Sticky Floor and Glass Ceilings in Academic Medicine: Analysis of Race and Gender

Purpose This paper examines the changes in the representation of women and racial minorities in academic medicine, compares the proportion of minorities in medicine and the general United States (US) population, and discusses potential explanations for observed trends. Methods A retrospective cross-sectional analysis of the Association of American Medical Colleges (AAMC) database was done and used to collect data on the gender and race of physicians in academic medicine. Data was collected for instructors, assistant professors, associate professors, full professors, and chairpersons from 2007 to 2018, and trends were presented. Results White physicians represented most academic physicians at every academic level, peaking in proportion at 82.74% of chairpersons and were lowest at the level of instructor at 59.30%. A similar distribution existed when gender was compared, with men comprising 84.67% of chairpersons and forming the majority at levels of full, associate, and assistant professors. However, most physicians at the level of instructors are women at 55.44%. Conclusions Though women and racial minorities have gained greater representation in academic medicine over the past decade, high-level academic positions are not as accessible to them. Existing efforts of advocacy for women and minority races have proven fruitful over the past decade, but much more work needs to be done.


Introduction
Gender and racial diversity in medicine improves workforce morale, reduces burnout, enhances patient care, and leads to a higher quality of scientific advancement [1,2]. Increasing diversity in medical academia also places a greater emphasis on serving populations with poor access to healthcare and increases research on improving healthcare access for marginalized populations [2,3]. Although recent trend analyses show that more women than men are matriculating in American medical schools, racial minorities are still not well represented amongst medical students. [4,5]. Additionally, while more women are entering the medical workforce, the same change has been slow to translate into academic medicine, particularly in leadership positions. Lack of progress in equity, diversity, and inclusion has therefore been referred to as an elusive dream and glass ceiling [6].
Given the high value of greater representation in medicine, trend analyses have been conducted for various trainee and physician groups in medicine. These analyses provide insights into the efficacy of representation advocacy and guide the goals of future advocacy efforts. Yu et al. showed that from 1998 to 2007, the greatest representation of women in academic medicine occurred at the lowest levels of the academic ladder, with "instructors" being almost 50% women, whereas only 14.7% of professors were women, with just 9.2% being chairpersons [7]. Racial minorities also follow a similar trend, in that the largest representation of racial minority groups is at the level of instructors or assistant professors at 23.3% and 23.5%, respectively. At the ranks of professors or chairpersons, the representation of racial minorities falls to 10.6% and 9.6%, respectively [7]. Across all specialties of academic medicine, the trend is that women and racial minorities are most prominently represented in the lowest rungs of the academic ladder [7,8]. These disparities have also been documented across medical schools, multiple specialties, professional societies, and editorial boards of medical journals [9][10][11][12][13][14].
Since 2007, many advocacy initiatives and social movements in America have been put into action to increase gender and racial representation in medicine, such as the efforts of the black lives matter movement. However, based on our literature review, to date, no study has analyzed trends in gender and racial representation in academic medicine since 2007. This paper examines recent trends in career advancement for women and racial minorities in academic medical from 2007 to 2018 to capture the effects of advocacy efforts in the past decade and a half. This study also aims to provide insights into current diversity gaps in American academics and therefore, guide future advocacy efforts.

Materials And Methods
This was a retrospective data analysis study that was exempt from the Research Ethics Board review. It was based on publicly available data, and the dissemination of results did not identify any individual or generate new forms of identifiable information. The methodology of this study has been validated in recent publications [15,16]. The data for this study was obtained from the Association of American Medical Colleges (AAMC) website. The AAMC database included demographic data on gender and race for academic positions in US medical school faculties from 2007 to 2018. A data collection tool was developed and tested by a senior author prior to data collection. Data collection was conducted from April 2021 to September 2021. Gender and race data was collected for each of the following academic positions in each medical specialty: instructor, assistant professor, associate professor, professor, and chairperson. Gender variables were defined as "male" and "female", based on the variables present in the AAMC demographic report categorization of the data. Data was collected for the following racial groups, as defined and categorized by AAMC demographic reports: White, Asian, Black, Hispanic, Multiple Race, unknown, and others. The "others" category includes data from American Indians, Alaskan Natives, Native Hawaiian, or other Pacific Islanders.

