Academic medical school faculty are challenged by demands for productivity and excellence.1 At the same time, they experience barriers to career development and promotion, including increased competition for grant funding, rising demands for clinical work, and a lack of effective mentoring to foster recognition and acquisition of skills such as scholarly writing.2–8
Many junior faculty are simultaneously managing significant family responsibilities at this time in their careers.3,9–13 Women faculty have been particularly affected, with documented needs for enhanced mentoring, flexible work environment, protected time, and administrative support to write grants and papers.2,14–20
Several academic medical centers and professional societies have evaluated the awarding of seed grants to foster faculty development. Most studies to date have been discipline specific, have focused on trainees, have considered educational faculty development programs, have evaluated awards that cover a significant portion of a faculty member’s salary, and have lacked controls.21–34 Although the programs studied heretofore have been small, intramural grants appear to yield positive outcomes in subsequent grant funding, publications, and retention.25,34–40 We sought to explore whether junior faculty fellowships awarded over 16 years had a demonstrable effect on promotion and retention for a large cohort of junior faculty compared with controls.
In 1996, Harvard Medical School (HMS) initiated a unique program of junior faculty grants in recognition of the 50th anniversary of admitting women to HMS. The goal was to support instructors and assistant professors to undertake research, clinical, or educational projects early in their careers. Although the fellowships were open to all junior faculty, the originators assumed that women faculty in particular would benefit because selection incorporated recognition of personal obstacles, such as managing child care, elder care, and other personal issues, which have been shown to be more frequently the responsibility of women faculty.10,11,41,42
Applicants to the 50th Anniversary Fellowship Program for Scholars in Medicine (later renamed the Eleanor and Miles Shore Fellowship) submitted a written project description, documented commitment of a mentor, and a letter of support from their department chair. Ultimately the program was managed by a 0.25-FTE staff member. Funding was solicited annually by the dean for faculty affairs from private donors, departments, foundations, divisions, affiliated hospitals, and schools at the level of $25,000 initially, with increases over the 16 years to the range of $30,000 to $50,000. Funds could be used for academic activities such as hiring a research assistant, offsetting clinical obligations, or supporting salary. Funds could not be used for nonacademic purposes such as child care. Each year the dean honored the recipients with a celebration which included family, mentors, and senior departmental and hospital leaders.
Fellowships were highly competitive and selected in one of two ways based on the preference of the funder. Initially, all fellowships were selected by a schoolwide committee that apportioned scoring of the applications to give two-thirds for the academic merit of the proposal and one-third for the level of personal obstacles, such as family responsibilities or health issues. Overall, 247 of the 652 fellowships (38%) were selected by this committee. As the program expanded over time, some departments, divisions, and hospitals that provided fellowships established their own selection process but maintained similar application requirements (Figure 1). Faculty received an annual announcement from HMS about the program and viewed available fellowships on a single Web site. Those fellowships evaluated by the schoolwide committee had a common application process. Those fellowships selected by local committees sometimes used the central application form and sometimes developed their own procedures. Faculty frequently applied for fellowships managed by the schoolwide committee as well as fellowships selected through local committees.
We undertook this study to evaluate whether in this era of scarce resources, a one-time investment of relatively small amounts of funding to instructors and assistant professors was associated with positive academic outcomes such as more rapid promotion or longer retention and whether there were differences by sex, rank, or presence of personal obstacles to career development.
Study subjects were members of the Harvard University Faculty of Medicine at the rank of instructor or assistant professor at some time between 1996 and 2011. Those who received fellowships were identified from the HMS Shore Fellowship database and were linked to the Medical Area Reporting Database (MARS) by university ID number. Controls were individuals who did not receive the fellowship and were identified from the MARS database, which includes information regarding age, gender, rank, date of appointment and promotion, education, training, department, and hospital affiliation.
The majority of HMS faculty (98%) are not governed by a tenure clock but are proposed for promotion when the department head agrees that the candidate has met criteria for promotion. Instructor is the entry-level appointment for 90% of the faculty.
