Andriole, Dorothy A. MD; Jeffe, Donna B. PhD
Medical schools throughout the United States have implemented a wide range of programs to support female faculty and promote their academic success1,2; however, although women now constitute 48% of U.S. Liaison Committee for Medical Education (LCME)-accredited medical school graduates3 and 46% of trainees in Accreditation Council for Graduate Medical Education–accredited training programs,4 they constitute only 33% of physicians with faculty appointments.5 Thus, as medical schools continue their efforts to recruit and support more women in academic medicine, we hypothesized that among contemporary (i.e., 1998–2004) U.S. medical graduates, women would be more likely than men to be appointed to full-time faculty positions. Because substantial differences exist between male and female medical graduates in their research career intentions6 and in their specialty choices,7,8 we also hypothesized that we would observe gender differences in the nature of associations both between faculty appointment and graduates’ research experiences and between faculty appointment and graduates’ specialty choices. We tested these hypotheses in a retrospective study (June 2011 to March 2012) using longitudinal data collected through July 2009 for the 1998–2004 national cohort of U.S. medical graduates.
Following institutional review board approval from Washington University School of Medicine, we constructed a database containing the individually linked, deidentified records for all 1994–2000 LCME-accredited U.S. medical school matriculants and their follow-up data (through July 2009). The database included selected items from the following databases of the Association of American Medical Colleges (AAMC): the Student Record System (SRS), the Matriculating Student Questionnaire (MSQ),9 the Graduation Questionnaire (GQ),10 Graduate Medical Education (GME) Track, and the AAMC Faculty Roster. We also used data from the National Board of Medical Examiners (NBME) and the American Medical Association (AMA) Physician Masterfile. We included graduates starting with the class of 1998 (most of whom had matriculated four years earlier, in 1994) because several variables of interest were not included on the MSQ before 1994. To allow sufficient time (at least five years) for most graduates in our study sample to complete GME requirements, we included only graduates through 2004.
We used “degree program of enrollment at the time of graduation” as listed in the SRS to identify and exclude MD-PhD program graduates from analysis, because of the inherent differences in MD-PhD graduates’ career intentions at graduation compared with their MD degree graduate peers.11 We used AMA Physician Masterfile data obtained from Medical Marketing Services, Inc. (Wood Dale, Illinois), a licensed AMA Masterfile vendor, to identify and exclude graduates still in GME at follow-up (the definition of full-time faculty in the AAMC Faculty Roster excludes residents and fellows).12 We searched the literature to identify factors potentially associated with physician academic medicine faculty appointment.13,14
The demographic variables from the SRS include each student’s gender, race/ethnicity, and year of graduation. Matriculants self-reported their race/ethnicity by responding to a list of options on the American Medical College Application Service questionnaire. We categorized race/ethnicity as Asian/Pacific Islander, underrepresented minority (URM) in medicine (including black, Hispanic, and American Indian/Alaska Native), other/unknown (including matriculants who self-identified as “other,” self-selected multiple races, or did not respond to this question), or white (reference group).
We included six research-related variables. We identified graduates who reported participating in a college laboratory research apprenticeship (yes versus no) based on their response to the MSQ item “Indicate any programs you participated in to prepare for a career in medicine or science.” We included responses to the MSQ item for extent (at matriculation into medical school) of planned career involvement in research (from 1 = not involved, to 5 = exclusively involved). The AAMC provided an indicator for matriculation into a medical school ranked among the top 40 in National Institutes of Health (NIH) funding,15 which we defined as a research-intensive medical school (yes versus no). We included responses to two research-related items on the GQ: medical school research elective with a faculty member (yes versus no) and authorship/coauthorship of a research paper submitted for publication during medical school (“research paper authorship,” yes versus no). We used AAMC GME Track data to distinguish between graduates who had completed one or more years of research during GME as indicated by their program director on the GME Census versus graduates for whom there was no program director indication of one or more years of research. We calculated the total number of research experiences (i.e., participation in college laboratory research apprenticeship, medical school research elective with a faculty member, and one or more years of research during GME [range 0–3]) in which each graduate had reportedly participated.
