In his 1910 report, Abraham Flexner1 articulated an ideal model of medical education that included a university-based, full-time, salaried faculty. Such an arrangement, he argued, would protect faculty members from the pressure to generate their own income through clinical practice and so allow them to pursue teaching and research free from that distraction. Over the next 100 years, this model underwent a metamorphosis to a concept of full-time faculty that is significantly different from what Flexner imagined. Problems that Flexner was attempting to solve with his model have reemerged in different forms, and words that he used now have different meanings. While there now is a full-time, salaried clinical faculty, there again is concern that pressures for faculty to engage in practice compete with time for teaching. Tracing the evolution of the “full-time” concept for clinical faculty and the factors that have affected its implementation permit us to understand how and why this situation has occurred.
Flexner was quite clear in what he was attempting to change in medical education. There was “no place” in medical schools for the
scientifically dead practitioner, whose knowledge has long since come to a standstill and whose lectures, composed when he first took his chair, like pebbles rolling in a brook get smoother and smoother as the stream of time washes over them.1(p57)
Flexner also was concerned that if a medical school did not own a hospital and control its medical staff, teaching would be the province of the “local profession.” Although he meant “no disrespect to the practicing profession,” these physicians were “not teachers; they have neither time for, or effective interest in, productive teaching.”1(p110)
To ensure that faculty had the time for scholarly pursuits, Flexner believed that faculty in both basic science and clinical departments should be salaried by the medical school or its parent university. This situation already existed in basic science departments at some medical schools.2 In fact, the presence of salaried professorships in the basic sciences was included as one of the requirements in the 1910 American Medical Association (AMA) Essentials of an Acceptable Medical College.3 However, this was not the case in clinical departments. Flexner recognized that salaries paid to clinical professors would be “inadequate” compared with their private practice income but that there was “no inherent reason why a professor of medicine should not make something of the financial sacrifice that the professor of physics makes.”1(p102) Although a clinical professor might conduct a consulting practice, such a practice, “developed in a professional or commercial, rather than in a scientific spirit—may prove quite as fatal to scientific interest as general practice.”1(p102) Although not explicitly stated by Flexner, other proponents of the concept of a salaried full-time clinical faculty expected that any income generated from practice would go to the institution.2
Flexner acknowledged that “medicine is expensive to teach” and could not be financed solely out of fees paid by students. Medical schools would need university support and philanthropy to fund a quality program of medical education, including a salaried full-time faculty.1 Flexner himself came to play a key role in providing philanthropic support for the implementation of the full-time system.
Initial Implementation of the Full-Time System
In 1913, the General Education Board (GEB) of the Rockefeller Foundation agreed to fund full-time clinical professorships at the Johns Hopkins University medical school.4 Preceding the gift, John Hopkins was visited by Abraham Flexner on behalf of the GEB. He found that the preclinical departments were more productive in research than the clinical departments, a fact he attributed to the consulting practices of the clinical faculty.2 The GEB gift was contingent on all clinical income going to the institution.2
The implementation of the full-time system for clinical professorships in medicine, surgery, and pediatrics at Johns Hopkins was not without controversy. William Osler, who had previously left Johns Hopkins, was concerned that the system would result in “a set of clinical prigs” whose sole concern would be the laboratory and whose only “human interest would be research.”2 Lewellys Barker, Osler's successor as professor of medicine at Johns Hopkins and an early supporter of the full-time system, resigned his chair for a private consulting practice rather than accept the system's limitations.3
Implementation of the full-time system for clinical faculty at other institutions was slow in coming. An AMA special committee on the reorganization of clinical teaching reported in 1915 that the full-time professorship plan was “now on the proving ground” and that there was, “with one exception... no medical school so financially situated as to institute the plan.”5 While the committee report elaborated the pros and cons of the full-time system and came to no formal judgment, it stated that it was “imperative” for clinical faculty to get away from an “extensive time-consuming private practice with the minimum of teaching responsibility.”5
Abraham Flexner, who had accepted a position with the GEB, soon found himself in a position to turn his model into a reality. Between 1913 and about 1920, he concluded agreements, accompanied by large gifts, for implementation of the full-time system at six medical schools. Funding for faculty support was provided to additional medical schools during the 1920s.6 Flexner himself wrote in 1924 that in about “one-half dozen” medical schools, “the material conditions for good teaching and productive clinical research” were “perhaps the best to be found anywhere in the world.”7 In the remaining 70 U.S. medical schools, clinical teaching was “partly professionalized in some, hardly at all in some, and not at all in the rest.”7 To a considerable extent, clinical teaching remained an “incident in the life of a busy practitioner.”7
Despite the enthusiasm of some reformers, during the 20 years after the Flexner Report the expectation that there be full-time teachers remained focused on the “laboratory branches.” The 1919 and 1927 versions of the AMA Essentials of an Acceptable Medical College8,9 stated that medical schools should have “at least eight expert thoroughly trained professors in the laboratory branches, salaried so that they may devote their entire time” to teaching and research. There is no mention in these standards of salaried clinical faculty.
