The future of academic medicine depends on the intellectual capital of its faculty, which remains unevenly distributed between men and women, creating a gender gap at all ranks and in leadership roles.1–7 Men’s domination of leadership positions8–10 has been attributed to a “pipeline” problem that would recede with women’s full participation in the workforce.8,11 Despite almost two decades in which women have composed nearly half of the labor force,12 a gender gap remains that is most marked at the full professor rank, and in prestigious research awards in science, technology, engineering, and medicine (STEM), even when women are well represented in the nomination pool.13 Dramatic increases in women graduating from medical schools have not been matched by a rise in women faculty,1 and during the last decade, the proportion of female full professors increased nationally by only seven percentage points, from 12% to 19%. The lack of women at the full professor rank may explain why, in 2010, only 13% of department chairs and 11% of medical school deans were women.1
Interventions at both institutional and individual levels are necessary to close the gender gap in faculty. Promising results from mentoring and career development programs in the Department of Medicine at John Hopkins University School of Medicine were published over a decade ago.14 The Universities of Michigan and Wisconsin ADVANCE programs found that education of search committees increased the hiring of women faculty.15,16 A growing body of evidence suggests that women’s advancement requires practices that establish organizational responsibility and accountability, thereby creating “organizational catalysts of change.”17,18 This study builds on prior work by examining the effectiveness of a multifaceted intervention at the Stanford University School of Medicine. We compare outcomes following the intervention with data from the official roster of accredited U.S. medical school faculty maintained by the Association of American Medical Colleges (AAMC). We have also evaluated faculty satisfaction prior to and several years after initiation of the intervention.
Assessment of gender-related problems at the School of Medicine
In 2001, Stanford University created the Provost’s Advisory Committee on the Status of Women Faculty (PACSWF) to enhance the representation and experiences of women faculty. PACSWF identified four areas of need: recruitment of more diverse faculty; resources that provide faculty time to conduct and publish research; career development programs that enhance professional success; and programs to decrease social isolation. PACSWF administered a university-wide quality of life (QOL) survey to assess faculty satisfaction in 2003 and 2008, providing one of our benchmarks.19
Interventions to combat gender disparities among medical school faculty
To address the four needs, the School of Medicine created the Office of Diversity and Leadership (ODL) in November 2004, and all subsequent interventions were executed under the auspices of this office. With few exceptions, the interventions were not deliberately targeted at women faculty because ODL recognized a need for professional development of both men and women.20 Below, we briefly describe each component of our intervention and participation rates.
Expand diversity in faculty recruitment.
The senior associate dean for ODL reviewed every search committee membership to ensure inclusion of women as a prerequisite for approval and presented the case for diversity to the committee—namely, arguments for how diversity is linked to excellence; current university and School of Medicine faculty demographic data; the demographic profile of the expected applicant pool, compiled from the AAMC faculty roster; processes for attracting diverse candidates, including strategies for avoiding cognitive errors in evaluating letters of recommendation, candidate selection, etc.21; and institutional financial resources to assist in the recruitment of women and underrepresented candidates. A total of 195 faculty search committees (95% of total) received the intervention.
Provide greater access to institutional resources.
The ODL developed the McCormick Faculty Awards, modeled after the Claflin Distinguished Scholar Awards,22 to provide women assistant professors with unrestricted funding for protected time to pursue research. This intervention addresses the critical barrier of competing demands faced by junior faculty in terms of clinical work, research, teaching, and service, which has been shown to lead to detrimental work–family conflict, especially for women. Three awards of $30,000 per year, each for two years, are made each year. A total of 12 awards were made during 2006–2010, serving approximately 8% of women assistant professors. All 12 recipients have been promoted, and 11 retained.
Provide structured professional development programs.