The two primary outcomes were time to promotion and length of retention (measured in years). Promotion to the next rank (instructor to assistant professor or assistant to associate professor) was based on date of promotion as noted in MARS. Retention in a given year was characterized as continuing to hold a faculty appointment on July 1 of that year.
To estimate the effect of the fellowship, we used propensity score matching to identify a suitable comparison group. Propensity score matching is used to minimize bias in comparisons of nonrandom samples.43 Here, propensity scores represented the conditional probability that a subject would receive a fellowship, and these were estimated using logistic regression models with the demographic and academic variables as predictors. Each fellowship recipient was matched without replacement to three nonrecipients with the closest estimated propensity score to create a comparison group that was similar with regard to educational degree, year of appointment, academic rank, hospital, department, training at Harvard, alumni/ae status, part-time faculty appointment, age, sex, and race/ethnicity as recorded in the Harvard database, thus controlling for confounding on the basis of these observed factors. Fellowship effects among all instructors, all assistant professors, instructors selected by the schoolwide committee, female instructors, and male instructors were estimated with separate propensity score modeling and matching procedures. Only the instructor pool provided a large enough sample to conduct separate analyses by sex.
Kaplan-Meier curves were used to compare fellowship recipients with the matched comparison group on time from appointment to promotion and time from appointment to leaving HMS. Times to promotion were censored if a subject left HMS or if the subject had not been promoted by July 1, 2012. Times to leaving were censored if a subject was still appointed on July 1, 2012. Group differences were assessed with log-rank tests. To ensure the most conservative estimate, only those evaluated by the selection committee that explicitly scored the level of personal impediment were included in the analysis of individuals with “personal obstacles.” All analyses were conducted using SAS statistical software, version 9.3 (SAS Institute, Cary, North Carolina) and S+ statistical software, version 8.2.0 (TIBCO Software Inc., Palo Alto, California). The study was approved by the HMS institutional review board.
Between academic years 1996–1997 and 2011–2012, the faculty grew from 6,993 to 11,268, and the total number of fellowships increased each year, from 10 in 1996 to 66 in 2011 (Figure 1). During this period, a total of 652 individuals received fellowships, including 480 (74%) instructors and 142 (22%) assistant professors; recipients accounted for less than 1% of eligible faculty at either rank. Thirty individuals (5%) were excluded from the analyses either because they did not hold an eligible faculty appointment when they received a fellowship (n = 27), they could not be matched with the MARS database (n = 2), or their rank had changed from assistant professor to instructor (n = 1). Table 1 describes the characteristics of the fellowship recipients, the population from which controls were drawn, and the matched controls for the three recipient groups: instructors, assistant professors, and those selected by the schoolwide committee. Data for hospital and department matching are not shown but did not differ across matched comparison groups.
Compared with the larger faculty from which the matched controls were drawn, fellowship recipients were likely to be younger and female. Instructors awarded fellowships were more likely than other instructors to be from groups underrepresented in medicine, to have trained at Harvard, and to hold full-time appointments. Instructors were awarded their fellowships on average 2.5 years into their faculty appointment, and assistant professors received theirs typically 2.3 years after their appointment as assistant professor. As shown in Table 1, the propensity score strategy successfully identified groups that were comparable to the awardees on the basis of these factors.
Instructors who received fellowships had a shorter time to promotion, with half being promoted by 7.3 years versus 12.2 years for their matched controls (P < .0001) (Figure 2). The subset chosen by the schoolwide committee (which apportioned one-third of their scores for personal challenges and potential impediments to academic success) were promoted earlier than their matched controls (P = .002). Female instructors were more likely than matched controls to be promoted earlier (P < .0001), while no difference was seen between male instructors and their matched counterparts (P = .17).
Assistant professors who received fellowships showed no differences in time to promotion (P = .15 for all assistant professor recipients versus controls). However, both instructors (P < .0001) and assistant professors (P = .0005) who received fellowships remained at Harvard for a longer period of time compared with controls (Figure 3). As demonstrated in Figure 3a, by 5 years only 13% of the instructors who had received the fellowship had left Harvard compared with 37% of their matched controls (P < .0001). The subset of instructors explicitly noted to have personal obstacles to promotion were also more likely to remain at Harvard (P = .03). Male and female instructors who received fellowships were similar in their likelihood to remain at Harvard compared with their matched controls (see Figure 3).
