Reliability and quality of colorectal cancer-related health information videos on social media platforms: a cross-sectional study

Abstract

Social media video platforms have become major sources of colorectal cancer (CRC) information, yet the reliability and educational value of CRC-related videos remain uncertain across platforms. This study aimed to compare the reliability, overall quality, and educational usability of CRC-related videos across Douyin, YouTube, and Bilibili. We conducted a cross-sectional study comparing CRC-related videos on Douyin, YouTube, and Bilibili, which differ in content format: Douyin and Bilibili are the dominant Chinese-language short-form and long-form video platforms, respectively, whereas YouTube is the largest global platform. Videos were searched on March 4, 2025. On YouTube, the English keyword “colorectal cancer” was used; on Douyin and Bilibili, the Chinese keyword “结直肠癌” was used. The first 100 eligible videos from each platform were included, yielding a total sample of 300 videos. Video characteristics, uploader type, presentation style, duration, and engagement indicators were extracted. Reliability and quality were assessed using the modified DISCERN (mDISCERN) instrument and the Global Quality Score (GQS), and educational usability was assessed using the Patient Education Materials Assessment Tool for Audiovisual Materials (PEMAT). Multivariable regression analyses were performed to examine whether platform differences were independent of uploader grouping and video duration. Uploader composition and presentation style differed substantially across platforms. In unadjusted comparisons, YouTube showed higher reliability, overall quality, and understandability, whereas Bilibili showed higher actionability. Douyin showed high engagement, but this was not accompanied by consistently higher reliability or educational quality. Across platforms, symptoms, diagnosis, and treatment were more frequently addressed than epidemiology, etiology, and prevention. Longer video duration was consistently associated with higher mDISCERN and GQS scores across platforms and with higher PEMAT scores on YouTube and Bilibili. In multivariable regression analyses adjusting for uploader grouping and video duration, the YouTube advantage remained significant for mDISCERN and PEMAT-Understandability but not for GQS or PEMAT-Actionability. By contrast, Bilibili was associated with significantly lower adjusted GQS, mDISCERN, and PEMAT-Understandability scores than Douyin, whereas its apparent unadjusted advantage in PEMAT-Actionability was no longer statistically significant after adjustment. CRC-related videos on social media vary markedly in quality and educational usability. Apparent platform-level differences are partly explained by uploader composition and video duration, and engagement alone is a poor proxy for educational value. After adjustment, YouTube retained an advantage in reliability and understandability, whereas Bilibili’s favorable unadjusted profile did not persist. Cross-platform evaluations of online health information should therefore account for both uploader structure and content format when interpreting differences in quality.