ODL created the Faculty Fellows Program, a yearlong leadership and career development program with a competitive nomination and selection process, involving individualized development, access to influential faculty and university leadership, and intensive mentoring. The program enrolls 16 to 20 male and female assistant and associate professors each year and comprises three complementary components: a monthly dinner meeting designed to demonstrate the variety of leadership styles exemplified by major institutional leaders, including the university president and provost, the dean of the School of Medicine, and research institute directors; small-group mentoring, each led by a senior faculty member with a successful track record in mentoring faculty; and an individualized career development planning process facilitated and coached by the School of Medicine’s director for organizational effectiveness (J.M.). In structuring each class, we used a rigorous nomination process, which explicitly emphasized the school’s priority for diversity and excellence in its future institutional leadership. During the six years spanning this report, 81 faculty members have participated in the program, 45% (43) of whom are women, representing approximately 14% (43) of the total eligible pool of 317 women assistant and associate professors.
Skills building workshops.
To provide faculty with guidance on professional and interpersonal topics relevant to career advancement, the ODL developed a series of skills building workshops that are open to all faculty and run by internal faculty and external consultants with the relevant expertise. The topics include scientific writing, grant writing, negotiation, delegation strategies, time management, managing information overload, work/life balance, recognizing and managing implicit bias, managing a laboratory, communication and presentation skills, and faculty reappointment and promotion criteria. Of the total 657 participants, 420 (64%) were women, and of these, 200 (nearly half) participated in multiple workshops. The greater participation by women is consistent with a recent study by Carter and Silva23 showing that women are more likely than men to participate in informal professional development programs.
Encourage professional networks.
To reduce women’s social and professional isolation, the ODL implemented the Women’s Faculty Networking Program, the only intervention component restricted to women. The program consists of monthly lunch meetings that are open to women faculty of all ranks. The sessions alternate between informal meetings that promote socializing and sharing experiences, to more formal gatherings featuring guest speakers who are renowned scholars in gender issues related to women’s career advancement. Two hundred twenty-two participants (approximately 55% of women faculty) attended at least one networking session.
Intervention outcome measures
Faculty demographic data.
In the spring of 2011, we obtained annual snapshot data for our school’s faculty demographics during the preceding decade, which included the number and percentage of women faculty (overall and by rank) for each year before and after initiation of the intervention (2000 and 2010). For comparison, we obtained AAMC Faculty Roster snapshot data from 2000 to 2010 for six peer institutions, randomly selected from the top-10-ranked research-intensive institutions. A second comparator group included the entire national cohort of faculty from the 126 U.S. medical schools accredited by the Liaison Committee on Medical Education at the time. Data for faculty at the instructor rank were excluded because some interventions were not offered to instructors.
Trends in promotion, hiring data, and candidate pools.
We reviewed faculty promotion and hiring data at Stanford to identify patterns related to faculty rank and track during the intervention period (2004–2010). We also analyzed candidate pools by reviewing all faculty search reports obtained after initiation of the intervention (2007–2011), and performed survival analyses to determine whether faculty in the investigator track, hired after the intervention was initiated, were more likely to remain at Stanford compared with those hired before the intervention.
Faculty satisfaction survey.
Two QOL surveys (2003 and 2008) were used to assess changes in global satisfaction, using a five-point Likert scale (“very dissatisfied” to “very satisfied”); for the current analysis, we combined responses for satisfied and very satisfied. Response rates for clinical and basic science faculty were 35% (221) and 68% (64) in 2003 and 43% (299) and 79% (80) in 2008, respectively. Respondent demographics from the two surveys were similar in distribution and representativeness across the medical school, allowing for comparisons of the two data sets. Because both surveys were anonymous, it was not possible to match responses from individuals. As the surveys used were part of the provost’s planned university-wide five-year assessment, the analyses are not exactly coincident with the intervention period, 2004 and 2010. The Stanford University institutional review board approved both surveys and exempted approval for their use in our study.
