Can routine EPA-based assessments predict OSCE performances of undergraduate medical students?

Background: Objective structured clinical examination (OSCE) is used worldwide. This study aims to explore potential alternatives to the OSCE by using entrustable professional activities (EPA)-based assessments in the workplace. Methods: This study enrolled 265 six-year undergraduate medical students (UGY) from 2021 to 2023. During their rotations, students were assessed using 13 EPAs, with the grading methods modified to facilitate application. Before graduation, they participated in two mock OSCEs and a National OSCE. We used generalized estimating equations to analyze the associations between the EPA assessments and the OSCE scores, adjusting for age and sex, and developed a prediction model. EPA8 and EPA9, which represent advanced abilities that were not significant in the regression models, were removed from the prediction model. Results: Most EPAs were significantly correlated with OSCE scores across the three cohorts. The prediction model for forecasting passing in the three OSCEs demonstrated fair predictive capacity (area under curve = 0.82, 0.66, and 0.71 for students graduated in 2021, 2022, and 2023, respectively all p < 0.05). Conclusions: The workplace-based assessments (EPA) showed a high correlation with competency-based assessments in simulated settings (OSCE). EPAs may serve as alternative tools to formal OSCE for medical students.