How to answer false health claims in your group without shaming anyone

Key takeaways

• In an April 2025 KFF poll, 24% of U.S. adults said a false claim that vitamin A prevents measles was probably or definitely true, and 70% sat in the unsure middle.
• A review of 65 studies found that corrections work, and work better on health topics than on politics, especially when they explain why a claim is wrong.
• A short video about manipulation tactics, shown to millions of YouTube users, improved viewers’ ability to spot those tactics by about 5%.
• Across 20 experiments, simple prompts to think about accuracy cut people’s intention to share false headlines by about 10%.
• Not every uncertain claim is misinformation, so label “not enough evidence” differently from “evidence says no.”

In March 2025, as measles spread through West Texas, Covenant Children’s Hospital in Lubbock found a second problem in some young patients. Children admitted for measles complications also had high vitamin A levels and abnormal liver function¹³.

The hospital said fewer than 10 children were affected. Texas Public Radio also cited reports that cod liver oil had become a big seller at area pharmacies¹³.

That April, a KFF poll found that 24% of U.S. adults, and 31% of parents, thought it was probably or definitely true that vitamin A can prevent measles infection¹¹.

Below, we look at the twist that matters for anyone who moderates a health group. The claim was not pure fiction.

A 2022 Cochrane review pooled six trials of young children in low- and middle-income countries. Supplements cut new measles cases by half, and the NIH Office of Dietary Supplements says that benefit applies to children at high risk of vitamin A deficiency¹². But the review found that vitamin A did not lower the risk of dying from measles¹².

There is a safety catch, too. The body stores extra vitamin A, mostly in the liver, so high doses taken over time can pile up and cause harm, including abnormal liver tests¹².

The false part was the leap, from “helped children at risk of deficiency” to “protects any child.” A lot of health misinformation looks like this: a half-truth, shared by caring people, stripped of its context. This playbook shows what research says works when one lands in your group.

Most health misinformation is a half-truth, not a lie

The U.S. Surgeon General’s 2021 advisory defined health misinformation as information that is “false, inaccurate, or misleading according to the best available evidence at the time”¹. Two parts of that definition deserve attention.

“Misleading” covers true facts used to suggest something false, like the vitamin A leap. “At the time” admits that evidence changes. A claim that is unproven today could be supported tomorrow.

So a fair moderator doesn’t sort claims into “true” and “misinformation.” This library uses a four-level evidence scale, and it works well as a response guide, too:

Evidence level	What it means	How a group can respond
Strong evidence	Consistent results from good human trials or reviews	Share freely, with a source
Promising but early	Small or few trials, mixed results, or mostly lab and animal work	Allow it, and add a note on what is still unknown
Not enough evidence	Little or no human research	Allow it as personal experience, not as fact, and ask for sources
Evidence says no, or unsafe	Tested and failed, or documented harm	Correct it promptly, and remove it if acting on it could hurt someone

“Not enough evidence” is not the same as false. Treating every unproven remedy as misinformation insults members and burns your credibility before the claims that matter.

The people reading are more persuadable than the person posting

When a false claim appears, the poster may never change their mind. But they are rarely the only audience.

KFF’s May 2026 poll asked about four common vaccine myths. Only 1% of adults were consistent believers, and another 6% leaned toward believing¹⁰. The largest single group, 31%, was a mixed middle¹⁰. Across the four myths, at least half of adults kept answering “probably true” or “probably false” rather than giving a firm answer¹⁰.

The vitamin A claim looked the same. In the 2025 poll, 70% of adults put it in the “probably” range¹¹.

So when you reply in a thread, write for the dozens of quieter members who are reading and unsure. Explain as you would to a curious friend. You are not trying to win a debate with one person.

The same poll carries a warning about tone. Only 38% of adults were confident that federal health agencies make decisions based on science rather than officials’ personal views¹⁰. “Because the CDC says so” will not persuade many readers. Showing the evidence, and where it comes from, might.

Why a false claim lingers after it’s corrected

Researchers call it the continued influence effect. People can accept a correction and still lean on the false idea later, when they reason about something related⁵.

The Debunking Handbook 2020, a consensus summary written by 22 misinformation researchers, puts it simply: misinformation is sticky⁵. Repetition makes a claim feel truer, so the handbook advises mentioning a myth only once, right before you correct it⁵.

The handbook also says the “backfire effect,” where a correction makes belief stronger, is less common than researchers once thought⁵. Fear of backfire is not a good reason to stay silent.

