Variation in Research Experiences and Publications During Medical School by Sex and Race and Ethnicity

Importance Diverse research teams are critical to solving complex health problems and producing high-quality medical research. Objective To examine the associations of student sex and racial and ethnic identity with publication rates during medical school. Design, Setting, and Participants This cohort study assessed individual-level data of US MD graduates from medical school who matriculated in academic years 2014 to 2015 and 2015 to 2016. Data were obtained from the Association of American Medical Colleges and analyzed from October 2021 to January 2022. Main Outcomes and Measures Outcomes of interest included students’ self-reported participation in unique research experiences, number of publications, and computed publications per research experience. Poisson regressions were constructed to determine the association of sex and racial and ethnic identity with research outcomes using adjusted rate ratios (aRRs). Results Among 31 474 graduates, 15 159 (48.2%) identified as women and 4344 (13.8%) identified as underrepresented in medicine by race and ethnicity (URIM; including American Indian, Alaska Native, Black, Hawaiian Native, Hispanic/Latinx, and Pacific Islander individuals). Students who attended National Institutes of Health (NIH) top 40 research-ranked schools reported higher number of research experiences and publication counts, resulting in a higher publication rate compared with students from non–top 40 schools (median IQR 1.60 1.00-3.00 vs 1.25 0.50-2.33; P lt; .001). Women reported a higher number of research experiences than men but a significantly lower number of publications (top 40 schools: aRR, 0.89; 95% CI, 0.87-0.90; non–top 40 schools: aRR, 0.93; 95% CI, 0.92-0.95). This resulted in a significantly lower publication rate among women (top 40 schools: aRR, 0.85; 95% CI, 0.83-0.86; non–top 40 schools: aRR, 0.91; 95% CI, 0.90-0.92). Compared with White students, Asian students had higher publication rates at both NIH top 40 schools (aRR, 1.10; 95% CI, 1.08-1.12) and non–top 40 schools (aRR, 1.07; 95% CI, 1.05-1.08), while lower publication rates were reported among Black students (top 40 schools: aRR, 0.83; 95% CI, 0.80-0.86; non–top 40 schools: aRR, 0.88; 95% CI, 0.85-0.95) and Hispanic students attending non–top 40 schools (aRR, 0.93; 95% CI, 0.90-0.95). Conclusions and Relevance These findings illustrate that inequities in the physician-scientist workforce began early in training and highlight key areas for intervention, such as funding support and mentorship training during undergraduate medical education, that may promote the future success of a diverse physician-scientist workforce.