Gender Differences in Clinical Performance Assessment of Internal Medicine Residents: A Longitudinal Analysis of the Influence of Faculty and Trainee Gender.

Purpose: Competency-based medical education relies on equitable assessment. This study examined the influence of faculty and trainee gender on assessments of internal medicine (IM) resident performance over time.

Method: A longitudinal analysis of clinical performance assessments from 7 U.S. IM residency programs (July 2014-June 2019) was conducted. Core competency scores (patient care [PC], medical knowledge [MK], practice-based learning and improvement [PBLI], systems-based practice [SBP], professionalism [PROF], and interpersonal and communication skills [ICS]) were standardized across programs. Cross-classified mixed-effects linear regression evaluated the relationship between gender and standardized competency scores within training programs, while adjusting for multiple variables including IM In Training Examination percentile rank.

Results: Data included 9,346 evaluations by 1,011 faculty (552 [55%] men, 459 [45%] women) for 664 residents (358 [54%] men, 306 [46%] women). Initially, women residents' scores were significantly lower than men's in PC (estimated difference [standard error ] -0.097 [0.033], P = .004), MK (-0.145 [0.034], P < .001), and PBLI -0.090 [0.040], P = .022). PC, MK, PBLI, and SBP scores increased more over time for women residents than men (PC: 0.050 [0.015], P = .001; MK: 0.052 [0.015], P = .001; PBLI: 0.036 [0.018], P = .048; SBP: 0.036 [0.016], P = .027). PROF and ICS scores were comparable across gender. There was a significant interaction between faculty gender and postgraduate year (PGY) across all competencies but none between resident gender, faculty gender, and PGY, indicating that men and women faculty rated residents differently over time but were consistent in how they rated men and women residents.

Conclusions: Gender-based assessment differences were variable across competencies and time. Women residents had lower scores initially but greater gains in "hard skill" (MK, PC, and PBLI) than in "soft skill" (ICS and PROF) competencies, suggesting assessment inequities. Efforts to ensure equitable assessment are needed.