Analysis of data
Percentages were used to report trends in gender and race in academic medicine. The percentage of faculty members was averaged over 12 years for each academic position (instructor, assistant professor, associate professor, professor, and chairperson). The absolute change in the number of physicians from 2007 to 2018 was calculated for each gender and racial group.

Results
Between the years 2007 and 2018, over 46,258 academic medicine positions were added in the United States (US) ( Table 1).  Others include American Indians, Alaskan Natives, Native Hawaiian, Pacific Islanders, etc.

Racial trends
The proportion of White physicians is lowest at 59.29% in the position of instructor and highest at 82.74% in the position of chairperson. This trend of increasing proportion at higher levels of academia is in sharp contrast to the other racial groups (Figure 1). Asians were in the greatest proportion at the level of instructor and assistant professor at 20.09% and 21.13%, respectively, but taper sharply at the ranks of full professorships and chairpersons at 9.88% and 6.25%, respectively. Hispanics had a relatively flat distribution throughout the different academic positions, going from 3.21% at the position of instructor to a slightly higher 3.64% at the position of chairperson. Comparing the number of members in each group at the start and end of the observed period provides an interesting snapshot of the changes that have occurred in academic medicine. Asian physicians have seen the largest relative increase at every academic rank (Table 3), with the largest increases occurring at the highest levels of the academic hierarchy.

All academic levels 2007, (%) 2018, (%) Relative change (%) Absolute change (%)
All   "+" denotes increase, and "−" denotes decrease From 2007 to 2018, there was a 105% increase in the relative proportion of Asian chairpersons; this was accompanied by an approximately 60% increase in the relative proportion of both Asian full professors and associate professors. This was in stark contrast to the proportion of Hispanic academic physicians with a relative decrease of approximately 5% at the levels of assistant professor and a relative decrease of approximately 6% at the level of instructor. While overall, the relative proportion of Hispanic academic physicians was higher by 6.23%, no other racial minority experienced a decline in numbers at any academic rank. This increase in the proportion of racial minority academic physicians was accompanied by a decrease in the proportion of White academic physicians at every academic rank; however, particularly in the upper echelons of academia, White academic physicians remain over-represented.

Trends in gender
The absolute change from 2007 to 2018 in every academic position was higher in female physicians when compared with male physicians. The proportion of men increased with increasing academic rank ( Figure 2).

FIGURE 2: Gender of academic physicians by position
The only deviation from that trend was at the level of an instructor; over a 12-year average, women represented 55.44% of the group. Beginning at the level of assistant professor, the proportion of men was higher at 56.26%, and this trend continued, peaking at 84.67% at the level of the chairperson. The change in the proportion of 33.75% women at associate professor to 21.10% women at professor showcases an absolute decrease of 12.65% but marks a relative decrease of 62%, the largest drop in proportion for women between academic levels.
The increase of 27,952 women in academic medicine between 2007 and 2018 resulted in a relative increase of 19.73%, while men in academic medicine saw a relative decrease of 10.22%. Women chairpersons, full, associate, and assistant professors, as well as instructors all, saw an increase, with the biggest relative increases being at the levels of chairpersons and full professors. Compared to 2007, in 2018 there was a relative increase of 53.94% in female chairpersons and an increase of 39.83% in female full professors; however, despite the increases, males comprised the majority in higher academic ranks.