To assess increases in faculty, we computed the relative change in the number of faculty by gender, rank, and year (2004 and 2010), and the ratio of the number of faculty in a given year to the number of faculty in 2000 (the “baseline” year), at Stanford and at peer and national cohorts. We estimated the rate of increase in the percentage of women faculty (with 95% confidence intervals) in each cohort for the time periods before (2000–2004) and after (2004–2010) the start of the intervention, overall and by rank. We used multiple linear regression to evaluate whether rates of increase in women faculty differed between Stanford and comparator groups in each period. Here, the dependent variable was the percentage of women faculty in a particular year; independent variables included year, group, and year-by-group interaction; year was modeled by a two-piece piecewise linear model in which the two linear pieces are connected at year 2004. This model allowed separate rates of increase in the two time periods. For example, if the percentage of women faculty in the national cohort increased at 1% per year based on the linear regression (e.g., 25%, year 1; 26%, year 2; and 27%, year 3), and Stanford’s percentage of women faculty increased at 1.56% per year, then the difference in the rate of increase is 0.56% per year. Lastly, we used chi-square and Savage tests to evaluate whether faculty satisfaction at Stanford improved from 2003 to 2008, for all faculty, and separately for men and women. For all statistical analysis, we used SAS version 9.2 (SAS Institute Inc, Cary, North Carolina).
When we compared women and men faculty in 2004 and 2010 at Stanford, peer, and national cohorts (Table 1), we observed that the number of women faculty at Stanford increased from 234 (27.7%) in 2004 (preintervention) to 408 (33.5%) in 2010 (postintervention), a 74.4% relative increase. Increases at the assistant (65.7%; 108 to 179), associate (86.5%; 74 to 138) and full professor (75.0%; 52 to 91) ranks were observed. Stanford also exhibited substantially greater increases in the percentage of women faculty compared with peer and national cohorts.
Figure 1 shows the increase in total faculty number per year relative to the year 2000 for Stanford, peer, and national cohorts (panel A) and by gender (panels B–D) for each of the three cohorts. The overall faculty increase during 2000–2010 was greater at Stanford (2010 ratio = 1.8) compared with peer and national cohorts (2010 ratio = 1.3). The increase in women faculty at Stanford was also larger (2010 ratio = 2.6) than in peer and national cohorts (2010 ratio = 1.8 and 1.6, respectively).
Figure 2 shows the percentages of women faculty for Stanford, peer, and national cohorts in 2000–2010. With all ranks combined (panel A), the percentage of women faculty increased at Stanford during 2000–2004, but it remained lower than in the comparator cohorts during this time period. However, post intervention (2004–2010), the percentage of women faculty increased to equal the percentages of women faculty in the peer and national cohorts. By rank (panels B–D), there was a gradual increase in women faculty at the assistant and full professor ranks, punctuated by a steep increase at Stanford after 2004, which is not present in the peer and national cohorts. Among full professors, the percentage of women faculty was lower at Stanford in 2000, but the rate of increase was larger, especially after the intervention in 2004–2010 (a statistically significant difference at P = .001; Table 2). In 2010, Stanford’s percentage of women full professors exceeded those of peers and the national cohort. Among associate professors, the relative increase in women faculty was smaller at Stanford than in peer and national cohorts during 2000–2004 (preintervention) but was similar during 2004–2010 (intervention period).
To determine the significance of gains in women faculty post intervention, we used multiple linear regression to compare the rate of increase between Stanford, peer, and national cohorts in the time periods 2000–2004 and 2004–2010. Table 2 shows the differences between Stanford and the comparator groups in the estimated rate of increase in the percentage of women faculty. For all ranks combined, Stanford exhibited a significantly higher rate. Post intervention, the rate of increase in the percentage of women faculty was 1.08 at Stanford compared with 0.80 in the peer cohort and 0.72 in the national cohort: difference, Stanford versus national, 0.36 (0.17–0.56), P = .001; Stanford versus peers, 0.29 (0.07–0.51), P = .015. Analyses of the preintervention period (2000–2004) at each rank revealed no significant differences for within- or across-group comparisons. However, higher rates of increase in women full professors at Stanford compared with peer and national cohorts (1.03 versus 0.66 and 0.64) were observed during the intervention (2004–2010): difference, Stanford versus national, 0.40 (0.18–0.62), P = .001; Stanford versus peers, 0.37 (0.14–0.60), P = .003. The rate of change for assistant and associate professors at Stanford did not differ from those in peer and national cohorts. At the rate of increase during 2004–2010, the number of years to achieve 50% women in faculty positions overall, and for assistant, associate, and full professors, respectively, is 15, 6, 17, and 28 years (Stanford); 20, 7, 19, and 44 years (peers); and 23, 12, 26, and 48 years (national cohort).
