Integrating self-assessment into feedback for emergency medicine residents

In 2013 the Accreditation Council for Graduate Medical Education (ACGME) introduced "Milestones" designed to nationally standardize the assessment of resident physicians. Previous studies compare resident self-assessment on milestones to faculty assessment, with varying degrees of agreement, but integration of self-assessment into the formative feedback process has not yet been directly studied. This study uses a conceptual framework of self-determination theory, integrated with concepts from adult learning theory, to compare the perception of the feedback quality given in semiannual reviews before and after the incorporation of resident self-assessment into the feedback process.

This was an interventional study conducted in a single emergency medicine residency program at a major academic hospital over 1 calendar year. Residents first engaged in a semiannual review without self-assessment. At subsequent semiannual reviews, residents completed a Milestone-based self-assessment that was provided to the faculty member assigned to conduct their semiannual review. Residents and faculty completed surveys rating perception of feedback quality. Two-sided Wilcoxon signed-rank tests were used in comparison analysis.

One resident did not self-assess prior to the semiannual review and was excluded leaving 25 paired surveys for analysis. Residents found feedback after the self-assessment more actionable (p = 0.013), insightful (p = 0.010), and better overall (p = 0.025). Similarly, faculty felt the feedback they provided was more actionable (p < 0.001), more insightful (p < 0.001), and better communicated (p < 0.001); led to improved resident understanding of milestones (p < 0.001); and were overall more satisfied (p < 0.001). Free-text comments explore pre- and postintervention perceptions of feedback.

Integration of self-assessment into semiannual reviews improves perception of feedback given to residents as perceived by both residents and faculty. Although limited by sample size, the results are promising for a simple, evidence-based intervention to improve feedback during an existing mandated feedback opportunity.