How to reverse the alarming trend of health misinformation

Health misinformation took center stage during the Covid-19 pandemic and continues to run rampant online. Falsehoods about evidence-based cancer treatmentsopens in new window, birth controlopens in new window, and other forms of medical care have caused confusion, conflict, and premature death (Gisondi, M. A., et al., Journal of Medical Internet Research, Vol. 24, No. 2, 2022opens in new window). Despite the growing threat to global health, social media companies have backtracked on promises to police misinformation on their platforms ( Big Tech Backslide , Free Press, 2023 opens in new window).

That has left users—especially young people, who frequently encounter antiscience views onlineopens in new window—vulnerable to being deceived. When they try to verify information, they often land on more inaccuracies (Aslett, K., et al., Nature, Vol. 625, 2024opens in new window). Turning to authority figures for clarity can also backfire: In February, Florida’s surgeon general shared advice that contradicted medical knowledge about measlesopens in new window.

“Many laypeople are just not familiar with how knowledge is produced in science and medicine, so these concepts can feel very abstract,” which is why research-backed health communication is so important, said Kurt Sampsel, PhD, the senior program manager for disinformation and community engagement at PEN America, a nonprofit organization focused on protecting free expression.

[Related: Misinformation and disinformation]

But as researchers work to address what some are calling an “infodemic” of health misinformation, they too face roadblocks—often politically motivated. Research groups such as the Stanford Internet Observatoryopens in new window, as well as advocacy organizations such as the Center for Countering Digital Hateopens in new window, are fielding lawsuits, public-records requests, and accusations of collusion with social media platformsopens in new window. Last August, the National Institutes of Health (NIH) suddenly halted its $154 million Advancing Health Communication Science and Practice initiativeopens in new window, which aimed to improve the accuracy of the medical information people receive and share.

“The term ‘misinformation’ has become very politicized, and that is a problem,” said psychologist William Klein, PhD, associate director of behavioral research at the National Cancer Institute (NCI), who was cochair of the now defunct NIH health communication initiative.

Despite the barriers, psychologists and other scientists are forging ahead with efforts to learn more about the threat of health misinformation and to put what’s already known into practice. In November 2023, APA released a consensus statement titled Using Psychological Science to Understand and Fight Health Misinformation that provides evidence-based guidance on the issue. The statement, edited by University of Cambridge psychologist and misinformation researcher Sander van der Linden, PhD, was funded by the Centers for Disease Control and Prevention (CDC).

APA will further tackle the topic on the Main Stage at APA 2024 in Seattle on August 9. Ultimately, a cross-disciplinary push that advances psychological interventions (including strategies such as debunking and prebunking), as well as health care access and broader systems-level changes in how people engage with scientific information, is what’s needed to drive change.

“When people mistrust science and scientists, there will be more of a rumor mill,” where false information spreads rapidly, Klein said. “Trust in science has gone down, but it’s still not low. Reversing that trend could make a big difference.”

Misinformation research took off in 2016 with a focus on false headlines shared on social media. But it’s now clear that organized disinformation campaigns (which deliberately intend to mislead) pose a significantly greater threat than individuals sharing false headlines via their personal social media accounts, said Jon Roozenbeek, PhD, an assistant professor in psychology and security at King’s College London who studies online misinformation.

“What has the most impact is a sustained, deliberate effort to make it look like the truth is irrelevant,” said Roozenbeek, who was one of the contributing authors on the APA consensus statement.

Bots that spread Covid -19 misinformation went dormant when Russia invaded Ukraine opens in new window, then resurfaced soon after with pro-Russia messaging, the British newspaper The Guardian reported in 2022. Those and other similar networks are now poised to spread falsehoods on a range of topics.

“There’s a whole industry that underlies the spread of a lot of this misinformation that is completely agnostic as to what it talks about,” Roozenbeek said.

Health crises such as the recent Mpox outbreakopens in new window are ripe for such exploitation, he added. When a disease breaks out, real uncertainty arises about how it spreads and how people can stay safe, leaving an opening where health misinformation can thrive.

Misleading health content can put people’s lives at extreme risk, such as when people take a toxic alternative cancer treatment or choose to follow a potentially dangerous health trend like drinking borax diluted with wateropens in new window with no supporting science. Covid-19 misinformation proved particularly deadly, with nearly 17,000 deaths estimated from the misguided use of hydroxychloroquine, a drug for the treatment of malaria and rheumatologic disease, to prevent or treat Covid-19 (Pradelle, A., et al., Biomedicine & Pharmacotherapy, Vol. 171, 2024opens in new window). Hydroxychloroquine was given to thousands of patients based largely on a nonrandomized study of 36 patients that gained popularity on YouTube before it underwent peer review (Saag, M. S., JAMA, Vol. 324, No. 21, 2020opens in new window).

“Without an alternative explanation, misinformation is more difficult to correct. It can rush in and fill all the blanks while science is still working out the answers,” said psychologist Briony Swire-Thompson, PhD, an assistant professor of political science and psychology who directs the Psychology of Misinformation Lab at Northeastern University and was one of the contributing authors on the APA consensus statement.

