The claim
“If it's all over my feed, there must be something to it.”
The short answer
How often you see a claim tells you how well it spreads, not whether it's right. False stories travel faster and further than true ones, a small number of accounts produce a large share of what circulates, and simply seeing a claim again makes it feel more believable.
Seeing a health claim everywhere doesn't make it true. It may mean the opposite.
On Twitter, false news was 70% more likely to be retweeted than the truth. Here's why health misinformation spreads so well, why corrections struggle to catch up, and a 60-second habit that helps you check a claim yourself.
Why "everyone's talking about it" feels like evidence
Using popularity as a shortcut isn't foolish. Most of the time it works. If every restaurant review says the same thing, the reviews are probably onto something. If all your friends are suddenly talking about a new symptom or a new supplement, it's reasonable to think something real is going on.
And some of what spreads online is real. People share honest experiences with illness, warn each other about genuine side effects, and sometimes notice problems before official bodies say anything.
The trouble is that a social media feed isn't a random sample of opinion. It's a sample of what spreads. And the research on what spreads is uncomfortable: being false, surprising and emotional helps a claim travel. Being accurate doesn't help much at all.
False news travels faster
The biggest study of this was published in Science in 2018. Researchers at MIT looked at about 126,000 stories that had been checked by six independent fact-checking organisations and then spread on Twitter between 2006 and 2017. Those stories were tweeted more than 4.5 million times by about 3 million people.
The results were lopsided:
- False stories were 70% more likely to be retweeted than true ones.
- True stories took about six times as long as false ones to reach 1,500 people.
- Falsehood spread further than the truth in every category they studied, from politics to science to urban legends.
It wasn't bots. When the researchers removed automated accounts, the gap between true and false stayed. People did the spreading.
Their best explanation was novelty. False claims tended to be newer and more surprising, and replies to them showed more surprise and disgust. A claim that has to match reality is limited by reality. A claim that doesn't can be built to be shocking.
That study wasn't only about health, and politics was its largest category. But the mechanism it found, that surprise beats accuracy, is exactly what health misinformation tends to have going for it.
How much of what you see is wrong?
There's no single answer, because it depends on the topic, the platform and how researchers define "wrong." But when researchers have looked at the most popular health content, the results are sobering.
- YouTube, COVID-19, 2020. Researchers analysed the most-viewed English-language YouTube videos about the coronavirus in March 2020. Of 69 videos, 27.5% contained non-factual information, and those videos had been viewed about 62 million times. Videos from governments and health professionals were all factual, but they made up only 11% of the videos and 10% of the views.
- TikTok, ADHD, 2022. Psychiatrists reviewed the 100 most popular TikTok videos about ADHD. They classified 52 as misleading, 27 as personal experience and only 21 as useful. Most of the misleading videos came from people who weren't healthcare providers.
- Across platforms. A 2021 systematic review pulled together 69 studies of health misinformation on social media. The share varied hugely by topic and study. Smoking and drug topics were the worst, with misinformation reaching 87% of posts in some studies, and vaccine misinformation was also common, at 43%.
These are snapshots of the most popular content, not of everything on a platform. The ADHD authors pointed out that they couldn't tell whether less popular videos were any better. But the most popular content is what most people actually see.
Repetition feels like truth
Here's the part that catches all of us, including doctors.
Psychologists have known since a 1977 study that people rate statements as more likely to be true when they've seen them before. It's called the illusory truth effect. A familiar statement is easier to process, and our brains mistake "easy to process" for "true."
Two findings make this more worrying than it sounds:
- Knowing better doesn't fully protect you. In a 2015 study, repetition made false statements seem more true even when participants demonstrably knew the right answer.
- One exposure is enough. In a 2018 study using real fake-news headlines formatted as Facebook posts, a single prior exposure made people rate a headline as more accurate, both in the same session and a week later. It happened even when the headline carried a "disputed by fact checkers" label.
