Values-Based Resident Selection in an Internal Medicine-Pediatrics Residency Program

Background There is no consensus regarding values important for medical resident success, and current methods for selecting residents correlate poorly with success in residency. Objective We developed and validated a set of values demonstrated by exemplary residents in the Internal Medicine-Pediatrics program at the University of Utah and used them to inform our resident selection process. Design We utilized a modified Delphi method to identify and internally validate values of successful residents. We implemented these values into the interview evaluation rubric. Participants Four members of the Internal Medicine-Pediatrics residency program leadership and eleven current residents aided in value generation. Nine faculty from leadership positions in the residency programs of Internal Medicine-Pediatrics, Internal Medicine, and Pediatrics formed a local expert panel for validation. Approach We performed a literature review and engaged local stakeholders in a semi-structured group interview to generate 107 values. After consolidation based on redundancy, two iterative cycles of expert review using a modified Delphi approach, and alignment with the Accreditation Council for Graduate Medical Education core competencies, eleven values achieved expert agreement and were integrated into an interview rubric to aid in resident selection. Key Results We identified eleven values important for resident success: academic strength, intellectual curiosity, compassion, communication, work ethic, teamwork, leadership, self-awareness, DEI (diversity, equity, and inclusion), professionalism, and adaptability. The rank list from 2021 was found to correlate with a score based on values, but not Step 2 score, as it did in 2017. Conclusions We applied a modified Delphi method to generate eleven observable values present in the ideal Internal Medicine-Pediatric resident at one academic health center in the Intermountain West. Higher Step 2 scores no longer correlated with higher ranking when we used these values to inform our rank list.