We included five professional development variables in our analysis. We obtained, with permission, first-attempt United States Medical Licensing Examination (USMLE) Step 1 results (pass versus fail) from the NBME; we used this variable as a measure of academic achievement during medical school. We included the following data from the GQ: medical school health education elective (yes versus no), specialty choice, planned career-setting preference, and total debt at graduation. On the basis of graduates’ responses to GQ specialty choice items, we assigned graduates to 1 of 12 specialty categories for analysis: (1) no specialty choice (graduates did not choose a specialty), (2) radiology, (3) internal medicine subspecialties, (4) family medicine, (5) pediatrics, (6) pediatrics subspecialties (7), all other nongeneralist, nonsurgical specialties (“other”), (8) surgery/surgical specialties (“surgery”), (9) dermatology, (10) obstetrics–gynecology, (11) psychiatry, and (12) internal medicine (reference group). We used a four-category variable for career intention: “full-time, non-university, clinical practice,” “undecided,” “other,” and “full-time faculty in basic science or clinical research/teaching” (reference group). We used a five-category variable for total debt at graduation: no debt (reference group), $1–$49,999, $50,000–$99,999, $100,000–$149,999, and $150,000 or more.
We derived the data for our outcome measure of interest, full-time faculty appointment at a U.S. LCME-accredited medical school, from AAMC Faculty Roster records for all 1998–2004 graduates who had held any full-time faculty appointments (active and inactive) through July 2009.16 We created a dichotomous variable for ever having a full-time faculty appointment (yes versus no).
We linked all the records for each graduate using a unique, AAMC-generated identification number. We used chi-square tests to describe associations between two categorical variables, and analysis of variance to describe associations between a continuous and categorical variable. We report adjusted odds ratios and 95% confidence intervals from three separate multivariate logistic regression models to identify independent predictors of full-time faculty appointment among (1) all graduates in our study sample, (2) only women in our study sample, and (3) only men in our study sample. We performed all tests using SPSS 17.0.3 (SPSS, Inc., Chicago, Illinois); we considered two-sided P < .05 to be significant.
Of the 113,522 matriculants in our database who entered medical school from 1994 to 2000, 103,597 had graduated between 1998 and 2004. From these, we excluded 1,232 MD-PhD graduates, 1 graduate without data for degree program at graduation, 17,122 graduates classified as residents, and 207 graduates who were missing these classification data according to AMA Physician Masterfile records, leaving 85,035 graduates eligible for inclusion in our study. Of these 85,035 eligible graduates, we excluded 17,611 graduates who had not responded to all MSQ and GQ items of interest and 535 graduates who had responded to all MSQ and GQ items of interest but were missing other data. Our final study sample of 66,889 graduates with complete data for all variables of interest thus included 78.7% of 85,035 eligible graduates.
Among all 85,035 eligible graduates, 14,964 (17.6%) had held faculty appointments, including 2,926 (16.1%) of the 18,146 whom we excluded, and 12,038 (18.0%) of the 66,889 included in the final sample (P < .001). Among the 85,031 eligible graduates with data for gender (missing for n = 4), 38,729 (45.6%) were women, including 7,815 (43.1%) of the 18,142 we excluded, and 30,914 (46.2%) of the 66,889 included in the final sample (P < .001).
Table 1 presents descriptive statistics for the entire study sample (N = 66,889) grouped by faculty appointment and grouped by gender. The gender gap in research paper authorship among all graduates in our study sample (as shown in Table 1) was also evident among only the 35,146 graduates who had participated in a medical school research elective; that is, 60.3% of men who had participated in a medical school research elective (12,299/20,380) and 51.9% of women who had participated in a medical school research elective (7,669/14,766) reported research paper authorship (P < .001).
Table 2 shows the characteristics of men and women medical school graduates grouped by full-time faculty appointment. As shown, findings for each variable examined were similarly associated with full-time faculty appointment among men and among women with the exception of debt, which was significantly associated with full-time faculty appointment among men but not among women.