The Medical School Faculty in the 1930s
In the 20 years following the Flexner Report, the number of medical schools decreased sharply, but the number of individuals identified as faculty rose. In his report, Flexner had reported that there were 148 U.S. medical schools with 8,032 faculty of all ranks.1 In 1932, the 76 medical schools listed in the annual report on medical education in the Journal of the American Medical Association included 14,387 faculty of all ranks.10 In earlier faculty counts, faculty were defined as individuals with appointments ranging from professor to instructor/assistant/demonstrator and included those considered to be full-time and other. By the mid-1930s, the most common faculty rank system included the categories of professor, associate professor, assistant professor, and instructor.11
The concept of tenure, defined as an indefinite period of appointment, existed in at least some universities since the early 1900s. This was coupled with the presumption that the faculty member who did not achieve tenure or promotion would be required to leave the university.12 A survey of four-year medical schools conducted in the early 1930s showed variable policies related to tenure for faculty. About three quarters of medical schools had the possibility of tenure for at least some academic ranks, and about one third of schools offered the possibility of indefinite appointment to all ranks.11 The concept of tenure in medical schools gained growing acceptance during the 1930s. For example, the 1944 revisions to the AMA Essentials of an Acceptable Medical College stated that “[R]easonable security of tenure must be assured in order that the personnel of the faculty may have adequate stability.”13
The Full-Time System Comes of Age (1940s to 1960s)
By the end of the 1940s, the concept of “full-time” had crystallized to mean faculty paid by the institution. However, by this time an explicit distinction between categories of full-time clinical faculty was being made. An Association of American Medical Colleges (AAMC) study of U.S. medical schools conducted in the late 1940s defined “strictly full-time” faculty members as deriving their income from the medical school, to which any practice income reverted. “Geographic full-time” faculty members were allowed to supplement their income from private practice, which was conducted in an office provided at the medical school or teaching hospital. Medical schools often set limits on the amount that could be earned from private practice.14
The geographic full-time system was lauded in the AAMC study as increasing the number of clinicians who would spend their entire day in the hospital and medical school while earning a large portion of their income from the practice of medicine.14 However, the report noted the danger that medical schools would bring pressure on faculty to increase their earnings “in order to help finance the overall activities of the institution.”14 The general acceptance of the concept that clinical faculty should be able to engage in medical practice was illustrated by the 1944 revision of the AMA Essentials of an Acceptable Medical College,13 which stated that appointment in the clinical departments “does not exclude men who are in the active practice of medicine and surgery.”
Perhaps stimulated by the flexibility permitted by the geographic full-time system, the number of full-time clinical faculty was steadily increasing. In 1951, there were 2,277 full-time clinical faculty (57.9% of total full-time faculty); in 1960, the number of full-time clinical faculty was 6,948 (64.9% of total full-time faculty).15
The enactment of Public Law 89-97 (Medicare) and the subsequent regulations as defined in intermediary letter 372 issued in 1969 allowed the collection of professional fee income by “full-time clinicians” when teaching physicians provided services to “private patients.”16 The availability of this funding resulted in a culture shift as, in the past, “little or no income was generated from ward patients.”17
Further Evolution of the Full-Time Concept (1970s to 1990s)
After the enactment of Medicare, clinical practice income came to be an explicit and increasingly important source of medical school revenue. Others, such as Medicaid and private payers, began following Medicare's lead.16 Over the next 30 years, the number of full-time clinical faculty grew in the context of relatively stable student enrollments (see Table 1). Clinical faculty came to constitute an increasingly larger percentage of all full-time faculty (see Table 1) in parallel with the increasing importance of clinical practice as a source of medical school revenue (see Table 2).
In addition, the enactment of Medicare allowed hospitals to obtain support for faculty involvement in teaching and supervising residents. Graduate medical education was in its infancy at the time of the Flexner Report. By 1990, there were almost 83,000 residents in accredited residency programs.18 The growth in graduate medical education likely contributed to the increase in the number of clinical faculty.
