Abstract
Introduction:
A diverse health care workforce can better serve diverse patient populations. This study characterizes the experience of sexual minority medical students, particularly those with multiple marginalized identities, to determine how these are related to intended specialty and satisfaction with medical education.
Methods:
Using data from 2018 to 2020 Association of American Medical Colleges Graduation Questionnaire (GQ) and matching Matriculating Student Questionnaire (MSQ), our study had three primary outcomes: any change in specialty from matriculation to graduation, a change in specialty from less welcoming to more welcoming specialty for sexual and gender minority (SGM) students, and overall satisfaction with medical education. Intersectional position was determined by cross-tabulation of sexual identity, race, ethnicity, and gender. We conducted multivariable logistic regressions to assess the association between intersectional position and all outcomes.
Results:
From 2018 to 2020, there were 49,509 unique respondents (82.7% response rate) to the GQ. Of this sample, 42,584 (86.0%) had at least one outcome of interest and 28,702 (58.0%) had responses to both GQ and MSQ items regarding intended specialty. Compared to White heterosexual male students, respondents with multiple marginalized identities were less likely to report being satisfied with their undergraduate medical education and more likely to have changed their intended specialty from a less to a more SGM-welcoming specialty.
Conclusion and Relevance:
Sexual minority medical students, especially those with multiple marginalized identities, are more likely to change their intended specialties from matriculation to graduation. This affects the diversity within specific medical and surgical specialties.
Introduction
Sexual and gender minority (SGM) patients, including lesbian, gay, bisexual, transgender, and queer/questioning populations, experience many barriers to care, including poor access to competent and compassionate clinicians.1,2 SGM patients have reported feeling more comfortable accessing medical care when their clinician is familiar with or a member of SGM communities.3–5 In addition to training clinicians in SGM health across undergraduate, graduate, and continuing medical education,6–9 a potential avenue to improving the care and outcomes of SGM populations would be to increase the representation of SGM clinicians in the health care workforce by supporting SGM trainees in their undergraduate and graduate medical education.
However, SGM persons face various barriers to pursuing careers in medicine.10,11 Sexual minority (e.g., lesbian, gay, bisexual) medical students are more likely to report stress, isolation, verbal harassment, mistreatment, and discrimination based on their sexual orientation12,13 and are about twice as likely to experience depression and related mental health comorbidities.13,14 Some sexual minority medical students feel the need to conceal their sexual identity and blend into the cisgender heterosexual population out of fear of discrimination. 15 In a recent study, graduating medical students who identified with sexual minority groups endorsed burnout at higher odds than their heterosexual peers, with a dose–response association between perceived frequency and intensity of mistreatment and the risk of burnout. 16 A recent systematic review evaluating factors that SGM medical students’ decision to pursue a career in a surgical subspecialty found that identity acceptance and instances of discrimination and bias during training were most influential. 10
Following medical school, students have a choice in pursuing a specialty. The culture, training environment, and inclusivity of a specialty can influence the specialty choice for medical students. A 2007 study of practicing physicians found that (in order) surgery, family medicine, and orthopedics were fields where physicians and surgeons expressed the most homophobic attitudes, whereas psychiatry, internal medicine, and pediatrics were fields that expressed the least homophobic attitudes. 17 Similarly, a 1998 study reported that surgeons were particularly likely to discourage sexual minority trainees from entering their specialty. 18 These survey data show a negative bias of some specialties that may be learned by medical students and potentially impact their specialty choice.
There are limited avenues to understand the broader experience of SGM persons in medicine. Recently, the American Medical Association embarked on a multiyear effort to confidentially collect sexual orientation and gender identity data from its medical student and practicing physician members. These data are not currently available for research.
The Association of American Medical Colleges (AAMC) has begun including sexual orientation and, later, gender identity questions in their surveys of matriculating and graduating medical students. Data from the AAMC may give clues to the experience of those entering the profession. In 2016, medical school graduates who identified as “gay or lesbian” were 3.1% of respondents with 2.1% identifying as “bisexual”; this increased to 4.1% and 5.3%, respectively, in 2023.19,20 The increasing recruitment and/or self-identification of sexual minority medical students provides an opportunity to explore their unique experience during training.