Table 3 shows the results of the three regression models. Among all graduates, women were more likely than men to have held faculty appointments. In all three models, each of the following variables was associated with a greater likelihood of faculty appointment: matriculation into a research-intensive medical school, participation in one or more years of research during GME, USMLE Step l first-attempt passing score, and medical school health education elective. In all three models, each of the following variables was associated with a lower likelihood of a faculty appointment: a more recent graduation year; Asian/Pacific Islander race/ethnicity; URM race/ethnicity; the specialty choices of family medicine, surgery, and dermatology; and GQ career intention of “other,” “undecided,” or “full-time (non-university) clinical practice.”
Among only male graduates, higher total debt at graduation from medical school was associated with a lower likelihood of a faculty appointment, whereas participating in a laboratory research apprenticeship during college, greater extent of planned career involvement in research at matriculation into medical school, research paper authorship during medical school, and pediatrics and psychiatry specialty choices were each associated with a greater likelihood of faculty appointment.
Among only female graduates, higher debt at graduation from medical school, participating in a laboratory research apprenticeship during college, greater extent of planned career involvement in research at matriculation into medical school, and research paper authorship during medical school were not independently associated with likelihood of faculty appointment. The specialty choices of pediatrics, obstetrics–gynecology, and psychiatry were each associated with a lower likelihood of faculty appointment for women.
Table 4 shows the characteristics of all faculty appointees in the study sample grouped by gender. Male and female faculty appointees differed significantly in their research experiences, specialty choices, and total debt at graduation from medical school, among other characteristics.
Full-time faculty positions were held by 18.0% of U.S. medical graduates in our cohort of 1998–2004 graduates followed to 2009, which is higher than the 9.3% of 1979–1993 graduates followed to 1997 who had held full-time faculty positions.17 Because annual numbers of U.S. medical graduates remained relatively unchanged from 15,113 in 1979 to 15,736 in 2004,12 this greater level of participation in academic medicine among a more recent cohort of U.S. medical graduates is likely due, at least in part, to growth in the number of clinical faculty in the academic medicine workforce.18
In several earlier studies of U.S. medical graduates, women were more likely than men to have held faculty positions in general19,20; women were also more likely than men to have held full-time faculty positions.17 In a study of the 1965–2000 cohort of graduates followed to 2004, a higher percentage of female graduates than male graduates from every graduation year held faculty appointments in 2004.21 The authors suggested that the higher percentage of female graduates in academic medicine may be attributable to “actual recruitment of high-achieving women” or to female graduates more actively seeking academic positions compared with male graduates.21 We similarly observed a positive relationship between female gender and faculty appointment even after controlling for research-related and professional development variables, including career intention at graduation, which these earlier studies did not examine.
Our observations that URM as well as Asian/Pacific Islander graduates were less likely than white graduates to be appointed to faculty positions (Table 3) suggest that continued efforts to promote a more racially and ethnically diverse academic medicine workforce are warranted.22–24 Because we made these observations in models that controlled for career intention at graduation from medical school, we speculate that URM and Asian/Pacific Islander graduates’ interest in academic medicine careers may decrease during GME, and thus these graduates may be less likely to seek faculty appointments; however, nonwhite graduates may also have been disproportionately less likely than white graduates to receive faculty appointment offers aligned with their personal and professional goals.
We observed positive associations between faculty appointment and each of several research-related variables. As participation in a college laboratory research apprenticeship was significantly associated with faculty appointment among only male graduates, we speculate that there may be gender differences in reasons that college students who subsequently enter medical school participate in college laboratory research apprenticeships. Perhaps male college students planning medical careers participate in college research experiences because of an early interest in research-related careers, whereas female college students planning medical careers may perceive participation in such experiences as a means to increase their competitiveness in the medical school admissions process, especially given the fact that female medical school applicants tend to have lower Medical College Admission Test scores and lower science grade point averages than their male counterparts.25
We also observed that men entered medical school with higher levels of planned career involvement in research than did women (Table 1). This gender gap was previously observed among earlier cohorts of U.S. medical school matriculants.6 Our observation that a higher level of planned career involvement in research at matriculation into medical school was associated with a higher likelihood of faculty appointment among male graduates—but not among female graduates (Table 3)—suggests that male graduates may be particularly interested in academic medicine careers for the research opportunities that such career paths provide.