Research indicates that people tend to share and engage more with false or low-quality information they encounter online, compared with factual information (Allen, J., Integrity Institute, October 13, 2022opens in new window; Vosoughi, S., et al., Science, Vol. 359, No. 6380, 2018opens in new window). That’s likely because it’s designed to be engaging and often has an emotional pull. Social media companies have little incentive to police content if it means removing users and advertising dollars from their bottom line.

“A big part of the social media business model is getting eyeballs on the platform, and there’s a large portion of the audience that’s quite hungry for misinformation of various types, including health misinformation,” said Deen Freelon, PhD, a computational social scientist and a professor at the University of Pennsylvania’s Annenberg School for Communication and a contributing author on the APA consensus statement. That’s partly due to motivated reasoning, which makes people more likely to accept a conclusion when it’s a desirable one, despite potential inaccuracies.

Artificial intelligence (AI) plays a role in the production and spread of false information about health and other topics, particularly by foreign governments. Malicious bots have generated millions of false social media posts—but the true scope of the problem is still unclear because it can be tough even for experts to tell what content is generated by AI (Hajli, N., et al., British Journal of Management, Vol. 33, No. 3, 2022opens in new window; Bontridder, N., & Poullet, Y., Data & Policy, Vol. 3, 2021opens in new window).

Once people start accessing and sharing misinformation online, it can be difficult for them to disengage because of the connections they form, Freelon said. Information transfer is a highly social experience that offers a rush of positive feedback—even well-intended people fall into echo chambers where others reinforce their inaccurate information or harmful beliefs.

“To some extent, it becomes a matter of social standing,” Freelon said. “Even if you want to get out, there are social pressures keeping you in.”

Health misinformation differs from other forms of inaccurate content, such as political misinformation, in a few important ways. For one, health misinformation purveyors often stand to benefit financially by selling a product or service, such as TikTok influencers who have no formal medical training but promote alternatives to hormonal birth control to millions of viewers.

[Related: Is it safe to get health advice from influencers?]

Young people face a high risk for believing false health information, partly because of their outsize use of social media. In one survey, 73% of Canadian 18- to 24-year-olds reported following at least one influencer who has shared antiscience views ( Youth Science Survey: Final Report , Canada Foundation for Innovation, 2021 opens in new window).

Trust also plays a central role in health communication, and research shows that formally trained health care professionals and scientists do garner a lot of public trust. According to 2023 data from the Pew Research Center, 77% of U.S. adults trust medical scientists to act in the public’s best interest—the highest rating given to any group or institution in society. By contrast, elected officials were the least trusted, with just 24% of U.S. adults expressing the same degree of confidence (Kennedy, B., & Tyson, A., Americans’ Trust in Scientists, Positive Views of Science Continue to Decline, Pew Research Center, 2023opens in new window).

“People expect politicians to lie, but they expect health professionals to give them the truth,” Swire-Thompson said.

That points to a vulnerability that can be exploited by health care professionals who share or recommend false health information, like the surgeon general of Florida, but also an opportunity: People trust doctors and other health care professionals, Klein said, and these health care professionals can help patients, families, and communities access science-backed and reliable information about their health.

One exception is when past abuses from medical providers, such as the repeated medical exploitation of African Americansopens in new window, place certain populations at risk. Peddlers of health scams can and do leverage justifiable mistrust to their own advantage, Freelon said, stoking fear of the pharmaceutical industry while promising a silver bullet.

“Folks in these populations are disproportionately vulnerable because they have been mistreated by the medical establishment,” he said. “That makes them apt audiences for misinformation related to health.”

Health misinformation also takes hold because not everyone has a firm enough understanding that the scientific process takes time to establish evidence. Efforts to address misinformation should expand science literacy, health literacy, and numeracy on a very large scale, Klein said. If more people had a stronger grasp of the amount of time required to develop a vaccine, for example, it could reduce the prevalence of belief in false information.

“We need to help people understand that when there are changes in the scientific findings, those findings are not problematic but the way science works,” he said.

Research by psychologists and others has already pinned down several strategies that effectively counter health misinformation, including debunking false information by providing a correction and explaining why the information is false, or inoculating people against misleading claims by teaching strong critical-thinking skills. Another effective strategy is leveraging trusted sources such as religious leaders or athletes to communicate with the public about health.

Those insights are being increasingly applied, including by the CDC, as it aims to regain widespread public trust following the Covid-19 pandemic. In 2022, one study found that 54% of U.S. adults said they trust their doctors a great deal, but only 37% said the same about recommendations from the CDC (Pollard, M. S., & Davis, L. M., Rand Health Quarterly, Vol. 9, No. 3, 2022opens in new window).

The CDC’s new approach involves using AI to monitor viral online health misinformation on topics such as vaccines, emergencies, and drug overdoses. Then, the agency aims to debunk or prebunk incorrect content as quickly as possible, offering alternative explanations when available.

