Answers are grounded in this paper and cite their sources.
Widely prevalent worldwide and increasingly noted in China, adjustment disorder (AD) is frequently featured on short-video apps; however, because platforms like TikTok, Bilibili, and RedNote lack robust regulation of health content, the potential for inaccurate information is high, motivating our systematic comparison of the quality of AD-related videos across these three platforms. We conducted a cross-sectional content analysis of publicly available Mandarin videos retrieved with the keyword “适应障碍” (adjustment disorder). From the top 100 search results on each platform, 228 unique videos were included (Tiktok 78; Bilibili 62; RedNote 88). Two raters evaluated overall quality (Global Quality Score, GQS 1–5), reliability (modified DISCERN, 0–5), source credibility (JAMA, 0–4), and educational usability (PEMAT-Understandability and- Actionability, 0–100%). Content completeness was coded across six domains (epidemiology, etiology, clinical manifestations, diagnosis, treatment, prevention) to create a Content Completeness Index (CCI, 0–12). Between-group comparisons used non-parametric tests; proportional-odds ordinal logistic regression examined independent factors associated with higher GQS and mDISCERN. Median video length was 123 s (IQR 66–254), with the longest on Bilibili (298 s). Platform differences were observed for overall quality ( P < 0.001), reliability ( P = 0.040), JAMA score ( P < 0.001), and actionability ( P = 0.021): TikTok and Bilibili scored higher on GQS than RedNote; Bilibili had higher reliability; TikTok showed the highest JAMA score and higher actionability than RedNote. Understandability did not differ across platforms. Licensed uploaders (physicians, licensed non-physicians, institutions) outperformed non-licensed accounts (health communicators, patient-caregivers) on GQS ( P < 0.0001), mDISCERN ( P < 0.05), JAMA score ( P < 0.0001), and actionability ( P < 0.01), with no difference in understandability. Quality, reliability, and usability increased stepwise with higher CCI (all P < 0.001). In adjusted models, CCI had the strongest association with higher GQS (OR 2.46 per 1-point increase; 95% CI 2.05–2.96) and higher mDISCERN (OR 1.26; 95% CI 1.08–1.47). Patient–caregiver uploads had lower reliability than institutional accounts (OR 0.09; 95% CI 0.01–0.70). Engagement metrics (likes, collections, comments) correlated strongly with each other but weakly with quality or reliability. AD short videos on Chinese platforms show mixed quality with limited source transparency. Content completeness—rather than platform, popularity, or length—was most strongly associated with higher quality and reliability. Encouraging structured coverage of core domains, improving source disclosure, and promoting collaborations between licensed professionals and skilled communicators could raise the baseline quality of online mental-health education.
No takes yet. Share an insight, caveat, or question.
Jiang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: