The Role of Implicit Bias in Surgical Resident Evaluations

Background: Implicit bias is a key factor preventing the advancement and retention of women and underrepresented minorities in academic surgery. Purpose: We examined the role of implicit bias in the technical component of the residency performance evaluation. The Fundamentals of Laparoscopic Surgery (FLS) score, an objective measure of technical performance, was compared to the subjective technical skills (TS) score given by attending surgeons. Procedures: FLS scores and the average TS scores from chief resident evaluations at a university program were analyzed from 2015 to 2019 (n = 29 residents; female 22%, underrepresented minorities 27%). The average TS score for each resident was calculated, scores dichotomized above and below the mean for the program and analyzed across gender and racial identity. Main Findings: There were no significant differences in FLS or TS scores between male and female trainees or racial identity. The Kappa correlation coefficient between the 2 dichotomized scores was significantly lower for female (-0.50) versus male (0.23) trainees (p < 0.01); it was not significantly different between racial groups (p = 0.34). Principal Conclusions: There was statistically significant difference in agreement between the FLS and TS scores of individual female and male trainees, suggesting the presence of implicit bias in our pilot study. Further research with a larger sample size is warranted. Objective: To investigate the presence of implicit bias against women and underrepresented minorities in the technical component of the residency performance evaluation. We hypothesized that women and underrepresented racial minorities would have lower subjective technical skills (TS) scores as compared to their objective FLS scores, relative to the mean for the training program. Design: FLS scores and the average TS scores from chief resident performance evaluations were analyzed from 2015-2019. Both FLS and the average TS scores were dichotomized above and below the mean for the program and analyzed across gender and racial identity. Research was approved by institutional IRB. Setting: This study was conducted at the University of Arizona General Surgery Residency Program at Banner University Medical Center in Tucson, Arizona. This is a tertiary care university training program. Participants: Educational records of graduated general surgery chief residents from 2015 to 2019 were accessed for the study. We analyzed 37 TS scores from attending performance evaluations and 29 FLS scores reported to the program during the study period (22% female, 27% underrepresented racial minorities). Results: There were no significant differences in FLS or TS scores between male and female trainees or racial identity. The Kappa correlation coefficient between the 2 dichotomized scores was significantly lower for female (-0.50) versus male (0.23) trainees (p < 0.01); it was not significantly different between racial groups (p = 0.34). Conclusions: There was a statistically significant difference in agreement between the FLS and TS score of individual female and male trainees, suggesting the presence of implicit bias in this pilot study. Further research with a larger sample size is warranted.