Corrections work, especially on health

A meta-analysis pools the results of many studies to find the overall pattern. A 2018 meta-analysis by communication researchers Walter and Murphy combined 65 studies and found that corrections had a moderate, positive effect⁷.

Corrections worked better on health than on politics⁷. The strongest ones offered a coherent alternative explanation, which beat simple fact-check labels and appeals to an expert’s credibility⁷. In that analysis, rebuttals after people saw a claim also did better than warnings given beforehand⁷.

A 2017 meta-analysis by Chan and colleagues, released by the University of Pennsylvania’s Annenberg Public Policy Center, pooled 20 experiments with 6,878 participants⁶. Those studies tested false social and political news, not health claims. Detailed counter-messages changed minds more than simply calling a claim wrong⁶. Inviting people to think through why the claim was wrong helped, too⁶.

The same team found that complete correction is very hard⁶. Their advice: don’t dwell on the arguments for the myth, and bring new evidence instead of just a label⁶.

Prebunking: warn people before the claim arrives

Prebunking, also called psychological inoculation, borrows its logic from immunization. You show people a weakened example of a manipulation tactic and explain how it works, so they recognize it later.

In 2022, researchers from the University of Cambridge, the University of Bristol and Google’s Jigsaw unit tested this at scale³. After six lab experiments with 6,464 people, they placed short animated videos in YouTube’s ad slot³.

About 5.4 million U.S. users saw a video, and almost 1 million watched at least 30 seconds³. A random 30% of viewers got a one-question quiz within 24 hours, and among the 22,632 who answered, recognition of the tactics rose about 5% on average. Each view of real length cost the team about a nickel³.

The videos covered five tactics: scapegoating, deliberate incoherence, false dichotomies, ad hominem attacks and emotionally manipulative language³. This is how each might show up in a health group. These examples are ours, not the researchers’:

• Scapegoating: “One company is behind every chronic disease.”
• Deliberate incoherence: “This herb is totally harmless, and also strong enough to cure anything.”
• False dichotomy: “Either you trust nature or you trust medicine.”
• Ad hominem: “Of course a doctor would say that.”
• Emotional manipulation: “If you love your kids, you’ll stop that today.”

A 5% gain is small. But a pinned post that teaches these patterns costs nothing, and it reaches every member before the next viral claim does.

A ten-second pause before sharing

Gordon Pennycook and colleagues reported in Nature in 2021 that most people say sharing only accurate news matters to them⁴. Yet whether a headline was true barely affected whether they wanted to share it⁴. The researchers argued that the problem is often inattention, not a lack of caring about truth⁴.

Subtle prompts to think about accuracy improved the quality of what people shared, including in a field experiment on Twitter⁴. In 2022, Pennycook and David Rand pooled 20 of their experiments with 26,863 participants. Accuracy prompts cut intentions to share false headlines by about 10% compared with no prompt⁸.

The effect held across political views and education levels, and it was larger among older, more reflective and more attentive participants⁸. One caution: this is the authors’ own analysis of their own studies, mostly online surveys.

For a group, the prompt can be as simple as a line in your posting rules: “Before you share a health claim, ask: is this accurate, and where is it from?” The Surgeon General’s advisory offers a shorter version: “If you’re not sure, don’t share”¹.

When a false claim appears: a seven-step routine

This routine is ours, built from the research above. Moderators can use it as written. Members can use steps 2 through 4 on their own.

1. Pause and sort. Ask two questions: Is it spreading? Could someone get hurt acting on it? The Debunking Handbook advises picking your battles, since a claim nobody is repeating may not be worth the spotlight⁵.
2. Check it. Look up the claim on NIH or Cochrane pages and place it on the four-level scale above.
3. Reply in public with a truth sandwich. Lead with the fact. Warn that a myth is coming and mention it once. Explain why it misleads. End with the fact again⁵.
4. Message the poster privately. The Surgeon General’s advisory suggests listening, asking questions, staying calm and not expecting one conversation to change a mind¹.
5. Act on harm. If a post could lead someone to skip care or take a toxic dose, hide or remove it, and say why in the thread.
6. Prebunk the repeats. If the same claim keeps coming back, pin a short explainer.
7. Listen for the question underneath. The World Health Organization notes that false claims can spread quickly when people’s questions go unanswered⁹. Ask members what they are worried about, and answer that.

