ABSTRACT
INTRODUCTION: The dramatic reduction of clinical and research activities within medical and surgical departments during COVID-19, coupled with the inability of medical students to engage in research, away rotations and academic meetings, have all posed important implications on residency match. METHODS: Using Twitter application programming interface available data, 83,000 program-specific and 28,500 candidate-specific tweets were extracted for the analysis. Applicants to urology residency were identified as matched vs unmatched based on 3-level identification and verification. All elements of microblogging were captured through Anaconda Navigator. The primary endpoint was residency match, assessed as correlation to Twitter analytics (ie retweets, tweets). The final list of matched/unmatched applicants through this process was cross-referenced with internal validation of information obtained from the American Urological Association. RESULTS: A total of 28,500 English language posts from 250 matched and 45 unmatched applicants were included in the analysis. Matched applicants generally showed higher number of followers (median 171 [IQR 88-317.5] vs 83 [42-192], p=0.001), tweet likes (2.57 [1.53-4.52] vs 1.5 [0.35-3.03], p=0.048), and recent and total manuscripts (1 [0-2] vs 0 [0-1], p=0.006); 1 [0-3] vs 0 [0-1], p=0.016) in comparison to the unmatched cohort. On multivariable analysis, after adjusting for location, total number of citations and manuscripts, being a female (OR 4.95), having more followers (OR 1.01), individual tweet likes (OR 1.011) and total number of tweets (OR 1.02) increased overall odds of matching into a urology residency. CONCLUSIONS: Our study of the 2021 urology residency application cycle and use of Twitter highlighted distinct differences among matched and unmatched applicants and their respective Twitter analytics, highlighting a potential professional development opportunity offered by social media in underscoring applicants' profiles.