INTRODUCTION
The United States Medical Licensing Examination (USMLE) will stop reporting a 3-digit numerical score and begin reporting a pass/fail outcome at the end of January 2022.1,2 The proposal for this change is to eliminate the major emphasis placed by program directors (PDs) on USMLE performance as a tool for residency screening.2,3 This change was made to allow medical schools to focus on educating physicians rather than teaching to standardized examinations, allowing PDs to use selection factors that truly represent potential in a given specialty.4 The purpose of this study was to assess the opinions of Dermatology residency program directors on USMLE examinations' current ability to predict clinical performance and specialty board examination performance, the relative importance of USMLE Step 1 in current selection of residents, and how the change of USMLE Step 1 to pass/fail will change the relative importance of other factors in residency selection.
MATERIALS AND METHODS
Medical students and physicians at our institution conducted a study to examine the predictive outcomes of the United States Medical Licensing Exam (USMLE) Step 1 transitioning to pass/fail under IRB-300007220 "Predictive Outcomes of USMLE Step 1 Transitioning to Pass/Fail". The study was conducted using an anonymous Qualtrics XM® online survey that was distributed via email and through the APD Program Director's listserv for a total of 6 weeks. These respondents were asked to rank 16 different factors, including Step 1 score, in terms of their importance for assessing applicants both before and after the change to a pass/fail system. The factors were ranked on a scale of 1 to 17 (1 being the most important) before the change and 1 to 16 after the change, with Step 1 score being removed as a factor to consider. The factors included in the ranking were: Step 1 score, publications and presentations, research experience, volunteer experience, membership in Alpha Omega Alpha, membership in the Gold Humanism Honor Society, the Dean's letter, personal statement, clerkship grades, preclinical grades, class rank, away rotations, letters of recommendation, graduate degree, Step 2 score, leadership experience, and attendance at a top 40 National Institutes of Health-funded medical school. Custom programming scripts in R and MATLAB were used to analyze the responses, and statistical tests were conducted to determine the significance of the results. Responses were analyzed using nonoverlapping 99.9% confidence intervals, which were used to determine significance for multiple choice questions, and independent t tests were used to analyze ranking questions with significance at P<.001.