Table 3 shows Stanford faculty responses to satisfaction questions in 2003 and 2008. Women’s satisfaction was significantly lower than men in 2003 (P < .01) but nonsignificantly in 2008. Men and women reported increases in overall satisfaction (P < .001 and P = .05, respectively); however, no significant change was observed among women basic science faculty (data not shown). Increases were greatest among clinical faculty in whom the percentage demonstrating satisfaction rose 13.4 percentage points in men, 68.3% (82) in 2003 to 81.7% (156) in 2008; and 23.2 percentage points among women, 48.1% (26) in 2003 to 71.3% (62) in 2008. These cross-sectional analyses in clinical faculty satisfaction were significant for both men (P = .01) and women (P = .01).
Assessment of faculty rank, track, promotion, and hiring on the net gains in women faculty
At the full professor rank, net gains occurred predominantly in the investigator tracks, accounting for 80% (31) of the 39 net gains in women at full professor rank. Of the 74 net gained women associate professors, 36% (27) were in the investigator track, and 64% (47) were in the clinician educator track. The net gain of 71 women assistant professors was evenly distributed between the investigator 46% (33) and clinician educator tracks (38). In the investigator track, 41% (32) of women were promoted to, whereas 19% (15) were hired as, full professors since the initiation of the intervention in 2004. Similarly, for associate professors, 66% (52) were promoted, and 20% (16) were hired into the rank since 2004. In the clinician educator track, 62% (7) of women full professors were hired, and 38% (5) were promoted. Hiring and promotion in the clinician educator track were equally distributed at the associate and assistant professor ranks, with new hires (25) and promotions (25) each accounting for 42% of women associate professors. At the assistant professor rank, all net gains in the investigator and clinician educator track resulted from new hires. Survival analysis of faculty in the investigator track revealed no significant difference in gender or rank regarding probabilities for departure (data not shown).
Analysis of candidate pools over a five-year period after the intervention was initiated
Analyses of search reports reveal that one-third (316) of the definitive pool of candidates (947) were women, mirroring the overall composition of the faculty as a whole. Of the women in the definitive pool, 20% (65) were hired, similar to the hiring rate for men. The applicant pool data were not analyzed because of inconsistencies in nonmandatory self-reporting.
We found that the Stanford University School of Medicine rose from below to above national benchmarks in the representation of women among faculty while making a visible commitment of resources to support faculty, with special attention to women. By comparing the Stanford cohort to peer and national cohorts, we have demonstrated three main findings: a relative increase in women faculty across the Stanford, national, and peer cohorts during 2004–2010 that was larger at Stanford; a rate of increase in the percentage of women faculty at Stanford that was significantly larger during the intervention period compared with preintervention; and rates of increases in women faculty during the intervention period that were greatest among Stanford full professors, and coincident with an increase in overall faculty satisfaction.
The larger increase in women faculty at Stanford may reflect that Stanford started with a lower proportion of women. However, this cannot explain the increases in women full professors at Stanford that exceeded the comparator groups by 2010. Recognizing that the relative increase is driven by the cohort size, we focused our analyses on the rate of change as a more robust approach to assessing changes in women faculty. Here we show a more rapid rate of increase, both overall and for full professors, at Stanford. Few studies have reported on the rate of increase in women in academic medicine—data that could provide important benchmarking for targeted interventions. One study of gender composition among faculty in the STEM fields indicates that the goal of women occupying 50% of faculty positions may require up to 100 years in the physical sciences.23 Our data on the rate of increases in women faculty during the intervention period reveal that the estimated time to achieve 50% occupancy of full professorships by women, even after the multifaceted intervention, is 28 years, with corresponding estimates of 44 years and 48 years in the peer and national cohorts. These findings underscore the need for even more effective interventions to bridge the gender gap in all faculty ranks across all institutions.
