The contribution of MD–PhD training to academic orthopaedic faculties

ELSEVIER

Journal of Orthopaedic Research 19 (2001) 505-510

Journal of Orthopaedic Research www.elsevier.nl/locate/orthres

The contribution of MD-PhD training to academic orthopaedic faculties John M. Clark *, Douglas P. Hanel University of Washington, School of Medicine, BB 1043, 1959 N E Pacific Street, Seattle, WA 98195-3565, USA

Received 9 March 2000; accepted 3 1 July 2000

Abstract

Little is known about the distribution of research-trained physicians across the various specialties. To document the extent to which MD-PhD programs are a source of research-trained faculty for orthopaedic departments, this study examined the specialty choices of graduates of the Medical Scientist Training Program (MSTP) from 1964 to 1994. The MSTP, a combined MD-PhD program supported by the National Institute of General Medical Sciences, (NIGMS), produces roughly 25% of all MD-PhDs in the

us.

Methods. Copies of the appendices from training grant applications containing information on MSTP graduates were obtained from the NIGMS. Also, a questionnaire was mailed to 116 university-affiliated orthopaedic surgery departments asking how many faculty were MD’s, PhDs or MD-PhDs. Results. Records were obtained for all MST programs. Information on postdoctoral training and/or a current position was reported for 1615 graduates who earned both MD and PhD. Of these graduates, 277 chose non-clinical paths. The other 1338 entered a residency or internship. Of these, 593 were still in residency training, 566 were academic faculty members and 130 were in private practice. In the records, 12 (0.9%) were listed as orthopaedic surgical residents (6) or faculty (6). At this time, all 12 have completed training, and 11 are in academic practice. Eighty-three departments replied to the questionnaire. In that sample of 1761 faculty positions, 1478 were MDs, 217 were PhDs and 36 (2.0%) were MD-PhDs. Conclusion. Despite robust support of MD-PhD programs, the number of dual degree recipients on orthopaedic faculties is small when compared to the relative size of the specialty. Other sources of research-trained staff should perhaps be developed. 0 2001 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved.

Introduction

In 1992, the American Academy of Orthopaedic Surgeons’ Council on Research published a report entitled “Strengthening Orthopaedic Research” [3]. The report concluded that: “The number of orthopaedic surgeons trained in techniques of basic musculoskeletal research and committed to career involvement in research is inadequate and probably is continuing to decline”. Generally, the number of MDs engaged in research has diminished for years, [16,19], yet it is uncertain whether this situation affects some subspecialties disproportionately. Such a shortage would place affected departments at a disadvantage when competing for research funding, laboratory space and access to graduate students, and could impede the advancement of entire * Corresponding author. Tel.: +1-206-543-3690; fax: + 1-206-6853139. E-mail address: [email protected] (J.M. Clark).

fields of medicine. If individuals with research training shun careers in academic orthopaedic surgery, the specialty should be aware of this, and try to understand why it occurs. The most widely accepted certification of research training is the PhD degree, and one way to judge the research capacity of a specialty is by the relative number of MD-PhD faculty. Through a variety of programs, over 100 medical schools provide financial support for medical students who wish to earn a PhD in the course of undergraduate medical training. The largest of these is the Medical Scientist Training Program (MSTP), a combined MD-PhD program funded by the National Institute of General Medical Sciences (NIGMS). The NIGMS states that the program was established to produce graduates “motivated to undertake a career in biomedical research and academic medicine”. The MSTP began in 1964, when it funded 17 students, and has grown steadily since. The current annual budget is $21 million, and roughly 800 trainees are now

0736-0266/01/$- see front matter 0 2001 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved. PII: 5 0 7 3 6 - 0 2 6 6 ( 0 0 ) 0 0 0 5 1 - 6

506

J.M. Clark, D.P.Hanel I Journal qf Orthopaedic Research 19 (2001) 505-510

supported. Trainees currently receive 60% of tuition costs plus a stipend and are supported through both phases of doctoral training for a maximum of six undergraduate years. The Program apparently has been successful in its objective to produce medical scientists. According to published reports, 80% of graduates ultimately hold academic appointments [2,6,12,13]. Therefore, the MSTP and similar programs should serve as a significant source of basic science-trained medical school faculty. The study presented here examines the contribution of the MSTP specifically to orthopaedic surgery faculties, both absolutely and in comparison to other specialities. The objective is to determine whether medical students with formal research training choose to enter the field.

