Does Implicit Bias Influence Resident Evaluations? A Single-center Review of Intraoperative Evaluations

Implicit bias influences feedback and evaluation in surgical residency programs, but the effects of gender, affinity, and leniency bias are varied in the literature. We hypothesized that affinity bias may affect assessment ratings, with performance rated superior for those trainees who share traits like those of the assessors regarding gender, training program, and seniority.

Intraoperative evaluations were collected from 06/2021 to 10/2024 using a modified Ottawa Surgical Competency Operating Room Evaluation including information on overall performance rating as below, at, or above level expected per resident year. Demographic information on trainees and faculty included gender, training program, and seniority level. The competence ratings were compared across groups using crosstabulations with a p < 0.05 considered significant.

A single general surgery residency program located primarily at an academic, university hospital.

A total of 858 evaluations were completed by 128 faculty members for 93 residents.

When comparing evaluator to resident gender, there was no difference in the scores between the groups (p = 0.53). However, male faculty had lower rates of rating resident's performance as "above expected for level" compared to female faculty (182/579 (31.4%) vs. 120/279 (43.0%); p < 0.01). There was no difference in ratings based on faculty's residency or fellowship training program. While the rate of evaluations for "below expected for level" was low, only senior faculty gave this rating compared to junior faculty (7/434 (0.5%) vs. 0/424 (0.0%); p < 0.01).

Male evaluators are less likely to rate residents as "above level of expected" and more senior evaluators are more likely to rate residents as "below level of expected." While it is reassuring that there were no overt differences based on resident gender, the findings suggest that leniency bias may play a role in resident feedback. Future studies should evaluate these biases and create systems to improve equity in resident evaluations.