We observed increases in women faculty despite the relatively small proportion of faculty who participated in many components of the intervention, raising questions about mechanisms of the observed changes. Research suggests that creation of the ODL at Stanford provided “institutional responsibility for change.” This element of responsibility and oversight is essential for advancement of women, beyond any individual program.17 Analogous to the University of Michigan’s ADVANCE program,15 creation of the ODL and integration of its leadership into existing academic leadership roles provide an “organizational catalyst” for change. This is consistent with a longitudinal study of gender and race composition in 700 institutions, where the strongest predictor of women’s representation among top management was institutional leadership and accountability, rather than individual programs for diversity training, social isolation, and mentoring.18
Consonant with the National Academy of Sciences’ recommendations, Stanford’s multifaceted intervention included specific actions to establish institutional responsibility for change, expand faculty recruitment, foster career development, provide mentorship, and reduce social and professional isolation.24,25 The relative impact of the intervention on hiring, advancement, and faculty tracks requires exploration. Our observations indicate that net gains in women faculty varied by rank and academic track, with the largest net gains at the full professor rank occurring in investigator tracks. Similarly, promotions accounted for the majority of the net gains in women associate and full professor positions, suggesting that Stanford’s advancement strategies in the investigator tracks were effective. The finding that the majority of net gains in women assistant professors were in the clinician educator track and resulted from new hires underscores the need for a multifaceted approach to increase both hiring and promotion of women faculty. The largest gain in women faculty was seen in the pool who were advanced to full professor, suggesting that the interventions described for faculty development primarily benefited senior research faculty. In contrast, the largest gain in new hires was among assistant professors in the clinician educator track, consistent with recruitment having a greater impact on new hires at the junior ranks. These findings raise the question of what strategies are most likely to sustain the observed trends in advancement, while also emphasizing a potential disconnect between recruitment and advancement. Further analyses of the impact of the individual and collective interventions are needed to address whether the reported trends can be sustained across all ranks and tracks. The observed acceleration in the rate of increase in women at the full professor rank, concomitant with increases in faculty satisfaction, provide indirect evidence for a link between the interventions and outcomes observed. Because women full professors represent the available pool of women to occupy leadership positions,26 their greater numbers contribute to the likelihood that more women will occupy leadership roles in the future.
We do not know what interventions were in place at the comparator schools and therefore cannot conclude that our intervention accounts for the differences observed. Because the faculty surveys were anonymous, we are unable to ascribe a link between respondents and participants in the intervention. The small sample size of the Stanford faculty and the relatively short follow-up represent significant limitations, especially in conducting survival analysis, which would be helpful in fully assessing the impact of the intervention. Despite these limitations, these promising findings suggest that a multifaceted intervention, driven by institutional responsibility for change, may ameliorate the gender gap in academic medicine.
Acknowledgments: The authors thank Patricia Jones, PhD, prior vice provost of faculty development; Deborah Rhode, JD, and Shelley Correll, PhD, past and current chairs of the Stanford University Equity Panel; and Corrie Potter for their work in designing, implementing, and analyzing the Faculty Satisfaction Survey. They also thank Stanford President John Hennessey and Stanford Provost John Etchemendy for providing support for the establishment of the Office of Diversity and Leadership (ODL). In addition, they are grateful to the members of the ODL Executive Committee for their assistance in developing the ODL strategic plan and for program implementation. They extend appreciation to Judith Cain, Kathleen Warmoth, and Jennifer Scanlin for assistance with data retrieval and analysis, and Jon Lang and his colleagues at the Association of American Medical Colleges (AAMC) for providing data from the AAMC Faculty Roster. Finally, the authors greatly appreciate review of the manuscript by Caroline Simard, associate director of the ODL; Philip Lavori, chair, Department of Health Policy and Statistics, and coinvestigator on the Pathfinder Award; and Science Editors Network for editorial assistance.
