Clerkship grading outcomes and the role of curricular and assessment structures: a multi-institutional study

Single institution studies have shown differences in clerkship grades by gender and race/ethnicity, which may harm students. Less is known about how modifiable curricular structures influence these differences. Using multi-institutional data, this study examined the prevalence of differences in third-year clerkship grading outcomes and whether Longitudinal Integrated Clerkships (LIC) and grading committees mitigate these differences.

The authors conducted a retrospective cohort analysis of data from five US medical schools, 2016-2023. The primary outcome was receipt of the highest grade (i.e. Honors) vs. other grades. A series of multilevel logistic regressions modeling the effects of gender, race/ethnicity, and socioeconomic status (SES) controlling for student medical knowledge and school were conducted. Augmented models evaluated curricular factors (Traditional block rotation (TBR) vs LIC and use of a grading committee) as moderators of grading differences.

The analysis included 17,513 grades from 3,012 third-year medical students. Grading differences related to gender, race/ethnicity, and SES were present and persisted after controlling for medical knowledge and curricular factors. Although students were as likely to receive Honors in LICs (40%) as TBRs (42%; Adjusted OR: 0.98, p = 0.22) overall, both under-represented in medicine (URM) (Adjusted OR: 1.73, p < 0.01) and Non-URM minority (Adjusted OR: 1.77, p < 0.01) students were more likely to receive Honors in LICs compared with TBRs. Honors rates were similar with and without grading committees (Adjusted OR: 0.98, p = 0.87). However, in clerkships without grading committees, men were less likely than women to receive Honors, whereas this difference was attenuated when grading committees were used (Adjusted OR: 1.29, p = 0.02).

Grading differences related to gender, race/ethnicity, and SES persist across institutions even after controlling for medical knowledge and curricular factors. Certain curricular structures partially mitigated some differences. Grading committees mitigated gender differences, while LICs mitigated racial/ethnic differences, suggesting structural reforms may reduce differences in assessment.