Sadeghpour M, Bernstein I, Ko C, Jacobe H. Role of sex in academic dermatology: results from a national survey. Arch Dermatol. 2012. doi:10.1001/archdermatol.2011.3617.
eTable 1. Promotion Timeline: Time Spent at Each Academic Rank by Sex
eTable 2. Adjusted Academic Productivity Measures Across Sexes
eTable 3. Professional Activities
Sadeghpour M, Bernstein I, Ko C, Jacobe H. Role of Sex in Academic DermatologyResults From a National Survey. Arch Dermatol. 2012;148(7):809–814. doi:10.1001/archdermatol.2011.3617
Author Affiliations: Department of Dermatology, Yale University (Dr Sadeghpour), and Yale University School of Medicine (Drs Sadeghpour and Ko), New Haven, Connecticut; and Departments of Clinical Sciences (Dr Bernstein) and Dermatology (Dr Jacobe), The University of Texas Southwestern Medical Center, Dallas.
Objective To determine whether there is an association between sex and academic rank and track, leadership, productivity, income, and career satisfaction.
Design National cross-sectional survey.
Setting Academic dermatologists across the United States.
Participants A total of 1263 full-time academic dermatologists.
Main Outcome Measures The association of sex with the following predictive variables: rank, promotion, academic productivity, leadership, salary, and career satisfaction.
Results Of the 343 respondents (27.2% response rate), 259 were full-time academic dermatologists, of whom 159 (61.4%) were men. Men held more senior positions (P < .001) even after adjustment for age (P < .02) and number of years since completion of residency (P < .05). Men were also more likely to occupy investigative career tracks (26.5% vs 11.1%), whereas women predominantly occupied clinical educator tracks (81.5% vs 50.0%) (P = .03). There was no significant difference in the hours worked between men and women (P = .052), and after controlling for academic rank, there was no difference in number of publications (P = .06) or grants received (P = .19). Difference in yearly salary became insignificant when adjusted for rank and other variables ($20 000 decrement for women; P = .12). Although most men (90.3%) and women (82.8%) were satisfied with their career, women were 24.6% more likely than men to consider leaving academia (P < .001).
Conclusions Sex-based differences in academic dermatology, including career track, academic rank distribution, leadership, and career satisfaction, persist. Measures that enhance the subjective rewards (eg, influence, collegiality, and mentorship) of academics and increased family-friendly measures for early-career academicians are important to close these gaps.
The past 4 decades have seen a dramatic increase in the number of women entering medicine, evidenced by current medical school graduates, 48.8% of whom are women compared with 6.9% in 1965.1 Women now represent the majority of dermatology trainees (61% of all dermatology residents in 2008).1 Prior studies2 report women are more likely than men to pursue a career in academic medicine, and currently women compose roughly 40% of the academic dermatology workforce.1,3
Numerous studies, including those by the American Association of Medical Colleges (AAMC),1,5 have demonstrated sex-based differences in advancement,1,2,4 leadership,5,6 productivity,7- 9 and salary10,11 in academic medicine. Studies12-15 in dermatology report similar findings. However, a more recent survey (2009) indicated that women in dermatology are highly “productive” academically, as measured by their prevalence as first and senior authors in prestigious dermatology journals.16 This raises the question of whether sex gaps are closing in academic dermatology. We hypothesized that men and women of equivalent age, rank, and career track have similar salary and productivity. To test this hypothesis, a national cross-sectional survey of academic dermatologists was performed to determine whether there is an association between sex and academic rank and track, leadership, productivity, income, and career satisfaction.
A cross-sectional, anonymous survey was approved by the institutional review board of The University of Texas Southwestern Medical Center and mailed to 1263 academic dermatologists (individuals holding an MD or a DO degree whose salary and benefits came, in large measure, from an academic institution) across the United States in January 2009. A preliminary list of academic dermatologists was provided by the American Academy of Dermatology, which was then cross-referenced with academic dermatology websites to capture an up-to-date and comprehensive list of academic dermatologists. Those included were full-time academic dermatologists holding an MD (or MD and other graduate) degree who reported their annual salary.
A 41-question survey was created, which included demographic characteristics, geographic region of practice, current academic appointment, number of years spent at each academic rank, American Board of Dermatology certification, training background (including date of completion of residency and the type and date of completion of fellowships), current academic track (clinical scholar/investigator, defined as a minimum of 50% of time devoted to patient-based research; clinical educator, defined as 50% to 75% of time devoted to patient care, teaching, and administrative duties; or basic/translational research, defined as greater than 75% of time devoted to laboratory-based research), and tenure status. Professional activities included total number of peer-reviewed publications, grants (federally funded and other peer-reviewed) received as a principal investigator and coinvestigator, and hours worked per week in total (and in patient care, teaching, research, administration, and writing for publication). In addition, respondents were asked about leadership positions. Career satisfaction with regard to both career in academia and promotion was assessed.
