‘Just a checkbox’: Growth mindset and feedback in resident experiences with EPA assessments

INTRODUCTION: There have been consistent reports of tension between the dual goals of assessment of learning and assessment for learning in competency-based medical education. This is exemplified by assessments of Entrustable Professional Activities where trainees report a focus on assessment of learning and engaging in performance-oriented behaviours that undermine meaningful assessment. It has been proposed that fostering a growth mindset amongst trainees may encourage use of EPA assessments for learning. This study aimed to explore how trainee mindset interacts with their perception of the role of EPAs in assessment of and for learning. METHODS: We conducted 19 semi-structured interviews of residents enrolled in 12 Canadian specialty post-graduate programmes. Interviews explored residents' experiences with EPAs and their views on how tasks are learned. Transcripts were analysed using reflexive thematic analysis to generate themes. RESULTS: Though most participants described attitudes of a growth mindset and a desire to learn from failure, a minority described attitudes associated with a fixed mindset contributing to avoidance of unsuccessful EPA assessments, suggesting mindset may impact the use of EPA assessments for learning. However, even amongst residents expressing views consistent with a growth mindset, the lack of meaningful feedback on EPA assessments prevented them from being a tool for learning. Participants described a variety of system factors that undermined provision of meaningful feedback and a view that programmes utilised EPA assessments only as a checkbox. DISCUSSION: Even if a growth mindset may promote assessment for learning, a system that results in poor-quality feedback turns EPA-based assessments into an assessment of learning. An assessment system that supports both a culture of growth mindset and high quality of feedback is needed if EPA assessments are to be utilised for assessment for learning.