Contributions
- Comprehensive systematic review of 57 peer-reviewed empirical studies on health misinformation spread via social media (database search from PubMed, Cochrane, Web of Science, Scopus, Google; 651 records screened)
- Taxonomy of health misinformation topics: vaccines (n=30, largest), communicable diseases (Ebola, Zika, influenza), chronic non-communicable diseases, diet/nutrition, and miscellaneous health topics
- Analysis of methodological approaches: content analysis (most common), social network analysis, sentiment analysis, psychological experiments, and rumor-spread modeling
- Identification of theoretical frameworks: psychology (epistemic belief, inoculation theory, conspiracy ideation) and network science (homophily, cascade models, community detection) as dominant explanatory lenses
- Evidence that health misinformation is abundant online and often more popular than factual messages; agents creating misinformation typically lack official institutional affiliation
Method
The authors conducted a systematic literature review following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. They searched PubMed, Cochrane, Web of Science, Scopus, and Google for records published between January 2012 and November 2018 using search terms combining misinformation/fake news keywords with online social media platforms and health/disease keywords. After title and abstract screening to exclude off-topic records (n=651 initially identified, 402 excluded), 131 full-text articles were assessed for eligibility. Of these, 74 were further excluded due to lack of misinformation as a main component of analysis, leaving 57 studies for qualitative synthesis. The analysis examined health issues, geographic focus, theoretical frameworks, study design, and empirical methodologies employed across included studies.
Results
Landscape of health-related misinformation research: The number of published studies on health misinformation grew substantially from 2012 (n=7) to 2018 (n=41, through November), indicating rising scholarly attention. Vaccine-related misinformation dominated (n=30 studies), far exceeding work on other health topics. The top topics were:
- Vaccines and communicable diseases (n=30): vaccination safety, anti-vaccine movements, HPV vaccine hesitancy, Ebola, Zika, influenza
- Chronic non-communicable diseases (n=3): cancer screening, gynaecological cancer, diet and nutrition
- Miscellaneous (n=9): diet, abortion, dialysis, multiple sclerosis, smoking, water quality
Methodologies: Content analysis was employed by most studies (n=38), analyzing text, images, Twitter or Facebook posts, YouTube videos, or websites to identify and classify health claims. Social network analysis was used to trace propagation (n=12), while psychological experiments (n=8), sentiment analysis (n=6), and various modeling approaches captured spread dynamics.
Theoretical frameworks: Psychology (n=9) and network science (n=7) were the most common disciplinary lenses. Psychological frameworks focused on individual cognition: conspiracy ideation, epistemic belief in truth, confirmation bias, and rumor theory. Network science employed homophily, cascade models, and clustering algorithms to characterize diffusion patterns and online communities.
Geographic and platform variation: Studies were geographically dispersed; vaccines were studied across multiple countries, while other health misinformation research concentrated in specific regions. Twitter was the most-studied platform, followed by Facebook and YouTube; emerging work examined WeChat, Sina Weibo, and Instagram.
Key findings: Health misinformation spreads due to multiple drivers—distrust of institutions, perceived credibility of peer networks, emotional appeal of health scares, and algorithmic amplification of sensationalized content. Belief in misinformation correlates with conspiracy ideation, low institutional trust, and epistemically-oriented thinking. Studies reveal that false information often achieves greater online reach and engagement than factual corrections.
Notes
This review is the foundational systematic map of health misinformation research as of 2019. Its main strength is comprehensive coverage of the health-domain landscape and careful taxonomy of research methods and theoretical approaches. A key insight is that psychology and network science provide complementary lenses: individual-level belief formation and collective diffusion patterns respectively. The concentration of work on vaccines reflects both real-world public health significance (vaccination campaigns and anti-vaccine movements) and the relative early adoption of computational methods in vaccine-discourse research.
Limitations acknowledged by the authors include publication language (English only; potentially missing relevant non-English work), difficulty in ascribing motives to misinformation agents, and challenges in comparing across studies using disparate methodologies and platforms. The review identifies critical gaps: a marked absence of large-scale longitudinal tracking of individual behavioral outcomes, limited work on interventions beyond fact-checking, and sparse examination of susceptible populations and demographic asymmetries in misinformation belief. Future work should address the causal impact of health misinformation on population health outcomes and test tailored, multidisciplinary interventions.