All surveys were anonymously coded and entered into a spreadsheet (Excel 2007; Microsoft), which was used as input to statistical software (SAS, version 9.2; SAS Institute Inc). In the survey, questions regarding age and salary were categorized. For the purposes of statistical analysis, the value assigned to the category was the midpoint. The mean (SD) was reported for continuous variables, and t tests were used for sex comparisons. We used χ2 tests of association to determine sex differences on categorical variables, adjusting for continuity in 2 × 2 tables. Regression analyses using sex and selected other covariates were used to evaluate sex differences in salary, publications, and grants. These were scaled so that negative values indicated that salary, publications, or grants of women were less than those of men. P values test the null hypothesis that the regression coefficient for sex did not differ significantly from zero. P < .05 was considered significant.
Of the 1263 surveys sent, 343 (27.2%) were returned. Of the 259 respondents who met inclusion criteria, 159 (61.4%) were men and 100 (38.6%) were women (P < .01); the mean age was 49.9 years for men and 42.6 years for women (Table 1). This age difference was significant (P < .001).
There was a sex difference in academic rank, despite adjustment for age and years since completion of residency (P < .02 and P < .05, respectively). Women were predominantly at the assistant professor level (50.0%) compared with men (24.3%); conversely, men were predominantly at the full professor level (47.4%) compared with women (14.9%) (Table 1). More men were in clinical research–based tracks (26.5%) than women (11.1%), whereas more women (81.5%) were in the patient-care oriented clinical educator track than men (50.0%) (P = .03) (Table 1). There were no women respondents in the basic science research track vs 17.6% of men. Reflecting this distribution, more men (39.1%) than women (17.1%) were tenured (P = .005) (Table 1).
Differences in the timing of promotion of men and women from instructor and assistant professor to associate professor did not differ significantly across sexes (eTable 1).
The difference in publication across sexes when controlled for both rank and years since completion of residency missed statistical significance by a small measure (P = .06) (eTable 2). When adjusted for rank, years out, or both, the difference in the total number of grants became nonsignificant across sexes (P = .65, P = .54, and P = .19, respectively) (eTable 2).
There was no significant difference in the total amount of time worked between men (52.9 hr/wk) and women (50.2 hr/wk) (P = .052) (eTable 3). The mean time spent in each type of service is reported in eTable 3.
There was representation of both sexes in most leadership positions (Table 2). However, men outnumbered women in the assistant or associate dean, department chair/division chief, and program director positions (P = .001, P = .001, and P = .01, respectively) (Table 2).
Regression analysis was used to determine whether observed sex differences in income could be attributed to potential confounders, such as rank and age. When rank was included as a covariate, the sex difference was approximately $20 000 and was not significant (P = .12); when rank was excluded, the difference was larger (Table 3).
Most men (90.3%) and women (82.8%) reported being either “very satisfied” or “somewhat satisfied” overall with their career; however, 55.5% of men reported being “very satisfied” in comparison to 36.4% of women (P = .02). Women were 24.6% more likely to have considered leaving academia, with 75.2% of women vs 50.6% of men reporting either a current or past desire to leave academia (P < .001). Men and women had different reasons for considering leaving academia. For example, “higher salary” was cited by 40.9% of men and 55.0% of women (P = .04), “institutional pressures” was cited by 34.0% of men and 56.0% of women (P < .001), and “desire to spend more time with family” was cited by 23.9% of men and 41.0% of women (P < .006).
Both men and women reported that “institutional support for academic activities” (reported by 35.8% of men and 43.0% of women), followed by a “more supportive and collegial institutional environment” (reported by 29.6% of men and 41.0% of women), would increase the likelihood of staying in academics (Table 4). Women reported a higher desire for “career development programs offered” by home institution (P = .04). Table 4 provides further details regarding career satisfaction.
The purpose of this study was to determine whether there is an association between sex and rank, productivity, leadership, income, and career satisfaction among full-time academic dermatologists. Our results, while encouraging, also underscore continued discrepancies in rank, with women predominating as assistant professors and men as professors, even after adjustment for age. Men also predominated in research-based tracks and leadership positions. Women considered leaving academia more frequently and had lower career satisfaction than men. Men and women respondents, however, did have similar timing to promotion, productivity (as measured by publications and grants), work hours (when comparing full-time faculty), and income (although there was a $20 000 difference in favor of men).
