RESUMO
OBJECTIVE: The neurosurgical match is a challenging process for applicants and programs alike. Programs must narrow a wide field of applicants to interview and then determine how to rank them after limited interaction. To streamline this, programs commonly screen applicants using United States Medical Licensing Examination (USMLE) Step scores. However, this approach removes nuance from a consequential decision and exacerbates existing biases. The primary objective of this study was to demonstrate the feasibility of effecting minor modifications to the residency application process, as the authors have done at their institution, specifically by reducing the prominence of USMLE board scores and Alpha Omega Alpha (AΩA) status, both of which have been identified as bearing racial biases. METHODS: At the authors' institution, residents and attendings holistically reviewed applications with intentional redundancy so that every file was reviewed by two individuals. Reviewers were blinded to applicants' photographs and test scores. On interview day, the applicant was evaluated for their strength in three domains: knowledge, commitment to neurosurgery, and integrity. For rank discussions, applicants were reviewed in the order of their domain scores, and USMLE scores were unblinded. A regression analysis of the authors' rank list was made by regressing the rank list by AΩA status, Step 1 score, Step 2 score, subinternship, and total interview score. RESULTS: No variables had a significant effect on the rank list except total interview score, for which a single-point increase corresponded to a 15-position increase in rank list when holding all other variables constant (p < 0.05). CONCLUSIONS: The goal of this holistic review and domain-based interview process is to mitigate bias by shifting the focus to selected core qualities in lieu of traditional metrics. Since implementation, the authors' final rank lists have closely reflected the total interview score but were not significantly affected by board scores or AΩA status. This system allows for the removal of known sources of bias early in the process, with the aim of reducing potential downstream effects and ultimately promoting a final list that is more reflective of stated values.