Materials and methods Twenty-seven medical schools have administered training programs long enough to have significant numbers of graduates. The postdoctoral careers of these graduates are voluntarily monitored by the schools through periodic questionnaires. This information is included as appendices to applications for renewal of existing training grants. Copies of these appendices were obtained through the Freedom of Information Officer at the NIGMS after confidential information, but not names, had been expunged. Although the data were presented in a variety of formats, the appendices all contained information on graduates who pursued postdoctoral clinical specialty training. Because the appendices included names, it was possible to confirm that graduates reported to have entered orthopaedic surgical training were indeed orthopaedic surgeons. A short confidential questionnaire was sent to those who could be located. The graduates were asked for descriptive data and opinions on the value of dual degrees. For comparison data, recipients of Orthopaedic Research and Education Foundation (OREF) Career Development Awards between 1980 and 1994 were asked for information regarding faculty appointments and publications. The NIGMS also provided general information on the MSTP, including the number of programs and trainees funded yearly since 1964. To confirm that the appendices were accurate, local directors of all active MST Programs were contacted by mail and asked how many trainees had completed their respective programs by the fall of 1997. To estimate how many surgeons with both MD and PhD degrees are in academic orthopaedic practice, surveys were sent to all civilian orthopaedic surgery departments with a medical school affiliation. To learn the degree mix in a scientific organization devoted to musculoskeletal research, the Orthopaedic Research Society (ORS) was contacted. Demographic data on the subspecialty training of practicing physicians were obtained through the AMA publications on this topic [15].

Results

The MSTP funds programs at 33 schools. Appendices were obtained from all 27 which had been in place long enough to matriculate graduates by 1993. The oldest of these programs enrolled trainees in 1964. The reports were submitted between 1991 and 1995, and contained data on students who graduated as recently as 1994. These reports provided information on 1833 individuals, of which 1658 had clearly completed both the MD and PhD degrees (Table 1). Specific career information was

given for 1615 (97%) of these graduates, and 277 of these held biomedical research positions with no evidence of medical specialty training. The remaining 1338 had entered or finished residency training in a field of clinical medicine. These graduates provide the basis for this report (Fig. 1). Usually, the specific type of residency or fellowship training and/or a departmental location was reported for each graduate. In some cases, the only medical specialty identification available was the category of internship chosen (i.e. medical or surgical). Five hundred and ninety-three of the graduates who entered clinical medicine were still in residency or fellowship when the report were submitted (Fig. 2). Insufficient information was given for 43 graduates. A few were in careers with no clear relation to medicine or biomedical research, e.g. “Freelance Writer”, but only one was unemployed. Five hundred and sixty-six graduates held the rank of Assistant, Associate or full Professor, or were listed as departmental chair or program director at a medical school. This group was considered to be academic medical faculty. Another 49 graduates held the rank of “clinical instructor” or “clinical assistant,” and could have been junior faculty or clinical fellows. Of the group with postdoctoral clinical training, 130 graduates were obviously in private practice. Those who held a “clinical” appointment at a medical school, e.g. “clinical assistant professor”, here were not counted as in private practice, unless clearly based in a private office. Thirty-one of the graduates in research positions were employed by (or headed) private commercial companies.

Table 1 Composition of the group of MSTP candidateslgraduates reported in the appendices” Total reported in appendices Undergraduates Undergraduates who withdrew Graduates with postdoctoral clinical training Still in residency or fellowship Medical school reg. faculty Clinical instructorlassistant In private clinical practice Entered orthopaedic training In orthopaedic academic practiceb

1876 155 23 1338 593 566 49 130 12 6

277 Graduates without postdoctoral clinical training and in research positions In postdoctoral position 50 Definite academic position 76 120 Other research position Private industry research 31 Insufficient information 49 Deceased 9 a This review analyzes the group which graduated and pursued further clinical training. ’At time of survey.

