JMIR Publications

ABSTRACT

Background:

Understanding how health-related rumor debunking evolves and spreads on social media is critical for public health communication and policy.

Objective:

This study aims to delineate the thematic evolution, demographic disparities, and engagement drivers of official health rumor debunking in China over a 15-year period.

Methods:

Using Sina Weibo as the empirical setting, we collected rumor-debunking posts published between 2009 and 2024. A two-stage classification pipeline using SetFit was constructed to identify health-related debunking content. We applied BERTopic to map the thematic landscape and utilized a large language model to extract demographic mentions (gender, age, and social roles) and health content referenced in posts. Finally, we employed XGBoost with SHAP explanations to quantify the drivers of engagement and virality.

Results:

Across 377,520 rumor-debunking posts from 24,742 blue-verified accounts, 93,799 posts (24.8%) were health-related. Health debunking volume surged during COVID-19 and remained structurally above pre-pandemic levels through 2024; however, its relative share declined markedly in 2024, suggesting a shift away from pandemic-centered attention. Topic distributions were highly long-tailed; prevalent themes included vaccines, debunking reports, personal care, and sleep hygiene. Temporal trajectories fell into three patterns: event-driven spikes, sustained growth (notably lifestyle-oriented themes such as vision/eye health and nutrition), and recurrent fluctuations. Demographic mentions were uneven: females were referenced more often than males, youth were the most frequently mentioned age group, and older adults were the least mentioned. Engagement was driven primarily by source-level authority signals—especially follower count and verification-related attributes—rather than topic features, with clear non-linear threshold effects for reposts, comments, and likes.

Conclusions:

This 15-year longitudinal analysis shows that China’s health-related debunking on social media shifted from crisis-driven mobilization to more routinized governance after COVID-19. Engagement patterns were driven primarily by source authority rather than thematic content, underscoring the value of leveraging high-credibility actors and platform distribution tools to amplify corrections. Additional offline or cross-platform strategies are likely required to better serve older adults who are less visible in social media discourse.