The growth in importance of clinical income was not without its threats. One commentator on surgical practice reflected in 1970 that “(W)e are bewildered by the rapidly changing suggestions, guidelines, and requirements of a great new political force in America, the third-party payer of the medical bill.”19 In order to cope, this commentator recommended “the abandonment of the rigidities of the full-time, full-salaried position.”19 Others at the time agreed that the geographic system was more appropriate for the current conditions. Such a system was held to be “less expensive to the school.” Salaries and fringe benefits could be set at a lower level because geographic full-time faculty would be expected to generate a significant portion of their salaries through clinical practice. Therefore, the geographic full-time system eliminated the need for the salary guarantees associated with tenure20 and foreshadowed the creation of non-tenure-granting employment tracks for clinical faculty.
Between 1960-1961 and 1990-1991, the percent of total medical school revenue from clinical faculty practice increased from 3% to 31.4% (see Table 2). Various accommodations involving faculty were made in the context of this growing dependence on clinical income.
Faculty practice organizations
Although group practice by medical school faculty had existed for decades, the appearance of Medicare and Medicaid in the mid-1960s stimulated renewed interest in organized systems of faculty practice.21 Between 1960 and 1985, the number of medical schools with faculty practice plans increased from 6 to 118.22 As managed care appeared in the late 1980s, faculty practice plans expanded from a focus on billing and collections to the operation of satellite centers and ambulatory care facilities.23 To staff these practice sites, physicians were recruited into the medical school whose main responsibility was clinical care and who were not intimately linked to its education and research missions.
Changes to faculty appointment and tenure policies
The growing importance of clinical practice resulted in a recognition that the traditional faculty appointment and tenure system was too rigid. Faculty who devoted their time to clinical care and teaching made valuable contributions to the institution but could not meet the criteria for promotion and retention in a tenure system that focused on research as a major criterion for advancement.24 Medical schools responded with the creation of non-tenure-granting clinician-educator tracks. Clinician-educator faculty obtained their income from clinical practice and so freed the university from the financial obligations of tenure.24 By the mid-1990s, about three quarters of medical schools had introduced a separate appointment track for full-time clinical faculty whose main responsibilities were teaching and research.25 The introduction of non-tenure-granting faculty tracks resulted in a significant shift in clinical faculty appointment status. By the mid-1990s, more than one half of full-time clinical faculty with the rank of assistant professor and above were in faculty tracks that did not grant tenure.26 In 2004, only about one quarter of new MD faculty in clinical departments were hired on a tenure-eligible track.27
Even tenure did not ensure full salary, as the salary guarantees previously associated with tenure no longer existed for most clinical faculty. In 2004, of the medical schools that did offer tenure to clinical faculty, about 50% included some form of salary guarantee; however, the vast majority did not guarantee the full salary.27
What Goes Around, Comes Around? (2000 and Beyond)
In the 100 years since the Flexner Report, clinical faculty have become “full-time” and “salaried.” However, instead of deriving their salaries from the nonclinical resources of the medical school, such as tuition and income from endowment, they are significantly contributing to institutional financing through their practices. They also may be contributing to the scholarly environment of the medical school, as well as to its financing, through their participation in research.
Medical schools are dependent on clinical revenue to support their other missions, such as research and education. In addition, the increasing number of clinical faculty who must engage in practice in order to generate their incomes are vulnerable to time pressures that mitigate their ability to participate in teaching and research. Individuals on the clinician-educator track have been found to have less mentoring and less time for academic pursuits. They also may progress more slowly through the academic ranks.28
Is the current situation different from the one deprecated by Flexner? What is being done to address Flexner's concerns about the limitations of using the busy practitioner as a teacher? Is the current clinical teacher in danger of becoming the “scientifically dead practitioner” that so concerned Flexner? Some illustrations follow of ways being used to ensure the ongoing availability and quality of clinical teaching.
