Differences in Pediatric Fellow Assessments Using the Entrustable Professional Activity Framework

This study analyzed whether assessments of pediatric subspecialty fellows, based on the Entrustable Professional Activity (EPA) framework, showed partiality related to gender, race, or ethnicity. Specifically, the researchers examined supervision levels assigned by Clinical Competency Committees (CCCs) for seven common subspecialty EPAs. The EPA framework measures a trainee’s ability to perform core professional tasks with varying levels of supervision, ranging from observation-only to independent execution.

Biases and partiality in trainee assessments can have lasting effects on career trajectories, creating gaps in professional development and opportunities. Previous studies have identified differences in assessment outcomes based on gender and race or ethnicity, raising concerns about fairness in medical education. This research aimed to determine whether EPA-based assessments, which are being increasingly adopted in medical training programs, are subject to such differences. Understanding this is crucial, especially as EPA-based evaluations are set to play a central role in certification decisions for pediatric subspecialty fellows by 2028.

The researchers conducted a secondary analysis of two linked deidentified datasets from pediatric subspecialty fellows across 14 subspecialties. Data included supervision levels assigned by CCCs to fellows from 2018 to 2022 and demographic information (gender, race, and ethnicity) collected during a prior well-being study. Fellows were categorized as underrepresented in medicine (UIM), Asian, or white based on self-reported race and ethnicity. Statistical models were used to analyze potential differences in EPA supervision ratings while accounting for factors such as year of training, subspecialty, and institution.

Among the 561 fellows analyzed, no significant differences in supervision levels were found based on gender. However, Asian fellows were more likely to receive lower supervision levels for the Consultation EPA, which involves activities such as providing expert advice to other medical professionals. This was the only significant difference identified across the seven common subspecialty EPAs evaluated. No differences were observed for fellows categorized as UIM compared to white fellows.

The findings suggest that EPA-based assessments by CCCs are largely free from bias related to gender, race, or ethnicity. This is reassuring for the medical education community, as EPAs are becoming integral to certification processes. However, the observed difference for Asian fellows in the Consultation EPA highlights the need for further investigation into the potential role of unconscious bias, particularly in interpersonal and communication skills, which are central to this EPA.

The authors recommend further research with larger and other types of cohorts to confirm these findings and to explore potential differences in subspecialty-specific EPAs. They also suggest ongoing efforts to refine structured group-based assessment processes to minimize the impact of bias and partiality in medical education.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the American Board of Pediatrics or the American Board of Pediatrics Foundation.