What the evidence says about these tools

Approach	Evidence level	What the research shows
Detailed corrections that explain why a claim is wrong	Strong evidence	Two meta-analyses found corrections reduce false beliefs, with explanations doing best67
Prebunking videos and posts	Promising but early	Lab experiments plus one large field test; average gain about 5%3
Accuracy prompts	Promising but early	About a 10% drop in intent to share false headlines across 20 experiments, mostly online surveys by one team8
The truth sandwich format	Promising but early	Recommended by a 22-author research consensus built on correction studies5
Corrections often backfire, so silence is safer	Evidence says no, or unsafe	Backfire effects are less common than researchers once thought5
Any of these tools inside small wellness groups	Not enough evidence	The studies used online panels and platforms; we found little research inside tight-knit communities

Skepticism is a natural-health value, and point it in every direction

Many natural-health readers learned to question drug ads, industry-funded studies and official statements. That habit is healthy. It works best when it applies to a viral supplement post as firmly as to a pharmaceutical commercial.

A good correction fits natural-health values. It shows its evidence instead of leaning on titles. It admits what is still unknown. It treats the person who posted as a neighbor, not an enemy.

The World Health Organization describes managing false health information as a community job: listening to concerns, explaining risks, building resistance to misleading claims and helping people act². A wellness group is already built for that work.

Questions readers ask

The bottom line

A false claim will reach your group, and it will probably arrive as a half-truth from someone who means well. The research offers a calm response: sort claims by evidence and harm, correct with an explanation rather than a label, warn people about manipulation tactics ahead of time, and remind everyone to pause before sharing.

Write your corrections for the unsure readers in the middle, not for the poster. Treat everyone with respect, admit what is uncertain, and point your skepticism in every direction.

Related reading

• What helps in a family health argument, since facts alone rarely win
• Learning health skills together works when the group checks its facts
• Online health groups offer reach, while in-person groups have stronger evidence
• Most health products aren’t scams, but many just haven’t been proven
• How to protect people who come to wellness groups at fragile moments

Sources

1. Office of the U.S. Surgeon General. “Confronting Health Misinformation: The U.S. Surgeon General’s Advisory on Building a Healthy Information Environment.” HHS, 2021. hhs.gov
2. World Health Organization. “Infodemic.” WHO, accessed October 2026. who.int
3. University of Bristol. “Experiment on YouTube shows potential to ‘inoculate’ millions of users against misinformation” (on Roozenbeek et al., Science Advances). University of Bristol, 2022. bristol.ac.uk
4. Pennycook G, Epstein Z, Mosleh M, et al. “Shifting attention to accuracy can reduce misinformation online.” Nature, 2021. MIT Media Lab record. media.mit.edu
5. Lewandowsky S, Cook J, Ecker UKH, et al. “The Debunking Handbook 2020.” 2020. sciencelearning.umd.edu
6. Annenberg Public Policy Center, University of Pennsylvania. News release on Chan MS, Jones CR, Hall Jamieson K, Albarracín D, “Debunking: A Meta-Analysis of the Psychological Efficacy of Messages Countering Misinformation,” Psychological Science. 2017. annenbergpublicpolicycenter.org
7. National Communication Association. “Does Correcting Misinformation Really Work?” (summary of Walter N, Murphy ST, “How to unring the bell,” Communication Monographs, 2018). NCA. natcom.org
8. Pennycook G, Rand DG. “Accuracy prompts are a replicable and generalizable approach for reducing the spread of misinformation.” Nature Communications, 2022. MIT DSpace record. pubmed.ncbi.nlm.nih.gov
9. World Health Organization. “Introducing rapid social listening and infodemic insights for action: WHO and UNICEF launch manual on 6 steps to build an infodemic insights report.” WHO, 2023. who.int
10. KFF. “New KFF Poll Examines Patterns of Belief Across Common Vaccine Myths.” The Monitor, 2026. kff.org
11. KFF. “KFF Tracking Poll on Health Information and Trust: The Public’s Views on Measles Outbreaks and Misinformation.” KFF, 2025. kff.org
12. NIH Office of Dietary Supplements. “Vitamin A and Carotenoids: Fact Sheet for Health Professionals.” NIH, accessed October 2026. ods.od.nih.gov
13. Davies DM. “West Texas children treated for vitamin A toxicity as medical disinformation spreads alongside measles outbreak.” Texas Public Radio, republished by High Plains Public Radio, 2025. hppr.org