Platform, creator, and engagement patterns in BPH health information on chinese short-video platforms: A cross-sectional study of TikTok and Bilibili

Patients increasingly encounter benign prostatic hyperplasia (BPH) information on short-video platforms, but the practical adequacy of this content for evaluation and treatment decisions may differ across platform and creator ecosystems. We compared TikTok and Bilibili to characterize structural quality, clinically relevant content coverage, and quality-engagement relationships. We conducted a cross-sectional search on March 12, 2026, retrieving the top 100 videos from each platform with the Chinese keyword for BPH. After excluding unrelated and duplicate content, 171 videos were analyzed (TikTok n = 98; Bilibili n = 73). Two urologists independently assessed structural reliability with modified DISCERN (mDISCERN), educational quality with the Global Quality Score (GQS), and content coverage with a study-specific Completeness Score (CS) spanning Definition, Symptoms, Risk factors, Evaluation, Management, and Outcomes. Inter-rater agreement was excellent for all scored domains (Cohen’s κ > 0.80). Creator composition differed substantially between platforms: professional individuals accounted for 96.9% of TikTok videos, whereas professional and non-professional institutions accounted for 32.8% of Bilibili videos (p < 0.001). TikTok videos had higher median engagement but shorter duration (62 vs. 116 s, p < 0.001). Platform CS and GQS distributions were similar, while TikTok had a slightly higher median mDISCERN score (3 vs. 2, p = 0.002). Institutional videos were longer and had higher CS and GQS values, although institutional subgroups were small. Risk factors and Evaluation were the least covered clinical domains. On TikTok, view count remained weakly associated with lower CS after adjustment for duration and uploader type (partial rs = -0.21, p = 0.032); no corresponding association was observed on Bilibili. BPH videos on both platforms provided incomplete clinical coverage, and engagement was not a dependable proxy for educational value. The underrepresentation of risk factors and evaluation, together with the greater completeness of longer institutional videos, identifies practical targets for clinician- and institution-led digital education in China. Because the data are cross-sectional, the observed associations do not establish algorithmic effects or patient outcomes.