Our results are similar to those reported across medicine and even in other scientific fields outside medicine. For example, prior studies1,2,14 also report that women are more likely to occupy junior-level academic positions than their male counterparts. The relative lack of women in investigative career tracks and leadership positions is also well documented and the focus of multiple task groups in organizations such as the National Institutes of Health and National Academies.5,6,17,18 Prior studies4,10,11 suggest income disparities are prevalent in academia. The income decrement of $20 000 in our study is comparable to the $22 347 average income decrement reported by McMurray et al.11
The results of this study have some important differences from prior reports. We found that the current cohort of women is progressing through academic ranks at the same rate as men. This is in contrast to prior reports4 in pediatrics that suggested women remain in junior positions rather than advancing to senior rank. We also found women are equally productive in terms of hours worked, number of peer-reviewed publications, and grants received after adjustments for seniority (when comparing faculty with full-time appointments). This is in contrast to Jacobson et al,13 who reported that women dermatologists work fewer hours. This discrepancy is likely due to differences in the surveys themselves. Our respondents were full-time academic dermatologists who reported their current work routine; Jacobson et al13 surveyed recent graduates at the American Board of Dermatology Certification Examination about their intended work hours.
This study was unique in that it examined career satisfaction among academic dermatologists. Dermatologists as a whole have high career satisfaction, and this was also true of academics.19- 21 Women, however, were less satisfied than men. Interestingly, we also found women more frequently considered leaving academia for a higher-paying position, perhaps reflecting a real or perceived decrement in income. The top reasons for leaving academia for both sexes were similar to prior reports,22,23 including more time with family, higher salary outside academia, and institutional pressures. But we also found dissatisfaction with mentorship, collegiality, and level of institutional support among respondents, particularly women. Our findings have implications for the future of the academic workforce. Lack of senior women mentors may have a reciprocal negative effect on junior women entering the field. Although men certainly may serve as mentors on an individual level, the lack of women in senior roles may discourage junior women on a more global level.24- 26 We found that women working full-time will proceed through the promotion process at a rate similar to their male counterparts, yet women were still more likely to consider leaving. This suggests that the disparity in senior positions may be due to sex-based differences in attrition, as has been reported to occur across science and medicine at all levels of training.18 Furthermore, we found that subjective rewards were important to retention across sexes, which is consistent with the “subjective rewards hypothesis,”27,28 which implies that despite other sex disparities, subjective job rewards (eg, intellectual stimulation, relationships, and influence at work) are key factors in retention. This is important because factors such as income differences between academia and private practice are fixed, but enhancing the subjective rewards of academia (particularly for women) is an obtainable goal for most academic centers. The higher proportion of women in part-time positions may also partially account for discrepancies in leadership positions and investigative career tracks, as these largely require a full-time appointment.29 Creating more family-friendly career tracks, particularly for early-career academicians, is important in addressing this disparity.
This study has several limitations. First, responses were based on self-report with no available means for validation, creating the possibility of sex-based difference in self-reporting. Second, P values might be inflated because of multiple comparisons producing some inflation of α. Third, the surveys were anonymous (to maximize honest reporting), so we were unable to contact nonresponders. This may account for our relatively low response rate. We were also unable to determine how nonresponders differed from responders.
In conclusion, we found that although gaps in rate of promotion and productivity in academic dermatology have closed, important differences persist. We hypothesize that some of these discrepancies are the result of women's increased likelihood to occupy part-time positions and attrition, emphasizing the need to better understand faculty turnover. Furthermore, family-friendly measures to increase flexibility in early investigative career tracks are needed.17,18 Although income is cited as the largest factor in choosing a career in private practice, our data highlight that non–salary-related measures (eg, mentoring, career development, and collegiality) may be just as important in an academic setting.
Correspondence: Heidi Jacobe, MD, MSCS, The University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390-9069 (Heidi.Jacobe@utsouthwestern.edu).
Accepted for Publication: December 20, 2011.
Author Contributions:Study concept and design: Sadeghpour, Bernstein, Ko, and Jacobe. Acquisition of data: Sadeghpour and Jacobe. Analysis and interpretation of data: Sadeghpour, Bernstein, and Jacobe. Drafting of the manuscript: Sadeghpour, Bernstein, and Jacobe. Critical revision of the manuscript for important intellectual content: Bernstein, Ko, and Jacobe. Statistical analysis: Bernstein and Jacobe. Obtained funding: Jacobe. Administrative, technical, and material support: Ko and Jacobe. Study supervision: Ko and Jacobe.
Financial Disclosure: None reported.
Funding/Support: This study was supported in part by Yale University School of Medicine Office of Student Research.
Role of the Sponsors: The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; or in the preparation, review, or approval of the manuscript.
Additional Contributions: We are indebted to Molly Carnes, MD, for her invaluable guidance and contribution to survey questions; Richard Edelson, MD, and Kim Yancey, MD, for endorsing and supporting our study; Merle Waxman, MA, for her expertise and editing of the survey; and Ardavan Akhavan, MD, for his expertise in manuscript preparation.