member teaching in an adult ambulatory (out—patient) clinic set- ting." From the pooled list generated by each group, each member selected up to ten problems which he felt were most important and distributed one hundred points among these problems in prOportion to his perception of their importance. After the group meeting, the points assigned to each problem by different group members were summed and the problems were rank ordered by the total assigned points. Each problem was designated by the investigator as relating to one of the variable groups, Patient, Faculty, Student, or Process, from the model presented in Chapter 1.1 A total of 116 problem statements was generated by the four groups. The list of the tOp ten problems from each group, the ranking of these problems, the total points assigned to them, and the variable groups to which they refer are presented in Tables 4.2 through 4.5.2 Categorical Grouping of Problems The problem statements generated by the four groups were sorted into categories. The total points assigned to 1The assignment of problems to variable groups necessarily was somewhat arbitrary. Some problems assigned to the Process Variable group related more to the environ— ment than to the process itself. One problem was assigned to both the Student and Faculty Variable groups. 2The complete lists of problem statements are in Appendix C, Tables D.l through D.4. 75 Table 4.2 Major Problems Listed by Faculty at School One Rank Points Variable Problem Group 1 126 Process The achievement of continuity of care between student and patient. 2 44 Patient The selection of patients to achieve a representation of a cross-section of the total popu— lation of patients. 3 40 Process The lack of identification of the physician primarily responsible for patient's care. 4 33 Process The difficulties in the scheduling of patients, preceptors, and facilities. 5 31 Process The patient spends an unreasonably long time in the out-patient clinic in order to receive care. 6 30 Process The integration of primary care with specialty care clinics. 7—8 25 Process Establishing an esprit de corps among faculty, students, and patients in the ambulatory care clinic. - 25 Process An impairment of the physician - patient relationship in the teach- ing clinic setting. 9-10 22 Patient Acquisition of an adequate patient population base for a teaching clinic. - 22 Process Establishment of a financial basis for the clinic operation ("Who pays?") 76 Table 4.3 Major Problems Listed by Students at School One Rank Points Variable Problem Group 1 103 Patient The "no—show" patient, or patient arriving late for an appointment. 2 30 Patient The low proportion of patients with significant organic disease. 3 28 Process An inadequate amount of learning for the time commitment involved. 4 18 Process Patients frequently have a long wait while the students consult with preceptors. 5 17 Patient Lack of adequate patient mix on socio—economic basis. 6-8 16 Process Patients hospitalized from the clinic are frequently admitted to a hospital other than the one where the student is primarily assigned. - 16 Process The hospital administration's opposition to birth control and abortion. - 16 Process Inappropriate social worker input. 9 15 Process Lack of continuity of patient follow-up due to the student tak- ing a clerkship outside the imme— diate geographic area. 10 12 Patient Community physicians are sometimes reluctant to excuse students from clerkship responsibilities to see patients in the clinic. 77 Table 4.4 Major Problems Listed by Faculty at School Two Variable Rank Points Problem Group 1 105 Process Who should pay for faculty teach— ing time in the ambulatory care setting? 2 60 Process Conflict between the teaching and service demands on the organi— zation. 3—4 40 Faculty Difficulty with time availability, quality, and interest of the teach— ing staff. - 40 Process Inadequate facilities for teaching adult ambulatory care. 5 30 Process Lack of acceptance of primary care as an entity. 6-8 20 Process Inadequate supporting services in the ambulatory care teaching set— ting. - 20 Process Poor continuity of patient care and student learning in the ambulatory care teaching setting. - 20 Process Lack of good teaching methods for ' ambulatory care instruction. 9 19 Student Student disinterest in learning in the ambulatory care setting. 10 18 Process Medical record systems in use were not designed for the ambulatory care situation. 78 Table 4.5 Major Problems Listed by Students at School Two Rank Points Variable Group Problem 10 45 43 25 25 25 22.5 20 20 14.5 11 Process Process Process Process Process Faculty Process Process Process Process The lack of adequate patient follow-up by the student results in the lack of feedback in a "trial and error" learning setting. The lack of orientation of the stu- dent to the support services available in the clinic and the community. A data review problem exists when a follow-up patient is seen by a new student. No available summary of medical records. Excessive use of subspecialty clinics leads to fragmentation of ambulatory care. Lack of faculty (as opposed to house staff) teaching in the clinics. Lack of adequate follow-up to establish optimal rapport between student and patient. Lack of adequate follow-up between clinic visits and hospitalization of patients. Emergency room is inappropriately used for primary care. Inadequate time in clinic setting for students to deal with the problems of patient education. 79 problems in each category were then used to rank order the categories. The ranks, total points, and categories are shown in Table 4.6. The top ranked category related to concern for the continuity of care both for the benefit of the patient and for the education of the student. Analysis of Problems by Variable Groups As indicated previously, each problem statement which was generated was assigned to a variable group based on the model presented in Chapter 1. Problem statements were assigned to a variable group for all problem identifi— cation groups, and for faculty and student groups separ- ately (see Tables 4.7 and 4.8). Even though the problems relating to the Process Variables constituted the great majority of problem state— ments, the average number of points per problem across variable groups was quite similar. Table 4.8 shows that although both students and faculty produced a predominance of problems in the Process Variable group, students formu- lated more problems in the Patient Variable group and faculty perceived more problems in the Student Variable group. Summary A problem identification process to identify areas of difficulty in adult ambulatory care teaching was carried out primarily to assist in the design of Phases Two 80 Table 4.6 Problem Categories by Rank and Total Points for All Groups Rank Points Category 1 341 Continuity of care; within the clinic, between clinic and hOSpital, and among various clinics. 2 207 Ensuring good patient care in the clinic. 3 177 Financial—~cost problems; especially cost to the patient. 4 172 Scheduling of patients, faculty, and stu- dents, including "no-shows." 5 134 Logistics, including transportation, park— ing, records, and efficiency of performing diagnostic procedures. 6 133 Acquisition and selection of appropriate patients for a teaching clinic. 7 113 Physician-patient relationships (rapport) in the clinic setting. 8 106 Esprit de corps in the clinic; student interest and academic atmosphere. 9 105 Faculty; including acquisition, time commitment, and adequacy for supervision of residents and students. 10 80 Nature of ambulatory care; lack of research base and proven value. 11 56 Definition of students' roles and responsi- bilities. 12 48 Communications between the clinic and referring physicians. 13 36 Utilization and training of allied health personnel in the clinic. l4 0 Planning of the ambulatory clinic's role in the medical curriculum. 81 Table 4.7 Number of Problems, Total Points, and Average Points per Problem for the Problems in Different Variable Groups . Number of . Average Points Variable Group Problems Total POints per Problem Patient 16 257 16 Faculty 6 97.5 16 Student 6 73 12 Process 89 1,272.5 14 Total 117a 1,700 15 aOne problem was assigned to both Student and Faculty Variable groups. Table 4.8 Number of Problems, Total Points, and Average Points per Problem for Different Variables by Student and Faculty Groups Students Faculty Variable % of Group Number Total Stu— Number Total % of Of Points dent Of Points Faculty Problems . Problems Points P01nts Patient 8 171 28 8 86 8 Faculty 2 27.5 5 4 70 6 Student 0 O 0 6 73 7 Process 32 401.5 67 57 871 79 Total 42 600 100 75 1,100 100 82 and Three of this study. One faculty group and one stu— dent group at each of two medical schools participated in the problem identification process. One hundred sixteen problem statements were produced. The problem statements and assigned priorities were presented for each participant group. Problems regarding concern for the continuity of patient care and continuity of the student—patient relationship received the largest number of points. Problems included among the Process Variable group of the model presented in Chapter 1, received the most attention from both students and faculty. Students demonstrated more concern than faculty with the Patient Variables, while faculty showed more concern with the Student Variables. Chapter 5 THE TEACHING OF ADULT AMBULATORY PATIENT CARE IN THREE MEDICAL SCHOOLS During March, 1973, the adult ambulatory care teaching programs at three medical schools were studied. The specific nature of the issues investigated was guided by the results of the problem identification process reported in Chapter 4. The main purpose of this second phase of the study was to clarify what areas were to be investigated in the final, questionnaire phase of the study. This chapter consists of brief case studies of the adult ambulatory care teaching at three medical schools and their ten programs, which were selected to demonstrate the diversity of approaches used in this teaching. Table 5.1 contains selected comparative data on the nine clinic programs. Adult Ambulatory Care Teaching at University A University A is an independent, nonsectarian institution whose medical school is more than fifty years old. The current ambulatory care teaching programs relate 83 84 .mwmmwcaaw vasouno msofluw> mo wmoaoflmxsmonumm 0:» mo :mwwvsum amaommma mo xmmB Hod when 03» haoumfiaxoummm.mwvsHocH Edumoum mwsao .mvnun menu we wanna dunno ca cocoaocfi >m>usm owfimccowumosv uawpsum Gnu co woman new pmumfla muscflao Hmzuo man on w>qumamu mum mmcflumu amonan .05 x x3\mmmm .m.w .mfiusofi mo amass: on» How xmoz mom mcoflwmom mo Hogans may no ooswfluwmxo oEHuIHHsm mo mausoe as we moswwummxm m.uswosum Gnu mo sowuwusp ones .oae on .02 NIH x x3 .nudd< “one: oeooaz caeaao oumuoooz now: ocoz mud Ina \.mmwm a .oz touso< lumsoq nomsso» .002 .Emm .GHE cm .05 a x x3 owsaao 30A 30a 0:02 Nua tom \.mmmm N wovnom oasoufio Hosea uwoao .Umz .suu .OE H x x3 HHmEm 30a oumumpoz 0:02 and nun: \mmmm mus >um> omxfiz some: omxfiz =.sfiHU .omm: o Nuwmuw>wco =.aaHo .omem nae: xuo> now: muo> omsoumsdz Nuana .ceE oma .moE N wma UHQOHEU uwsoq umpao .mmom .00: .cas oNH =.:aao .ommm soar mumnoooz .x3\a msmua :om .moE N womuoa oesounu uwzoq Hooao .mmom .>H:D: m Nuwmum>wco .05 m x x3 Home: owcfiao 304 mumuoooz .mmwm\a NHH .sfle om \.mmmm a wow .noxmm matte: momma .ommaood .06 m x x3 Hose: Eduoowm swam oumnmvoz 0:02 and .swE om \.wmmm a woe .ousod ooxaz Umxwz .pmz .Emm .06 m x x3 . owsflHU mumumcoz mumwmpoz mcoz mud .CHE om \.mmom a mom oaconsu meson umoao Hmowomz .oE m x x3 wumuocoz swam .mmmm\a NHH .sHE om \.mmwm a wmmnom encouno umsoq umoao =.cwao .oomm: e Nuwmuo>wso uaosaonsm msam> mwmwwz mwmwwww mmwww msouu ucwpsum Hosowuuosom oaumm musmwuom .Homxm u 4 u. m mod . . no w noun nofloom owcflao msoo sum 3&2 nuwz no a mudmum>aca nowswao 0:» .00h mafia sum msoaumudo manoeumm no mocwuum acousum medumoum owsflao so sumo o>euwummsoo H.m wanna 85 back to the late 1950's when the school instituted a com— prehensive ambulatory care teaching program that was a required experience for all students. A special clinic was established for this program. It was coordinated by the Department of Preventive Medicine and the faculty was obtained from virtually all clinical departments. Medical students devoted one half-day per week to this experience during their senior year. The teachers were predominantly from the Department of Medicine, but consultants in psy— chiatry, gynecology, and surgery were available in the clinic. The clinic was organized so that each student saw a new patient one week and follow-up patients the next week. This program, demonstrating a genuine concern for ambulatory care teaching, had the support of the chairmen of clinical departments and was considered to be an impor- tant experiment in medical education. The physicians actually involved in the clinic—based teaching were some— what less enthusiastic, but the "experimental atmosphere" was maintained and the program worked very well for about five years. During the next ten years the program per- sisted, but enthusiasm was less vigorous. The consultants in psychiatry, gynecology, and surgery were "on call," and later, available by appointment. Other clinics, such as rheumatology, hematology, community-based clinics, and preceptorships with individual physicians became available 86 as options to fulfill the required comprehensive ambulatory care experience. The enthusiasm of working in an experi— mental clinic seemed to have run down. In 1972 the requirement of a prolonged (one year) experience in ambulatory medicine remained, but the oper— ation of the special clinic had been assumed by the Department of Medicine and two—thirds of the students were having their ambulatory care experiences in settings other than this clinic. Table 5.2 indicates where the students were obtaining their continuity ambulatory care experiences during the academic year 1972—73. Nearly 10 percent of the students were completing this require— ment with a more intensive, shorter duration, block time experience rather than the original half—day per week for an academic year. This block time experience is con- sidered by program supervisors to be less desirable, but it is considered necessary for some students who Spend much of their senior year away from the university medical center. The faculty is considering revising the curricu— lum so that the students have more experience in history taking and physical examination during their second year, thus facilitating the introduction of ambulatory care experiences in the students' third year of medical school. The "Special" Clinic This clinic was developed and operates as a medi— cal student teaching clinic. Approximately five students 87 and two or three faculty members participate in each clinic session. There are four half—day sessions per week, each with different students and faculty. Table 5.2 Location of Ambulatory Care Teaching for University A Senior Students, 1972-73 . Percentage Location of Class "Special" Clinic 32 Preceptorships with Community or University Physicians 22 Specialty Clinics and University Health Service 14 Family Medicine Program 12 Medical Clinic 9 Adolescent Clinic 7 Neighborhood Health Center 4 Total 100 Patients. The patient population using this clinic is predominantly from the lower socio-economic classes. The patients seen by the students undergo no formal screening or selection process, however, one observer stated that they tend to be the patients that are not selected for continuing care by the house staff (interns and residents). The students described the patients as mostly elderly, obese women with