Prognostic Value of the MULTIPLE MINI-INTERVIEW for Emergency Medicine Residency Performance.

Background: The Multiple Mini-Interview (MMI) uses short, structured contacts, and is known to predict medical school success better than traditional interviews and application materials. Its utility in Emergency Medicine residency selection is untested. Objectives: We investigate whether it provides additional information regarding future first-year resident performance that can be useful in resident selection. Methods: From three Emergency Medicine residency programs, 71 interns in their first month completed an MMI developed to focus on desirable resident characteristics. Application data were reviewed. First-year resident performance assessments covering the American Council for Graduate Medical Education (ACGME) core competencies, along with professionalism and performance concerns, were obtained. Multiple logistic regressions were employed and MMI correlations were compared with program rank lists and typical selection factors. Results: An individual's score on the MMI correlated with overall performance (p < 0.05) in single logistic regression. MMI correlated with ACGME individual competencies patient care and procedural skills at a less robust level (p < 0.1), but not with any other outcomes. Rank list position correlated with the diagnostic skill competency (p < 0.05), but no others. Traditional selection factors correlated with overall performance, disciplinary action, patient care, medical knowledge, and diagnostic skills (p < 0.05). MMI was not correlated significantly with the outcomes when included in multiple ordinal logistic regression with other selection factors. Conclusions: MMI scores correlate with overall performance, but are not statistically significant when other traditional selection factors were considered. The MMI process seems potentially superior to program rank list at correlating with first-year performance. The MMI may provide additional benefit when examined using a larger and more diverse sample. (C) 2015 Elsevier Inc.