The gender difference in research interest at medical school matriculation appears to persist during medical school: a higher proportion of men than women participate in research electives and author research papers during medical school (Table 1). Medical school research electives likely vary widely in both scope and duration (some programs offer as much as a full year of intensive research training).26 We speculate that the finding (which we observed among men, but not women)—that a productive research experience during medical school (i.e., resulting in research paper authorship), rather than a medical school research elective experience per se, predicted an increased likelihood of faculty appointment—might reflect the wide range in scope and duration of elective research experiences during medical school. The authorship gender gap during medical school that we observed extends previous observations of authorship gender gaps among physicians27,28 and suggests that gender differences may exist in the quality and scope of the research experience itself or in the extent to which medical students’ mentors recognize their research contributions.
Participation in at least one year of research during GME was among the strongest predictors of faculty appointment for both men and women. This variable reflects a substantive commitment to research and is the most temporally proximate research-related variable to our outcome of interest: full-time faculty appointment. These findings align with a previous report of positive relationships between a period of research after medical school graduation and physicians’ academic medicine career interests.13 In our study, the relationship between at least one year of research during GME and faculty appointment was somewhat stronger among women than among men (Table 3), suggesting that a substantive period of research during GME might be particularly important to female graduates’ decisions about academic medicine careers. Other researchers have recently noted that GME and fellowship training seem to be critical points in female graduates’ consideration of academic medicine career paths.29
The gender differences we have observed in research-related variables among faculty appointees suggest that men might enter academic medicine better prepared than women to engage in productive research activities (compared with female faculty appointees, male faculty appointees had greater cumulative research experiences, and a higher proportion had authored research papers during medical school; Table 4). We speculate that such differences might contribute to gender disparities in academic medicine promotion, as men have been promoted at greater rates than have women17,30,31 and research productivity may be considered in promotion decisions.30,32
Professional development variables
Among all graduates, first-attempt passing score on USMLE Step 1 and participation during medical school in a health education elective were each associated with a greater likelihood of faculty appointment (Table 3). The latter finding, for demonstrated interest in teaching, is consistent with previous research.13,14 Among faculty appointees, a higher proportion of women than men had participated in a medical school health education elective (Table 4). Thus, female graduates may be particularly interested in academic medicine for the opportunities afforded to pursue teaching activities. Indeed, in a recently reported study of 53 female physicians in academic medicine, 62.3% of these female physicians rank-ordered teaching as the aspect of academic medicine that most interested them when they decided to pursue an academic medicine career, whereas only 7.5% rank-ordered research as the most attractive aspect.29
Choice of any non-full-time-faculty career intention at graduation was associated with a lower likelihood of faculty appointment, which is consistent with the behavioral theory that intention predicts future behavior.33 The recent addition of an academic medicine focus to the AAMC Careers in Medicine program34 may help medical students make better informed decisions about potential careers in academic medicine.