hypertension, diabetes, and/or arteriosclerotic heart disease. In about half of the patients the primary problem is a 88 psychosomatic or emotional disorder. Twenty to 25 percent of the patients fail to keep their scheduled appointments. Faculty. All of the faculty who participate in this clinic volunteer for it, and three-quarters of them are full-time university faculty. There is a core of faculty participants who are full—time faculty primarily interested in ambulatory care teaching. Clinic Operations. At the beginning of the year, the students are given an orientation to the clinic oper- ation by the clinic nurse. Each student sees a new patient and, perhaps, a follow—up patient, or three to four follow—up patients each clinic session. The students spend about one and one—half hours with new patients and forty-five minutes with follow—up patients. The typical student sees twenty to thirty different patients over the year, and on the average sees each patient about three times. A faculty member sometimes observes part of the student's history taking and physical examination, but always reviews the patient's problems with the student and talks with the patient. During a typical clinic session, each student Spends about forty-five minutes with a faculty member. The student may review a new patient with any one of the faculty members present, but with follow—up patients he sees the same faculty member 89 that previously saw the patient. In this way, students are exposed to different faculty members, but with a specific patient the student-faculty relationship is consistent over the entire year. If a patient telephones or comes to the clinic when his student—faculty team are not there and the problem cannot wait until the next regular clinic session, the student and/or faculty member are contacted. On about half of these occasions the student sees the patient. The faculty members are also available on call during nights and weekends to handle these "out—of—clinic hours" patient coverage situations. Either prior to, or following, each clinic session a conference is held. These are lecture/discussion sessions conducted by one of the faculty participants or another faculty member. The majority of the con— ferences are on medical conditions common in ambulatory care. Some conferences are devoted to discussions of common psychosomatic problems, the economics of office practice, and other related subjects. When the conference is at the end of the clinic session there is some diffi- culty created by the fact that the students do not all finish with their patients at the same time. Student Reactions. The students' main complaints about the clinic eXperience concern the lack of variety among the patients with regard to age, socio—economic 90 class, and type of medical problems. The students rated this clinic experience as moderately enjoyable and of significantly greater educational value than the average of their clinical experiences. Medical Clinic In contrast to the "Special" Clinic, which operates primarily to teach medical students, the Medical Clinic's mainfunctions are to provide patient care and teach interns and residents. Although the clinic functions throughout the week, medical students are assigned to it for only four half—day sessions per week. There are usually two students assigned to each of these sessions. In addition to the student there is a nurse practitioner, an intern, and several residents working in the clinic. They are all supervised by the one or two faculty members assigned for that session, but the faculty Spend more time with the students than with the house staff. The patients in the Medical Clinic are from the same population as those in the "Special" Clinic, but since there are several types of "practitioners" in the Medical Clinic it is possible to select the appropriate types of patients for each type of "practitioner." Patients with common, chronic medical problems who have not been seen in the clinic for more than one year are usually assigned to the medical students. Patients with 91 obvious psychosomatic complaints are assigned to the nurse practitioner, and the patients with more complex or unusual problems are referred to the residents. Since the program goals, patient p0pulation, student characteristics, and participating faculty are the same, or nearly the same, in both clinics, it is not surprising that the clinics' operations are very similar. The only major differences in the clinics' operations are that the Medical Clinic program does not include conferences or the "out of clinic hours" patient coverage system. Student reactions to the Medical Clinic experience are nearly the same as their reactions to the "Special" Clinic. Their complaints are similar and their ratings of the enjoyment and educational value of the program were not significantly different. Family Medicine Program The Family Medicine Program, although an integral part of the university's programs, is geographically located at an urban community hospital. The continuity ambulatory care teaching is only one of several teaching programs carried out by this family medicine education group. Students in the continuity program spend one half“ day per week in the Family Medicine clinic. 92 Patients. The patients in the Family Practice Program approximate a cross section of the community population. In fact, there is a slightly higher percentage of the upper socio-economic classes than in the general population. Only 15 percent of patients have emotional or psychosomatic problems as their chief complaints, and only 10 percent of patients fail to keep their appointments. Patients present more acute, but relatively minor, problems in this clinic, most of which can be managed during a single clinic visit. The patients include children as well as adults. Clinic Operation. Each student in the continuity program is assigned to work with one staff physician. Eight or nine patients are scheduled to see this team during a half-day. Suitable patients are assigned to the student. The staff physician sees patients while the student is taking medical histories and doing physical examinations. The student usually sees about half of the patients, spending approximately one hour with each new patient. The student and staff physician confer for about ten minutes regarding each of the student's patients. Although conferences are held for trainees in this clinic, they do not necessarily coincide with the times that the continuity program students are in the clinic. 93 Student Reactions. Students commented that the relatively simple nature of the patients' problems in this clinic resulted in few follow—up visits. This lack of a continuing relationship with the patients resulted in a level of rapport with the student that was less than his expectations. Students also complained of the lack of an arrangement to see their patients when problems occurred outside the student's clinic hours. Students rated the Family Medicine experience as more enjoyable than the "Special" Clinic and Medical Clinic, but of equal, or slightly lower, educational value. Adolescent Clinic The Adolescent Clinic was founded about ten years ago in response to the need for a referral clinic to deal with problems of adolescents. Since that time the clinic has become increasingly involved in teaching. Currently participating in the clinic are trainees from nursing, social work, and clinical psychology, as well as five medical students. Patients. The patients are predominantly upper middle class adolescents referred by private physicians or school counselors. The primary diagnosis is an organic disease in only one—quarter of cases, the remainder being psycho-social disorders. Approximately 20 percent of patients fail to keep their appointments. 94 Faculty. The participating faculty, drawn from the area of pediatrics, psychiatry, psychology, and social work, are, in general, very enthusiastic about the clinic program. Clinic gperation. This clinic is currently held one evening each week. It previously met on Saturday mornings. Each medical student sees a new patient and a follow-up patient, or two to three follOWvup patients during each clinic session. The student spends about one hour with new patients and one—half to one hour with follow—up patients. During the year each student sees about fifteen to twenty different patients, and sees each patient an average of five to six times. The ratio of trainees to faculty in the clinic is two to one. The students devote considerable time to this program outside of regular clinic hours. A few days before the clinic session the student receives a list of his appointments and the referring diagnosis and/or a brief history. He is expected to do some reading on the subject prior to the clinic session. The student's records of clinic visits are usually written outside of clinic hours, and he is expected to visit the home and/or school of many of his patients. Prior to each clinic session a one—hour conference is held on a subject related to problems of adolescence. 95 Student Reactions. Students seem to view this clinic as a unique opportunity to deal with adolescents, but some students find their own proximity to this age group and its problems somewhat stressful. Some students complained that the advanced information sheets on new patients were not adequately complete and accurate to be useful in studying or planning prior to seeing the patients. It was pointed out that the varied backgrounds and interests of the different types of trainees in this clinic presented some difficulty in making the conferences valuable to all trainees. Students rated this program as having educational value equivalent to the programs pre- viously described, but, on the average they rated their enjoyment of this program somewhat lower than students did in the other clinics.l Other Programs Half of the students at University A have their required ambulatory care experiences in a variety of settings other than the programs described previously. These settings were listed in Table 5.1 (page 84) but were not included in this study. In addition to the required ambulatory care experience, each student must spend one month in the 1The reliability of these observations can be questioned because of the small number of students involved in this program. 96 university hospital Emergency Department. Approximately half of this time is Spent in dealing with acute, minor problems in the fields of internal medicine and pediatrics. An intensive series of teaching sessions are included in the Emergency Department program. Adult AmbulatoryyCare Teaching at University B University B is a privately endowed institution whose School of Medicine is over one hundred years old. In the 1950's the School of Medicine instituted a three- part program for the teaching of ambulatory care. During their first two years all students participated in the Family Study Program. This Program consisted of each student becoming acquainted with a woman late in pregnancy and following the development and medical care of the infant. The second component of the ambulatory care teaching was a Continuity Program in which the student participated in the care of a group of selected, chroni- cally ill patients during one-half day per week for one and one—half years. The third part of the overall program was a two-month experience for fourth-year students in a special ambulatory care clinic. Since the inauguration of the three—part program, the Family Study Program and the "Special" Clinic experience have undergone only minor modifications. After about fifteen years the Continuity Program was 97 .made an elective experience. Since that time student participation in the Program has gradually decreased, and it is anticipated that the Program will be discon— tinued after the current year. The demise of the Continuity Program appears to have been due to two factors. Although the segment of the faculty who were intimately involved in the Program were quite enthusiastic, a sizable pr0portion of the faculty, including some key departmental chairmen questioned its value relative to other clinical exper— iences. The second factor, which was no doubt related to this faculty attitude, was that students found the Continuity Program competing for time and effort with their other, concurrent assignments. Since the other assignments consumed the vast majority of their time, it is not surprising that students tended to assign second priority to the Continuity Program. Although many members of the faculty and admin- istration value ambulatory care teaching highly, the general attitude of the faculty seems to be that student experiences with hospitalized patients are more important and that teaching students on the wards is more pres- tigious than ambulatory care teaching. The Family Study Program will not be discussed in detail since it involves students in their "pre- clinical" years working in the areas of pediatrics and 98 obstetrics, which is beyond the scope of this study. The remainder of this section deals with some of the details of the organization and operation of the two major "Special" Clinics, one at "University Hospital" and the other at "County Hospital." "Special" Clinic at "Uni— versity Hospital" "University Hospital Special Clinic" was designed and operates primarily for the teaching of medical stu— dents. Approximately two-thirds of the senior students have their "Special Clinic" experience at "University Hospital." During the two months that students spend on this rotation they devote five mornings per week to seeing patients in the clinic, four afternoons per week in Ophthalmology, otolaryngology, and dermatology clinics, and one afternoon per week in the "Special" Clinic Con— ference. The students may elect to spend one day a week in a Specialty clinic, rather than the "Special" Clinic. Patients. The patients served by this clinic consist predominantly of the medically indigent, inner- city black population. Patients are referred to this clinic from other clinics on the basis of the need for a thorough evaluation of a condition which is commonly seen in ambulatory patients. About 10 to 20 percent of the patients fail to keep their appointments. 99 Faculty. The faculty in this clinic are described as "loyal, dedicated semievolunteers." (The majority of the participating faculty select this clinic to fulfill the teaching requirement for maintaining hospital admit— ting privileges.) The faculty is composed of one-quarter full-time faculty, one or two senior residents, and phy— sicians in private practice (volunteer faculty). In addition to internists, there is a psychiatrist present at each clinic session and consultants in surgery, gynecology, and dermatology are available "on call." Clinic Operation. A student sees one new patient, or two or three follow—up patients, each morning. One and one-half to two hours are spent with each new patient, plus about thirty minutes discussing the patient with a faculty member. The student performs a white blood cell count, hematocrit, blood cell morphology, and urinalysis on each new patient. On alternate days the student sees follow-up patients. If a student's schedule is not filled, he may see a patient with an acute, minor illness referred by the "Screening Clinic." The scheduling of patients is arranged so the student sees patients for follow—up on the same day of the week as he saw them the first time. This permits the same faculty member to follow the patient with the student. During two months the student sees fifteen to twenty new patients and sees each patient an average of about three times. 