The effect has limits. In that same 2018 study, repetition didn't make absurd claims like "the earth is a perfect square" more believable. But a claim only needs to be slightly plausible for repetition to work on it.
Now think about a feed. The same claim arrives as a video, then a screenshot of the video, then a friend's comment about it, then a "doctor reacts" clip. Every one of those is another repetition. By the fifth, it doesn't feel like a claim any more. It feels like something you already knew.
A few accounts, a lot of reach
Part of the reason a claim seems to be everywhere is that a small number of very active accounts can make it so.
The clearest evidence comes from politics. A 2019 study in Science linked Twitter accounts to registered US voters during the 2016 election. Just 1% of people accounted for 80% of exposures to fake-news sources, and 0.1% accounted for nearly 80% of the shares.
For health, the most-quoted figure comes from the Center for Countering Digital Hate (CCDH), an advocacy group. Its March 2021 report, The Disinformation Dozen, analysed anti-vaccine content shared or posted about 812,000 times on Facebook and Twitter over six weeks. It concluded that up to 65% of that content could be traced to 12 prominent anti-vaccine figures.
That figure was disputed, and the dispute is worth knowing. In August 2021, Meta's vice president of content policy, Monika Bickert, wrote that there "isn't any evidence" for the claim, that the report "analyzed only a narrow set of 483 pieces of content over six weeks from only 30 groups," and that the 12 people were responsible for "about just 0.05% of all views of vaccine-related content on Facebook." Meta also said it had removed more than three dozen pages, groups and accounts linked to them.
The two numbers aren't measuring the same thing. CCDH's figure is a share of anti-vaccine content in its sample. Meta's is a share of all views of vaccine-related posts, true and false. Neither side's figure settles the question, and you should treat both with some caution. What the stronger evidence from other research does show is that online misinformation tends to be concentrated: a small number of people produce and share a large share of it.
That matters for you because it means "I keep seeing this" can really mean "a few people keep posting this, and the people I follow keep passing it on."
The feed rewards emotion, not accuracy
Platforms don't set out to promote false health claims. But most of them rank posts largely by engagement: what people click, like, comment on and share. And what people engage with isn't the same as what's true.
- A 2017 study of more than 560,000 posts about polarising issues found that each additional moral-emotional word in a post increased its spread by about 20%.
- A 2021 experiment published in Nature found that whether a headline was true had little effect on whether people said they'd share it, even though it strongly affected whether they thought it was accurate. Most people said accuracy mattered to them. In the moment of sharing, they just weren't thinking about it.
- A 2025 audit of Twitter's ranking algorithm found that, compared with a simple chronological feed, the engagement-based feed amplified emotionally charged, hostile content, and that users didn't actually prefer the political posts it chose for them.
Put those together and you get a system where the content that provokes the strongest reaction gets shown to more people, whether or not it's right. Health claims built on fear ("what they're not telling you about…") or outrage are well suited to it.
Why corrections lag behind
If false claims spread quickly, why don't corrections catch up?
Partly it's speed and novelty. A correction is, by nature, less surprising than the claim it corrects. "This turns out to be wrong" rarely goes viral.
Partly it's memory. A major 2012 review in Psychological Science in the Public Interest described the continued influence effect: after people hear a correction, they often keep relying on the original misinformation anyway. Retractions are surprisingly weak at updating what people believe, and the review used the false link between vaccines and autism as a leading example of a myth that persisted and did harm.
The good news from the same research is that corrections aren't useless. They work better when they explain why the claim is wrong and offer something to replace it, rather than simply saying "false."
When misinformation does real harm
It's tempting to think of this as a debate about opinions. It isn't. There are well-documented cases where misinformation was followed by serious illness and death.