J . M . Clark, D.P. Hand I Journal of Orthopaedic Research 19 (2001) 505-510 PREDOCTORAL WORK

POSTDOCTORAL TRAINING

CAREER t

* PostdoctoralSludents (50)

Pvt. Industry (31)

MD-PHD GRADUATES ANALYZED HERE

Fig. 1. Diagram showing the numbers of MSTP graduates in each stage of training or career. This report describes a group that has graduated with both degrees, and the boxes on the right represent the components of this group. The upper box indicates the group of 277 graduates who took no postdoctoral clinical training and are in some non-clinical position, usually involving biomedical research. Fifty of these were still in postdoctoral fellowships, 31 held jobs in industry and 76 were university faculty members in basic science departments. One hundred and twenty others in that group could not be classified, because their title was listed as “fellow”. The lower boxes represent those graduates who pursued postdoctoral clinical training. Of these, 593 were still in residency, 566 held tenure-track faculty positions and 130 were in private practice at the time of the survey. Forty-nine held the position of “assistant” or “instructor” in a clinical department and are represented by the box labeled “I”. It is unclear how many of these were junior faculty and how many were in advanced clinical training, when fellows are often listed as instructors and assistants.

The distribution of graduates by subspecialty is shown in Table 2 and Figs. 1 and 3. Of the 1338 graduates who undertook postdoctoral clinical training, six were listed in the appendices as orthopaedic surgeons and six as orthopaedic surgery residents. Because five years have passed since the last graduated, all have completed residency by this time. Ten now hold regular faculty positions in orthopaedic departments. One is in

507

private orthopaedic practice and one declined to provide informa tion, The ORS reported that 771 of its members are MDs, 455 are PhDs, and 195 (14%) hold both degrees. Eightythree of 116 orthopaedic surgery departmentslsections (72%) responded to the questions on faculty composition. In that sample of 1751 orthopaedic faculty positions, 1498 (86%) were held by MDs, 217 (12%) by PhDs and 36 (2.0%) by MD-PhDs. Of these MD-PhDs, nine were MSTP graduates. All 12 MSTP graduates in orthopaedic practice could be located and 11 were contacted. Of these, 11 said that more MD-PhDs are needed in orthopaedic surgery, but only three view the PhD as essential to their careers. Two had received major grant support from the NIH and six had been funded by the OREF. One is in private practice. Another, who did not respond to the survey, is clearly in an academic position and has NIH funding. Thus, of the entire group of 12, 1I are in full-time academic practice and three have been recipients of NIH funding. Information was available for all 30 OREF Career Development Award recipients. This group, which includes three MSTP graduates, compared favorably to the MD-PhDs, having authored an average 43 peer-reviewed papers and 14 book chapters. Twenty-six (87%) are in full-time academic practice, and a11 are engaged in funded research programs. Eighteen (60%)) have been investigators on one or more NIH grants.

Discussion This study asked the question whether medical students with formal research training go on to academic

6 00 5 21 5 00

4 00

3 00

2 00

100

0 00

Fig. 2. This graph compares the percentage of MSTPs entering each subspecialty to the percent of all physicians in that subspecialty. For example: 32% of MSTP graduates enter Internal Medicine and its specialties, while about 33% of all physicians are in that field, and the ratio is 0.97.

