Evaluating the quality and reliability of Kawasaki disease–related content on TikTok and Bilibili: a cross-sectional study

TL;DR

This study examines the quality and reliability of Kawasaki disease content on TikTok and Bilibili, finding that most videos are suboptimal and often contain misinformation.

Contribution

The study is the first to evaluate the quality and reliability of Kawasaki disease content on TikTok and Bilibili using standardized medical criteria.

Findings

TikTok videos showed higher quality and popularity compared to Bilibili, but overall quality on both platforms was suboptimal.

Five major misinformation themes were identified, including symptom oversimplification and incorrect claims about treatments.

Pediatrician-uploaded videos had higher quality and engagement compared to those from individual users.

Abstract

As public interest in health and immunology grows, short video platforms have become an increasingly important source of medical information. Kawasaki disease, a pediatric immune-mediated vasculitis with potential cardiovascular complications, has attracted substantial attention; however, the accuracy and quality of related content on these platforms remain unexamined. This study aimed to evaluate the overall quality of Kawasaki disease–related videos on TikTok and Bilibili.

On February 25, 2025, newly registered accounts were used to search the term “川崎病” (Kawasaki disease) on TikTok and Bilibili, and the top 100 videos from each platform were collected. Video quality was evaluated using the JAMA benchmark criteria, a modified DISCERN, and PEMAT, while user engagement metrics (likes, comments, saves, and shares) were analyzed for correlations.

A total of 146 videos were included. Although TikTok videos demonstrated higher quality and popularity than those on Bilibili, overall video quality on both platforms remained suboptimal. Median JAMA scores were 2.00 and 1.00, modified DISCERN scores were both 3.00, intelligibility was 70% versus 64%, and operability was 67% on both platforms. Most videos were monologue-based and symptom-focused, with pediatricians and individual users as the main uploaders. Pediatricians and individual users were the two largest groups of content creators. Pediatrician-uploaded videos showed higher quality and engagement, whereas individual-user videos were more often misleading and less interactive. Five major misinformation themes were identified, including symptom oversimplification, incorrect etiological claims, promotion of non–evidence-based home treatments, misunderstanding of diagnostic criteria, and misleading statements about immunoglobulin therapy. Video quality was positively correlated with popularity, while longer duration was negatively associated with both quality and engagement. Heterogeneity was observed across platforms.

The quality and reliability of Kawasaki disease–related videos on short video platforms remain suboptimal, highlighting the need to address misinformation, refine evaluation tools, and promote high-quality content creation.

Figures