In this large study of junior faculty intramural grants, we found that, compared with matched controls, recipients were promoted from instructor to assistant professor earlier and remained at Harvard longer. Although those who received fellowships as assistant professors showed no difference in time to promotion, they, too, had longer retention rates. Outcomes were also positive for the subset who noted personal challenges that could impede career advancement. Faculty with noted obstacles might have been expected to have a slowed trajectory; that they were promoted earlier than matched controls suggests that fellowships may provide counterbalance to such obstacles and support academic achievement. That the effects were most pronounced for instructors suggests that a fellowship very early in a faculty member’s career will have its greatest impact.
Of note, female, but not male, instructors were more likely than their matched controls to be promoted earlier. Why male instructors who received a fellowship experienced no measurable improvement in time to promotion compared with controls would benefit from further study. Men in academic medicine are promoted at higher rates than women; it is possible that the impact of this relatively short and modestly funded fellowship will not augment factors that already support the gender differential in promotion rates.8,19,44,45 For women, however, the fellowship was associated with promotion despite the documented obstacles faced by women.8,44,45 The smaller sample size for men (n = 126; 26.2%), despite three controls per fellowship recipient, may also have provided inadequate power to detect a difference.
The Eleanor and Miles Shore 50th Anniversary Scholars in Medicine Program serves as an umbrella for a heterogeneous group of fellowships: some are for $25,000 and some for $50,000; some are selected by a schoolwide committee which takes into account personal obstacles in addition to scientific merit, while others are selected by departments or divisions with criteria that may include being a primary caregiver to family or dependents or other locally established requirements. Regardless of selection process, all fellowships have been consistently restricted to junior faculty, required written project descriptions, and been supported by a mentor and department chair. A strength of this study is the 16 years’ experience with the fellowship program, resulting in a sample large enough to establish robust findings. Although it is not possible in this analysis to identify the optimal or minimally effective dollar amount, selection criteria, use of funding, or other elements associated with advancement or retention, there are factors that likely contribute to the different outcomes for recipients compared with controls, including the actual funding, per se, the recognition by the dean and department, the celebration with families and mentors, the clarifying experience of writing a proposal to provide direction to a faculty member’s activities, and the explicit attention and commitment of a mentor.20,35,37,46–50
Given the emphasis of the fellowship program on both the 50th anniversary of the enrollment of women at HMS as well as on helping faculty overcome personal obstacles, it is not surprising that women were represented in higher proportions among the recipients than nonrecipients. Despite explicit publicity inviting applications from all faculty, the majority of the submissions were from women. Why recipients were more likely to be Harvard alumni/ae and trainees was not directly explored but may indicate preexisting mentorship relationships or greater familiarity with opportunities in the community. Faculty who obtained Shore awards by definition were early in their career; hence, awardees were younger than the overall population. Given all these differences between the recipients and the pool of eligible faculty, it was critical to match through propensity score modeling so that the impact of these variables could be neutralized.
We successfully identified a pool of nonrecipients through propensity score modeling who were as close as possible to the recipients on the factors we could measure so that the primary distinction would be exposure to the fellowship. The Harvard faculty offered a large pool of potential controls, thereby enhancing the reliability of the matching. Although propensity score modeling cannot account for all forms of potential confounding, it does permit controlling for a range of variables, in this case race/ethnicity, sex, department, prior training at Harvard, age, and years at rank, all of which could be associated with promotion and staying at Harvard.