100 There are two or three students per faculty member in the clinic with a different group of faculty each day of the week. The weekly clinic conference is of about two hours duration and is attended by the students, two internists, and a psychiatrist. The session is organized around one of the student's patients, or a group of patients with similar problems. Frequently a student role—plays his patient and the remainder of the group obtains the patient's medical history from the student as though he were the patient. Student reactions. The student's reactions to this experience were, in general, quite favorable. Some students stated that they would have liked to see more patients, and that the quality of the faculty input was quite variable. Some also complained of prolonged wait— ing to discuss a patient with a faculty member. The students rated the experience as being of slightly greater educational value than the average of their clinical experiences, and as being moderately enjoyable. "Special" Clinic at "County HOSpital"—‘ Approximately one—third of the medical students at University B have their required ambulatory care experience at the "County Hospital Special Clinic." This clinic is held in a specially designed area with 101 a large library—conference room, extensive laboratory space, and specially designed examination rooms. The students' experiences in this clinic are of three types. Two mornings per week are devoted to clinical sessions during which the students see patients and make manage— ment decisions. Approximately four half—days per week are spent in Ophthalmology, otolaryngology, and der— matology clinics. The remainder of the student's time is devoted to special study sessions, which are described below. Special Study Sessions. These sessions center around the student's studies of selected patients with chronic diseases. During the two-month experience, sessions are held for the study of allergy—immunology, gastroenterologic physiology and pharmacology, cardio— vascular physiology, pulmonary physiology, neurophy— siology et cetera. Patients demonstrating various abnormalities of the system under study are selected and a patient is assigned to each student. In addition to a complete history and physical examination, the student performs the appropriate physiologic studies and other laboratory procedures. The student prepares a case report, including a literature review, and pre— sents his case to the other students and the appropriate specialists. A variety of teaching conferences is integrated into this program. 102 Clinical sessions. During the first six weeks of the program each student sees two new patients a week plus follow—up patients. The clinical sessions during the last two weeks are devoted to completing the workup and management of patients seen earlier in the program. The patients in this clinic are mostly medically indigent, and are selected for study on the basis of the presence of a common chronic condition such as diabetes mellitus, hypertension, or arteriosclerotic disease. Approximately 15 percent of patients fail to keep their appointments. An internist devotes nearly full time to the supervision of this clinic. The faculty are half full— time faculty/staff of the county hospital and half volunteer physicians. These participating faculty members are described as "energetic and eager teachers." There is a psychiatrist and a surgeon in attendance during clinical sessions, as well as internists. The students Spend about two hours with new patients and one hour with follow—up patients. Approxi- mately thirty to forty—five minutes are devoted to dis- cussing each new patient with a faculty member. The faculty frequently observe students performing histories and physical examinations. Including the psychiatrist and surgeon, there are almost as many faculty as students 103 in the clinic. During this program each student sees nine to twelve new patients, and sees each patient two ortMeethms. Laboratory studies, including a twelve-channel blood chemistry screen, hemogram, urinalysis, serologic test for syphilis, chest X-ray, and electrocardiogram, are performed on new patients prior to being seen in the clinic. Student reactions. Students seemed quite satis- fied with the "County Hospital Special Clinic" program. They rated it higher on both educational value and enjoyment than any of the other programs described in this chapter. The only complaint mentioned by a student was that the clinic functioned in isolation from the health care delivery system and did not provide the students with a realistic picture of ambulatory care. Adult Ambulatory Care Teaching at Universipy C University C is a state—supported institution whose medical school was established more than fifty years ago. A recent curriculum revision at this insti- tution included a sizable increase in the students' exposure to ambulatory care programs. Students must have four months of ambulatory care experiences and a month of community medicine as part of their twenty months of clinical experiences. 104 For the typical student, the ambulatory care program consists of a month in medical clinics, a month in pediatric clinics, a month in miscellaneous clinics, and a month of a preceptorship or an additional month of community medicine. During his month in medical clinics the student usually spends two mornings per week in the General Medicine Clinic, one or two afternoons per week in the "Faculty Clinic," and the remainder of his time in medical subspecialty clinics. The miscellaneous clinics experiences include family medicine, emergency room, and surgical specialty clinics. The remainder of this section is devoted to brief descriptions of the "Faculty Clinic," the General Medicine Clinic, the Family Medicine Clinic, and the Community Medicine Program. Conferences are not a part of the activities in any of the clinics described. The "Faculty Clinic" In the "Faculty Clinic" students are assigned to Department of Medicine faculty members on a one—to—one basis. The clinic is housed in a setting where each student-faculty team is assigned to a suite of two examination rooms and one consultation room (office). Patients. The patients in this clinic are the private patients of the "geographic full—time" faculty. These patients are mostly from the upper and upper—middle 105 sociOveconomic classes with problems suitable for manv agement by a medical subvspecialist. Virtually all patients keep their appointments. Faculty. All full—time Department of Medicine faculty members participate in this clinic. For about eight months out Of the year each faculty member is available to teach in this clinic. Clinic gperation. Since the student's role in the clinic is determined by the faculty member to whom he is assigned, it is difficult to describe the typical student's activities. With one faculty member the student may be only an observer, while with a different faculty member the student may play a significant role in the patient's care. A "typical" session might be that the student would examine three of the faculty member's six patients and discuss them with the faculty member, but have little, if any, participation in manage— ment decisions. The student rarely sees a patient on more than one occasion in this clinic. Different stu- dents may see anywhere from ten to one hundred patients. Student reactions. Because student experiences in the "Faculty Clinic" depend upon individual faculty members, student reactions to this clinic are quite variable. Some students complain that they are only permitted to be observers. On the average students 106 rated this clinic experience as not being particularly enjoyable, but of slightly greater educational value than the average of their clinical experiences. General Medicine Clinic This clinic operates (a) to provide care to non- private patients, and (b) for the teaching of interns and residents, as well as medical students. Only about 20 percent of the patients seen in this clinic are seen by medical students. Patients. Most patients in this clinic are medically indigent. A large proportion of them are "chronically dissatisfied" with their medical care and drift from one health care facility to another. Approxi- mately 60 percent of the patients present with problems which are primarily psychological or psychosomatic in nature. Many patients come to this clinic for medical workups to qualify for welfare programs. Thirty to 40 percent of patients fail to keep their appointments. Faculty. The students in this clinic are super- vised by four faculty members during each session. Half of the participating faculty are full—time university faculty and half are volunteer physicians. If a regularly assigned faculty member is unable to attend a clinic session, a medicine resident may be called upon to assist 107 in supervising the students. The faculty members are not eager to teach in this clinic. Clinic operation. During a clinic session each student spends thirty to ninety minutes performing a history and physical examination on a new patient and an average of fifteen to thirty minutes reviewing the patient with a faculty member. The student may also see one or two follOWvup patients. The typical student sees half of his patients on only one occasion, and sees half of them twice. There are two students per faculty member in the clinic, but students are not assigned to a specific faculty member. In this clinic the student makes patient care decisions under supervision and feels that he is the patient's physician. Student reactions. Both faculty and students find the patients in this clinic to be less than optimal for use in teaching. Students appreciate the responsi- bility they are given, but cOmplain that there are too many students in the clinic each session, thus diluting the benefits to be gained from both patients and faculty. They also complain about frequently having to perform medical workups which are more extensive than the patients' complaints would justify. On the average 108 students rated this clinic as not particularly enjoyable, and as being of slightly less educational value than the average of their clinical experiences. Family Medicine Clinic This is an evening, nonappointment clinic designed primarily to manage minor, acute illnesses. Most of the patients seen in this clinic are seen by medical students. Patients. The patients in this clinic are younger than those in the other clinics described, with the highest age frequency being in the twenties and thirties. Most patients are in the lower or lower-middle income groups, but are not indigent. Many of the patients in this clinic are graduate students and their families. Faculty. The clinic is directed by a full-time faculty member from the Department of Medicine, but the supervision of the students in the clinic is done by local family physicians who attend the clinic. These physicians seem to enjoy teaching and the students appear to appreciate the opportunity to work with them. Clinic operation. Students spend fifteen to thirty minutes evaluating each new patient and about ten minutes consulting with a physician about the patient. The students are encouraged to perform selective, abbreviated histories and physicals and to keep their 109 workups to no more than thirty minutes in most cases. \Typically a student sees five to ten patients per evening session. One—third of the patients are seen more than once. The studenteto~faculty ratio in the clinic is Student reactions. Although the students sometimes complain about the clinic's evening hours, they seem to appreciate the sense of accomplishment produced by seeing a number of patients with problems amenable to short—term therapy. 'Students rated this clinic higher than either the "Faculty Clinic" or the General Medicine Clinic. It was rated as having significantly more educational value than their typical clinical experience, and as being moderately enjoyable. Community Medicine Prggram The Community Medicine Rotation affords the student an opportunity to Spend a minimum of one month, full-time, at a location away from the College of Medicine to participate in and study in—depth a selected aspect of the community health system.2 After discussing the types of experiences available with a faculty advisor, the student submits a report of his objectives and plans for this one-to-two-month experience. Half of the students elect to work in, and study, a private medical practice. The other students 2Unpublished data provided by the Coordinator of the Community Medicine Rotation. 110 take a variety of experiences ranging from participating in a public health study to working at Indian Health Service facilities. Approximately half of the students receive some stipend support during this Off—campus rotation. At the end of the rotation the students sub- mit an evaluation of their experience and a report on a community health topic studied during the rotation. The student's participation in the program is concluded by a seminar session in which the students report on their experiences. A study of student reactions to this program showed that 80 percent of students found their exper— iences to be good or excellent, and 85 percent found them to be relevant or very relevant.3 Cost of Adult Ambulatory Care Teaching Good ambulatory care teaching is expensive. This point was made at all three medical schools studied. Experts stated that if a medical student is working with a physician on a one-to—one basis and the physician pro— Vides good teaching, the student—physician team can only see half as many patients as the physician alone could have managed. The long period of time that the student takes in evaluating a patient leads to high overhead costs for space and ancillary personnel. Also a student 3Unpublished data provided by the Coordinator of the Community Medicine Rotation. 111 tends to order more laboratory studies than does an experienced physician. These added cost factors account for the fact that the cost per patient visit at one uni— versity was nearly $48 in clinics used for teaching medical students, while this cost was only $17 in clinics staffed by interns, residents, and faculty.4 At another university the comparison between costs for teaching with hospitalized patients was contrasted with the costs of teaching clinics. The university hospital's Operating costs for in-patients were 3 percent greater than the patient care revenues, but in the teaching clinics costs were 22 percent greater than revenues. Summary University A requires students to have a con— tinuity experience in adult ambulatory care. The student spends one half—day per week during his senior year in one of several different ambulatory care settings. The "Special Clinic" was specifically designed for teaching medical students. Here each student follows and manages an average of twenty to thirty patients with chronic diseases. The Medical Clinic operates in a very Similar manner, but its primary purposes are the delivery of care and the teaching of interns and residents, rather than the teaching of medical students. In the continuity 4This cost per patient visit included laboratory and pharmacy costs. 