A Systematic Review Of COVID-19 Misinformation Interventions: Lessons Learned.

Governments, public health authorities, and social media platforms have employed various measures to counter misinformation that emerged during the COVID-19 pandemic. The effectiveness of those misinformation interventions is poorly understood. We analyzed fifty papers published between January 1, 2020, and February 24, 2023, to understand which interventions, if any, were helpful in mitigating COVID-19 misinformation. We found evidence supporting accuracy prompts, debunks, media literacy tips, warning labels, and overlays in mitigating either the spread of or belief in COVID-19 misinformation. However, by mapping the different characteristics of each study, we found levels of variation that weaken the current evidence base. For example, only 18 percent of studies included public health-related measures, such as intent to vaccinate, and the misinformation that interventions were tested against ranged considerably from conspiracy theories (vaccines include microchips) to unproven claims (gargling with saltwater prevents COVID-19). To more clearly discern the impact of various interventions and make evidence actionable for public health, the field urgently needs to include more public health experts in intervention design and to develop a health misinformation typology; agreed-upon outcome measures; and more global, more longitudinal, more video-based, and more platform-diverse studies.

How Americans encounter guns: Mixed methods content analysis of YouTube and internet search data.

Firearm-related injury and death is a serious public health issue in the U.S. As more Americans consume news and media online, there is growing interest in using these channels to prevent firearm-related harms. Understanding the firearm-related narratives to which consumers are exposed is foundational to this work. This research used the browsing behavior of a representative sample of American adults to identify seven firearm-related content "ecosystems" (defined as naturally occurring networks of channels watched by the same users) on YouTube; we then described the demographics and internet search patterns of users affiliated with each ecosystem. Over the 9-month study period, 72,205 panelists had 16,803,075 person-video encounters with 7,274,093 videos. Among these, 282,419 were related to firearms. Using fast greedy clustering, we partitioned users and channel interactions into seven distinct channel-based content ecosystems that reached more than 1/1000 YouTube users per day. These ecosystems were diverse in reach, users, and content (e.g., guns for self-protection, guns for fun). On average, 0.5% of panelists performed a firearm-related internet search on a given day. The vast majority of searches were related to mass shootings or police-involved shootings (e.g., "active shooter"), and virtually none were about more common firearm harm such as suicide. Searches for firearm safety information were most common among panelists affiliated with the "Hunting & Fishing" and "Guns & Gear" ecosystems, which were watched primarily by older, white men. These findings identify an opportunity for analyzing firearm-related narratives and tailoring firearm safety messaging for users affiliated with specific online content ecosystems.