Simply being chosen for the fellowship may suggest that the recipients’ promotion success and retention are not attributable to the fellowship but, rather, to other features that were valued by the selection committees. Our analysis strategy accounts for such features insofar as they are captured (either directly or by proxy) by the characteristics used to construct the propensity-matched controls, but unmeasured factors may remain. Other factors that could influence our outcomes, such as publication rates, would require further study. That those awarded fellowships by the schoolwide committee (which was charged to show preference to those who had self-identified personal obstacles) still demonstrated earlier promotion rates and longer retention than controls suggests that these findings do not simply indicate that the selection committees were good at picking those who were likely to succeed. Even when they picked those seemingly less likely to succeed, the recipients did better than their controls. Moreover, that male instructor fellowship recipients and assistant professors of both genders were not likely to be promoted earlier than matched controls argues against the notion that it was the bias of the selection process that was associated with the promotion outcome. There is no reason to think that the committee would be particularly good at identifying women instructors likely to be promoted but not as effective at identifying male instructors or assistant professors who would be promoted. At the instructor level, the divergence of the promotion curves (Figure 2) at the point when faculty typically receive the fellowships (2.5 years) further suggests a possible direct effect of the fellowship on promotion.
This study had several limitations. The relatively smaller pool of 36 male assistant professors and those 28 selected by the schoolwide committee precluded detailed analyses of those groups by sex. The Harvard community consists of 16 different hospitals and research institutions. Nonetheless, it is a single faculty of medicine with shared explicit promotion criteria. Whether these findings would be generalizable to other institutions would require further investigation, although smaller studies have shown positive benefits from seed grants and fellowships in terms of subsequent funding, improved scholarship, faculty satisfaction, and retention, suggesting that these findings likely have salience at other institutions.25,34–38 Although the fellowships included under the Shore umbrella and the selection processes were not identical, the process of applying did share the common requirements of a written project description, identification of a mentor, and explicit support from a department chair. The analysis of those with personal obstacles was limited to those reviewed by the schoolwide committee even though many of those selected locally also were chosen on the basis of these criteria. We took the most conservative approach to ensure uniformity in the “personal obstacles” subset. We opted to measure impact of the fellowship on promotion and retention, although understanding the implications for future grant support would be of interest for further investigation.
In summary, we sought to evaluate the impact of a junior faculty fellowship on academic advancement and retention. Of the most junior faculty members who were selected for the fellowship, women were significantly more likely to be promoted earlier than matched controls. Both instructors and assistant professors were more likely to remain at Harvard longer than their comparison groups. Given the challenges of establishing gender parity on medical school faculties, such relatively small investments could contribute to sustained diversity in the faculty for years to come.8,19,44,51 In light of the costs of recruiting faculty, the return on investment for the program is substantial. The Eleanor and Miles Shore 50th Anniversary Fellowship Program for Scholars in Medicine is a model for supporting faculty at a vulnerable point in their careers.
Acknowledgments: The authors wish to thank Louisa Bloomstein for her exceptional contributions as the study’s research assistant and Dr. Kenneth Kleinman for his invaluable advice regarding statistical methods.
1. Pololi LH, Krupat E, Civian JT, Ash AS, Brennan RT. Why are a quarter of faculty considering leaving academic medicine? A study of their perceptions of institutional culture and intentions to leave at 26 representative U.S. medical schools. Acad Med. 2012;87:859–869.
2. Alexander H, Lang J. The Long-Term Retention and Attrition of U.S. Medical School Faculty. 2008.Washington, DC: Association of American Medical Colleges.
3. Cropsey KL, Masho SW, Shiang R, et al. Why do faculty leave? Reasons for attrition of women and minority faculty from a medical school: Four-year results. J Womens Health (Larchmt). 2008;17:1111–1118.
4. Ries A, Wingard D, Gamst A, Larsen C, Farrell E, Reznik V. Measuring faculty retention and success in academic medicine. Acad Med. 2012;87:1046–1051.
5. Taljanovic MS, Hunter TB, Krupinski EA, Alcala JN, Fitzpatrick KA, Ovitt TW. Academic radiology: The reasons to stay or leave. Acad Radiol. 2003;10:1461–1468.
6. Hromas R, Abkowitz JL, Keating A. Facing the NIH funding crisis: How professional societies can help. JAMA. 2012;308:2343–2344.
7. Weinert CR, Billings J, Ryan R, Ingbar DH. Academic and career development of pulmonary and critical care physician–scientists. Am J Respir Crit Care Med. 2006;173:23–31.
8. Jena AB, Khullar D, Ho O, Olenski AR, Blumenthal DM. Sex differences in academic rank in US medical schools in 2014. JAMA. 2015;314:1149–1158.
