Radiology Content on TikTok: Current Use of a Novel Video-Based Social Media Platform and Opportunities for Radiology

Purpose: TikTok, the fastest growing social media application worldwide, has been infrequently studied in medicine. We analyzed the top radiology-related posts on TikTok in order to describe opportunities for radiology engagement. Materials and Methods: We retrieved the top 300 posts meeting the search criteria "radiology." User- and post-related data were categorized based on a prespecified coding system. Descriptive statistics were reported. Kruskal-Wallis H and Mann-Whitney U tests were performed to assess for differences in followers, plays, likes, and comments among posts and users. Results: 284 working posts were broadcast by 187 unique users with median 119 followers (interquartile range [IQR]: 31-1,206) and 20 posts (IQR: 7-49). Most (81%, 151/187) were nonphysician radiology personnel, while only 5% (9/187) were radiologists. Posts by radiologists had more plays than those by nonphysician radiology personnel (median 3643 vs 1282, P = 0.001). The 284 posts had median 1520 plays (IQR 429-4374), 60 likes (IQR 18-272), and 2 comments (IQR 0-9). Most posts were work-related (184/284, 65%), followed by clinical (68/284, 24%), personal (30/284, 11%), or promotional (2/284, 1%). However, posts by radiologists were mostly clinical (65%, 31/48) and represented a large majority of posted imaging cases (29/33, 88%). Posts about COVID-19 represented 38% (107/284) of the study sample and 48% (93/193) of posts after the first U.S. COVID-19 case COVID-19 posts had significantly more comments (3 vs. 2, P = 0.034) and more likes approaching significance (89 vs 51, P = 0.134) than non-COVID-19 posts. Conclusions: Though radiologists represent a minority of TikTok users their post represent the majority of this platform's clinical content. This presents an important opportunity for radiologists to utilize TikTok for contemporary, unique content creation and engagement with nonphysician radiology personnel. (C) 2020 Elsevier Inc. All rights reserved.