1. Leadley J, Sloane R Women in U.S. Academic Medicine and Science: Statistics and Benchmarking Report 2009–2010. 2011 Washington, DC Association of American Medical Colleges https://www.aamc.org/members/gwims/statistics/
. Accessed February 19, 2014
2. Nonnemaker L. Women physicians in academic medicine: New insights from cohort studies. N Engl J Med. 2000;342:399–405
3. Hamel MB, Ingelfinger JR, Phimister E, Solomon CG. Women in academic medicine: Progress and challenges. N Engl J Med. 2006;355:310–312
4. Reed V, Buddeberg-Fischer B. Career obstacles for women in medicine: An overview. Med Educ.. 2001;35:139–147
5. Bickel J. Women in academic psychiatry. Acad Psychiatry. 2004;28:285–291
6. 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
7. 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
8. Eagly AH, Karau SJ. Role congruity theory of prejudice toward female leaders. Psychol Rev. 2002;109:573–598
9. Eagly AH, Karau SJ. Gender and the emergence of leaders: A meta-analysis. J Pers Soc Psychol. 1991;60:685–710
10. Valian V Why So Slow? The Advancement of Women. 1999 Cambridge, Mass MIT Press
11. Kanter RM Men and Women of the Corporation. 1977 New York, NY Basic Books
12. U.S. Department of Labor, Bureau of Labor Statistics. Women in the Labor Force: A Databook. 2004 http://www.bls.gov/cps/wlf-databook.htm
. Accessed February 19, 2014
13. Lincoln AE, Pincus S, Koster JB, Leboy PS. The matilda effect in science: Awards and prizes in the US, 1990s and 2000s. Soc Stud Sci. 2012;42:307–320
14. 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
15. Stewart AJ, LaVaque-Manty D, Malley JE. Recruiting women faculty in science and engineering: Preliminary evaluation of one intervention model. J Women Minor Sci Eng. 2004;10:361–375
16. Sheridan JT, Fine E, Pribbenow CM, Handelsman J, Carnes M. Searching for excellence and diversity: Increasing the hiring of women faculty at one academic medical center. Acad Med. 2010;85:999–1007
17. Sturm SP. The architecture of inclusion: Advancing workplace equity in higher education. Harv J Law Gend. 2006;29:247
18. Kalev A, Dobbin F, Kelly E. Best practices or best guesses? Assessing the efficacy of corporate affirmative action and diversity policies. Am Sociol Rev. 2006;71:589–617
19. Stanford University. . Report on the Quality of Life of Stanford Faculty. January 2010 https://facultydevelopment.stanford.edu/sites/default/files/documents/fqol-report-jan2010.pdf
. Accessed February 19, 2014
20. Burke R. Do managerial men benefit from organizational values supporting work–personal life balance? Gend Manag. 2010;25:91–99
21. Isaac C, Lee B, Carnes M. Interventions that affect gender bias in hiring: A systematic review. Acad Med. 2009;84:1440–1446
22. 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
23. Carter NM, Silva C Mentoring: Necessary but Insufficient for Advancement. 2010 New York, NY Catalyst
24. Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering. 2007 Washington, DC Committee on Maximizing the Potential of Women in Academic Science, Engineering (U.S.); Committee on Science, & Public Policy (U.S.)
25. Stewart AJ, Malley JE, LaVaque-Manty D Analyzing the problem of women in science and engineering: Why do we need institutional transformation? In: Stewart AJ, Malley JE, LaVaque-Manty D, eds. Transforming Science and Engineering: Advancing Academic Women. 2007 Ann Arbor, Mich University of Michigan Press
26. Nattinger AB. Promoting the career development of women in academic medicine. Arch Intern Med. 2007;167:323–324