9. Lowenstein SR, Fernandez G, Crane LA. Medical school faculty discontent: Prevalence and predictors of intent to leave academic careers. BMC Med Educ. 2007;7:37.
10. Carr PL, Ash AS, Friedman RH, et al. Relation of family responsibilities and gender to the productivity and career satisfaction of medical faculty. Ann Intern Med. 1998;129:532–538.
11. Jolly S, Griffith KA, DeCastro R, Stewart A, Ubel P, Jagsi R. Gender differences in time spent on parenting and domestic responsibilities by high-achieving young physician–researchers. Ann Intern Med. 2014;160:344–353.
12. Strong EA, De Castro R, Sambuco D, et al. Work–life balance in academic medicine: Narratives of physician–researchers and their mentors. J Gen Intern Med. 2013;28:1596–1603.
13. National Academies Collection. Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering. 2007.Washington, DC: National Institutes of Health.
14. Blood EA, Ullrich NJ, Hirshfeld-Becker DR, et al. Academic women faculty: Are they finding the mentoring they need? J Womens Health (Larchmt). 2012;21:1201–1208.
15. Kass RB, Souba WW, Thorndyke LE. Challenges confronting female surgical leaders: Overcoming the barriers. J Surg Res. 2006;132:179–187.
16. Levine RB, Lin F, Kern DE, Wright SM, Carrese J. Stories from early-career women physicians who have left academic medicine: A qualitative study at a single institution. Acad Med. 2011;86:752–758.
17. McGuire LK, Bergen MR, Polan ML. Career advancement for women faculty in a U.S. school of medicine: Perceived needs. Acad Med. 2004;79:319–325.
18. Speck RM, Sammel MD, Troxel AB, et al. Factors impacting the departure rates of female and male junior medical school faculty: Evidence from a longitudinal analysis. J Womens Health (Larchmt). 2012;21:1059–1065.
19. Carr PL, Gunn CM, Kaplan SA, Raj A, Freund KM. Inadequate progress for women in academic medicine: Findings from the National Faculty Study. J Womens Health (Larchmt). 2015;24:190–199.
20. DeCastro R, Griffith KA, Ubel PA, Stewart A, Jagsi R. Mentoring and the career satisfaction of male and female academic medical faculty. Acad Med. 2014;89:301–311.
21. Young KD; 2005–2006 Society for Academic Emergency Medicine Grants Committee; 2006–2007 Society for Academic Emergency Medicine Grants Committee. Productivity and career paths of previous recipients of Society for Academic Emergency Medicine research grant awards. Acad Emerg Med. 2008;15:560–566.
22. Crockett SD, Dellon ES, Bright SD, Shaheen NJ. A 25-year analysis of the American College of Gastroenterology research grant program: Factors associated with publication and advancement in academics. Am J Gastroenterol. 2009;104:1097–1105.
23. McDaniel A, Fullen DR, Cho KR, et al. Funding anatomic pathology research: A retrospective analysis of an intramural funding mechanism. Arch Pathol Lab Med. 2013;137:1270–1273.
24. Wallner PE, Ang KK, Zietman AL, et al. The American Board of Radiology Holman Research Pathway: 10-year retrospective review of the program and participant performance. Int J Radiat Oncol Biol Phys. 2013;85:29–34.
25. Aziz SR, Ziccardi VB. Survey of faculty educator development award recipients. J Oral Maxillofac Surg. 2011;69:3–10.
26. Boris C, Lessin SR, Wintroub BU, Yancey KB. A retrospective analysis of the Dermatology Foundation’s Career Development Award Program. J Am Acad Dermatol. 2012;67:969–974.
27. Bragg EJ, Warshaw GA, van der Willik O, et al. Paul B. Beeson career development awards in aging research and U.S. medical schools aging and geriatric medicine programs. J Am Geriatr Soc. 2011;59:1730–1738.
28. Butts GC, Johnson J, Strelnick AH, Soto-Greene ML, Williams B, Lee-Rey E. Diversity in academic medicine no. 4 Northeast Consortium: Innovation in minority faculty development. Mt Sinai J Med. 2008;75:517–522.