112 program in Family Medicine each student works with a family practitioner and sees patients whose problems are usually acute, minor illnesses. Students in the Adolescent Clinic gain experience in dealing with the medical and behavioral problems of middle—class adolescents. At University B all students spend two months in a special teaching clinic during their final medical school year. These clinics are interdepartmental with faculty from internal medicine, surgery, and psychiatry, plus other specialists available on call. In the "Uni— versity Hospital Special Clinic," each morning the student performs extensive evaluations of patients with chronic diseases and the afternoons are devoted to ambulatory care conferences or experiences in ophthalmology, otolaryngology, and dermatology. The "County Hospital Special Clinic" includes the basic clinical activities of the "University HOSpital Special Clinic" but at the "County Hospital" the equivalent of two days per week are devoted to the clinical and laboratory study of the pathophysiology of chronic diseases of various organ systems. University C requires students to spend five months in a wide variety of ambulatory care and community medicine experiences. Students devote one month each tCD (a) general and sub-Specialty medical clinics, (b) pediatric clinics, (c) surgical and miscellaneous 113 clinics, (d) a community medicine experience, and (e) a private practice preceptorship or a second month of community medicine. In the "Faculty Clinic" the student is a participant—observer as the faculty member cares for his private patients. In the General Medicine Clinic the student acts under supervision as the physician to patients with a variety of problems. The student gains experience in the management of acute, minor illnesses in the Family Medicine Clinic. The Community Medicine Program provides the student with experiences outside the university medical center. All of the universities studied demonstrated the fact that good ambulatory care teaching is expensive. Chapter 6 RESULTS OF THE QUESTIONNAIRE SURVEY Extensive analyses were carried out on the data obtained from the national survey of adult ambulatory care teaching programs. A profile of clinic programs is followed by a comparison of clinic and preceptorship programs. Comparisons between different groups of pro- grams and between student and faculty responses are pre- sented. The relationships between variables, particularly those factors relating to program goals and success, are reported and discussed. Three research hypotheses were accepted. Questionnaire Responses Of the sixty-two clinic programs selected for further study, adequate data for use in analysis were obtained on fifty programs.l Twenty-three of the twenty- six preceptorship programs were used in the analysis. Table 6.1 contains the questionnaire response rates by 1One additional "clinic program" was eliminated when it was determined to be an in-patient clerkship. 114 115 source.2 Of the fifty clinic programs used in the analy- sis, faculty responses were lacking for two programs and nurse responses were lacking for seven programs. Table 6.1 Questionnaire Response Rates by Source Number of . . Number of Percentage Source Quesfiggpggires Responses Responding Faculty 62 49 79 Nurses 62 43 69 Students 487 280a 57 Preceptorship Coordinators 26 23 88 aBecause of multiple questionnaires per student at one institution, a total of 320 responses was received from 280 students. Missing Data Elements The consolidated data on each of the fifty clinic programs consisted of forty-eight data elements. A total of 211 data elements was missing, which represents a missing data rate of less than 9 percent. The only variables on which data were missing on 20 percent or more of the programs were concerning the "no-show" rate of patients in the clinics. Because of these missing data the number Of programs varies in the analysis of different variables. 2Student response rates for individual programs are presented in Appendix E, Table E.l. 116 Data on the twenty-three preceptorship programs were complete except that one respondent did not rate the outcome variables, and another respondent failed to indi- cate whether the program was required, selective, or elective. Profile of Clinic Programs Questionnaire results were analyzed to characterize the clinic programs on all of the variables studied. Patient Variables The profile of clinic programs on six of the patient variables is presented in Table 6.2. Sixty-five percent of respondents indicated that patients seen in the clinic by medical students were selected or screened. The commonest criteria reported as being used in this selection/screening process were (1) patients with a variety of common "out-patient type" problems, and (2) patients with "learning value" or "potential" for the student.3 In 54 percent of the clinics, the students saw only adult patients and in the remaining clinics the students saw both children and adults. 3Appendix E, Table E.2 contains a tabulation of all of the patient selection criteria reported. 117 Table 6.2 Profile of Clinic Programs on Patient Variables Variable Median (%) Coded Valuesa Standard Mean . . DeViation Percentage of patients medically indigent. Percentage of patients over age 60. Percentage of patients with primarily psycho- logical or psychosoma- tic problems. Percentage of patients with problems not pre- viously diagnosed. Percentage of patients failing to keep appointments. Percentage of patients cancelling appointments and failing to schedule new appointments. 41-60 21-40 21-40 21-40 11-15 41 41 45 46 38 30 a Responses were coded Variables l 4 Code Percentage 2 3 1-20 OH Variables 5 & 6 Code 1 2 Percentage 0-5 6-10 Code 6 7 Percentage 26-30 31-35 21-40 11-15 8 36- 4 41-60 4 16-20 40 41+ for analysis as follows: 5 6 61-80 81-99 5 21-25 100 118 Faculty Variables The willingness of the participating faculty to teach in the clinics was rated on a five-point scale (1 = Teach in clinic only because it is required; 3 = Willing, but not eager; 5 = Extremely eager and enthu- siastic). The mean of responses was 3.8 with a standard deviation of 1.0, indicating that the typical faculty were not only willing, but somewhat eager to teach in the clinic. In only six of the forty-eight programs were the faculty participants rated as less than "willing" to participate. The seniority of the faculty teaching stu- dents in the clinics was rated as equivalent to, or slightly greater than, the seniority of faculty teaching students on the wards. (Mean of 3.2 and standard deviation of 0.7, on a five-point scale.) Table 6.3 presents data on two of the faculty variables. Student Variables In 50 percent of the clinic programs all students xvere in their fourth year of medical school. All of the participating students were in their third year in 20 per- <3ent of the programs. In the remaining 30 percent, both ‘third- and fourth-year students participated. 119 The clinic programs were almost equally divided among required, selective, and elective programs.4 The number of students per year who participated in each program varied from four to over one hundred with a mean of forty-nine and a standard deviation of thirty- five. The career choice preferences indicated on the 320 student responses are presented in Table 6.4. Table 6.3 Profile of Clinic Programs on Faculty Variables Coded Valuesa Variable Median (8) Standard Mean Deviation N 1. Percentage of faculty who are full—time faculty, rather than part-time or volunteer faculty. 41-60 4.6b 1.9 48 2. Percentage of teaching done by house staff. 1-20 1.8 0.8 48 a Code 1 2 3 4 5 6 7 Percentage 0 1-20 21-40 41-60 61-80 81-99 100 bThe distribution on this variable was skewed with 30 percent of the responses being 100 percent (coded value of 7) 4Required (student must take it without other Options) - 17 programs. Selective (student must take it or "an equivalent") — 18 programs. Elective (an optional experience) - 13 programs. 120 Process Variables Results of responses to questions regarding the amount of time students spent per patient visit in various clinic activities are presented in Table 6.5. In 70 per- cent of the programs the students usually wrote new patients' medical records during clinic hours, in 15 per- cent this was done after clinic hours, and in another 15 percent some record writing was done during, and some after, clinic hours. Table 6.4 Career Choice Preferences of Student Respondents Percentage Career Choice Eggbeisgg of po Responses General Practice or Family Practice 48 15 Internal Medicine 115 36 Pediatrics 21 7 Surgical Specialties 94 29 Other Specialtiesa 29 9 Undecided 13 4 aOther specialties were predominantly psychiatry, radiology, and pathology. Number of patients seen and number of visitsyper patient. In 60 percent of the programs studied the stu- dents saw a total of thirty patients, or less. In over half of the programs the students saw less than 20 percent of their patients three or more times, and students saw —: ,‘~" 121 more than 40 percent of their patients three or more times in only one—third of the programs. Purpose of clinics, and conferences. Tables 6.6 and 6.7 contain data on the primary purposes of the clinics, and the subjects of conferences included in the clinic programs. Two-thirds of the programs studied had conferences or seminars as part of the clinic teaching program for medical students. Table 6.5 Student Times Per Patient Visit Time (minutes) Average Time per Patient Visit Mean Standard N DeViation 1. With new patients 72 29 49 2. With follow-up patients 28 12 47 3. Conferring with a faculty member regarding the patient. 22 14 49 4. Writing a new patient's medical record. 29 19 46 Student time in the program. In twelve programs the students spent one half-day per week in the clinic, and in fifteen programs the students spent nine or more half—days per week in the clinic. In the remaining nineteen programs the number of half—days per week was quite evenly distributed between these extremes. Half of the programs were of six weeks or less duration, and 80 per- cent were twelve weeks or less in length. 122 Table 6.6 Primary Purpose of the Clinics ] Percentage Primary Purpose Hes 3:32: of p Responses 1. Teach medical students 18 38 2. Provide patient care and/or teach house staff 27 56 3. Both 3 6 Total 48 100 Table 6.7 Subjects of Conferences Percentage of Conferences on Conferences Organized Thls Subject Around Coded Valuesa Median (%) Standard Mean Deviation N l. Diseases or medical problems 21-40 2.9 2.1 31 2. Discussion of current clinic patients. 21—40 2.4 2.0 31 3. Psychological, psy- chosomatic, environ- mental, or clinical practice problems 21-40 3.3 2.4 31 a Code 1 2 3 4 5 6 7 Percentage 0 1-20 21-40 41-60 61-80 81-99 100 123 Students' patient care responsibility. In over 75 percent of the programs the students made patient care decisions consistent with their expertise under super- vision. In most of the remaining programs there was a mixture of student responsibilities including observation only, performing medical workups, and making patient care decisions. In only one program were the students described as observers only, and in only two programs did the stu— dents do only medical workups with little input in patient management decisions. Faculty-student ratio and nonteaching activities of the faculty. In nearly 80 percent of the programs there were no more than two students per faculty member involved in teaching in the clinic. In only four programs were there three or more students per faculty member. In 46 percent of programs, however, faculty members saw patients without students during the same clinic session that they were teaching students. In these programs, an average of approximately half of the patients were seen by these faculty members without student participation.5 5In 50 percent of clinic programs, 20 percent, or less, of the patients seen in the clinic were seen by medical students. The other 80 percent of the patients were seen by faculty, staff physicians, or house staff. 124 Program Goals Program directors rank ordered the goals of their programs. The mean ranking of the goals with the goals listed in order of their average ranking is presented in Table 6.8. Outcome Variables All clinic programs were rated by program directors on three outcome measures; (a) the educational value of the clinic experience to the typical student as compared with all other clinical experiences; (b) the students' enjoyment of the clinic experience; and (c) the degree to which the programs' goals were achieved. Table 6.9 contains the means and standard deviations of responses. Comparison of Preceptorship and Clinic Programs Data obtained on preceptorship programs were analyzed to characterize these programs on the variables included in the study. The analagous data for clinic programs are also presented to permit comparisons. The clinic programs studied were almost equally divided among required, selective, and elective programs, but over half of the preceptorship programs were elective.6 6Preceptorship programs: Required - 6; Selective — 3; Elective - l3. v ‘ClF-FFJ.’ 7: —" . —-—-- H —i———cv . e .r I.- . 125 Table 6.8 Average Ranking of Clinic Program Goals in Order of Mean Ranksa (N = 47) Mean Standard Goal Rank Deviation 1. Demonstrate the continuity of the doctor-patient relationship. 6.9 2.2 2. Provide experience with diseases which do not commonly lead to hOSpitalization. 6.4 2.6 3. Demonstrate the natural history of chronic diseases. 5.6 2.7 4. Demonstrate the pathophysiology of disease at a given point in time. 5.1 2.6 5. Goal statements added by indi- vidual program directors.b 4.2 4.3 6. Involvement in a health care delivery system other than a university medical center. 3.4 3.1 7. Provide student with experience to assist in career choice decisions. 3.1 2.5 8. Provide students with instruction and experience in abbreviated workups and seeing a volume of patients. 2.7 2.9 aTop-ranked goal = 9; Second-ranked goal = 8; . . . Nonranked = 0. bThese goal statements are listed in Appendix E, Table E.3. 126 Program Goals Table 6.10 lists program goals in the order of their average ranking in preceptorship programs, the mean and standard deviation of the rankings for both preceptorship and clinic programs, and the level of significance of the difference between the two groups of programs. Table 6.9 Faculty Ratings of Clinic Programs on Outcome Variables (N = 50) Standard Variable Mean of DeViation Responses of Responses 1. Educational Value 5.5 1.1 2. Student Enjoyment 5.4 1.0 3. Goal Achievement 5.2 1.1 Note: The rating scales were as follows: Educational Value--l = A waste of time; 4 = As good as other clerkships; 7 = An extremely valuable educational experience. Student Enjoyment--l = Hate it; 4 = So-SO; 7 = Extremely enjoyable. Goal Achievement-- l = Goals achieved much less than in typical clerk- ship; 4 = Same as typical clerkship; = Goals achieved much more than in typical clerkship. Outcome Variables Preceptorship program coordinators rated their programs on two outcome measures; (a) the educational value of the program to the typical student; and (b) the degree to which the goals of the program were achieved. a-p—w-. fly“; _ . 127 .AmEmHmoum mesmuoummowumv v.m pSm AmEmHmOHm OHSSHOV m.m magma .m xHoSmmmS SH woumfia mum muSmEmumum Hmom mmonen Hmom poxSmnSoz . u Hmom OmxSmHIUSOOOm “m u Hmom OmxSmnlmOBo .muSmeom mo mESHO> m mSfimmm USS was Am.mv Am.mv Ixuos ooumw>mnbbm SH OOSOHHmmxm oSm ow. S.m m.