Entrustable professional activities for cardiovascular medicine residents in China: a cross-sectional study

Abstract Background The entrustable professional activity (EPA) framework for resident physicians in cardiovascular medicine rotation in China is still in its exploratory stage. No quantitative studies have explored gaps in residents’ self- and supervisor-rated perceived competence across training cohorts and gender subgroups. Methods This cross-sectional study was conducted at Xiangya Hospital of Central South University. From January 2024 to November 2025, 156 internal medicine resident physicians conducting rotations in cardiovascular medicine were enrolled and grouped by postgraduate year (PGY1, PGY2 and PGY3). Twenty-two trained supervisor physicians were selected as evaluators. A 15-item EPA questionnaire was used to obtain the scores of the residents’ self-assessment and the supervisor assessment. Results This cross-sectional study revealed notable intercohort differences in supervisor assessment scores across the PGY cohorts. In contrast, a significant difference in residents’ self-assessment scores across PGY cohorts was observed only for basic clinical tasks, whereas self-rated performance for high-complexity competencies did not differ significantly between the PGY2 and PGY3 cohorts. Radar chart analysis revealed that the self-assessment scores for all 15 EPAs in the PGY1 cohort were significantly higher than those obtained in the supervisor assessment, and this gap narrowed in the PGY2 cohort. In the PGY3 cohort, supervisor assessments had significantly higher scores across multiple EPAs than self-assessments did. In the PGY1 and PGY2 cohorts, the self-assessment scores did not differ significantly between male and female residents. However, in the PGY3 cohort, the self-assessment scores for specific diagnostic and technical EPAs were higher for male residents than for female physicians. Conclusion The study findings suggest that integrated self- and supervisor-rated EPA-based assessment tools can be used to distinguish among groups of cardiology residents at distinct training stages. Moreover, exploratory subgroup analyses revealed differences between the self-assessments of male and female senior residents. These findings support the potential utility of multi-source EPA-based assessment in evaluating perceived competence and informing individualized feedback during residency training. Owing to the limitations of the cross-sectional design, the score disparities between different cohorts in this study cannot be interpreted as longitudinal improvements in individual residents. Longitudinal, multicenter investigations are needed to verify the reliablity, responsiveness, and educational value of this EPA-based tool in assessing the perceived competence of cardiovascular medicine residents.