29. Adler SR, Chang A, Loeser H, Cooke M, Wang J, Teherani A. The impact of intramural grants on educators’ careers and on medical education innovation. Acad Med. 2015;90:827–831.
30. Viggiano T, Strobel H. Cole T, Goodrich T, Gritz E. The career management life cycle: A model for supporting and sustaining faculty vitality and wellness. In: Faculty Health in Academic Medicine: Physicians, Scientists, and the Pressures of Success. 2009:Totowa, NJ: Humana Press; 73–82.
31. Smith MA, Barry HC, Williamson J, Keefe CW, Anderson WA. Factors related to publication success among faculty development fellowship graduates. Fam Med. 2009;41:120–125.
32. Searle NS, Hatem CJ, Perkowski L, Wilkerson L. Why invest in an educational fellowship program? Acad Med. 2006;81:936–940.
33. Moses AS, Skinner DH, Hicks E, O’Sullivan PS. Developing an educator network: The effect of a teaching scholars program in the health professions on networking and productivity. Teach Learn Med. 2009;21:175–179.
34. Paller MS, Cerra FB. Investing in research: The impact of one academic health center’s research grant program. Acad Med. 2006;81:520–526.
35. Jagsi R, Butterton JR, Starr R, Tarbell NJ. A targeted intervention for the career development of women in academic medicine. Arch Intern Med. 2007;167:343–345.
36. Jagsi R, DeCastro R, Griffith KA, et al. Similarities and differences in the career trajectories of male and female career development award recipients. Acad Med. 2011;86:1415–1421.
37. Emans SJ, Goldberg CT, Milstein ME, Dobriner J. Creating a faculty development office in an academic pediatric hospital: Challenges and successes. Pediatrics. 2008;121:390–401.
38. Albanese M, Horowitz S, Moss R, Farrell P. An institutionally funded program for educational research and development grants: It makes dollars and sense. Acad Med. 1998;73:756–761.
39. Domino SE, Smith YR, Johnson TR. Opportunities and challenges of interdisciplinary research career development: Implementation of a women’s health research training program. J Womens Health (Larchmt). 2007;16:256–261.
40. Gardner AK, Sudan R, Sidhu R, Mann BD, Scott DJ. The Association for Surgical Education CESERT grant program: The first 15 years. Am J Surg. 2015;209:115–119.
42. Fried LP, Francomano CA, MacDonald SM, et al. Career development for women in academic medicine: Multiple interventions in a department of medicine. JAMA. 1996;276:898–905.
43. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70:41–55.
44. Association of American Medical Colleges. The State of Women in Academic Medicine: The Pipeline and Pathways to Leadership, 2013–2014. 2014.Washington, DC: Association of American Medical Colleges;
45. Tesch BJ, Wood HM, Helwig AL, Nattinger AB. Promotion of women physicians in academic medicine. Glass ceiling or sticky floor? JAMA. 1995;273:1022–1025.
46. Chung KC, Song JW, Kim HM, et al. Predictors of job satisfaction among academic faculty members: Do instructional and clinical staff differ? Med Educ. 2010;44:985–995.
47. Daley S, Wingard DL, Reznik V. Improving the retention of underrepresented minority faculty in academic medicine. J Natl Med Assoc. 2006;98:1435–1440.
48. DeCastro R, Sambuco D, Ubel PA, Stewart A, Jagsi R. Batting 300 is good: Perspectives of faculty researchers and their mentors on rejection, resilience, and persistence in academic medical careers. Acad Med. 2013;88:497–504.
49. Feldman MD, Arean PA, Marshall SJ, Lovett M, O’Sullivan P. Does mentoring matter: Results from a survey of faculty mentees at a large health sciences university. Med Educ Online. 2010;15. doi: 10.3402/meo.v15i0.5063.
50. Files JA, Blair JE, Mayer AP, Ko MG. Facilitated peer mentorship: A pilot program for academic advancement of female medical faculty. J Womens Health (Larchmt). 2008;17:1009–1015.
51. Sege R, Nykiel-Bub L, Selk S. Sex differences in institutional support for junior biomedical researchers. JAMA. 2015;314:1175–1177.