Abstract
The COVID-19 pandemic and subsequent health crises, such as the Mpox outbreak, underscored the volatile intersection of digital communication and public health. This study evaluates the evolution and impact of social media platform policies on health misinformation between 2019 and 2023. Utilizing a longitudinal mixed-methods approach, we analyzed policy frameworks across major platforms—including Meta, X (formerly Twitter), and TikTok—against public engagement metrics and health information-seeking behaviors. Results indicate that while early pandemic responses relied heavily on reactive content removal, policies shifted toward algorithmic demotion, proactive labeling, and the promotion of authoritative sources from 2021 onwards. Our analysis of data from 2020–2023 reveals a significant correlation between the implementation of 'authoritative source nudges' and a decrease in the velocity of misinformation spread. However, the study also highlights the emergence of 'misinformation seeking' behaviors, where users actively bypass moderation through coded language or decentralized networks. Logistic regression analysis demonstrates that policies were most effective when combined with multi-lingual community-based communication strategies, particularly for culturally and linguistically diverse populations. We conclude that while platform governance improved significantly during the 2019–2023 period, the persistence of vaccine hesitancy and the rise of generative AI-driven disinformation pose ongoing challenges for global health security in 2024. Future pandemic preparedness must integrate platform accountability with behavioral interventions to sustain public trust in institutional health messaging.