~ SowuosuumSw Saws muSmOSum OCSSOSS .m .oEHu Ao.mv Am.mv SH SSHOQ Sm>wm m as mmmmmep mo mooo. H.m m.m mmoaowmmSmosumm mSu mpmuumSofima .s Am.vv Am.mv b.muSm©SommmH HSSOH> oa. N.v n.N IfiOSfl ma poops mSSmEmumum Hmow .o An.mv aa.mv .mmmmmmflo OflSOHSO mo mmo. m.m o.v SHObmHS HonsumS map ouwuumSosmo .m .mSoflmHomO Am.mv Ah.mv OOSOSO Hmonmo SH umfimmm Op mooo. H.m h.m OOSOHHOQSO Suez muSmosum mofl>onm .v .SOHSmNflHmuHQmOS Am.mv Am.av on puma SHSOEEOO SOS Op SOHSB ow. ¢.m m.m mommomflp Sues OOSmHHmmxm mow>oum .m .HmuSmo Hmoepmfi muflmum>fiSS Aa.mv Am.mv m SSS» HmSuo Emanm mum>wamo mooo. ¢.m m.m mumo SpammS m SH uSmEm>HO>SH .N Am.mv an.mv .mwSmSowumamu “Smwummluouooo ow. m.w h.m mSu mo muHSSquOO OSS mumnumSOEmo .H fluv my AwmmmvSmmz A.mwmvSmmz m Ox wwmwwmwwmemo leauzc manumoue Ammuz. msmumoum Hooo oaaeHo mfiSmHoummomHm 0H.m OHQMB mamumoum OflSHHU pSm mflnmuoummomum How mamoo Eoumoum mo mSflxSmm —: 7' 128 The means and standard deviations of these ratings are presented in Table 6.11, as well as the comparable data on clinic programs and the level of significance of the differences. Discussion Preceptorship programs usually involve students working with practitioners away from the university medical center. The inclusion of such activities in post—Flexnerian medical education is a relatively recent development. It is to be expected, therefore, that the majority of these programs are elective experiences for students. Such programs can provide students with experiences in a health care delivery system other than a university medical center and, therefore, expose the students to a wider perspective of the practice of medi— cine. This can be of value to them in making career choice decisions. That these factors are reflected in the goals of preceptorships indicates a recognition of these merits. The more traditional goals of medical education, an understanding of the pathophysiology of diseases and the natural history of chronic diseases, are not sur- prisingly more frequently associated with clinic programs, which are more closely tied to the university medical centers . 129 .QHSmwaHO HMOHmhu SH Smsu mHoE SOSE om>masom mHmoo u n umHSmmeHo HMOHQMH mm oEmm n v «QHSmmeHO HSOHmhu SH Swap mmoH SooE UO>OHSOM mHmow HIIHSOEO>OHS0¢ Hmoo .OOSmHummxm HMSOHHSOSOw memon> mHmEmHuxm SS n S «mmHSmxHOHO HmSHO mm poom ma I amEHu mo mumms S I HIIOSHM> HMSOHUSOSOM I v I "mBOHHOH mm mums memom mSHumu OSB "wuoz AH.HV AN.HV ov. m.m m.m uSmEm>mHSom Hmow .N AH.HV no.HC mH. m.m m.m OSHm> HMSOHHMOSOM .H A.v my mmmSommmm mo mmwSommmm HO O C O O O m m OOSOHOMHHQ NC A o mwommmvSmwz A o mwmwmzvSm z memHHm> OOSMOHMHSmHm mamumoum OHSHHU mEmumoum mHSmHoummomHm mmHQMHHm> mEoouso So mEmumOHm OHSHHU pSm mHSmuoumoomHm mo quHumm thSomm HH.m OHQMB 130 The supervisors of clinic and preceptorship pro- grams rate their programs as being equally successful. That they consider their programs, on the average, to be more successful than typical student clerkships may represent a bias on the part of the respondents. CompariSon of Clinic Programs With (SQ) and Without (NSQ) Student Questionnaire Data There may be limited general interest in the differences between the clinic programs in which student questionnaire data were obtained (SQ programs) and pro- grams without this data (NSQ programs). However, rather striking differences were found so it was considered appropriate to separate these two groups of programs for some of the subsequent analyses. For other analyses only the SQ programs could be used. It is appropriate, therefore, to describe these differences and thus facilitate an understanding of the results of the analyses in later sections of this chapter. Most of the clinic programs on which student questionnaire data were obtained were selected because more than 50 percent of the schools' students partici- pated in the program. This bias in favor of larger pro- grams would be expected to result in differences between programs with and without student questionnaire data on several variables. This was demonstrated in most 131 variable groups.7 No significant differences between the two groups of programs were detected among the patient variables. Faculty Variables In SQ programs the median proportion of partici— pating faculty who were full—time faculty was 40 to 60 per— cent, and in NSQ programs the median was 60 to 80 percent. This difference was statistically significant.8 No impor- tant differences were detected on other faculty variables. Student Variables In SQ programs the median percentage of partici— pating students who were in their fourth medical school year was 40 to 60 percent, while in NSQ programs the median was 100 percent. The average number of students per year in SQ programs was sixty-eight and in NSQ programs, thirty-three. There was a wide range of program size in both groups (standard deviations of thirty—three and thirty), but the difference was significant (p < .0005). 7Appendix E, Tables E.5 and E.6, contains the basic data and results of the statistical tests carried out on the differences presented in this section. 8The terms "significant" or "statistically sig— nificant" are used throughout this report to indicate a difference which is shown to be significant at the 95 per— cent confidence level (p < .05) by an accepted statistical test. 132 Ninety-five percent of SQ programs were required or selective, but only 55 percent of the NSQ programs were in these categories. Process Variables In over half of the SQ programs the primary pur- pose of the clinic was reported as being to teach medical students; but this was the reported purpose in only a quarter of NSQ programs.9 Students spent less time per week and more weeks in the SQ programs than in the NSQ programs. Students Spent two half-days per week or less in the clinic in over 50 percent of the SQ programs, but only 16 percent of the NSQ programs. The median duration of the students' experience was ten to twelve weeks in SQ programs and four to six weeks in NSQ programs. In clinics associated with SQ programs, as com- pared with clinics associated with NSQ programs, more of the patients were seen by medical students and fewer programs had faculty members seeing patients independent of students during the time they were supervising students. Program Goals The ranking of goals of SQ and NSQ programs dif- fered significantly only for the goal, "involve students 9A Chi-square analysis of the reported purposes of clinics in the two groups was not significant at the 95 percent confidence level. 133 in a health care delivery system other than a university medical center." The average ranking of this goal was about fifth in NSQ programs and seventh in SQ programs. Outcome Variables The faculty ratings of the educational value of NSQ programs were significantly higher than this rating of SQ programs. Although the ratings of student enjoy- ment and program goal achievement were also somewhat higher for NSQ programs, these differences were not sta- tistically significant. Discussion The criteria used to select programs for study by student questionnaires produced a difference between the SQ and NSQ programs. The SQ programs were larger, more frequently required, and more frequently involved the student for less time per week over more weeks. Required programs are expected to be larger, and more likely to involve third-year students because of the large proportion of elective experiences in the fourth year at most schools. Larger programs involve more par- ticipation by part-time or volunteer faculty and clinics that are organized for teaching. Elective programs are usually "block time" experiences, therefore more inten- sive and of shorter duration. Required programs are usually held in university medical centers so that 134 involvement in other health care delivery systems is a less common component of these programs. One of three outcome measures differed signifi- cantly between the two groups. Smaller, elective programs might be expected to be more satisfying to both faculty and students, thus accounting for this difference. Comparison of Student and Faculty Responses As was reported in Chapter 3, there was no mean- ingful difference between faculty responses and the indi- vidual program means of student responses on the descrip- tive, or independent, variables. On the outcome or dependent, variables there was no significant correlation between the faculty ratings of program success and the means of student ratings.lo Table 6.12 shows the relationship between student and faculty responses on the outcome variables as reflected by the assignment of programs to the Faculty and Student Success Groups. Discussion The lack of a significant relationship between student and faculty rating of program success has importance for two reasons; one involves this study, and the other is of more general importance. Regarding 10 Table C.4. These correlations are presented in Appendix C, 135 this study, it indicates that the factors associated with "successful" programs must be explored using student and faculty ratings separately. The more general conclusion is that students and faculty judge programs differently and presumably use different criteria for these judgments. Student and faculty perspectives should, therefore, be considered separately in program evaluation procedures. Table 6.12 Chi-square Test of Numbers of Programs in Student Success Group by Faculty Success Group Matrix Faculty Success Student Success Results of Analysis Group Group Low High Low 4 6 Chi-square = 1.61 Middle 4 5 df = 2 High 0 2 p < .50 Clinic Programs' Strengths and Weaknesses as Reported by Students Students, who had participated in fourteen dif- ferent programs, were asked to indicate the major area of strength and the major area of weakness of their programs from among the areas of (a) the type and number of patients, (b) the faculty participation, and (c) the clinic and/or curriculum organization. Responses from students were consolidated for each program such that 136 areas were rated as "weak," "strong," or "in the middle." Table 6.13 presents the results of this analysis. Table 6.13 Numbers of Clinic Programs in Each "Strength-Weakness Category" Based on Student Responses Type and Clinic and/or Number of Pariigilzzion Curriculum Patients p Organization Weak Area 5 3 8 "Middle" Area 4 6 6 Strong Area 5 5 0 Total 14 l4 l4 Even though a Chi-square analysis of Table 6.13 failed to demonstrate significance in the distribution of these student ratings,ll the "weak and middle areas" ratings of the "clinic and/or curriculum organization" suggest that this is a major area of weakness from the students' perSpective. This impression is substantiated by student com- ments on questionnaires. Each student was asked to comment on the clinic program's strengths and weak- 12 . . nesses. These comments were categorized as being l1A Chi square analysis of Table 6.12 revealed p of less than .50 and greater than .30. 12Of 320 student responses, 106 (33%) included such comments. 137 (a) general, (b) regarding the patients, (c) regarding the faculty, or (d) regarding the clinic and/or curriculum organization. Each comment was also categorized as reflecting a program strength or weakness. The results of this analysis of student comments are presented in Table 6.14. Table 6.14 Student Comments on Clinic Programs' Strengths and Weaknesses Comment Implied Category of Comment Weakness Strength No. of % of No. of % of Comments Comments Comments Comments General 4 2 19 12 Patients 32 20 8 5 Faculty 28 17 15 10 Clinic/curriculum organization 53 32 4 2 Total 117 72a 46 28 aPercentage of total, not sum of columns. Discussion This documentation of the "clinic and/or cur- riculum organization" as an area of major concern cor- roborates the finding of the problem identification phase of this study that process variable problems 138 account for a large proportion of the perceived problems in adult ambulatory care teaching.13 Relationships Between Variables-- Clinic Programs The relationships between the variables included in the study were examined using three methods. Based on program goals and the Faculty Success Index each program was assigned to two goal groups and one Faculty Success Group. Programs with student questionnaire data were also assigned to a Student Success Group based on the Student Success Index.l4 Each group was charac- terized on each variable, and the apprOpriate groups compared to determine relationships between the patient, faculty, student and process variables, and program goals or faculty estimates of program success. The relationships between most of the variables in the study were examined by forming an intercorrelation matrix for all programs combined and separate matrices for programs with and without student questionnaire data.15 13See Chapter 4. 14The method Of forming these groups is described in Chapter 3. lsOnly correlations with programs' goals and out- come variables will be reported in this chapter. All significant correlations are presented in Appendix E, Tables E.7 through E.10. 139 Those variables which appeared to have the more meaningful correlations with the outcome variables were then used in a series of stepwise regression analyses using the Faculty and Student Success Indices as depen- dent measures. Relationships Between Program Goals and Other Variables Each program was assigned to one of a set of primary goal groups, titled Continuity, Clinical, and Practical, based on its top-ranked goal. It was also assigned to one of a comparable set of tertiary goal groups based on goals which were ranked as being of low priority or not applicable. By comparing the charac- teristics of the programs in the primary goal group with those in the comparable tertiary group, it was possible to identify characteristics which might be expected to be related to the goals used in assigning the programs to these groups, i.e. continuity goals, clinical goals, or practical goals.l6 Variables related to continuityygoals. Signifi- cant differences were found between the Primary and Ter- tiary Continuity Goal Groups on four variables. The Primary Continuity programs had: (1) A larger percentage l . . 6General information on the comparison of appro- priate groups follows. More detailed data are contained in Appendix E, Tables E.ll and E.12. 140 of patients with problems which were primarily psychologi- cal or psychosomatic; (2) A larger percentage of patients seen three or more times by the same student; (3) Each student in the clinic fewer half—days per week; and (4) Each student in the program for a greater number of weeks. Variables related to clinical goals. Program characteristics on five variables were found to be sig- nificantly different between the Primary and Tertiary Clinical Goal Groups. A focus on clinical goals was related to: (1) Having a larger percentage of patients whose primary problems had not been previously diagnosed; (2) Having only adult patients in the clinic, rather than adults and children; (3) Having faculty who were rated as less willing to teach in the program; (4) Having stu- dents conferring with faculty members for a longer period of time per patient; and (5) Having a smaller percentage of conferences organized around psychological, psycho- somatic, environmental, or clinical practice problems. Variables related to practical goals. Significant differences were found between the Primary and Tertiary Practical Goal Groups on six variables. Programs in the Primary Practical Group had: (1) Fewer patients over age sixty; (2) Faculty rated as more willing to teach in the program; (3) Students spending less time per 141 patient conferring with faculty members; (4) Students spending less time per new patient writing the medical record; (5) Students seeing more patients during their clinic experience; and (6) A larger proportion of con- ferences dealing with psychological, psychosomatic, environmental, or clinical practice problems. Relationships between Goal Groups and Faculty Success Groups. Matrices were formed of (a) the Primary Goal Groups and Faculty Success Groups and (b) the Ter- tiary Goal Groups and Faculty Success Groups. No sig- nificant relationship was demonstrated between goal groups and outcome as measured by the Faculty Success groupings.l7 Relationships between individual goals and other variables. Table 6.15 contains the significant cor- relations between the rankings of individual program goals and other variables when all programs were studied. Two goals had positive correlations with outcome variables. Two other goals had no significant correlations with any of the other variables. 