Research Question
What is the experience of medical students with multiple marginalized identities, specifically including sexual, racial, and ethnic identity, with regard to their satisfaction with medical education, changing their intended specialty from matriculation to graduation, and changing from a less to a more SGM-welcoming specialty from matriculation to graduation?
Methods
Data Source
We analyzed cross-sectional data from 2018 to 2020 AAMC medical school Graduation Questionnaire (GQ); data were available from 141 AAMC-accredited medical schools in 2018, 142 in 2019, and 145 in 2020. The GQ is an annual national survey of graduating medical students that ask questions about preclinical, clinical, and elective medical education experiences. Students complete the GQ at the end of medical school, prior to graduation. Responses are confidential, and participation is voluntary. In 2016, the AAMC added questions to the GQ about sexual orientation and gender identity. We included data from GQ respondents who also completed a Matriculation Student Questionnaire (MSQ). The MSQ is administered annually by the AAMC and collects information about entering medical students for the purpose of helping medical schools accredited by the U.S. Liaison Committee on Medical Education improve medical education. GQ and MSQ responses were matched by AAMC staff and provided as one dataset to researchers. GQ and MSQ items included in the analysis are listed in Supplementary Appendix in the Supplement. We conducted study analyses according to the Strengthening the Reporting of Observational Studies in Epidemiology reporting guideline. 21 This study analyzes data from an already existing AAMC database; no separate data collection was conducted, and no participant consent was obtained for our analysis, although respondents do consent to take the survey. Participant consent for data analysis was waived as AAMC deidentified all data made available to the research team. As this study analyzed deidentified, retrospective data, the study is not human subject research.
Measures
Outcomes
Our study had three primary outcomes: Any Specialty Change, Specialty Change from Non-Welcoming to Welcoming, and Overall Satisfaction with Medical Education. On the MSQ, students were asked to indicate which general area of study they were considering from a list of 29–39 medical school specialties (some specialty categories changed throughout the years). On the Graduating Student Questionnaire (GQ), students were to indicate their intended area of practice from the same list.
Students were classified as changing intended specialties if they reported a different specialty selection on their GQ than their MSQ. Specialties were group as “welcoming” and “nonwelcoming” of SGM persons based on Sitken et al. 2016: “Nonwelcoming” specialties included: Anesthesiology or subspecialty, Obstetrics and Gynecology, Neurological Surgery, Colon and Rectal Surgery, Orthopedic Surgery or subspecialty, General Surgery, and Thoracic Surgery. “Welcoming” specialties included: Family Practice, Internal Medicine, Pediatrics, Preventive Medicine, Psychiatry or subspecialty, and Internal Medicine/Pediatrics. 22 If respondents reported changing specialties between surveys and their MSQ specialty was not considered “welcoming,” but their GQ specialty selection was, they were classified as changing from “nonwelcoming” to “welcoming” specialties. Overall satisfaction with the quality of their medical education was measured as combining strongly agree and agree responses as “Agree” and neutral, disagree, and strongly disagree as “Disagree.”
Covariates and additional measures
Sexual orientation was included as a survey question with three possible responses: Gay or Lesbian, Bisexual, or Heterosexual. We combined any sexual minority identity into “LGB” (Lesbian, Gay, or Bisexual) vs. Heterosexual. Gender identity was originally categorized as Female, Male, or Unknown; those responding “Unknown” were excluded from our sample due to low reporting frequencies.
Race response categories were: Asian, Black or African American, Indigenous (American Indian and Alaska Native), Pacific Islander (Native Hawaiian or Other Pacific Islander), White, Multiple Races Identified, and Other. Ethnicity was reported as Hispanic (Hispanic, Latino, or of Spanish origin) and non-Hispanic. For analysis, a mutually exclusive composite race and ethnicity measure was created by collapsing response levels and retaining four categories: Asian, Black, White, and Hispanic. Each of the first three categories (Asian, Black, and White) is defined as having the indicator for only that race and not having indicators for any other race category.
