IntroductionEfforts have been made to increase the number of women and physicians who are underrepresented in medicine (UIM). However, surgery has been slow to diversify, and there are limited data surrounding the impact of intersectionality.ObjectiveTo assess the combined association of race and ethnicity and sex with rates of promotion and attrition among US academic medical department of surgery faculty.Design, Setting, and ParticipantsThis was a retrospective cohort study using faculty roster data from the Association of American Medical Colleges. All full-time academic department of surgery faculty with an appointment any time from January 1, 2005, to December 31, 2020, were included. Study data were analyzed from September 2022 to February 2023.ExposuresFull-time academic faculty in a department of surgery with a documented self-reported race, ethnicity, and sex within the designated categories of the faculty roster of Association of American Medical Colleges.Main Outcomes and MeasuresTrends in race and ethnicity and sex, rates of promotion, and rates attrition from 2010 to 2020 were assessed with Kaplan-Meier and Cox time-to-event analyses.ResultsA total of 31 045 faculty members (23 092 male [74%]; 7953 female [26%]) from 138 institutions were included. The mean (SD) program percentage of UIM male faculty increased from 8.4% (5.5%) in 2010 to 8.5% (6.2%) in 2020 (P < .001), whereas UIM female faculty members increased from 2.3% (2.6%) to 3.3% (2.5%) over the 10-year period (P < .001). The mean program percentage of non-UIM females increased at every rank (percentage point increase per year from 2010 to 2020 in instructor: 1.1; 95% CI, 0.73-1.5; assistant professor: 1.1; 95% CI, 0.93-1.3; associate professor: 0.55; 95% CI, 0.49-0.61; professor: 0.50; 95% CI, 0.41-0.60; all P < .001). There was no change in the mean program percentage of UIM female instructors or full professors. The mean (SD) percentage of UIM female assistant and associate professors increased from 3.0% (4.1%) to 5.0% (4.0%) and 1.6% (3.2%) to 2.2% (3.4%), respectively (P =.002). There was no change in the mean program percentage of UIM male instructors, associate, or full professors. Compared with non-Hispanic White males, Hispanic females were 32% less likely to be promoted within 10 years (hazard ratio [HR], 0.68; 95% CI, 0.54-0.86; P <.001), non-Hispanic White females were 25% less likely (HR, 0.75; 95% CI, 0.71-0.78; P <.001), Hispanic males were 15% less likely (HR, 0.85; 95% CI, 0.76-0.96; P =.007), and Asian females were 12% less likely (HR, 0.88; 95% CI, 0.80-0.96; P =.03). Non-UIM males had the shortest median (IQR) time to promotion, whereas non-UIM females had the longest (6.9 [6.8-7.0] years vs 7.2 [7.0-7.6] years, respectively; P < .001). After 10 years, 79% of non-UIM males (13 202 of 16 299), 71% of non-UIM females (3784 of 5330), 68% of UIM males (1738 of 2538), and 63% of UIM females (625 of 999) remained on the faculty. UIM females had a higher risk of attrition compared with non-UIM females (HR, 1.3; 95% CI, 1.1-1.5; P = .001) and UIM males (HR, 1.2; 95% CI, 1.0-1.4; P = .05). The mean (SE) time to attrition was shortest for UIM females and longest for non-UIM males (8.2 [0.14] years vs 9.0 [0.02] years, respectively; P < .001).Conclusion and RelevanceResults of this cohort study suggest that intersectionality was associated with promotion and attrition, with UIM females least likely to be promoted and at highest risk for attrition. Further efforts to understand these vulnerabilities are essential.