• Michael Van Meter, Michael Williams, Rosa Banuelos, Peter Carlson, Jeffrey I Schneider, Bradley D Shy, Christine Babcock, Matthew Spencer, and Yashwant Chathampally.
• University of Texas Health Science Center McGovern Medical School, Houston, Texas.
• J Emerg Med. 2017 Jan 1; 52 (1): 77-82.e1.

BackgroundEmergency medicine (EM) residency programs use nonstandardized criteria to create applicant rank lists. One implicit assumption is that predictive associations exist between an applicant's rank and their future performance as a resident. To date, these associations have not been sufficiently demonstrated.ObjectivesWe hypothesized that a strong positive correlation exists between the National Resident Match Program (NRMP) match-list applicant rank, the United States Medical Licensing Examination (USMLE) Step 1 and In-Training Examination (ITE) scores, and the graduating resident rank.MethodsA total of 286 residents from five EM programs over a 5-year period were studied. The applicant rank (AR) was derived from the applicant's relative rank list position on each programs' submitted NRMP rank list. The graduation rank (GR) was determined by a faculty consensus committee. GR was then correlated to AR using a Spearman's partial rank correlation. Additional correlations were sought with a ranking of the USMLE Step Score (UR) and the ITE Score (IR).ResultsCombining data for all five programs, weak positive correlations existed between GR and AR, UR, and IR. The majority of correlations ranged between. When comparing GR and AR, there was a weak correlation of 0.13 (p = 0.03).ConclusionOur study found only weak correlations between GR and AR, UR, and IR, suggesting that those variables may not be strong predictors of resident performance. This has important implications for EM programs considering the resources devoted to applicant evaluation and ranking.Copyright © 2016 Elsevier Inc. All rights reserved.

22 keywords...

hide…