These three variables were then cross-tabulated to create a composite intersectional position of Race by Sexual Orientation by Gender measure that was used as the primary independent variable. We use the term, “intersectional position” to denote the combination of social identities that situate individuals within intersecting, multilevel, sociostructural systems of power, such as systemic racism, cissexism, and heteronormativity. 23 The 16 categories were as follows: Asian LGB Female, Asian LGB Male, Asian Heterosexual Female, Asian Heterosexual Male; Black LGB Female, Black LGB Male, Black Heterosexual Female, Black Heterosexual Male; White LGB Female, White LGB Male, White Heterosexual Female, White Heterosexual Male; Hispanic LGB Female, Hispanic LGB Male, Hispanic Heterosexual Female, and Hispanic Heterosexual Male.
Age was also used as a covariate in analyses and used the original five-level age category provided by the AAMC. Clerkship satisfaction was measured for all reported clerkships as Good, Poor, and Not Applicable. For required clerkships (i.e., Pediatrics, Surgery, Psychiatry, Obstetrics and Gynecology, and Internal Medicine) responses of “Not Applicable” were excluded. Questions inquiring about the influence that specific factors had on specialty choosing for the student included: work/life balance, competitiveness of specialty, content of specialty, high level of educational debt, future family planning, and salary expectations. For responses, Moderate Influence and Strong Influence were combined, as were Minor Influence and No influence.
Analyses
Analyses were conducted utilizing complete responses for each outcome of interest. For outcomes regarding specialty change, GQ responses were linked to MSQ responses. For satisfaction, only GQ responses were utilized. For each outcome, survey respondents who completed the survey were included in the analyses. We calculated means, standard deviations, medians, and interquartile ranges (25th and 75th percentiles) for all variables included in the primary analyses. We then stratified all variables by the main independent variable (intersectional positions of race by gender by sexual orientation). Spearman correlations were calculated for all independent variables and covariates, and no pair of variables included in subsequent regression models had a correlation >0.40.
For our analysis, we conducted multivariable logistic regressions to assess the association between composite race/gender/orientation and all outcomes (i.e., any specialty change, changing from a less welcoming to more welcoming specialty, and overall satisfaction with medical education). Potential confounders included age, clerkship experiences, and specialty choice factors. We report adjusted odds ratios and 95% confidence intervals for the multivariable logistic regression models. We performed all analyses using SAS 9.4 (Cary, NC).
Results
From 2018 to 2020, there were 49,509 unique respondents (82.7% response rate) to the GQ. Of this sample, 42,584 (86.0%) had at least one outcome of interest including measures of satisfaction with medical education (41,514, 83.9%) and responses to both graduation and matriculation questionnaire items regarding specialty choice (28,702, 58.0%) (Supplementary Fig. S1 in the supplement). Most respondents were younger than 30 years old at graduation (82.8%), White (64.7%), heterosexual (92.7%), and female (50.2%) (Table 1).
Characteristics of Graduating Medical Students by Intersectional Position
Hetero, heterosexual; LGB, lesbian, gay, bisexual; SGM, sexual and gender minority.
Most respondents were satisfied with their medical education (89.6%). Most respondents reported that work/life balance (77.3%), content of their chosen specialty (98.1%), and future family plans (57.4%) have moderate/strong influence on their specialty choice. Most respondents had changed their reported specialty from time of matriculation to graduation (71.5%). A minority of all respondents who changed their specialty reported at matriculation from a less SGM-welcoming specialty to a more welcoming SGM specialty (6.8%).
In fully adjusted multivariable analyses accounting for respondents’ intersectional position, age, clerkship experiences, and specialty choice factors, White LGB female, Asian hetero female and male, Black heterosexual male, Black LGB female, Hispanic heterosexual male, and Hispanic LGB male respondents were less likely to report being satisfied with their medical education compared to their White heterosexual male counterparts (Table 2).