Providing explicit compensation for teaching
Flexner was concerned that concentration on clinical practice would limit the amount of time faculty devoted to teaching and research. In the 1990s, systems were introduced at some medical schools to explicitly track the time faculty devote to teaching, as a means to compensate departments, if not faculty, for this effort.29 The attempts at individual institutions to quantify and reward faculty teaching activity have been variably successful in their implementation. There are no comprehensive national studies documenting whether such a system, in general, improves the availability of clinical teachers to support the educational program. However, the process of designing a system and conducting the requisite data gathering has focused attention on the educational mission.29
Enhancing the image of education and educators
A number of medical schools have created “academies of medical educators” as a means of recognizing faculty who are excellent in teaching and educational scholarship. Academies have been defined as “a formal organization of academic teaching faculty recognized for excellence in their contributions to the teaching mission of the medical school.”30 As structural units with schoolwide membership, the academies have institutional funding to support mission-related activities and to provide protected time for faculty participation in educational development and scholarship.31 The academy movement serves as a focus for the educational mission, for example, through supporting excellent and committed educators by providing mentorship and faculty development.31,32 In recent years, there has been a rapid increase in the number of schools with academies.32
Ensuring Clinical Faculty Remain Knowledgeable
In Flexner's day, opportunities for continuing education were focused on repairing the deficiencies in medical school education. By the 1930s, the emphasis of continuing education shifted to keeping physicians up-to-date with new medical knowledge, and the suggestion was made that ongoing education become a requirement to ensure competent medical practice.33
Currently, continuing education is an expectation. For example, 62 of a total of 68 U.S. medical licensing boards require continuing medical education for renewal of a medical license.34
Summary and Comment
A number of circumstances led to the inability to fully implement the strict full-time system envisioned by Flexner. Most important was the difficulty in funding this model. Only institutions that already had large endowments or could gain additional financial resources from philanthropy were able to make Flexner's recommended changes. Instead, the number of full-time clinical faculty increased nationally only in parallel with the ability to generate revenue from clinical practice earnings. The advent of this funding source led to a spiraling growth in both faculty numbers and in the percent of total medical school revenue that clinical practice contributed to medical schools.
Nationally, we now have a large, full-time, salaried clinical faculty. However, our definitions have changed from those used in Flexner's recommended model. The current “salaried” full-time clinical faculty member is highly dependent on clinical practice for his or her income. It is not even the case today that being full-time “faculty” means significant participation in education. Flexner's model aimed to ensure that medical school faculty members were dedicated to teaching and available to students on a regular and ongoing basis. His concern about the “distractions” of clinical practice potentially interfering with involvement in education seems very valid today.
Despite these concerns, there does not seem to be evidence that the clinical education program is, in general, in disarray. In Flexner's time, there was no process for evaluating the quality of clinical teaching and little opportunity to independently assess educational program outcomes. Today, there are multiple measures used. In general, these measures indicate that medical students are satisfied with their clinical training and feel confident in their clinical skills.35 U.S. medical students also perform well on the United States Medical Licensing Examination, a national assessment of clinical knowledge and skills.36
Flexner's strategy to address concerns about the distraction of clinical practice included the need to provide explicit financial support for and to recognize the teaching mission. Expanding on this concept, we make the following recommendations based on the history of clinical teaching in the past century.
Clinical faculty members should not be academically or, as far as possible, financially disadvantaged on the basis of their participation in education. This requires that there be explicit funding available to support release time for teaching. There also should be recognition of teaching in other ways, such as in retention and promotion criteria and decisions.
Faculty who teach should have access to faculty development, to assist them in enhancing their teaching skills, and access to continuing education, to ensure that they remain current in their areas of expertise. Participation in faculty development should be recognized for purposes of advancement or retention.
In summary, the introduction of various employment tracks has permitted clinical faculty to follow specific areas of career emphasis. Physicians who pursue teaching and clinical care have an explicit and important place in the faculty. This has proved to be a positive step, in that it has allowed the mission-specific growth of clinical departments. However, resources of various kinds are needed to support these physicians as they, in turn, contribute to clinical education. As in Flexner's day, it is important to remember that “medicine is expensive to teach.”1
1 Flexner A. Medical Education in the United States and Canada: A Report to the Carnegie Foundation for the Advancement of Teaching. Bulletin No. 4. Boston, Mass: Updyke; 1910.
2 Fye WB. The origin of the full-time faculty system: Implications for clinical research. JAMA. 1991;265:1555–1562.
3 AMA Council on Medical Education. Essentials of an acceptable medical college. JAMA. 1910;54:1974–1975.
4 Bryan CS, Stinson MS. The choice: Lewellys F. Barker and the full-time plan. Ann Intern Med. 2002;137:521–525.
5 Vaughan VC. Reorganization of clinical teaching. JAMA. 1915;64:785–790.
6 Bonner TN. Iconoclast: Abraham Flexner and a Life in Learning. Baltimore, Md: The Johns Hopkins University Press; 2002.
7 Flexner A. Medical education, 1909-1924. In: Proceedings of the Annual Congress on Medical Education, Medical Licensure, Public Health and Hospitals. Chicago, Ill: American Medical Association; 1924:7–12.
8 AMA Council on Medical Education. Essentials of an acceptable medical college. JAMA. 1919;73:517–520.
9 AMA Council on Medical Education. Essentials of an acceptable medical college. October 1927. [Accessed from the archives of the AMA Council on Medical Education.]