- Measles in Minnesota, 2017. Before 2008, more than 90% of Minnesota-born Somali-American 2-year-olds in Hennepin County had their first MMR dose. By 2014 it was 35.6%. The CDC's report on the outbreak attributed the fall to concerns about autism, a perception that autism was rising in the community, and "the misunderstanding that autism was related to the measles-mumps-rubella (MMR) vaccine." In spring 2017, measles spread through the community: 65 confirmed cases by the end of May, 95% of them unvaccinated, and 20 people hospitalised. The CDC called it a demonstration of "the challenge of combating misinformation about MMR vaccine." Studies have consistently found no link between vaccines and autism.
- Methanol in Iran, 2020. Early in the pandemic, false claims spread that drinking alcohol could prevent or treat COVID-19. Iran saw a large outbreak of methanol poisoning, which Iranian toxicologists linked to those claims, with bootleggers selling toxic industrial alcohol as drinkable. Ministry of Health records showed 5,876 hospitalisations between late February and early May 2020, and Iran's Legal Medicine Organization recorded 800 deaths.
- Vaccine intentions. In a randomised trial in the UK and US in September 2020, showing people recent COVID-19 vaccine misinformation lowered intent to vaccinate by about 6 percentage points among those who had said they would definitely take a vaccine, compared with people shown factual information. Misinformation that sounded scientific was the most strongly linked to that drop.
Measles is a useful warning because it's so contagious. The US reported 285 measles cases in 2024, then 2,289 in 2025. As of 12 November 2026, the CDC has counted 5,412 cases this year. Many things affect vaccination rates, so I'm not claiming online misinformation alone explains that rise. But Minnesota showed how quickly a community's protection can fall once a false belief takes hold.
How to check a claim in 60 seconds
You don't need a medical degree to filter out a lot of bad information. You need a couple of habits, and these have evidence behind them.
1. Read sideways, not down. Stanford researchers compared how PhD historians, university students and professional fact-checkers judged unfamiliar websites. Historians and students tended to stay on the page and judge it by how it looked: logos, layout, an official-sounding name. Fact-checkers did something different. They left the page almost immediately and opened new tabs to find out who was behind it and what other sources said about it. They reached better conclusions in a fraction of the time. This is called lateral reading, and it can be taught: in a later field study in high school government classes, fewer than six hours of lessons significantly improved students' ability to judge online sources.
In practice, when you see a health claim:
- Search the person or organisation, not just the claim. Are they a clinician in the relevant field? Do they sell a product related to the claim?
- Search the claim itself and see what established sources say: a medical society, a major hospital, a public health agency, or a fact-checking site.
- If a study is mentioned, look for it. Does it exist? Was it in people or in mice? Does it say what the post says it says?
2. Pause before you share. Ask yourself one question: "Is this accurate?" In the 2021 Nature experiments, simply nudging people to think about accuracy improved the quality of the news they went on to share, including in a field experiment on Twitter.
3. Learn the tricks, not just the facts. A 2022 study tested short videos that taught people common manipulation techniques: emotionally loaded language, false either/or choices, scapegoating and personal attacks. Across experiments, including one with more than 22,000 people on YouTube, viewers got better at spotting these techniques and at telling trustworthy content from untrustworthy content. If a health post is trying very hard to make you afraid or angry, that's a reason to slow down.
4. Notice how often you've seen it. If a claim feels true mainly because it's familiar, that feeling is the illusory truth effect at work. Familiarity is not evidence.
As an emergency physician, I'd much rather a patient walk in and say "I saw this online, is it true?" than act on it quietly. Asking your own doctor is a perfectly good form of lateral reading.
The bottom line
Seeing a health claim everywhere tells you that it spreads well, not that it's true. False stories travel faster than accurate ones, a small number of accounts can produce a large share of what circulates, engagement-driven feeds reward emotion over accuracy, and every repeat makes a claim feel a little more believable, even to people who know better.
That doesn't mean everything popular is wrong, or that everything official is right. It means popularity is the wrong test. Before you believe or share a health claim, spend a minute finding out who's behind it and what independent sources say. It's the same thing professional fact-checkers do, and it works.
This is general education, not personal medical advice.
Sources
- Vosoughi S, Roy D, Aral S. The spread of true and false news online. Science. 2018;359(6380):1146–1151.