Discussion
The aim of this study was to identify trends in the gender and race of physicians in academic medicine. The most important findings of this study showcased high-level positions in academia have a greater homogeneity in both gender and racial representation, with the highest proportions being represented by males and Whites, respectively. These findings are in keeping with those seen in analysis conducted across all medical specialties, which found that women and racial minorities are generally under-represented in the upper echelons of academic medicine [19,20]. The one exception to this is Asian academic physicians, who are better represented at all levels of academia compared to other racial minority populations [7,19].
While the reasons for the observed trends are multi-factorial, several compelling arguments have been put forth in the literature which aim to explain the differences seen. While medicine has made great strides in increasing inclusivity among different specialties and at all levels of the academic ladder, these changes have been relatively recent. Thus, more gender and racially diverse cohort of physicians is currently under training and, therefore, potentially decades away from holding positions in the upper rungs of academia such as full professorship or chairperson [21]. However, even at the level of undergraduate medical training, racial diversity can be improved as per a recent study showing the lack of racial diversity over the past four decades in medical students [22]. Secondly, the lack of visible role models for both female and underrepresented minority groups has also been proposed as a potential barrier to diversity [23,24]. Interestingly, while there is an under-representation of females at the level of professors, Kapoor et al. [20] examined factors influencing career advancements such as first author publications or grants from organizations like the National Institutes of Health (NIH) found that women were less likely to have those characteristics and the rates of professorship were not significantly different between men and women after these factors were accounted for. The same study suggested that the under-representation could, therefore, be due to differences in access to research resources or support [20]. Additionally, women reported higher rates of gender discrimination in all specialties except for obstetrics and gynecology, where males reported higher rates of gender discrimination [25]. It showed the predominance of women physicians in the obstetrics and gynecology workforce yet a predominance of White male physicians in tenure positions, senior academic ranks, and leadership positions [25]. These perceptions of discrimination could contribute to certain groups having decreased participation in academic medicine.
To increase diversity in academic medicine and address implicit biases in access to research and leadership positions, current research in the field suggests that blinded reviews be implemented in the research review process. Additionally, improving female representation on editorial boards could be a potential means to prevent bias and improve first authorship among women [26,27]. In the United Kingdom (UK), the Athena Scientific Women's Academic Network (SWAN) has approached this issue by recognizing institutions that have focused on improving gender parity in high-level academic positions. This recognition has successfully increased the number of women in leadership positions and the rate at which gender parity is being achieved in academia [28]. Underrepresented racial groups also face barriers to attaining high-level academic positions. Research suggests that promoting mentorship and professional development programs specific to racial and ethnic minorities support more diverse authorship in academic medicine. This leads to more opportunities for members of these populations in the form of promotions and research grants [29]. The success of these interventions could serve as a model for academic institutions as they choose their next steps in improving gender and racial representation in academic medicine. Since succession planning in leadership is vital to continued improvement and lasting excellence in healthcare organizations, a lack of transparency in succession planning is also an important factor that will need redressal [30] Limitations This study has its share of limitations. Firstly, only specific genders were included in the study (female and male). Representation of other genders (transgender, non-binary, and others) was not studied. Future studies should assess the representation of all genders as well. There is a limited amount of research on the role that gender and race play in the selection of a career in academic medicine by under-represented groups. This is a relevant area for future research, particularly on the identification of perceived and real barriers to entry and progress in academic medicine, such as access to grant money, resources, and supports needed to publish or the ability to participate in clinical trials. The data did not include indicators of academic productivity such as h-index, number of citations, and the total number of publications. Furthermore, this study did not explore any additive effects of being both a gender and racial minority, which may place them at a further disadvantage.

Conclusions
Analysis of the AAMC data examining the past 12 years revealed that at higher ranks of academia, there is gender and racial disparity, with a disproportionately high representation of males and Whites. However, between the years 2007 and 2018, under-represented minorities have made large strides at all levels of academia, and the proportions of gender and race in academic medicine are trending towards greater diversity and representation, matching their proportions in the general population. While the causes of the disparities in representation are likely multifactorial, several compelling arguments seek to explain these differences. These results demonstrate a cohort of academic physicians that is increasing in both gender and racial diversity at all levels of academic medicine.

Additional Information Disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: The study was partially supported by a grant from the Arrowhead Neuroscience Foundation and a grant from the Disparities Research Institute. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.