508

J.M. Clark, D.P.Hand I Journal o j Orthopaedic Research 19 (2001) 505-510

Table 2 Distribution of MSTP graduates by specialty Specialty 1

# Of MSTP

YOOf MSTP

graduates in specialty

graduates in specialty

MSTPs on medical faculty in specialty

Estimated # of MDs in specialty

% Of practicing physicians

# Of match program positions each year

All Internal Medicine specialties Pathology Pediatrics

462

31.7

176

194,617

32.7

7944

200 196

15 14.4

93 90

17,149 44,960

2.9 7.6

522 2710

All surgery specialties

164

12.3

50

137,862

General surgery Neurosurgery OBIGyn Orthopaedics Plastic surgery ENT, headlneck Urology Neurology Psychiatry Anesthesia Ophthalmalogy Radiology Dermatology ER medicine Family medicine Radiation oncology Rehab medicine

23.2

5914

46 24 12 12 0 10 7

3.5 1.8 0.98 0.8 0 0.8 0.5

15 7 6 6 0 2 0

36,661 4424 34,908 20,374 4797 8132 9212

6.2 0.7 5.9 3.4 0.8 1.4 1.5

2213 34 1180 547 49 51 850

103 69 61 53 39 27 6 5 5 1

7.7 5.2 4.6

49 29 28 13 6 9 0 1 5 1

9822 35,069 29,042 16,283 7341 7846 15,856 69,491 3164 4582

1.7 5.9 4.9 2.7 1.2 1.3 2.7 11.7 0.5 0.8

43 1230 1176 12 1226 20 1047 3234 136 325

4

2.9 2 0.5 0.4 0.4 0.1

Fig. 3. Bar graph showing the percentage of MSTP graduates in each specialty of clinical medicine and the percentage of graduates who hold faculty positions in that specialty. Line indicates the percentage of all practicing physicians in each specialty [16].

careers in orthopaedic surgery. The information collected here reveals that in its first 30 years the MSTP produced 1 1 orthopaedic faculty members. Orthopaedic surgeons comprise 3.1% of the physicians in the US, [15] but only 0.7% of MSTP graduates are orthopaedic facul ty (Fig. 3). This trend has continued during a period when orthopaedic residency became an increasingly popular career choice for medical students [17].

The MSTP funds about one-quarter of all MD-PhD training [9], and 25% of MD-PhDs in orthopaedic surgery departments were MSTP graduates in the survey done here. Thus, orthopaedics draws MD-PhDs from all sources, and the reluctance to enter the specialty is a general characteristic among dual degree students, not specific to the MSTP. As a consequence, the number of these formally trained physician-scientists in

J.M. Clark, D.P. Hand I Journal of Orthopuedic Research I 9 (2001) 505-510

orthopaedic surgery is low. At the University of Washington approximately 9%, and at Johns Hopkins 10% of the Medical School faculty hold dual degrees [13], but in our survey only 2.0% of orthopaedic faculty positions are filled by MD-PhDs. This maldistribution is more significant in light of the fact that orthopaedic departments typically are small, both absolutely and in terms of faculty-to-resident ratio [1,5]. The annual budget for the MSTP is now $21 million, and perhaps $100 million is spent yearly on MD-PhD training when all costs are included. This study suggests that the investment does not “trickle down” to most surgical specialties, and raises the question whether MD-PhD programs could ever serve as a significant source of academic orthopaedic surgeons. It is unlikely that MSTP graduates entered orthopaedic surgery during the time covered by this survey and were not identified by this review. The information about the 566 graduates who had finished clinical training was quite complete, with rank and department listed for 97% of faculty. All nine of the MSTP graduates reported in the survey of orthopaedic surgery faculties were among the 12 orthopaedic residents/surgeons identified in the appendices. Because five years have passed, the six in residency when the appendices were produced have now finished clinical training. For these reasons, this study should provide a reliable picture of mature career choices made by orthopaedic surgeons who graduated from a MST Program between 1970 and 1994. In the time span covered by the reports, there is no apparent change in patterns of subspecialty choice. In 1986, when Duke University reported the career choices of its MSTP graduates, 21 of 75 entered pathology, and only six became surgeons [2]. Recent surverys of MD-PhD candidates and other medical students with interest in academic careers found 15% planning to enter Pathology and 2% choosing surgery [8,10]. Our study augments these smaller samplings by describing how these trends affect specific surgical subspecialities. Within this survey, the specialty choices of graduates still in residency training were similar to those who had finished (Table 2, Fig. 3). For these reasons, one can predict that the relative shortage of MD-PhDs in surgery will persist into the foreseeable future. The capacity to perform research is a valuable, if not essential function for an academic department. Without

