Gender Disparities in Surgical Resident Operative Performance Evaluations
Yao Liu*, Jasmine Gibson, Kenneth Lynch, Ashlie Haas-Rodriguez, Carla Moreira, Rachel Beard, David Harrington, Thomas J. Miner
Surgery, Brown University, Providence, RI
Objective: Does general surgery resident gender influence their operative performance rating?
Design: Retrospective analysis of de-identified resident operative evaluations
Setting: Single academic general surgery residency program from 2019 to 2022
Participants: All surgical resident operative evaluations at a single institution from 2019 to 2022 were included in the analysis, totaling 626 evaluations across 50 residents and 43 attendings. Resident gender was balanced (50% female/50% male) in the overall sample, though there were fewer female resident evaluations at the PGY-5 level (33% female/67% male). 29% of evaluating attendings were female.
Interventions: For each evaluation, attending surgeons rated resident operative performance after a given surgery on a 5-point Likert scale (1 = worst, 5 = best) in seven categories: positioning/draping, exposure/dissection, instrument handling, tissue handling, time/motion, operative flow, degree of prompting/direction, overall performance. Resident factors (gender, race, PGY level, annual operative volume); attending factors (gender, years in practice); and case difficulty were recorded for each evaluation. Univariate/multivariate linear regressions identified predictors of performance ratings.
Main Outcome Measure: Mean resident operative performance rating across all categories
Results: Female residents received lower performance ratings than male peers in both univariate (p < 0.001) and multivariate analysis (p = 0.037), after controlling for resident, attending, and case-level factors (Table 1). Though non-white resident race was associated with lower ratings in univariate analysis, race was not a significant predictor in multivariate analysis (p = 0.374). Difficult cases (p < 0.001), higher PGY level (p < 0.001), and mid-career attendings (p = 0.015) were associated with higher ratings; resident race (p = 0.065), attending gender (p = 0.580), annual resident operative volume (p = 0.395) were not predictive of ratings. Notably, the gender rating disparity for women was equivalent to the rating increase expected with an increase in PGY level (βgender -0.271 vs. βPGYlevel 0.271).
Conclusions: We find female surgical residents receive lower operative performance ratings than their male counterparts, independent of operative volume. Our findings point towards persistent evaluator biases in academic surgery against female residents and suggest the need for further faculty interventions to address this bias.
Table 1: Univariate/Multivariate Regressions on Predictors of Operative Performance Ratings
| Operative Performance Rating (Likert scale 1-5) | Operative Performance Rating (Likert scale 1-5) | Operative Performance Rating (Likert scale 1-5) | |
| Female resident gender | -0.301*** (0.000) | -0.271* (0.037) | |
| Non-white resident race | -0.244** (0.002) | -0.0650 (0.374) | |
| PGY level | 0.271*** (0.000) | ||
| Annual resident operative volume | -0.00109 (0.292) | ||
| Female attending gender | 0.0515 (0.482) | ||
| Case difficulty (Likert scale 1-5) | 0.0987*** (0.000) | ||
| Mid-career attending (5-10 years in practice) | 0.285* (0.015) | ||
| Senior attending (11+ years in practice) | -0.0172 (0.815) | ||
| Interaction terms between female resident gender + PGY level | N.S. |
p-values in parentheses. N.S. = Not significant
* p < 0.05, ** p < 0.01, *** p < 0.001
Back to 2023 Abstracts