The quality and reliability of Kawasaki disease-related short videos on TikTok and Bilibili: a cross-platform content analysis using GQS, mDISCERN, and JAMA benchmarks

Abstract Background Kawasaki disease (KD) represents the leading cause of acquired heart disease in children, with timely diagnosis and treatment being crucial to prevent severe coronary complications. The widespread use of short-video platforms such as TikTok and Bilibili has made them increasingly important sources of health information for the public. However, the quality and reliability of KD-related content on these platforms remain largely unassessed. This study aimed to evaluate the educational quality and informational reliability of short videos about KD on TikTok and Bilibili, two of the most popular social media platforms in China. Methods We conducted a cross-sectional analysis of the top 100 videos retrieved for KD on each platform. A total of 186 videos were assessed using the Global Quality Scale (GQS), modified DISCERN (mDISCERN), and JAMA benchmark criteria. Uploaders were categorized as professionals or non-professionals. Video characteristics, engagement metrics, and content coverage were analyzed. Results Overall quality was suboptimal (median scores: GQS = 2, mDISCERN = 2, JAMA = 2). Videos from professionals scored significantly higher than those from non-professionals (p < 0.001). TikTok videos were shorter but had higher engagement and GQS scores than Bilibili videos (p < 0.001). Content coverage was skewed: clinical manifestations were discussed in > 80% of videos, while prognosis and prevention were covered in < 25%. Engagement metrics showed negligible correlation with quality scores. Conclusions KD-related short videos on popular Chinese platforms exhibit notable quality deficiencies and content gaps. Incomplete or low-quality online information may delay caregiver recognition and timely medical evaluation, potentially increasing the risk of coronary complications in children with KD. Although professional involvement is associated with better quality, overall reliability remains limited. These findings highlight the need for improved content governance and tailored strategies for accurate health communication on social media.