Family medicine specialty choice was associated with a markedly lower likelihood of faculty appointment compared with internal medicine (Table 3). This finding might reflect lower levels of interest in research-related academic careers among U.S. medical graduates choosing family medicine,35 but it might also reflect the low proportion of clinical faculty departmental appointments that are in family medicine compared with the proportion of physicians that are family medicine specialists overall. About 12% of our cohort of graduates (regardless of full-time faculty appointment) had chosen family medicine, and about 10% of board-certified physicians self-designate family medicine as their primary specialty,36 but only 3.9% of clinical department faculty appointments are in family medicine.37
Among female graduates, the specialty choices of psychiatry, pediatrics, and obstetrics–gynecology were each associated with a lower likelihood of faculty appointment, whereas, among male graduates, specialty choices of psychiatry and pediatrics were each associated with a greater likelihood of faculty appointment, and specialty choice of obstetrics–gynecology was not associated with either a greater or a lower likelihood of faculty appointment (Table 3). We speculate that gender differences in associations between specialty choice and faculty appointment might reflect, at least in part, efforts to recruit men to academic medicine in the specialties of psychiatry, pediatrics, and obstetrics–gynecology, which are all currently characterized by a predominance of female trainees. Although women accounted for 46.0% of all GME trainees in 2010, women accounted for 54.9% of psychiatry trainees, 73.2% of pediatrics trainees, and 79.7% of obstetrics–gynecology trainees.4
We also observed differences in specialty choice associated with gender among only full-time faculty appointees. The largest gender gap was evident in obstetrics–gynecology, the specialty choice of 10.1% of female faculty appointees, but only 2.8% of male faculty appointees (Table 4). Retention at their school of medicine among full-time clinical MD faculty at U.S. medical schools varies by specialty, and faculty in obstetrics–gynecology departments have the lowest 5-year retention rate among 14 clinical departments examined.38 Although clinical departmental affiliation has not been included in some previous studies that described the higher 5-year and 10-year attrition (leaving their school of medicine or leaving academic medicine) rates in academic medicine among women compared with men,39,40 our findings—and those of Corrice and colleagues38—suggest that specialty or clinical departmental affiliation might be considered in future studies of gender differences in academic medicine faculty career paths.
Finally, previous studies on the relationship between debt and academic medicine careers have yielded mixed results.14 Our finding that debt at graduation from medical school was not independently associated with faculty appointment among all graduates (Table 3) extends the observations of a previous study.41 However, men, but not women, with higher levels of debt in our cohort were less likely to have held a faculty appointment, indicating that debt as a disincentive to graduates’ pursuit of academic careers13,14 differs between men and women. Although higher debt did not appear to deter women in our sample from entering academic medicine, we hypothesize that lower financial rewards, which disproportionately affect female faculty even after controlling for specialty,42 might differently impact the retention of more heavily indebted faculty and be a greater detriment to retention of women than men.
Our study has several strengths and limitations. Two strengths of our study are our use of a national cohort of MD degree graduates from LCME-accredited U.S. medical schools and our use of data for numerous variables associated with faculty appointment. Our outcome measure is actual full-time faculty appointment rather than interest in or intention to pursue a full-time faculty career. However, our study’s observational design precludes making causal inferences. The higher proportion of women than men included in our sample is consistent with generally higher survey response rates among women, a possible source of selection bias. Our results cannot be generalized to MD-PhD graduates, to graduates of other types of medical schools (i.e., osteopathic or international), or to PhD degree holders, who constitute about 25% to 30% of full-time faculty at U.S. medical schools.37 We have also not accounted for other unmeasured factors that reportedly may be associated with academic medicine careers, such as influence of a mentor or role model.13,14,23 Finally, our observation that more recent graduation year was associated with a lower likelihood of faculty appointment is not unexpected; faculty appointments among medical graduates in our study sample are likely to continue to accrue with longer follow-up.
Despite these limitations, our findings inform the evidence base regarding factors associated with contemporary U.S. medical graduates’ entry into academic medicine. Our observations on the characteristics of men and women who were appointed to full-time faculty positions suggest that male and female physicians enter academic medicine with different research experiences, debt loads, and professional activity preferences, which may impact their subsequent academic medicine career trajectories.
Acknowledgments: The authors thank Paul Jolly, PhD, Gwen Garrison, PhD, and Franc Slapar, MA, at the Association of American Medical Colleges, Washington, DC, for their provision of data and their assistance with coding; Robert M. Galbraith, MD, MBA, at the National Board of Medical Examiners for assistance with data; and Mr. Jim Struthers in the Division of Health Behavior Research at the Washington University School of Medicine for data management services.
Funding/Support: By grant R01 GM085350 from the National Institute of General Medical Sciences (Dr. Jeffe and Dr. Andriole).
Other disclosures: None.
Ethical approval: The study was approved by the institutional research board at Washington University School of Medicine.
Disclaimer: The conclusions made by the authors are not necessarily those of the Association of American Medical Colleges, the National Board of Medical Examiners, the National Institutes of Health, the American Medical Association, or their respective staff members. The National Institute of General Medical Sciences was not involved in the design or conduct of the study; the collection, management, analysis, or interpretation of data; or the preparation, review, or approval of the manuscript.
