Abstract
BACKGROUND: The Accreditation Council for Graduate Medical Education (ACGME) introduced milestones for Emergency Medicine (EM) in 2012. Clinical Competency Committees (CCC) are tasked with assessing residents on milestones and reporting them to the ACGME. Appropriate workflows for CCCs are not well defined. OBJECTIVE: Our objective was to compare different approaches to milestone assessment by a CCC, quantify resource requirements for each and to identify the most efficient workflow. DESIGN: Three distinct processes for rendering milestone assessments were compared: Full milestone assessments (FMA) utilizing all available resident assessment data, Ad-hoc milestone assessments (AMA) created by multiple expert educators using their personal assessment of resident performance, Self-assessments (SMA) completed by residents. FMA were selected as the theoretical gold standard. Intraclass correlation coefficients were used to analyze for agreement between different assessment methods. Kendall's coefficient was used to assess the inter-rater agreement for the AMA. RESULTS: All 13 second-year residents and 7 educational faculty of an urban EM Residency Program participated in the study in 2013. Substantial or better agreement between FMA and AMA was seen for 8 of the 23 total subcompetencies (PC4, PC8, PC9, PC11, MK, PROF2, ICS2, SBP2), and for 1 subcompetency (SBP1) between FMA and SMA. Multiple AMA for individual residents demonstrated substantial or better interobserver agreement in 3 subcompetencies (PC1, PC2, and PROF2). FMA took longer to complete compared to AMA (80.9 vs. 5.3 min, p < 0.001). CONCLUSIONS: Using AMA to evaluate residents on the milestones takes significantly less time than FMA. However, AMA and SMA agree with FMA on only 8 and 1 subcompetencies, respectively. An estimated 23.5 h of faculty time are required each month to fulfill the requirement for semiannual reporting for a residency with 42 trainees.