Catching a Big Fish with a Small Net: Factors Associated with First-Choice Match from Urology Residency Applicants' Self-Reported Data.

Abstract

We sought to determine the accuracy of self-reported urology applicant match data and determine which factors were most influential on successful application outcomes. A publicly accessible Google spreadsheet entitled "Urology Residency Applicant Spreadsheet" containing self-reported urology residency applicant characteristics and match outcomes was analyzed for differences across the years 2017+2018 (pre-aggregated)-2020. These data were compared to published data from the American Urological Association and the Association of American Medical Colleges. Statistical modeling of the self-reported data was performed to determine which applicant characteristics were predictive of match outcomes. Averages of self-reported data were similar to published match data with a bias towards more competitive applicants. The factors associated with increased interview offer rate were: Step 1 score, Step 2 score, number of research items, class quartile, and Alpha Omega Alpha membership. Logistic regression modeling correctly predicted an applicant matching to their first-choice program with 74.7% accuracy, with significant negative predictors being the number of programs to which the applicant applied and interviews offered from waitlist or cancellations, and positive predictors being the number of interview offers received. Many applicants "apply broadly" with the goal of improving their match outcomes, but we found that applying to more programs is associated with a decreased likelihood of the applicant matching to their first-choice program. Applicant characteristics such as United States Medical LicensingExamination® scores were not related to first-choice match, suggesting that program selection, among other factors, is vital in the match process.