10 AMA Council on Medical Education. Description of medical colleges, 1932-1933. JAMA. 1933;101:691–697.
11 Weiskotten HG, Schwitalla AM, Cutter WD, Anderson HH. Medical Education in the United States, 1934-1939. Chicago, Ill: American Medical Association; 1940.
12 Peterson MR. Academic tenure and higher education in the United States: Implications for the dental education workforce in the twenty-first century. J Dent Educ. 2007;71:354–364.
13 AMA Council on Medical Education and Hospital. Essentials of an acceptable medical school. June 1944. [Accessed from the archives of the AMA Council on Medical Education.]
14 Deitrick JE, Berson RC. Medical Schools in the United States at Mid-Century. Evanston, Ill: Association of American Medical Colleges; 1960.
15 Medical school faculty staffing patterns. J Med Educ. 1968;43:943–944.
16 Harvey JC. An institutional approach in the funding of full-time clinical faculty salaries. J Med Educ. 1974;49:219–228.
17 Siegel B. Medical service plans in academic medical centers. J Med Educ. 1978;53: 791–798.
18 Rowley BD, Baldwin DC, McGuire MB. Selected characteristics of graduate medical education in the United States. JAMA. 1991;266:933–953.
19 Ballinger WF. The “full-time system” revisited. Surgery. 1970;68:919–923.
20 Cohn R. Full-time or geographic. Arch Surg. 1974;108:763–764.
21 Hardy CT Jr. Group practice by medical school faculty. J Med Educ. 1968;43:907–911.
22 MacLeod GK, Schwarz MR. Faculty practice plans: Profile and critique. JAMA. 1986;256:58–62.
23 Bentley JD, Chusid J, D'Antuono GR, Kelly JV, Tower DB. Faculty practice plans: The organization and characteristics of academic medical practice. Acad Med. 1991;66:433–439.
24 Parric M, Stemmler EJ. Development of clinician-educator faculty track at the University of Pennsylvania. J Med Educ. 1984;59:465–470.
25 Jones RF, Gold JS. Faculty appointment and tenure policies in medical schools: A 1997 status report. Acad Med. 1998;73: 212–219.
26 Jones RF, Gold JS. The present and future of appointment, tenure, and compensation policies for medical school clinical faculty. Acad Med. 2001;76:993–1004.
27 Bunton SA, Mallon WT. The continued evolution of faculty appointment and tenure policies at U.S. medical schools. Acad Med. 2007;82:281–289.
28 Thomas PA, Diener-West M, Canto MI, et al. Results of an academic promotion and career path survey of faculty at the Johns Hopkins University School of Medicine. Acad Med. 2004;79:258–264.
29 Mallon WT, Jones RF. How do medical schools use measurement systems to track faculty activity and productivity in teaching? Acad Med. 2002;77:115–123.
30 Dewey CM, Friedland JA, Richards BF, Lamki N, Kirkland RT. The emergence of academies of educational excellence: A survey of U.S. medical schools. Acad Med. 2005;80:358–365.
31 Irby DM, Cooke M, Lowenstein D, Richards B. The academy movement: A structural approach to reinvigorating the educational mission. Acad Med. 2004;79:729–736.
32 Searle NS, Thompson BM, Friedland J, et al. The prevalence and practice of academies of medical educators: A survey of U.S. medical schools. Acad Med. 2010;85:48–56.
33 Richards RK. Continuing Medical Education. New Haven, Conn: Yale University Press; 1978:24–31.
34 Donini-Lenhoff F, ed. State Medical Licensure Requirements and Statistics, 2009. Chicago, Ill: American Medical Association; 2009:53–56.
36 National Board of Medical Examiners. 2008 Annual Report. Philadelphia, Pa: National Board of Medical Examiners; 2009. References Cited in Tables Only
37 American Medical Association. Medical education in the United States, Section II. Undergraduate medical education. JAMA. 1968;206:1993–2022.
38 American Medical Association. Medical education in the United States, 1977-1978. JAMA. 1978;240:2819–2836.
39 Jonas HS, Etzel SI. Undergraduate medical education. JAMA. 1988;260:1063–1071.
40 Barzansky B, Etzel SI. Medical schools in the United States. JAMA. 1998;280:827–834.
41 Barzansky B, Etzel SI. Medical schools in the United States, 2007-2008. JAMA. 2008;300:1221–1227.
42 Association of American Medical Colleges. AAMC Data Book, Table D2. Revenues of U.S. Medical Schools. Washington, DC: AAMC; 2005.
43 Association of American Medical Colleges. AAMC Data Book, Table D2. Revenues of U.S. Medical Schools. Washington, DC: AAMC; 1994.