In one very public example, when rapper Nicki Minaj tweetedopens in new window that her cousin’s friend became impotent after getting a Covid-19 vaccine, triggering protests outside the CDCopens in new window, the agency responded quickly on social media with a fact-check about the vaccine and fertility. During respiratory virus season, the CDC now releases weekly social media videos where CDC Director Mandy K. Cohen, MD, MPH, addresses the public’s top concerns in an approachable way.

Preliminary findings from several research groups point to other strategies that may help people resist health misinformation. Learning-based interventions seek to arm people with skills or knowledge that make them less likely to believe false information. Swire-Thompson reports testing whether discrediting a source, such as by highlighting a conflict of interest or a track record of inaccuracy, could help correct cancer and vaccine misinformation; her findings have not yet been peer-reviewed ( PsyArXiv Preprints , 2024 opens in new window).

“We found that discrediting a source helped people resist misinformation, but track-record discreditations were less effective when the source was a media outlet, compared to a person,” she said.

Sampsel’s team at PEN America applies some of the same insights at the community level. They fund counter-disinformation efforts in Miami, Dallas–Fort Worth, and Phoenix. They also run media literacy webinars on topics such as how cognitive biases affect the way we process information online, and they provide public guidance on how to have productive conversations with family members, friends, and acquaintances about divisive topics such as public health and immigration.

Other psychological interventions aim to change online behavior—for example, reducing the amount of misinformation people share. Research led by psychologists Gordon Pennycook, PhD, of Cornell University, and David Rand, PhD, of the Massachusetts Institute of Technology, suggests that people share less misinformation when platforms nudge them to consider the accuracy of their posts, but the group’s latest findings have not yet been peer-reviewed (working paper, 2024opens in new window).

Effect sizes are generally small, and some psychologists remain cautious of how applicable such findings will be. “I still haven’t seen work that reliably and over time changes behavior in exactly the intended manner—people sharing less unwanted content, more desired content, or both,” Roozenbeek said.

One intervention, the Spot the Troll Quizopens in new window, developed by psychologist Jeffrey Lees, PhD, of Princeton University, and his colleagues, did reduce participants’ total retweets on all topics for about a week and helped people learn to identify fake accounts ( PNAS Nexus , Vol. 2, No. 4, 2023 opens in new window).

Some research even suggests that correcting health misinformation won’t change health behavior. Psychologist Dolores Albarracín, PhD, a Penn Integrates Knowledge Professor and director of the communication science division at the University of Pennsylvania’s Annenberg Public Policy Center, suggests focusing instead on interventions that do reliably change behavior, such as increasing access to health care.

“Misinformation is very salient, so people think it matters a great deal, but the impact of interventions in the information space is negligible,” said Albarracín, also a contributing author on APA’s consensus statement. In other words, time spent correcting misinformation might be better spent elsewhere, such as increasing health care access.

In a review of meta-analyses on behavior change interventions, including those targeting health behaviors, Albarracín and her colleagues found that interventions that provide general knowledge or skills training—a category that includes many psychological efforts to correct misinformation—had little to no effect on behavior.

On the other hand, approaches that improved access to care (for example, by providing free vaccines or walk-in appointments) had the largest effects on behavior. Interventions that helped people build habits (performing and automating a behavior, such as dental cleanings) or provided social support (such as pairing people up to get vaccinated together) had medium effects ( Nature Reviews Psychology , 2024 opens in new window).

“At this point, we have enough information to say: Yes, correcting misinformation is important, because people are entitled to the truth. But in order to change behavior, do something else,” Albarracín said.

In the relatively new field of misinformation research, many questions remain.

“We know less about how information spreads through social networks, especially online,” Klein said.

One key research priority is to better understand the role of social media influencers, including how they gain influence, how their influence magnifies or runs out, and on which platforms they prove most persuasive. Videos, including those shared on YouTube and TikTok, are also increasingly important to monitor and study because they are by far the most popular social media content by many metrics.

“We know narratives make a difference, but we don’t yet understand all the nuances. How are narrative videos affecting health decision-making and behavior?” Klein said.

Amid their focus on the subtleties of health communication, researchers also acknowledge how influential systems-level change can be. Sampsel points to the way societal stigmas around smoking and sexual harassment led to widespread behavior change over time.

“It’s going to take more than media literacy, platform reform, and government policy,” he said. “This will require social changes, such that sharing misleading, false, or inaccurate information is something that becomes more and more socially condemned.”

While those shifts can take years, change can also happen quickly, Freelon said. For example, if a popular figure like Elon Musk made statements about the value of sharing accurate information, his followers would likely listen. That suggests leveraging the voices of thought leaders and other influential figures could be a key part of shutting down the spread of false information and saving lives.

“People really take cues from folks they trust, whether it’s politicians, celebrities, or business leaders,” Freelon said. “Those kinds of steps can change the social context all at once.”