Bias Reporting in the Clinical Learning Environment: A National Survey of Internal Medicine Clerkship Directors

Purpose Bias exists in the internal medicine (IM) clinical learning environment; however, it is unclear how often bias is identified by clerkship directors (CDs), how bias is addressed, and whether best practices exist for identifying or mitigating bias. This study investigated how IM CDs receive and respond to bias reports in the clinical learning environment. Method In May 2021, the Clerkship Directors in Internal Medicine (CDIM) created an 18-question survey assessing the frequency of bias reports, macroaggressions and microaggressions, and report outcomes. Of the 152 U.S. medical schools that met study accreditation criteria, the final survey population included 137 CDs (90%) whose medical schools held valid CDIM membership. Results Of the 137 surveys sent, 100 were returned (survey response rate, 73%). Respondents reported a median of 3 bias events (interquartile range, 1-4; range, 0-50) on the IM clerkship in the past year. Among 76 respondents who reported 1 or more event, microaggressions represented 43 of the 75 total events (57%). No mechanism emerged as the most commonly used method for reporting bias. Race/ethnicity (48 of 75 [64%]) and gender (41 of 75 [55%]) were cited most as the basis for bias reports, whereas the most common sources of bias were student interactions with attending physicians (51 of 73 [70%]) and residents (40 of 73 [55%]). Of the 75 respondents, 53 (71%) described the frequency of bias event reports as having increased or remained unchanged during the past year. Only 48 CDs (49%) responded that they were "always"aware of the outcome of bias reports. Conclusions Bias reports remain heterogeneous, are likely underreported, and lack best practice responses. There is a need to systematically capture bias events to work toward a just culture that fosters accountability and to identify bias events through more robust reporting.