Association of Intersectional Position with Being Satisfied with Medical Education
Model 1: adjusted for age at graduation.
Model 2: adjusted for age at graduation, clerkship experiences, specialty choice factors.
Hetero, heterosexual; LGB, lesbian, gay, bisexual.
In fully adjusted multivariable analyses, compared to their White heterosexual male counterparts, Asian heterosexual female and male, as well as Asian LGB male respondents, were more likely to have changed their specialty choice from matriculation to graduation (Table 3).
Association of Intersectional Position with Any Change in Specialty Choice from Matriculation to Graduation
Model 1: adjusted for age at graduation.
Model 2: adjusted for age at graduation, clerkship experiences, specialty choice factors.
Hetero, heterosexual; LGB, lesbian, gay, bisexual.
In fully adjusted multivariable analyses, compared to their White heterosexual male counterparts, White LGB female, Asian LGB female, Black heterosexual female, Black LGB female, and Hispanic heterosexual male respondents were more likely to have changed their specialty choice from a less SGM-welcoming specialty to a more SGM welcoming specialty (Table 4). Conversely, Asian heterosexual female respondents were less likely to have changed their specialty choice from a less SGM-welcoming specialty to a more SGM-welcoming specialty (Table 4).
Association of Intersectional Position with Changing from a Less SGM Welcoming Specialty to a More Welcoming SGM Specialty from Matriculation to Graduation
Model 1: adjusted for age at graduation.
Model 2: adjusted for age at graduation, clerkship experiences, specialty choice factors.
Hetero, heterosexual; LGB, lesbian, gay, bisexual; SGM, sexual and gender minority.
Discussion
In exploring changes in specialty choice reported by students at matriculation and graduation, we found a series of differences by intersectional position. Notably, sexual minority persons with intersecting marginalized identities by race, ethnicity, and gender are less satisfied with their medical education than White heterosexual male counterparts. Similarly, by utilizing recent research that found specialties medical students perceive to be more welcoming (Psychiatry, Family Medicine, Pediatrics, Preventative Medicine, and Internal Medicine) or less welcoming of their identities (Neurology, Anesthesiology, Orthopedics, Neurosurgery, Thoracic Surgery, Colorectal Surgery, and General Surgery), we found that medical students with multiple marginalized identities, specifically including sexual, racial, and ethnic identities, are more likely to switch specialty choice from matriculation to graduation.22,24 These findings build on prior research noting disparate experiences of medical training among marginalized groups in medical training.16,25,26
In addition to the positive and negative biases toward specialties that have persisted over time, medical students who identify as a sexual minority, particularly if they have multiple marginalized identities, may also experience frank harassment during educational time in the classroom or wards, potentially by a member of these specialties.16,27 Previously published findings do not reveal how often sexual minority medical students experience this harassment, particularly sexual minority persons with multiple marginalized identities including race, ethnicity, and gender, and to what extent the occurrence of these events may affect the choice of specialty.
Worth noting is that the specialties that have been grouped as more welcoming versus less welcoming of sexual minority trainees share commonalities with regard to potential compensation, administrative burdens, perceived prestige, and proportion focused on the provision of ambulatory care, particularly primary care. 28 The factors that push or attract sexual minority trainees toward welcoming specialties face the chance of limiting their earning potential as well as exposing them to further burnout during the course of their career. 29 Improving the experience of sexual minority medical students, particularly those with multiple marginalized identities, will require further assessing and addressing the culture of each specialty.
Beyond specialty and institution culture, further consideration of the state and national political climate regarding SGM persons should be considered. 30 While this study is limited to graduating medical students from 2018 to 2020, the political climate has become decidedly more negative toward SGM persons since 2020 and is expected to worsen following 2025.31–33 As such, state-level policies may affect where and in which specialties medical students choose to pursue training. Future research exploring the effects of state policies on the health care workforce would be illuminating and could potentially reveal how negative policies affecting SGM persons may deprive certain specialties and states of an adequately staffed health care workforce.
