Radiology Residency Match: The Cost of Being in the Dark

RATIONALE AND OBJECTIVES: The Electronic Resident Application Service (ERAS) publishes monthly statistics before the match and the National Resident Matching Program publishes the match outcomes. We sought to determine whether early ERAS data influences applicant behavior and correlates with match outcomes. MATERIALS AND METHODS: We searched the 2007-2017 ERAS archives for the applicant pool size (PS), the average number of applications per program (AP), and the average number of applications per applicant (AA) in November, before radiology match, and the 2007-2017 National Resident Matching Program archives for the average number of ranked applicants needed to fill each position (ANRA) and the number of unfilled positions (UP) in radiology match. Correlation coefficients were calculated for each pair. RESULTS: PS correlated very strongly with AP (r = 0.80, p = 0.001708), UP (r = -0.92, p = 0.000063) and ANRA (r = -0.90, p = 0.000164). UP correlated strongly with ANRA (r = 0.76, p = 0.006349) and AP (r = -0.77, p = 0.005339). A trend to moderate correlation between AP and ANRA (r = 0.58, p = 0.062686) and AA (r = 0.53, p = 0.074395) did not reach statistical significance. There was no correlation between AA and PS in the same (r = -0.05, p = 0.878585) or the following year (r = 0.35, p = 0.297166), and AA and UP in the same (r = 0.13, p = 0.701983) or the following year (r = 0.32, p = 0.336136). CONCLUSION: The real-time data reported by ERAS in November, before match, is a predictor of radiology match outcomes and can be used by all participants to limit their application and recruitment costs. Medical students applying to radiology do not consider either the real-time or historic data when submitting ERAS applications