- Dizikes P. Study: On Twitter, false news travels faster than true stories. MIT News, 8 March 2018.
- Li HO, Bailey A, Huynh D, Chan J. YouTube as a source of information on COVID-19: a pandemic of misinformation? BMJ Glob Health. 2020;5(5):e002604.
- Yeung A, Ng E, Abi-Jaoude E. TikTok and Attention-Deficit/Hyperactivity Disorder: A Cross-Sectional Study of Social Media Content Quality. Can J Psychiatry. 2022;67(12):899–906.
- Suarez-Lledo V, Alvarez-Galvez J. Prevalence of Health Misinformation on Social Media: Systematic Review. J Med Internet Res. 2021;23(1):e17187.
- Hasher L, Goldstein D, Toppino T. Frequency and the conference of referential validity. J Verbal Learning Verbal Behav. 1977;16(1):107–112.
- Fazio LK, Brashier NM, Payne BK, Marsh EJ. Knowledge does not protect against illusory truth. J Exp Psychol Gen. 2015;144(5):993–1002.
- Pennycook G, Cannon TD, Rand DG. Prior exposure increases perceived accuracy of fake news. J Exp Psychol Gen. 2018;147(12):1865–1880.
- Grinberg N, Joseph K, Friedland L, Swire-Thompson B, Lazer D. Fake news on Twitter during the 2016 U.S. presidential election. Science. 2019;363(6425):374–378.
- Center for Countering Digital Hate. The Disinformation Dozen. 24 March 2021.
- Bickert M. How We're Taking Action Against Vaccine Misinformation Superspreaders. Meta, 18 August 2021.
- Brady WJ, Wills JA, Jost JT, Tucker JA, Van Bavel JJ. Emotion shapes the diffusion of moralized content in social networks. Proc Natl Acad Sci U S A. 2017;114(28):7313–7318.
- Milli S, Carroll M, Wang Y, Pandey S, Zhao S, Dragan AD. Engagement, user satisfaction, and the amplification of divisive content on social media. PNAS Nexus. 2025;4(3):pgaf062.
- Pennycook G, Epstein Z, Mosleh M, Arechar AA, Eckles D, Rand DG. Shifting attention to accuracy can reduce misinformation online. Nature. 2021;592(7855):590–595.
- Lewandowsky S, Ecker UKH, Seifert CM, Schwarz N, Cook J. Misinformation and Its Correction: Continued Influence and Successful Debiasing. Psychol Sci Public Interest. 2012;13(3):106–131.
- Loomba S, de Figueiredo A, Piatek SJ, de Graaf K, Larson HJ. Measuring the impact of COVID-19 vaccine misinformation on vaccination intent in the UK and USA. Nat Hum Behav. 2021;5(3):337–348.
- Hall V, Banerjee E, Kenyon C, et al. Measles Outbreak — Minnesota April–May 2017. MMWR Morb Mortal Wkly Rep. 2017;66(27):713–717.
- CDC. Measles Cases and Outbreaks (data as of 12 November 2026).
- Hassanian-Moghaddam H, Zamani N, Kolahi AA, McDonald R, Hovda KE. Double trouble: methanol outbreak in the wake of the COVID-19 pandemic in Iran—a cross-sectional assessment. Crit Care. 2020;24(1):402.
- Wineburg S, McGrew S. Lateral Reading and the Nature of Expertise: Reading Less and Learning More When Evaluating Digital Information. Teachers College Record. 2019;121(11):1–40.
- Wineburg S, Breakstone J, McGrew S, Smith MD, Ortega T. Lateral reading on the open Internet: A district-wide field study in high school government classes. J Educ Psychol. 2022;114(5):893–909.
- Roozenbeek J, van der Linden S, Goldberg B, Rathje S, Lewandowsky S. Psychological inoculation improves resilience against misinformation on social media. Sci Adv. 2022;8(34):eabo6254.