Source-Dependent Quality Variation in Shoulder Dislocation Videos on YouTube

Abstract

Purpose: To assess the quality of YouTube videos for patient education on shoulder dislocation. Methods: A standard YouTube search was performed in March 2023 using the terms “shoulder dislocation,” “dislocated shoulder,” and “glenohumeral joint dislocation” to identify eligible videos. Multiple scoring systems, including DISCERN (a validated tool for analyzing the quality of health information in consumer-targeted videos), Journal of the American Medical Association (JAMA) Benchmark Criteria, and the Global Quality Score (GQS) were used to evaluate the videos. Video quality scores from various sources were compared using the Kruskal-Wallis test for initial analysis, followed by Dunn's post-hoc test with Bonferroni correction, and the strength of relationship between variables was assessed using Spearman's rank correlation coefficient. Results: A total of 162 eligible videos were identified. The mean video duration was 11.38 ± 3.01 minutes, the median number of views was 653. Median number of days since upload was 1,972, the median view rate was 0.343, and median number of likes was 66.12. Based on the DISCERN classification, a substantial proportion of videos were classified as insufficient quality, with 19.4% as “very insufficient” and 42.1% as “insufficient”; 24.1% were classified as “average” quality, whereas only 13.1% were classified as “good” and 1.2% were “excellent.” Videos from academic and professional sources showed a significant positive correlation with DISCERN scores (rho: +0.784, P < .001) and greater scores on all 4 scoring systems compared to health information websites. Conclusions: This study reveals that the majority of YouTube videos on shoulder dislocation lack sufficient quality for patient education, with content quality significantly influenced by the source. Clinical Relevance: Examining the accuracy of information that patients encounter on YouTube is essential for health care providers to direct individuals toward more reliable sources of information. © 2024 The Authors