Fact-checking COVID-19 health misinformation on social media

Theoretical Background and research questions/hypothesis: Politics, commercial interests, rumors, and selective news reporting contributed to an ongoing COVID-19 “infodemic.” Fact-checking health misinformation involves filtering, analyzing, correcting, and transforming public knowledge, but fact-check organizations struggle to keep up with a rapidly evolving pandemic. Unchecked COVID-19 misinformation sidelined and suppressed science, politicized public health and sparked confusion, societal disruptions, and deadly health consequences. Three research questions guided the analysis: RQ1) Which social media platforms were flagged the most often for sharing misinformation about COVID-19 prevention and treatment? RQ2) Which sources were frequently attributed in flagged posts? RQ3) What COVID-19 remedies and treatments were recommended in flagged posts?

Methods: This content analysis examined 236 fact-check articles that evaluated false prevention and treatment health advice shared online during the infodemic. All fact-check articles that evaluated posts about COVID-19 treatment or prevention were coded. A set of 236 fact-check articles were analyzed from seven fact-check websites that ran between 3/13/20, the day COVID-19 was declared a national emergency, until 2/25/22, the last day CDC recommended masking.

Results: On a 6-point scale, the mean score of 180 fact-checked posts was 5.1 (mostly false). Facebook shared a disproportionately high number of flagged posts, compared with the other three social media platforms. The fewest number of flagged posts appeared on Instagram. Authoritative source types – including doctors, scientists, CDC, WHO, etc. – frequently were cited in flagged posts. Viral misinformation posts often promoted inexpensive and readily available remedies such as water, bleach, lemon juice, salt, baking soda, orange peels, etc. Some posts also tried to appeal to partisan beliefs by citing Trump, other politicians, and TV/radio personalities almost five times as frequently as the top official Covid source (Fauci).

Conclusions: Many posts attributed false information to credible sources, to improve their credibility and virality – an imposter tactic. Although many posts provided bad advice, none overtly tried to sell bogus cures. Across the four social media platforms was a continuum of COVID-19 misinformation, ranging from fake news, conspiracy theories, tentative medical conclusions, propaganda, rumors, and inaccurate content falsely attributed to credible sources.

Implications for research and/or practice: Fact-check sites need more tools to flag and counter health misinformation with facts more quickly. This misinformation includes myths, fake news, and other fabricated content, as well as clickbait, misleading framing, false context, imposter content, deep fakes, and other manipulated images. Most flagging apps are based on Twitter API databases, since most tweets are publicly accessible. However, user-friendly flagging apps may be needed for Facebook since that platform has been spreading far more viral health misinformation. Automated searches across social media platforms for mentions of relevant authoritative sources frequently used in pandemic misinformation posts might help fact-checkers more efficiently identify problematic posts before they can go viral. Since COVID-19 posts have often recommend the use of common household items for treatment or prevention, perhaps health agencies could develop campaigns that debunk myths about dangerous home remedies – especially since many people are substituting them for vaccinations.