Quality, content, and engagement of pulmonary tuberculosis related videos on Douyin: a cross-sectional study

Pulmonary tuberculosis (PTB) remains a major public health concern worldwide despite being preventable and curable. China remains one of the 30 high tuberculosis (TB) burden countries, and short-video platforms such as Douyin (the Chinese mainland version of TikTok) have become important sources of health information, yet the quality of PTB-related content remains unclear. This study aimed to evaluate the content completeness, reliability, quality, and engagement of PTB-related videos on Douyin.

We conducted a cross-sectional content analysis on October 23, 2025. Using the hashtag "#pulmonary tuberculosis" under the default "comprehensive ranking", we initially retrieved 220 videos. Videos ranked 201-220 were reserved as a priori for pilot reviewer calibration, and the remaining top 200 videos underwent formal screening, of which 191 met the inclusion criteria. We extracted video characteristics, uploader type, and engagement metrics. A PTB-specific checklist covering six dimensions (definition, symptoms, risk factors, evaluation, management, outcomes) was scored, and reliability/quality were assessed using Journal of the American Medical Association (JAMA) benchmarks, Global Quality Score (GQS), modified DISCERN instrument (mDISCERN), and Medical Quality Video Evaluation Tool (MQ-VET). Spearman correlation and multivariable linear regression were performed.

Of 191 videos, 150 (78.5%) were posted by health professionals. Most were uploaded in 2024-2025 (86.4%); median duration was 74 s [interquartile range (IQR) 48-110]. Content completeness was limited (median 3.0, IQR 2.0-4.5), with risk factors and outcomes rarely addressed; social-context coverage was also limited. Health professionals and science communicators achieved higher content scores than general users and news agencies (P<0.001). Overall quality was modest (median MQ-VET 46, IQR 42-48), lowest among general users and highest among science communicators (P<0.001). Engagement indicators were strongly intercorrelated but weakly or negatively associated with quality, and high-quality videos were not concentrated in top-ranking positions. In regression models, higher comment counts were associated with lower quality, whereas collections were associated with higher quality.

PTB-related videos on Douyin showed substantial heterogeneity in content completeness and overall quality. Higher-quality videos were not consistently prioritized by platform ranking or user engagement, highlighting the need to promote trustworthy and comprehensive PTB information.