17The matrices and results of the Chi-square analyses are presented in Appendix B, Table E.13. 142 Table 6.15 Correlations Between the Ranking of Goal Statements and Clinic Program Characteristics Goal and Other Variable r p < 1. Goal: Demonstrate the continuity of the doctor-patient relationship. a. Percentage of faculty who are full- time. .33 .05 b. Percentage of students who are fourth year. -.36 .05 c. Student hours per week in the clinic. -.44 .01 d. Duration of program for typical student. .32 .05 2. Goal: Demonstrate the pathOphysiology of disease at a given point in time. a. Student time per patient conferring with a faculty member. .35 .05 3. Goal: Provide experience with diseases which do not commonly lead to hospitali- zation. a. Educational Index. .34 .05 4. Goal: Provide students with experience to assist in career choice decisions. a. Percentage of faculty who are full- time. .49 .01 b. Percentage of teaching by house staff. -.33 .05 c. Students' role involves making patient care decisions under supervision. -.43 .01 d. Students' role is mainly to do workups. .48 .01 5. Goal: Involvement in a health care delivery system other than a uni— versity medical center. a. Student hours per week in the clinic. .49 .01 b. Duration of program for typical student. -.46 .01 c. Goal achievement. .39 .05 Notes: Correlations are Pearson's product-moment cor- relation coefficients. The ranking of goals was coded as follows: Top ranking = 9; second ranking = 8; . . . Unranked = O. 143 Relationships Between Outcome Variables and Other Variables The relationships between the outcome variables and other variables were studied by (a) examining the characteristics of the High, Middle, and Low Faculty 8 and the High and Low Student Success Success Groups1 Groups;19 (b) inspecting the correlations between the outcome variables/indices and other variables; and (c) performing stepwise regression analyses with the Faculty and Student Success Indices as dependent variables. Contrasts between Faculty Success Groups. The High, Middle, and Low Faculty Success Groups were char— acterized on each of the variables included in the study. Comparing the High Faculty Success Group with the Low Faculty Success Group revealed that the High Faculty Success Group had significantly: (1) Fewer patients over 18The Faculty Success Index, an average of faculty ratings of the program's educational value, enjoyment by the students, and goal achievement, was used to assign each program to the High, Middle, or Low Faculty Success Group. The composition of these groups is reported in Appendix C, Table C.5. 19The Student Success Index, an average of the students' mean ratings per program on educational value and student enjoyment, was used to assign each program to the High or Low Student Success Group. The compo- sition of these groups is reported in Appendix C, Table C.5. The relationship between Faculty and Student Success Groups was reported in Table 6.12. 144 age sixty;20 (2) Greater willingness on the part Of faculty to participate in the program; (3) More of the program's conferences organized around diseases or medical problems; and (4) The typical student in the clinic more hours per week.21 Contrasts between Student Success Groupg. Only one of the independent variables differed significantly between the programs rated highly by students and those programs that students rated lower. The highly rated programs had an average of about one and one-half stu- dents per faculty member in the clinic, while the lower rated programs had between two and two and one-half stu- dents per faculty member.22 Correlations between outcome variables and other variables. Intercorrelation matrices were formed between faculty ratings of outcome variables and independent variables for (a) all programs, (b) programs without student questionnaire data (NSQ), and (c) programs with 20The Middle Faculty Success Group had a lower percentage of patients over age sixty than either the High or Middle Groups, but the difference between the High and Middle group was not significant. 21More detailed results of these analyses are presented in Appendix E, Table E.14. 22High Student Success Group (code units) ii= 2.3; _§.D. = 1.2; N = 12; Low Student Success Group (code units) X = 3.5; S.D. = 1.8; N = 8; p < .05. 145 student questionnaire data (SQ). The significant cor- relations are presented in Table 6.16. NO significant correlations were identified between student ratings on the outcome variables (Student Success Index) and other variables.23 Regression analyses on the Faculty Success Index (FSI) and the Student Success Index (851). The cor- relations between the independent variables and the FSI for all clinic programs, NSQ programs, and SQ programs were examined and those variables with larger cor- relations were used in stepwise multiple regression analyses. Different combinations of variables and dif- ferent orderings of the variables were used to identify those variables which accounted for the greatest amount of the variance of the FSI. In an analysis of all clinic programs, two variables accounted for 26 percent of the variance of the FSI (Table 6.17). The rating of faculty willingness to participate in the program accounted for 21 percent of the variance. In NSQ programs this variable con- tributed 28 percent out of the total variance accounted for of 39 percent. No variables were identified which significantly accounted for the variance of the FSI in SQ programs. 23Because of the small sample size (N=15) used in this intercorrelation matrix, a correlation coefficient of .51 was required to achieve significance. , "1"‘— v ' 146 .lmanzc om “lamuzo omz “lmvuzo HHS "nmNHn mHesom «macaw mum mSoHHMHOHHOU Amo. v do uSMOHuHSmHm >HSO “uSOHOHmmooo SOHUMHOHHOU uSmEOE uosooum m.SomHmom mum mSOHumeunoo “mmuoz .mEmHmoum OHSHHO HHS u HHS «sumo OSHMSSOHummso usmosum SSHS mamumoum moose u Om «pump OHHMSSOHummSo uSoosum uSOSqu mamumoum muons u Omzw I I am. I I I I I I .HOSSOO HMOHOOE muHmHm>HSS m SSS» Hmnuo Emumhm >H¢>HHOU wumo SHHMOS 8 SH HSOEO>HO>SH I I ~m.I I I I I I I .Edumoum on» SH mpSmmm . uSwosum HMOHQMH may mxmw3 mo HwQESZ I ma. mv. I I I I I I .OHSHHO was SH mpsomm uSwp Isuw Sumo xmmz Ham mHSOS mo HOQESZ I I mm. I I I I I I and» n N .OS u H "ompoov .mmoSmummSoo mSHnummu mo 00S0mmum mm. I I oo. I I I I I .uSwHumm Hem mEHu SOHuomuquH hquomquSoosum I I I I I mm.l I I I .Emumoum 0:» SH Show mom muSmpsum Ho umnesz I I I mm. I I I I I .pmuomHm u m .pmpomem n N .UOHHSUOH u H "Owooo .EMHDOHQ Owuome HO .omuoonm .OmuHsowm I I I I I I I mv.I I .Hmmw SHHSOM HHwSu SH mum 0S3 muSmosum Ho mmmuSmonm I I I am. I em. I I I .muHSomm ummuSSHo> Ho wEHquHmm Sosa HmSumH .huHsomw wEHu IHHSM who 0S3 SHHSOMM mo mmmuSmoumm I I mv. I I I I we. vv. .OHSHHO 0S» SH Scamp Ou MHHSOMH mo mmwSmSHHHHB 0m Omz HH< 0m Omz HH< Om Omz HH< mmEmumoum mmsdeoum Mmfimumoum uSoso>wHSo< Hmow uSofimoflSM uSmpsum OSHM> HMSoHumospm mHSMHSm> uSOUSoomoSH mmHSMHum> msoouso mo mmSHuwm auHsomm masouo Emumoum wmuSa new meQMHHm> USOOSOQOUSH USM mmHQMHHw> oEoouSo So mmSHumm zuHSoom Swm3uom mSoHHMHOHuOU mH.m anoB re ~ fiwgi—n- . A __ - ..- _ v — F 147 Table 6.17 Results of Analyses Of the Predictors of Success of Adult Ambulatory Care Teaching Programs Multiple Regression Analysis Program Group and Variables r % of Variance Accounted for p < A. All pgograms; (N=47) FSI and: 1. Rating of willingness of faculty to participate in the program. .45 21 .002 2. Ranking of Goal: Involvement in a health care delivery system other than a university medical center. .32 5 .085 Total 26 .002 B. NSQ programs; (N=27) FSI and: 1. Rating of willingness of faculty to participate in the program. .53 28 .005 2. Ranking of Goal: Involvement in a health care delivery system other than a university medical center. .45 ll .05 Total 39 .003 C. SQyprograms; (N=l9) FSI and: No significant results. D. SQgprograms; (N=l9) $81 and: l. Ranking of Goal: Demon- strate the pathophysiology of disease at a given point in time. .49 24 .035 2. Number of students per year in the program. -.41 19 .04 Total 43 .02 Notes: The "percentage of variance accounted for" is from a‘ stepwise multiple regression analysis performed in the order in which the variables are listed. The percentage assigned to the second variable in each group is, therefore, the additional variance accounted for by adding that variable. r = Pearson's product moment correlation coefficient. FSI = Faculty Success Index (the average of the .faculty ratings of the program on three dependent variables). $81 = Student Success Index (the average of stu- dents' ratings of the program on two dependent variables). NSQ = programs with no student questionnaire data. SQ = programs with student questionnaire data. 148 Multiple regression analysis of SQ programs identified two variables which accounted for a total of 43 percent of the variance of the Student Success Index. The ranking of the goal, demonstrate the pathophysiology of disease at a given point in time, accounted for 24 per- cent of the variance. The number of students in the pro- gram per year was negatively correlated with the S81, and accounted for an additional 19 percent of the variance. Relationships between "faculty willingness" and other variables. Faculty willingness to participate in the program was shown to be significantly related to outcome variables by comparison of Faculty Success Groups, by correlation coefficients, and by regression analysis. Since the rating of faculty willingness is a subjective measure, it was appropriate to carry out additional analyses in an attempt to identify objective measures related to the rating of faculty willingness. When all programs were analyzed, no significant cor- relations were found between "faculty willingness" and other variables. However, when the SQ and NSQ programs were analyzed separately, it was discovered that in both groups "faculty willingness," correlated Significantly with one variable, the number of students per faculty member in the clinic, but the correlations were in Opposite directions (SQ programs, r = +.58; NSQ pro- grams, r = -.39). q—n—H‘-_——v, ~ -. . , ._....-— — m— 149 Discussion Only two goal statements and none of the goal groupings demonstrated significant relationships to faculty ratings of success. This suggests that there is no strong relationship between the degree of program success and the nature of the program's goals. Perhaps this is due to program goals being selected on the basis of the chance of succeeding with those goals. The presence of an enthusiastic faculty is undoubtedly a key element in the success of any edu- cational program. It is not surprising, therefore, that the rating of faculty willingness to participate in the program accounts for a large percentage of the variance of the Faculty Success Index. The fact that "faculty willingness" and the FSI are both subjective, and were rated by the same individual, raises some question as to the validity of this finding. The high correlation is, perhaps, only a result of the raters' biases, but the findings cannot be discounted without more concrete evidence. The search for objective correlates of "faculty ‘willingness" was disappointing. The conflicting, large correlations in the SQ and NSQ programs cannot be explained with the information available. The variables of value in predicting students' ratings of success suggest that students feel that 150 experiences in smaller programs and dealing with the more traditional aspects of clinical medicine are most valuable and enjoyable. Another possible explanation for the high correlation between students' ratings Of program success and the program's emphasis on the pathophysiology of diseases is that faculty are more adept at teaching in this area than in the more practice- oriented aspects of medicine. Relationships Between Variables-- Preceptorship Programs The relationships between the variables used in studying preceptorship programs were examined by two methods. Each program was assigned to a Goal Group, based on the nature of the program's top-ranked goal, and an Educational Group, based on the program's Edu- cational Index (an average of the ratings on educational value and goal achievement).24 The distribution of pro- grams assigned to these two sets of groups was analyzed using the Chi-square test. In addition, an intercor- relation matrix was formed using all of the variables studied to identify the relationships between individual variables.25 24The method of forming these groups is described in Chapter 3 and Appendix C, Table C.5. 25The correlations between the rankings of indi— vidual goal statements are of limited value and there were no significant correlations between the "required—selective— elective" status of programs and their goals or outcome measures. These correlations will, therefore, not be reported. 151 Goal Group by Educational Group Matrix Programs were assigned to a High or Low Edu- cational Group and a Continuity, Clinical, or Practical Goal Group.26 Because only one program was in the Clinical Goal Group, this group was eliminated from the analysis. A Chi-square test of the distribution of programs in the two sets of groups failed to Show . . . 27 Significance. Correlations Between Rankings of Individual Goals and Outcome Measures The rankings of the goal, demonstrate the natural history of chronic diseases, correlated negatively with the preceptorship coordinators' ratings of the programs' degree of goal achievement (r=-.55, p < .01).28 No other significant correlations between goal rankings and out- come measures were identified. 26This set of goal groups is analogous to the Primary Goal Groups for clinic programs. The lower priority goals were not in the "clinical" area for only two preceptorship programs. The "tertiary goal grouping" was, therefore, not used in analyzing these programs. 27This analysis is presented in Appendix E, Table E.17. 28Top-ranked goal coded as 9; Second ranking = 8; . . . Unranked = O. 152 Tests of Hypotheses Three of the seventeen research hypotheses were accepted after performing the appropriate statistical tests.29 Medical students planning to enter the primary care specialties did rate adult ambulatory care clinic teaching programs as being of greater educational value than did students planning to enter other Specialties. Faculty ratings of the success of clinic programs (Faculty Success Index) were (a) positively correlated with the percentage of participating faculty who were full-time faculty (r = .31), and (b) positively cor- related with the rating of the willingness of faculty to teach in the clinic (r = .48). Discussion Students planning to enter primary care specialties were expected to rate the clinic programs as more valuable than other students since the clinics studied were pre— dominantly devoted to primary care. This finding raises the issue as to whether or not ambulatory care teaching programs should be required for all students. EXperiences in these programs can provide the students with a greater understanding of the practice of medicine in general, 29The list of hypotheses, the tests used, and the results of the tests are contained in Appendix E, Table E.18. , 153 and may well be vital for all students regardless of their career plans. If all students are required to participate in clinic programs, however, it would appear to be essential that the program emphasize the factors considered Vital for all students and that the importance of these factors be adequately demonstrated to the stu- dents. If students do not see the program as being relevant to their personal futures, it is doubtful that the program will succeed in its intended purpose. Summary Adequate data for analysis were obtained on fifty of the sixty-two clinic programs surveyed and twenty-three of the twenty-six preceptorship programs. Questionnaire results were analyzed and a pro- file of clinic programs was presented characterizing the programs on each of the variables studied. A comparison of clinic and preceptorship programs revealed that: (a) preceptorships were more commonly elective exper- iences; (b) they had a higher priority on two goals, student involvement in a health care delivery system other than a university medical center and experience to assist the student in making career choice decisions; (0) they had a lower priority on goals relating to knowledge of the pathophysiology of diseases and the 154 natural history of chronic diseases; and (d) there was no difference in the faculty ratings of success of the two types of programs. A comparison of the programs with and without student questionnaire data revealed that those programs selected for study with student questionnaires (a) had a smaller percentage of full-time faculty teaching in the clinic, (b) involved more third-year students, (0) were larger, (d) were more commonly required, and (e) were rated lower by faculty on the educational value of the programs. There was no meaningful difference between the faculty and student responses to questions on the descriptive, or independent, variables. On the outcome variables, however, there was no significant relationship between faculty and student responses. Students' ratings of program strengths and weaknesses confirmed the earlier finding that clinic and/or curriculum organization was a major problem area. The relationships between variables were studied (a) by comparing the characteristics of groups of pro- grams with different goals and different success ratings, (b) by correlations between variables, and (c) by multiple regression analyses using the success indices as depen- dent variables. A number of program characteristics was found to be significantly related to program goals. 155 The ratings of the willingness of faculty to participate in the programs bore the strongest relationship to the faculty ratings of program success, accounting for over 20 percent of the variance of the Faculty Success Index for all clinic programs and programs without student questionnaire data. Student ratings of success were positively correlated with high ranking of the goal, demonstrate the pathOphysiology of disease at a given point in time, and negatively correlated with the number of students in the program. Three of the seventeen research hypotheses were accepted. (1) Students planning careers in the primary care specialties rated clinic programs as being of greater education value than did other students. The Faculty Success Index was positively correlated with both (2) faculty willingness to participate in the program and (3) the percentage of participating faculty who were full—time faculty. Chapter 7 SUMMARY AND CONCLUSIONS This chapter provides a general overview of the study, including brief descriptions of seven models of ambulatory care teaching and two suggested areas for further research. Summary Approximately half of the student's time in medical school is devoted to clinical instruction, learning through supervised eXperiences with patients. The current rapid increase in the number of medical stu- dents without an accompanying increase in the number of university hospital beds suggests that new or modified approaches to clinical instruction are going to be needed. One way that this impending shortage of hos- pitalized patients might be alleviated would be to increase the utilization of ambulatory patients for clinical instruction. This seems particularly appro- priate since the vast majority of clinical instruction in U.S. medical schools involves experiences with hos- Pitalized patients, but the greater proportion of medical Inmactice involves the care of ambulatory patients. 156 157 As a first step in stimulating research in this area, a descriptive and exploratory study of the teach- ing of adult ambulatory patient care in U.S. medical schools was conducted. The information produced is expected to be valuable to medical schools in planning and conducting ambulatory care teaching programs, and to researchers by suggesting topics for further study. The research was intended to produce information with which to answer the following questions: 1. What are the major problems which faculty and students perceive in current adult ambulatory care teaching? 2. Which characteristics of current adult ambulatory care teaching programs are associated with the success or failure of the program? 3. Does the degree of success of programs vary with differing program goals? 4. Are preceptorship programs more successful than clinic programs for some types of goals? A model was proposed for use in the study Of clinical instruction. Patient, Student, and Faculty Variables are largely beyond the control of program mo Hmuu < mmouuzm mmmmu .mcuuumm menu xuoumuznsm msu cu mcouumgmao mcu ocuccwucou mcuwccouummsc ucogm m ou ucoammc nuaou oz: .aucmxuopu muzu $0 couugon Aucmuuma-u=ov AccumuamEm emu zuuz Lauuue~u amuse a we mmocuum mauuume van mam: ozu umuu .m zzzuou .um>Po>:u mucoucoaxm memo scououaaEm mzu mo mpumumu :mu m acuucammc muumccouummac ugonm m ou ucoammg ou muse mp nuaoz on: .augmuco_u mugu we couucoa wuou >couo~=osm wad Lou muWFWQumcoammL cur: tongue Aupaum» a we mmmcucm ccuuume wen mam: mgu umuu .5 zzsuou .mxmmz cu nuzmxcmuu mzu uo cumcop mgu new nuzmxumpu muzu cu mucmuuma xcoumuaaEm guuz mcuxcoz ou mmuo> -mv ucmnsum mcu umcu xmwz can muse; do guess: mzu oumuuuCH .o zzzuou .woou umuou nuaogm mmcmumu_ mmugu ozu .m—o5mxw Lo» .mumuuamo: ucmcmm -muu mwcuu um vocmwuo nusmucouu nmguaamc a co; .awnmuumpu muzu mxmu on! mumpu 6 gr mucwtaum yo ucwugwa wquFXOLQnm mzu mumuvaM .m 222.50 . .m.w;u co 1 «cu m=u_ucuu >9 o>uuumpo Lo vmcuacmc mu augmxcmpu wsu umzumzz 33.6:H .e zzauou .uoguacwu Lo umcomuo mu auzmem—u mugu sous: cu “sue .vgm .ccw .umuv meow» Lo gum» Poocum quuume mcu mumuuncu .m zzauou .mausmgouamu -mcq Lou xcaun macaw mugu w>aou .pouunmos mxu xu »_mumgmaom uu umup .uauuamo; mco cagu egos um umcmwwo m. augmxcmuu mama ogu u~ .mocmmmo mu auzmxumuu ugu mums: pauuamo; ogu we use: «nu ouauuocu .~ zzauou .mucowzum Lao» uo once Lo umm An cmxmu mum sous: o>ona N aoum c. eaesuucu muuo_u wgu cu maunmxua_u asuuumuo pun umuu .m amum .mupmum Pauuouam Lo .xguuucuama .muugumuvma cu maunmxcopo ou:_ocu uoc oo .Amucouuua-u=ov mucouuoa zcoumuanEm upscm cum; mmucmwumaxm ucouaum meow mvauucp sous: .ocuu_uos o>umcmzmgaeou co xuwcaesou .xqume was .mgmu xcmeucq .mmuu -umuumamozm muu ecu ocuuuume unecoucu .ocuuuume pacmcmm mo mcumuu mnu cu .m>onm _ amum cu emuzuucu uo: maugmxcm_u cmcusmmc pun umuu .N amum .Amaunmucmpu menu accumuane< .mcuuuumz xqumu .mauzmcouamumca “mmuamemv .mupaum ou umuueup apuuom -mwumc uoc use mzuoauucu .mucmuuma xgoumpauEM cuuz .xpm>um3_uxm cam: co .>_m>_m=puxm Pmou umsu new ouaa_9wucoa mucmuaum ozu mo uo— ummmp um nous: cu .m>uuumpo so umcuaamu .msmcmoua Lo maugmxcopu upm umuu .P nwum .cm>mzoz .muco mpco umumrhlon wwsocm maucmuouamuouni .uw>uo>cw .u..mmn. Luau to» muwuocnmmm I I ..I . IL." -oum Loguo co mauzmxumuu .uouvaum maumu mt when can maucmemPU __n weapon? puuz so» .3opwo umumu. mamum mmuuu m u uruzoupow Am .F zzauoo umumup ma opzosm umuWQmo; mco cmcu menu w umcmmmo mum cows: macaw I .sgou ogu co aou mgu ua auau acuacuueoc_ ogu muwuasou mmmmpa .muoocUm quuvwe .m .2 Fun ou umuumucu do an ou you -uwaxw mu aumv ommzu mo mumzumcm vcm cowumpwanu och .mpoogum ——m Eocw vocumuuo an ou aumv munogmnsoo uracon Fur: wrap—om nous: Ego» «Pusan umuo~n50u mzu wan zopmn umuuup mcouuuacumcu ozu ou cowucouum usumuuu mcouuozgumcu >k~mmw>~zz whmo oz< zu¢mohm>m=m 1777 1.7E3 2 m 0 m m A m 2 (.m vmmufi muaum .csounuz vmwuu oumum .c3oucuz hmuu mmm xmo Hmvn oscm>< xmo Hm¢> Hmuwamox cmuauomouuoz . amuHQmo: cmufiuomouumz .2.m .zomaLoze .h .4 .o.: .couu¢ .<.¢ mu m u ow m AHV cum advanomouuox luaduduunlu vmmaa oumum .c3ouowz vmmHH wumum .c3oucwz w>uuo umuuamox Humw w>uuo uauuamom Humu Hmuuamo: .>fi:D Joucufiu .vmz Hmuwmmom >ufimuw>uca .z.m .nqum .2 .m .o.z .mucon .o .o «H m u om m AHV nun .Nuumuo>uca ocuoucmx vmmuu mumum .naoucux vmmua mumum .czoucuz m>uuo Hmuummox HHNN o>uuo Hmuummo: HHNN Hauwmmom .>uca .oflcfiao .n&< Hmuflmmo: >uwmum>fico .z.m .mocon .m .n .o.z .nuusm .m .4 m on a mu nu u. sue uuumuosuno ouou auoumusnsa vmmuu cumum .csoucuz .oo: uo .uou ..>uc: Hmoua>u< acuoucwz mHuEam mo .umwa .a.z .mxmoo .m .n w on a ow Amv a sue auuauuaqququuul mom~>zma2m mom~>mmasm mxz do u x2\mm: mmuhumum Acuv .vcm zouhbu=u uoozum .ouz no uzmmm2m wmmaz no mmwmoo< az< uzamaam >huzuuuuuuu mzuu wz~=umu _N.z A_a.uv mo.m “Nuz Amm._v wL.N mmuz Aom.mv om.o mmnz Amm.mmo ¢N.mm LNuz Amm._o ee.m “Nuz AON.NV m_.m mNum mamEmm new .A.o.mv «.cmmz msocmocm omz m_uz ANN.NV em.e FNuz A.N.mo _N.e umuz Amm.mv mm.m quz Am¢.mmv om.mo Pmnz 85$ om.m anz Ame.uv om.m wuwm mpasmm new .A.o.mv e.cmmz msocmoca om .mucmuzum xn comm mucmuuma oucupu we ucmocma .Amxmmz mo consacv oucupu cw msuu ucmuzum .Axmmz can mace my ouCFFU cu meuu ucmczum .cmm> emu seem -ocg cu mucmnzum mo conszz .me» zuczoe cumzu cw mew oz; mucmuzum mo ucmucma .Au_=umm msuuup_:w mew 0:3 au_:um$ co ucmocma ._ a_aauca> Amumau uv Aomzo usoguuz use Aom m.m mpnob uma acumccowummzo ucmuzum guuz mamcmoca uwcupo mo comucmaeoo .Ap.o m—nMH .o xuvcwaa< mmmv .mmspm> nmoou mgu mum muucz k nmuz quz o_. Amo.uv Amu.uv w¢.m mm.¢ .ucmEm>mu;om umom mo mmcmmo .ou w Ruz mmuz a. om. Aou._v Amm.v .aocmueaaxa mm.m om.m co ucae»0w=a ucmusum .a nmuz mmuz mo. Amo._v Amo._o .Eaumoea acu Lo m~.m m_.m mucmuaum ou mzpm> :ouumuzuu .m .cmucmu Pmuuvme nmuz omuz zuwmcm>ucz m :mzu cmguo Emumzm Fo. AFN.NV Am_.NV seasuuaa memo :uuma; a Cu mo.m mm._ usaEa>_o>=H "Paco Lo mcwxcam .N chmgu mmmP av mNWm munsmm wNum wFQEmm mpamwcm> mucmuumucmum co Fm>m4 use .A.o.mv «.cmwz new .A.o.mv k.camz mswgmoca omz msmcmoca om Aeascuucouo m.m aunau 220 Fo. op. Po. Aucmgu mmmpmo mucmoueucmwm eo Fw>mu m o msmcmocalomz mo .02 msmcmoLm am we .02 oz mm> guom mLmo “cm wpmm mucousum gummh w>wuumum m>uuumme umcuzcmm mmcommmm .uwcuuo ocugummu acuczv mucwuzum uaozuuz mucmuuma mum zupzumw om .m .mcmo ucmuuma co mucwuaum ocucummuv oucu—u mo mmoacza xgmsuca .N .Emcmoga umuompm co .umuompwm .umcwzcmm ._ aunauaa> Amummu mcmzcm ucov mums wgumccouummso ucmuzum Aomzv usocuuz use Aomv guuz usacmoza au:u_u Lo camuaaaeou o.m wrnmp 221 Table E.7 Key to Variable Names for Intercorrelation Matrices Patient Variables (PV) . Percent of patients medically indigent. Percent of patients over age 60. Percent of patients with primarily psychological or psychosomatic problems. ' Percent of patients with problems not previously diagnosed. Percent of patients failing to keep appointments. Percent of patients cancelling appointments and failing to schedule new appointments. Selection or screening of patients assigned to students. Patients are adults only or adults and children. . Willingness of faculty to teach in the clinic. . Percent of faculty who are full time faculty, rather than part-time or volunteer faculty. . Seniority of faculty. . Percent of teaching done by house staff. tudent Variables (SV) . Percent of students who are in their fourth year. . Student's career choice. . Required, selected, or elected program. Coded: 1 = required, 2 = selected, 3 = elected. 1 2 3 4 5 6 7 8 Faculty Variables (FV) l 2 3 4 S 1 2 3 Process Variables (PrV) Student time per new patient. Student time per followup patient. Student-faculty interaction time per patient. Student time per new patient spent in writing the medical record. Record written during or after clinic hours. Number of patients seen per student during program. Percent of patients seen three or more times. Primary purpose of clinic (Coded: 1 = teaching students, 2 = both, 3 = patient care and/or teaching house staff). 9. Presence of teaching conferences. (Coded: l = no; 2 = yes) 10. Types of teaching conferences lla. Number of half-days per week each student spends in the clinic. llb. Number of hours per week each student spends in the clinic. llc. Number of weeks the typical student spends in the program. 12a. Percent of time that student's role is making patient care decisions consistent with his expertise. 12b. Percent of time that student's role is observing and discussing patients with little input in decision-making. 12c. Percent of time that student's role is doing workups, with little input in patient management decisions. CDNO‘IU‘l-th—I 222 Table E.7 (continued) 13. Number of students per faculty member in the clinic. 14. Percent of clinic patients seen by students. 15. 00 faculty see patients without students during teaching clinic. (Coded: 1 = No; 2 = Yes) 16. Percent of patients seen by faculty without students. 17. Percent of patients on whom a complete medical workup is performed. Program Goals (Coded: 9 = top ranked goal; 0 = non-ranked) 1. Demonstrate the continuity of the doctor-patient relationship. 2. Demonstrate the natural history of chronic diseases. 3. Demonstrate the pathophysiology of disease at a given point in 4 time. Provide experience with diseases which do not commonly lead to hospitalization. 5. Provide student with experience to assist in career choice decisions. 6. Provide student with instruction and experience in abbreviated workups and seeing a volume of patients. 7. Involvement in a health care delivery system other than a university medical center. Write-in goals. utcome Variables (0V) Education value to students of the program. Student enjoyment of experience. Degree of goal achievement. Faculty Success Index Student Success Index Educational Index 03thme co >.m mFQMF c? umcumucoo mu moan: mFQmwcm> mcu ou xmx meek N>aa >>La e>m m>m >>m e>4 N>L N>L >>L m>a >>a - - - - - - - - 2. - - a - - - - - - - >m. Ne. - - . a - - - - - - - - m5. - - M N NN.- - - - - om. - - - W N - - - - - - - - ea. - - 1 _ >0 mm.. - me. - - - L m e. - - - - - - mm. - - - w ... - - - - mm.- - - - - - - > we - - - - - - Ne- - - - - 52 - - - - - - - mm. - - - nN> B 2 1 - - - - .. - Fm... - - .. m3 - - - - - - - - - Nm. - o>> - - - - oe. - - - - - - . n>> - - - - - - - - mm. - - w m - No. - - - - - - - - - m >24 - - - mm.- - - - NN.- - - - e - - - - em. - - Nm. - - - N - - - - - Nm. - - - - - > >m - - - - - - - - - oe. - m >4 - - - - - - - - - - Ne. N >4 45>aauea> msmcmoca oucuuo >>< . xucumz couumpwccoocquH w.m mpnmh “ ' fugx fl 1‘ .y 224 N.N aunau cu em:_mucou mu mmsm: m>nmucm> wcu ou xmx meek ewe mam Nwa >wa UN>>24 DN>>aa MN>>24 o>>>La n>_>ea m>ca N>La em. - - - - - - - me. mm. - - - - - - - - - oe. - - - - - - - - - N.m- me. mm. - - - om.- - - - - - - - - 04. Ne.- - - - - - 54.- me. - - - - - - Ne. - m¢.- - - - - - mm. - mm.- - - - - - - - - - - - - - - - 45.- - - Nm.- - - - - 04.- - - - - - - - - Nm.- - - - - - - - - - - - em. - k0 >o <1' mmoomq- P on mp up— n~P>La .munauea> Avaseuucouo N.N w>nm> 225 .mucmuouemmou couumumccou ucmsos1uuznoca 0.0000000 0L0 mcouumumgcoo Auv .xucume Facumuco mcu mauscow cu com: 003 mwpnmuc0> mm mo quou < Ago Ape u zv .czogm mum Amo. :mzu mmmp 0v meowumumccou ucmoueucmwm xpco Amy ”mmuoz >.m munch cu vacuoucoo mu muse: m>00w>0> mzu ou max mgpa 0>0 m>0 N>0 u>0 N00 000 >0. 00. 00. um. - - 0 1 40. >0. 00. - - 0 - - >0. >0. 00. 1 0 - - - 04. - - N >0 1 1 1 1 1 0m. m on 10>0auam> >0000uucoav 0.0 «>0au .mucmuouwemou cowu04occou ucmsoe1uozco>0 0.0000000 mcm1mcovum>mceoo Adv .xvcume chwmwco msu mewscow :4 now: mm: mm—nmw>0> 4m 40 Pmuou < Any Aem u zv .czcgm mum Amo. :msu mmmp 0v mcowumpwccou ucmuwewcmwm cho Amv ”mmuoz Nd wrap—1 Cw Uwcwmucoo mw mam: ownmem> 9.3. on hmx 9:104. 226 4>0 0>0 N>0 _>0 0>>>04 0>00 >>La N>0 >>m >>4 m>00 >0. 00. 00. 00. 1 1 1 1 - N0. 1 0 1 00. N0. >0. 1 1 1 1 1 04. 1 4 1 1 04. N0. 04. 