509

a documented research capacity, access to outside funding - especially grants that pay indirect costs - is limited. To be promoted, academic physicians are expected to participate in a research program, and need the skills to do so. Orthopaedic residency accreditation requires that residents be exposed to a research experience. Judging from the composition of the ORS and the authorship of papers given at its annual meetings, PhDs serve a prominent role in musculoskeletal research, but most of the 145 MD-PhDs in the ORS are not members of orthopaedic faculties. At 12%, the number of PhDs on orthopaedic faculties is comparatively low (Table 3). One could argue that, in a time when the supply of PhDs is generous, hiring more non-clinical scientists is good policy. Conversely, the driving force behind MD-PhD programs is the belief that physicianscientists make a unique contribution and are in short supply [4]. In the last few years, numerous articles have documented the continuing decline of the physicianscientist and have proposed measures to reverse the trend [ 14,16,18]. The relevance of this situation to orthopaedic surgery was presented last year in an editorial in the Journal of Orthopaedic Research [7]. One must ask why research-trained students have an aversion to orthopaedic residency, and time in training is an obvious factor. Because the length of time required to obtain the PhD has steadily increased, many MSTP trainees are well over 30 years of age when they finish medical school. Six years of postdoctoral training - an internship, four years of residency and a fellowship - are usually necessary to qualify for advertised orthopaedic faculty openings. A serious scientist cannot afford to spend years closed off from any modern discipline, and residency provides little time for maintenance of research interests. The financial support provided by the MSTP may protect students from large debts, but this must be balanced against the delay incurred before they can benefit from a specialist’s income. It is also likely that orthopaedic surgery does not attract MD-PhD students because orthopaedic research is inadequately funded. According to the AAOS report, the number of NIH grants awarded to orthopaedic surgeons is small. Only three of the 11 MSTP-trained orthopaedic faculty surveyed had ever obtained NIH support. Because the pool of practicing orthopaedic physician-scientists is very small, the chance that a medical

Table 3 Comparison of degree mix in membership of various groups Sample of orthopaedic departments All medical schools (clinical departments) University of Washington medical school ORS

MDs

PhDs

MD-PhDs

985 (85%) 71Yo 632 (58Yo) 771 (54%)

144 (12%) 14% 353 (33%) 455 (32%)

29 (2.5%) 4.5% 91 (9%) 195 (14%)

510

J.M. Clark, D.P.Hand I Journal of Orthopaedic Research 19 (2001) 505-510

student will happen upon a laboratory or encounter a physician-scientist role model in an orthopaedic surgery department is low. Postdoctoral students gravitate towards NIH-funded laboratories with the knowledge that this improves their own prospects for funding. In a large survey of clinical researchers, Lee et al. found that having PhDs (as opposed to MDs) in one’s training environment was correlated to future success in federal funding awards [l 11. Conversely, for an MD, time spent with patient care, teaching and administration correlate negatively with likelihood of federal funding. On average, the orthopaedic MSTP graduates devoted 72% of their time to patient care, and those without NIH support devoted significantly less time to academic pursuits. Without solid grant support, a surgeon must always balance time in the laboratory against clinical activity. Any effort to increase the numbers of effective orthopaedic clinician scientists must break this cycle, primarily by improving the proportion able to qualify for federal funding. Given the low numbers of graduates entering orthopaedic residency, and the time required to earn a PhD, it is likely that dual degree programs will never serve as an important source of such faculty. Even were the observed trend to be reversed now, the salutary effect would be years away. One alternative to the MD-PhD is intensive research experience for orthopaedic residents, fellows and junior faculty. For example, between 1984 and 1994, the OREF funded 30 Career Development Awards, using private contributions. At $100,000-150,000, these awards are roughly equivalent to the federal support given to an MSTP trainee, but are awarded to orthopaedic surgeons in junior faculty positions. Although a direct comparison is impossible, the recipients of the career development awards have been more successful in achieving further research funding, especially from the NIH. Among the 10 MSTP graduates on orthopaedic faculties, the three who also had Career Development Awards reported the highest percentage of time devoted to research. In far less time, the OREF has trained many more orthopaedic faculty than the MSTP. Federal support of such postdoctoral orthopaedic training appparently is minimal. In 1991, there were 265 funded graduate students in “Anesthesiology”, 169 in “Radiology”, 46 in “Psychiatry”, 1138 in “Pathology” but none in orthopaedics and only 83 in “Surgery”, generally [9]. Although the primary interest of this study was to review the impact of the MSTP program on orthopaedic surgery, this paper serves as the first published review of the entire MSTP. The sample is comprehensive and provides a view of what the program graduates had accomplished by 1994. Given the difficulties in tracing people with questionnaires, the grant appendices upon which this report is based were remarkably complete.