4. Brotherton SE, Etzel SI. Graduate medical education, 2009–2010. JAMA. 2010;304:1255–1270
6. Guelich JM, Singer BH, Castro MC, Rosenberg LE. A gender gap in the next generation of physician–scientists: Medical student interest and participation in research. J Investig Med. 2002;50:412–418
7. Lambert EM, Holmboe ES. The relationship between specialty choice and gender of U.S. medical students, 1990–2003. Acad Med. 2005;80:797–802
8. Dorsey ER, Jarjoura D, Rutecki GW. The influence of controllable lifestyle and sex on the specialty choices of graduating U.S. medical students, 1996–2003. Acad Med. 2005;80:791–796
11. Andriole DA, Whelan AJ, Jeffe DB. Characteristics and career intentions of the emerging MD/PhD workforce. JAMA. 2008;300:1165–1173
12. Association of American Medical Colleges.AAMC Data Book: Medical Schools and Teaching Hospitals by the Numbers.. 2006 Washington, DC Association of American Medical Colleges
13. Straus SE, Straus C, Tzanetos KInternational Campaign to Revitalise Academic Medicine. . Career choice in academic medicine: Systematic review. J Gen Intern Med. 2006;21:1222–1229
14. Borges NJ, Navarro AM, Grover A, Hoban JD. How, when, and why do physicians choose careers in academic medicine? A literature review. Acad Med. 2010;85:680–686
15. Moy E, Griner PF, Challoner DR, Perry DR. Distribution of research awards from the National Institutes of Health among medical schools. N Engl J Med. 2000;342:250–255
17. Nonnemaker L. Women physicians in academic medicine: New insights from cohort studies. N Engl J Med. 2000;342:399–405
19. Jolly HP, Larson TA. Datagram: Women physicians on U.S. medical school faculties. J Med Educ. 1975;50:825–828
20. Jolly P. Women physicians on U.S. medical school faculties. J Med Educ. 1981;56:151–153
22. Law M Striving Towards Excellence: Faculty Diversity in Medical Education.. 2009 Washington, DC Association of the American Medical Colleges
23. Daley SP, Palermo AG, Nivet M, et al. Diversity in academic medicine no. 6 successful programs in minority faculty development: Ingredients of success. Mt Sinai J Med. 2008;75:533–551
24. Nivet MA, Taylor VS, Butts GC, et al. Diversity in academic medicine no. 1 case for minority faculty development today. Mt Sinai J Med. 2008;75:491–498
26. Fang D, Meyer RE. Effect of two Howard Hughes Medical Institute research training programs for medical students on the likelihood of pursuing research careers. Acad Med. 2003;78:1271–1280
27. Jagsi R, Guancial EA, Worobey CC, et al. The “gender gap” in authorship of academic medical literature—A 35-year perspective. N Engl J Med. 2006;355:281–287
28. Reed DA, Enders F, Lindor R, McClees M, Lindor KD. Gender differences in academic productivity and leadership appointments of physicians throughout academic careers. Acad Med. 2011;86:43–47
29. Borges NJ, Navarro AM, Grover AC. Women physicians: Choosing a career in academic medicine. Acad Med. 2012;87:105–114
30. Fang D, Moy E, Colburn L, Hurley J. Racial and ethnic disparities in faculty promotion in academic medicine. JAMA. 2000;284:1085–1092
32. Palepu A, Carr PL, Friedman RH, Amos H, Ash AS, Moskowitz MA. Minority faculty and academic rank in medicine. JAMA. 1998;280:767–771
35. Senf JH, Campos-Outcalt D, Kutob R. Family medicine specialty choice and interest in research. Fam Med. 2005;37:265–270
36. Smart DK Physician Characteristics and Distribution in the US, 2011.. 2011 Chicago, Ill American Medical Association
42. Ash AS, Carr PL, Goldstein R, Friedman RH. Compensation and advancement of women in academic medicine: Is there equity? Ann Intern Med. 2004;141:205–212