Professional content and quality assessment of cervical spondylosis prevention on Chinese short video platforms: a cross-sectional comparative study

Read full text Summarise with VIRAT
Authors: Guo Yu, Shuang Lu, Long-Xian Liu, Hao Wang, Qing-Hua Lai, Meng Wu
Cited by: 0
References: 41

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study aimed to evaluate the content characteristics, reliability, and educational quality of cervical spondylosis prevention videos on Douyin and Bilibili and to explore their associations with user engagement. A cross-sectional analysis was conducted from January 27 to 30, 2026, using four Chinese search terms related to cervical spondylosis prevention. The top 50 non-advertisement videos for each keyword on each platform were screened. After applying predefined exclusion criteria, 384 videos were included. Video characteristics, uploader identity, content category, and engagement metrics were recorded. Two trained reviewers independently assessed video reliability using the modified DISCERN scale and educational quality using the Global Quality Score. Inter-rater reliability for mDISCERN and GQS scores was assessed using intraclass correlation coefficients based on a two-way random-effects model with absolute agreement. Group differences and correlations were analyzed using non-parametric tests and Spearman correlation analysis. Douyin videos were significantly shorter than Bilibili videos and had higher numbers of likes and shares, whereas comments and favorites did not differ significantly between platforms. Most videos were uploaded by specialist doctors (37.5%) and non-professionals (32.6%), with practical techniques being the most common content category (53.1%). Overall, the included videos showed moderate reliability and educational quality, with median mDISCERN and GQS scores of 3.00 and 3.00, respectively. Post-hoc Dunn’s tests with Bonferroni correction indicated that specialist doctors, rehabilitation therapists, other professionals, and social organizations scored significantly higher than non-professionals in both mDISCERN and GQS. Spearman correlation analysis showed weak positive correlations between reliability or educational quality scores and several engagement metrics, such as GQS and likes (r = 0.360, P < 0.001), and a weak negative correlation between video duration and GQS (r = − 0.395, P < 0.001). In general, user engagement metrics and video duration should not be interpreted as direct proxies for educational quality or reliability. Inter-rater reliability was good for both scales, with ICC(2,1) values of 0.805 (95% CI: 0.681–0.872) for mDISCERN and 0.853 (95% CI: 0.823–0.879) for GQS, indicating substantial agreement between the two reviewers. Douyin and Bilibili may serve as accessible sources of cervical spondylosis prevention information; however, the reliability and educational quality of the videos were generally moderate and varied by uploader type. User engagement metrics and video duration should not be interpreted as direct indicators of information quality. Greater participation of healthcare professionals, clearer citation of evidence sources, and the use of understandable, safety-oriented language may improve the value of short-form health education content.

Fields of Study & Subjects

Social Media in Health Education Health Literacy and Information Accessibility Health Education and Validation

Publication details

Journal / source
Scientific Reports
Volume
16
Issue
1
Publication date
20 August 2026
Language
EN