1 1 1 1 1 1 m 1 1 1 N4. 1 1 1 1 1 1 1 N 1 1 1 1 1 1 1 1 04.1 40. 1 > >0 1 1 1 1 1 04.1 1 1 1 1 1 0> 1 1 1 1 1 1 1 1 1 1 00. 0N> 1 1 1 1 04.1 1 00. 1 1 1 1 0>> 1 1 1 1 1 1 1 1 44. 1 1 0>4 1 1 1 1 1 1 00. 1 1 1 N4. 0 >00 1 1 - 1 1 1 1 04.1 1 1 1 4 >0 1a>04400> mumo acumccouummzo ucmcaum usoguwz msmcmoca uwcho 1 chumz cowumpwccoucwucu m.m wpnmp 4.0 04004 :0 00:00ucoo 00 0050: m>00000> mcu ou 004 0:44 227 0>0 >>0 4>4 0>4 N>4 >>4 4>4 4>4 0>4 N>4 >>4 N0. 1 1 1 1 1 1 1 1 1 1 4 00. 1 1 11 40. 1 1 1 1 1 1 N >0 1 - 1 1 >0. 1 1 1 1 1 1 0> 1 1 1 1 N0. 1 1 1 1 1 1 0> 1 1 1 1 1 N0.1 1 1 00.1 1 1 0> 1 1 1 >0.1 1 1 1 1 1 1 1 0N> 1 1 1 1 >0. 1 1 1 1 1 1 0N> 1 1 1 1 1 1 1 40. 1 1 1 0 >04 1 1 1 1 1 1 1 1 N0.1 1 1 4 00. 1 1 00. 1 1 1 1 1 1 0 1 1 >0. 1 N0. 1 1 40. 1 1 1 > >0 1 1 - 1 - 1 1 1 1 00.- - N 1 1 1 1 1 1 00.1 1 1 1 1 > >4 1 1 1 1 1 1 1 1 1 1 >4. N >4 10>00>00> mums 0000:000u0mso ucmozum guwz 050cmoca 040040 1 chumz co>u0>mgcoucmu0> o_..m £0.21 .u00000emmoo cowu0>m00oo ucmeoe1u000000 0.:o00004 000 0co>u0>o0coo Auv .x00ume >000m00o ecu mC>E0o4 :0 0000 003 0w>00000> mm yo >0uou < Anv A04 u zv .03000 000 Amo. cmgu 00m> 0v 0cowu0>000oo ucmu>mvcmw0 >>co Amv “0muoz >.m m>00> 00 umc>0ucoo 00 0050: m>00>00> mcu ou 000 0044 “w 4>0 0>0 N>0 >>0 0N>>04 0N>>04 0>04 0>04 >>04 4>0 2 00. 40. 00. 00. 1 1 1 40. 1 1 0 1 44. >0. 04. 1 1 1 00. 1 1 4 - 1 1 1 1 - 1 00. 1 1 0 1 1 1 40. 1 1 1 00. 1 1 N >0 1 1 1 1 04. >0.1 1 1 1 1 0> 1 1 1 1 1 1 44.1 - 1 1 4> 1 1 1 1 1 1 00. 1 - - 0N> 1 1 1 1 1 1 1 1 1 00.1 0>> 1 1 1 1 1 1 1 1 00. 1 0>> 1 1 1 1 1 1 1 1 >0. 1 0 >04 10>00>00> >000000000V 0>.0 0>00> 229 mmo. mo. mNo. mmo. moo. mmo. «000u 000> 0V 000000000000 00 >0>00 0Nuz 000.0 0N.4 0Nuz >40.>0 0>.0 0~>m 040000 000 .>.0.0v 1.0000 00000 4000 0000u004 0>1z 00>.>0 44.0 0>1z 044.>0 N0.0 d'F-Z MNZ sz V V v 0~00 0>0000 000 .A.o.mv 4.000: 00000 >000 >00E>00 .0>00>0 00u 04 0000u 0u au>0000 00 0000000>>03 .000000000 >>000>>000 u00 0E0> 10000 0u>z 0u000u00 00 u000000 .> 0nmo0o Fmoo quwcwpo .0000000 00u 0> 000000 u0000u0 >0000>u 00u 00003 00 000002 .0>0>>0 00u 00 000000 u0000u0 0000 0003 000 000010400 00 000002 .00E>u 0005 00 0000u 0000 0u00wu00 00 u000000 .0E0>0000 00u0000000 1000 00 >00000>000000 0>000s 1000 guuz 0u000u00 00 u00u000 .4 000000 >000 mu0000u000 .< 00>00>00> 000 000000 >000 _—.m 0>000 >< u000v >0u00p uv 000000 >000 0u 00u0>00 x>u0000000000 00>00>00> mmo. moo. mo. mo. 230 mo. moo. 1~000u 000> 0v 00000>0>0m>m 00 >0>010 0>1z >N0.0N0 04.00 0>1z >00.>0 00.0 0>1z >00.>0 >0.0 4>1z 044.0 0>.0 0>1z >00.N0 N0.0 0Nuz 000.0 00.> 04 >0 0 E0000 000 .>.o.0v 4.000: 00000 >000 000>u00> 0>1z 400.00 00.>N 0Nuz >00.o 00.> >Nuz 040.0 04.0 0>1 >0N.>0 00.N 4>1z >40.>0 04.> 0>1z >00.>0 00.0 0N .5 0 30.00 400 .>.0.00 1.0400 00000 >000 >00E>00 .000000 >00>00e 00u 00>u>03 0> u0000 u00>u00 300 000 0E>u u0000um .4 .u00>u00 000 0E>u 00>u000000> 0u>00001u000000 .m .0>0>>0 00u 0> 00000 ou 0u>0000 00 000000>>>>3 .N .00 000 00>0 0u00>u00 00 0000000 .> wmso0o >000 >00>u0000 .o .0E0>0000 00>u0000 >0000>>0 00 .>0u000000>>00 .0>u0000000000 .>00>00>000000 000000 00~> 100000 00000000000 00 u000000 .4 .000>u00 000 0E>u 00>u0000u0> 00>:0001u0000um .0 00>00>00> 000 000000 >000 >0000>00000 >>.0 4>00> 231 .mw#::wE om mm UmUOU 0000000 00 0000 0000000 000000000 .00000>e 000 0000000 000 00 00000 0000000 00 000500 >0000< 00000 0>0E000x0 mm .mm 0>00w00> A00 .m n owgmwwmchcm Ucm 00000 000 000 000>>03 m> u 00000000 00 00 0000000 >>00 0000>0 00 00000 .No 000 mm 0>00>00> AOV 0mumm >N-0> w>u0> m>um> N>no> 0-0 0-0 muo 00002 w m 0 0 0 m N > 0000 .0< 0>00000> >00 .000000 >00000 00 00000 - m< 0>00000> ADV 000> N00u>w 0ows>0 0o0u>0 000u>m 0omu> 00 0000000 0 0 0 0 0 N > 0000 .>0 000 .0 .0 .>0 .0 .>0 00>00000> 000 ”030>>00 00 000 00>00000> 000000000 000 00000 0000 .0000: 0000 00 000 0000000>00 00000000 000 000020 mo. 0N0. ~0000 000> av 000000000000 00 >0>00 .0E0>0000 00000000 >00000>0 00 0>nz .>000000000>00 .0000000000000 A00.NV .>00000>000000 000000 0000 00.0 -00000 00000000000 00 0000000 .0 0>uz A00.Nv .0000000 000000 0000000 00.0 000 0000 00000000 00 000002 .0 0000 0>0E0m 0~0m 0>0E0m 00>00000> 000 000000 >000 000 .A.0.0V «.000: 000 .A.0.0v «.0002 00000 >000 00000000 00000 >000 0000000 00000000000 _>.0 00000 232 No. 00000 000000 000000000000 00 >0>00 >> 0 0 >> 0m000 >000 00000 >000 >00000>0 00000000 >00000>0 0000000 05000000 00 000002 0000>000 000 00>000 00>000 ”000 00000000 00000000 Am 00000 00000 000000-0000 000000 >000 00 0000>00 x>00000000000 00>00000> N>.m 0>00H 233 000000000 .0 quwcwpo .N 0003500500 .0 00. 5050 0000 0 0 m 0 0 u 00 0 00 5 00.0 n 000300-050 q 0 0 00. 50:0 0000 a 0 00 0 0 u 00 0 N 0 00.5 n 000300-050 0 0 0 0003000 50.: 000002 .000 003000 0000030 zw_3000 00m000 P000 00000005 _00000000 .0 P0000050 .0 0003500500 .0 003000 5000 0005000 .0 .< .wm3000 F000 00000002 003000 0000030 >0F3000 00 003000 _000 :0 05000000 00:0po 00 0000532 00 000xpmc< 000300-050 mp.m 0—005 +00 00-00 00-00 0N-FN 00-00 00-0_ NF-0 0-0 0-0 03\00: 0 0 0 0 m 0 0 N F 0000 ”0 0_00000> A00 .0 u 000000030000 000 00000 0005000x0 00 u 00000 00: 030 0000003 ”F u 00003000 00 00 0030000 0000 000000 :0 00000 ”N 0000000> A00 000_ 000-00 000-F0 000-F0 000-FN 000-0 00 0000000 0 0 m 0 0 N — 0000 ”0 000 F 00_00_00> A00 ”030_000 00 000 00000000> 000000000 000 00003 0000 .0000: 0000 00 000 0000000>00 00000000 000 000020 mruz 0~u2 000. A00.0v Amm.mv .000000 000 :0 000000 0000300 00.0 00.0 0000 0003 000 00300 00 000532 .0 0Pu2 mpuz .0500 00. 000.00 000.00 -0000 0000005 00 00000000 003000 m 00; 00.0 03.200000 00000000000 00 0000000 .0 0_u2 0_u2 0000. 000.0 A00.v .000000 000 :0 00.0 00.0 00000 00 00_3000 00 00000000003 .N m—nz 0Fu2 00. 000.00 000.0 00.0 00.0 .00 000 00>0 00000000 00 0000000 .F 00000 0000 mm 0~00 000500 0~00 000500 0—00000> 000000000000 000 .W.0.0~ «.0002 000 .A.0.m~ 0.0002 00 _0>00 000 300 000 0 0: A00000 00 00000000> 00000F0m 00 A0000 003000 0000030 0003000 300 0:0 000: 00 0000000500 00.0 00005 235 Table E.l5 Stepwise Multiple Regression Analysis of All Clinic Programs on the Rating of Faculty Willingness to Participate (N = 44) Percent Level of of Variance Significance Variable Direction* Accounted For (p less than) l. Presence of con- + 6.2% .lO ferences. 2. Student's hours/ + 6.3% .09 week in the clinic. 3. Required, selective, + 4.0% .l8 or elective pro- gram.** 4. Ranking of Goal: + l.5% .40 Provide students with experience to assist in career choice decisions.*** Total of the four 18.0% .095 variables. *Indicates the directionality of the relationship between the varia- ble listed and the rating of faculty willingness to participate in the program. **Coded as: Required = l; Selective = 2; Elective = 3. ***Top ranking = 9; . . . ; Unranked = 0- 236 Table E.l6 Chi-square Analysis of Numbers of Preceptorship Programs in the Goal Groups by Educational Groups Matrix Educational Groups Goal Group§_ Low_ High Results l. Continuity 5 4 Chi-square = l.03 df = l 2. Practical 4 8 p less than .30 237 .0: 000000 00 0000 000. 0000 0000 0 0 n 00 00.0 n 000300-000 - 003000 0000 0000000000000 .00 000000 00 0000 N00. 0000 0u00 0 0 n 00 0.0 u 000300 -000 - 003000 0000 00000000 000 .0: 000000 00 0000 000. 0000 0000 0 0 u 00 0.0 u 000300 -000 - 003000 0000 0000000 A00 00 000000 00 0000 00. 0000 0000 0 0000. u 0 mm": 0: 000000 000. 0000 0000 0 000.0 u 0000 000 u 0 00.0 - 00 00.0 - 00 000 u c 00.0 - 00 00.0 - 0m .00 000000 00 0000 00 u 0 000. u 00 .000000000 0000 00 00000000 000000000 000. u 00 0003000 .00000000 0000000000000 000 000000 000 003000 0000000 -00300V 0000030 000 003000 0000 00 00000000 000300-000 .0000000> 00 00000000 .00000 000000-000 .0000000 000000000000 00005 0003000 0000000000 00 0000-0 .00000000 -0000 00000000000 030 0003000 0000000000 00000000000 00 0000 0003 0000 00000000000 .00000000 0000000000000 000 00000000 000000 0000 000 003000 0000 000000 0000 -003 000 00 0000030 0000000 .0 .000000000 0000300 000 0300> 00000000300 0000 00 00000 -000 00003000 0000 00000300 00 000000 00000 000 00000000 000000 0>000000 000 0>0000000 .0 .000000 0000000 00000 00000 00 00000000 00000300 00 0000 03003 00000000300 000000 00 00000 00 00000000 000000 0000 00000000000 00 .00000000 00000000 .00000000 000000 .0000 -0000 00000000 000000 0000 0000 -000 00000 00 00000000 00000300 .0 .03_0> 00000000300 00 0000000 0000300 0000 0000000 -00 0000300 00 0000000 0000300 000E0x00000 0000000 0000 0300> 00000000300 00 0000000 0003000 .0 00000000»: 00000000 0000000000 00 00000 00.0 00000 238 00. 0000 0000 0 .00 000000 N? u : PM. u L .00 000000 00 0000 NW": M_.0IHL .00 000000 00 0000 NWHC NO.IHL .00 000000 00 0000 00 u 0 00.- u 0 .00 000000 00 0000 N¢ n : FF. n L .00 000000 00 0000 N0 " 0 00. u 0 0003000 cowpm—mLLoo 00000000000 cowpmrmLLoo 00000000000 00000000000 00000000000 00 $0 00 O 0 .00000000000 000000000000 .00000000000 000000000000 .00000000000 000000000000 .00000000000 000000000000 .00000000000 000000000000 .00000000000 000000000000 0000 0000 00000000000 .0003000 0000-0030 000 003 0003000 0000000000000 00 0000000 000 0003 0000000000 000>000000 .0 .000 -000 000 00 000090 0003000 000 00000300 00 000030 000 0003 0000000000 000>000mmz .0 .00000000 000000000 000 003 00000000 00 0000000 000 0002 0000_00000 000>000m0z .0 .00 000 00>0 00000000 00 0000000 000 0003 0000000000 000>000m0z .0 .00000 -0000 000300>000 0000 000 0>00 00000000 0000000 00003 00000000 00 0000000 000 0002 0000000000 000>000000 .0 .000000000 0000 0000000 00 0000000000000 0000300 0003 0000000000 000>000000 .0 ”000 mx0000 0000030 00030000 0000 000 0000—0 00 0000030 000 00 0000000 0003000 00030000000 00.0 00000 0000000 0 00000000 .m 239 00. 0000 0000 0 .00 000000 00 0000 00 u 0 00.- u 0 .00 000000 00 0000 00 u 0 00.- u 0 .00 000000 00 0000 N? H C 0P. N L .00 000000 NV H C mfi. H L .00 000000 00 0000 00 u 0 00. u 0 0003000 .00000000000 00000000000 0 0o 00000000000m .00000000000 00000000000 0 mo 000000000m0m .00000000000 00000000000 0 mo 00000000000m .00000000000 00000000000 0 00 00000000000m .00000000000 00000000000 0 mo 00000000000m 0000 0000 00000000000 .000020000000 000 000000 00 00 00005 -0000000 0000 00 0000000 00000000 00 0000000 000 0003 0000000000 000>000m0z .0000500000000 00 0000mo0o00000 000005000 000 00003 05000000 0003 00000000 00 0000000 000 0003 0000000000 000>000mmz .00000300 0000 000300 000 003 5000000 000 00 00000300 00 0000000 000 0003 0000000000 000>000000 .000000 000 00 00000 00 0003000 00 0000000000: 000 0003 0000000000 000>000000 .0003000 @00000 -0000000 mo 000000000 000 0003 0000000000 000>000000 .m 00000000»: 00000000 00000000000 00.0 00000 24() .00 000000 00 0000 00. 0000 0000 0 000. u 000 00 u 0 00.0 n W0 00.0 n 00 00 u 0 00.0 n W0 00.0 n 00 .00 000000 00 0000 N0 - 0 00. n 0 .mmammmm .00000 000000-000 .0000000 000000000000 00002 0003000 0000000000 00 0000-0 .00000000000 00000000000 0 mo 00000000000m 0000 0000 00000000000 .00000 -000 0000005 00000 00 0000 000000 000000000 00000000 00 0000 0000000 000>o00 00 00 000000 000 00 0000 -0:0 0005000 000 00 00000: .0 .00000000000 00 00000000 000 0003 0000000000 000>000000 .0 0000000000 00000000 00000000000 00.0 00000 SELECTED BIBL IOGRAPHY SELECTED BIBLIOGRAPHY A. REFERENCES CITED A.A.M.C. Curriculum Directory. Washington: Association of American Medical Colleges, 1972. Anlyan, William G. "1985," Journal of Medical Education, 46:917-26, November, 1971. Bogdonoff, Morton D., Stanley W. Elwell, and Julian M. Ruffins, "Medical Out-Patient Department Teaching," Journal of Medical Education, 38:885-89, October, 1963. Colmore, John P., and Stewart WOlf. "A New Design for Service and Teaching in the Out—Patient Department," Journal of American Medical Association, 156:830-33, October 30, 1954. Delbecq, A. L., and A. H. Van de Ven. "A Group Process Model for Problem Identification and Program Plan- ning," Journal of Applied Behavioral Science, 7:466-92, 1971. Flexner, Abraham. Medical Education in the United States and Canada. New York: The Carnegie Foundation for the Advancement of Teaching, 1910. Goodrich, Charles H., Margaret C. Olendzki, and Anne- marie F. Crocetti. "HOSpital-based Comprehensive Care: Is It a Failure?" Medical Care, 10:366, July-August, 1972. Hammond, Kenneth R., and Fred Kern, Jr. Teaching Com- prehensive Medical Care. Cambridge: Harvard Uni- versity Press, 1959. Lee, Peter V. Medical Schools and the Changing Times. Evanston, Ill.: Association of American Medical Colleges, 1962. 241 242 Merton, Robert K., George Reader, and Patricia Kendall (eds.). The Student Physician: Introductory Studies in the Sociology of Medical Education. Cambridge: Harvard University Press,’l957. "NIH Funds To Be Less in Fiscal 1974 Than in Fiscal 1972." Bulletin of the Association of American Medical Colleges, Vol. VIII, No. 2 (February, 1973). Reader, George G., and Mary E. W. Goss (eds.). Compre- hensive Medical Care and Teaching. Ithaca, N.Y.: Cornell University Press, 1967. Reed, David E. "Twelve Years' Experience With a Compre— hensive Ambulatory Care Program," Journal of Medical Education, 45:1043, December, 1970. Sanazaro, Paul J., and Barbara Bates. "A Joint Study of Teaching Programs in Comprehensive Medicine," The Journal of Medical Education, 43:777-89, July, 1968. Schulman, John, Jr. "The Role of Hospital Ambulatory Service in Medical Education," Journal of Medical Education, 46:246, March, 1971. Shank, Robert E. "Three Years' Experience in the Coordi- nated Outpatient Program at Washington University," Journal of Medical Education, 31:283-93, May, 1956. Snoke, Parnie S., and E. Richard Weinerman. "Compre- hensive Care Programs in University Medical Centers," The Journal of Medical Education, 40:625-57, July, 1965. Somers, Anne R. Health Care in Transition: Directions for the Future. Chicago: Hospital Research and Educational Trust, 1971. Sussman, Marvin B., et al. The Walking Patient: A Study in Outpatient Care. Cleveland: The Press of Western Reserve University, 1967. "U.S. Medical School Enrollments, 1968-1969 Through 1972— 1973." Journal of Medical Education, 48:293-97, March, 1973. Van de Ven, A. H., and A. L. Delbecq. "The Nominal Group as a Research Instrument for EXploratory Health Studies," American Journal of Public Health, 62:337-42, March, 1972. 243 Wasserman, Edward, et al. "Medical Student Involvement in Comprehensive Health Care," Journal of American Medical Association, 215:2098, March 29,41971. Weinerman, E. Richard. "Yale Studies in Ambulatory Medical Care," The New England Journal of Medicine, 272:947-54, May 6, 1965. White, Kerr L., and William L. Fleming. "Improving Teaching on Ambulant Patients," Journal of Medical Education, 32:30-36, January, 1957. B . GENERAL REF ERENCES Analyn, William G. "Chairman's Address," Journal of Medical Education, 46:917-26, November, 1971. Beloff, Jerome S. "Medical Student Response to a Program for Teaching Comprehensive Care," Journal of Medical Education, 45:1047-59, December, 1970. Caughey, John L., Jr. "Clinical Teaching During Four Years," Journal of Medical Education, 31:530-34, August, I956. Deitrick, John E. "Organization of Outpatient Depart- ments," Journal of Medical Education, 41:710-23, July, 1966. Glass, Gene V., and Julian C. Stanley. Statistical Methods in Education and Psychology. Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1970. Gregg, Lucien A. "Outpatient Program for Senior Medical Students," JQurnal of Medical Education, 31:757-64, November, 1956. Kerlinger, Fred N. Foundations of Behavioral Research. 2d ed. New York: Holt, Rinéhart and Winston, Inc., 1973. Pozen, Michael W. "A Family Health Care Center--An Ongoing Student Endeavor," American Journal of Public Health, February, 1972. 244 Smith, Hugo Dunlap. "Essays in Medical Education," American Journal of Diseases of Children, 10:l85—88, August,—1965. "The Teaching of Comprehensive Patient Care." Editorial, American Journal of Public Health (March, 1970), pp. 429-32. ”'Tfix'fiximnflimjufulMilliilifillllhliimm” 061710