Limitations
This study has several noteworthy limitations. First, AAMC GQ and MSG utilized limited categories for sex, gender, and sexual orientation, as well as simplified categories for race and ethnicity; the experience of a Black Caribbean or African National student may not be the same as an American Black student. Future research exploring intersectional positions would benefit from more refined and expanded demographic categories. Furthermore, additional categories of identity and status (including first generation, disability, and veteran) are absent from AAMC data and warrant future research.34–36 Second, this study focuses on sexual minority students; the experiences of transgender (inclusive of nonbinary, genderqueer, and additional persons who are not cisgender) trainees are not captured in our analyses due to lack of data availability. Notably, the experiences of transgender persons in health care training remain understudied. Third, while it is a common practice to combine transgender and sexual minority populations into a single group for statistical convenience, conceptually, these populations experience different structural and interpersonal forces. Similarly, the experiences sexual minority groups (e.g., lesbian, gay, bisexual) face are not uniform and are shaped by different forces (e.g., misogyny). As such our choice to collapse these groups into a single sexual minority sample is an analytic concession made to accommodate statistical modeling. Furthermore, the psychology of specialty selection is complex and only recently garnered attention. The experience of sexual minority people is a recent focus, largely due to the availability of new data 37 but is only one facet that impacts specialty selection and career choice. Additional research is needed to assess the complex interplay of multiple inputs influencing specialty selection. 10 Fourth, the taxonomy of specialty welcoming status is derived from the article that was published in 2016, and while the data in this study are from 2018 to 2020, it is possible that specialty subcultures have evolved in the interims between the 2016 study and data collection, and to present day. However, we note that research from 1996 38 to 2016 22 showed a similar grouping of welcoming and nonwelcoming specialties, with little difference over 20 years. We would expect little difference in these specialty groupings between 2016 and data collection for this study. Lastly, as we utilized surveys from matriculation and graduation, we are not able to assess when medical students change their intended specialty and therefore cannot pinpoint exact experiences or stages in training when students alter their career plans.
Conclusion
In this national sample of medical students, sexual minority students, especially those with multiple marginalized identities, are changing their intended specialties from matriculation to graduation. Biases expressed by physicians and surgeons toward or against SGM people likely affect the diversity within specific medical and surgical specialties.6,10,11,25 Future research would benefit from exploring a number of factors that may influence intended specialty among sexual minority students, especially those with multiple marginalized identities, including board exam performance, clinical clerkship grades, perceived prestige of intended specialty, perception of the need to serve one’s community, and state-level policies targeting SGM persons.
Footnotes
Acknowledgments
The authors would like to thank the medical students who completed the AAMC matriculation and GQs.
Authors’ Contributions
C.G.S. and S.R.C. conceived of the study design, approved study methods, and reviewed study findings; E.L. provided significant input on study methods and interpretation of results; A.L.M. and E.S. addressed data management, performed statistical analyses, and reviewed study findings with additional input by C.G.S. and E.L.; J.H. provided input on conceptualizing data analyses; C.G.S., A.L.M., E.S., E.L., and M.N. prepared the first draft, and all authors were involved in revising the article and approving the final submission.
Disclaimer
This material is based upon data provided by the “AAMC.” The views expressed herein are those of the authors and do not necessarily reflect the position or policy of the AAMC, AHA, DDCF, US DOJ, NIH, NHLBI, or author employers.
Author Disclosure Statement
Dr. Streed was partially supported through the American Heart Association career development grant (AHA 20CDA35320148), National Heart, Lung, and Blood Institute career development grant (NHLBI 1K01HL151902-01A1), Doris Duke Charitable Foundation (Grant #2022061), the Boston University Chobanian and Avedisian School of Medicine Department of Medicine Career Investment Award, and the Boston University Learn More Research Grant. Dr. Streed received consultation fees from EverlyWell, L’Oreal, the Texas Health Institute, the Research Institute for Gender Therapeutics, and the United States Department of Justice unrelated to this work. The remaining authors have no conflicts of interest.
Funding Information
No funding was received for this article.
Abbreviations Used
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