The results show that very few medical students with the PhD degree elect to be orthopaedic surgeons. If our specialty is indeed serious about development of future clinician scientists, it cannot depend on mechanisms that better serve the non-surgical disciplines.

Acknowledgements

The authors thank the office of Senator Patty Murray of Washington State for assistance in obtaining records from the NIGMS. References [l] Barzansky B, Jones HS. Educational programs in U.S. medical schools 199C97. JAMA 1997;278:744-9. [2] Bardford WD, Pizzo S, Christakos AC. Careers and professional activities of graduates of a Medical Scientist Training Program. J Med Educ 1986;61:915-8. [3] Buckwalter JD, Goldberg VM, Peacock-Heath N. Building the future of orthopaedics: strengthening orthopaedic research. Park Ridge, IL: American Academy of Orthopaedic Surgeons; 1992. [4] Culotta E. Doctordoctor: growing demand for MD-PhDs. Science 1993;261:17847. [5] Dunn MR, Miller RS. U.S. graduate medical education, 199G97. JAMA 1997;278:7504. [6] Frieden C, Fox BJ. Career choices of graduates from Washington University’s Medical Scientist Training Program. Acad Med 1991; 3: 16 2 4 . [7] Hurwitz SR, Buckwalter JD. The orthopaedic surgeon scientist: an endangered species?. J Orthop Res 1999;17:155-6. [8] Kassebaum DG, Szenas PL, Ruffin AL, Masters DR. The research career interests of graduating medical students. Acad Med 1995;70:848-52. [9] Kennedy Jr TJ. Graduate education in the biomedical sciences: critical observations on training for research careers. Acad Med 1994;69:779-99. [lo] Lee JS. MD-PhD students in the 1990s: finding a niche for the dual degree. JAMA 1995;274:173&7. [Ill Lee TH, Ognibene FP, Schwarts JS. Correlates of external research support among respondents to the 1990 American Federation for Clinical Research survey. Clin Res 1991;39:1354. [I21 Martin JB. Training physician-scientists for the 1990s. Acad Med 199l;66: 123-9. [I31 McClellan DA, Talalay P. M.D.-Ph.D. training at the Johns Hopkins University School of Medicine 1962-1991. Acad Med 1992;67:3641. [14] Nathan DG. Clinical research-perceptions reality and proposed solutions. JAMA 1998;280:1427-31. [15] Roback G, Randolph L, Seidman B, Pasko T, editors. Physician characteristics and distribution in the U.S. Chicago: American Medical Association; 1994. [I 61 Rosenberg LE. Physician-scientists-endangered and essential. Science 1999;283:331-2. [I71 Simon MA. Symposium: orthopaedic workforce in the next millennium. Workforce issues and perspective of the Academic Orthopaedic Society. J Bone Joint Surg [Am] 1998;80A:1 5 4 1 4 [18] Thompson J, Moskowitz J. Preventing the extinction of the clinical research ecosystem. AMA 1997;278:2414. [19] Wyngaarden JB. The clinical investigator as an endangered species. N Engl J Med 1979;301:1254-9.