The Attention Machine

← Symptoms

Health Misinformation

Health advice travels on the same engagement signals as everything else. The feed cannot tell a remedy from a rumour — only what holds attention. Alarming and contrarian claims hold it well.

How to read this page. Tap any underlined word to see the precise term and a short definition. Expand any "Deeper" box for the evidence and contested points. The main text works on its own — you can skip both and still get the whole argument.

It started before the feed

The clearest case study predates social media. In 1998 a paper in The Lancet claimed a link between the MMR vaccine and autism. It was wrong. Investigation later showed the data had been manipulated and that the lead author had undisclosed financial interests; The Lancet fully retracted the paper in 2010 and the author was struck off the UK medical register. There is no link between the MMR vaccine and autism — this is settled, not contested.

But the retraction took twelve years, and by then the claim had travelled the world and depressed vaccination rates. This is the pattern the Journalism and Science pages describe: a false claim moves fast, the correction arrives slowly and travels less far. Health misinformation did not need an algorithm to do harm. The algorithm made the asymmetry worse.

The infodemic

When COVID-19 arrived, the World Health Organization described the parallel wave of false and misleading claims as an infodemic. It was not a metaphor for inconvenience. The WHO had already, in 2019, listed vaccine hesitancy among its top ten threats to global health — at a time when vaccination prevents an estimated two to three million deaths a year. Engagement-optimised feeds amplified the most emotionally charged health content, and a frightening or contrarian claim reliably outperforms calm, accurate guidance.

What the evidence shows it does

This is one of the few places where the causal evidence is strong rather than correlational. In a randomised controlled trial in the UK and USA, researchers showed people either factual information or recent online misinformation about a COVID-19 vaccine. Brief exposure to the misinformation lowered intent to vaccinate by about six percentage points in both countries, among people who had said they would "definitely" accept a vaccine. Notably, misinformation that sounded scientific — citing made-up studies and statistics — was the most persuasive.

The route is not only the social feed. A controlled experiment manipulating search-engine results found that changing which websites a search surfaces — leaning the results pro- or anti-vaccination — measurably shifted people's beliefs and attitudes, and that users were largely unaware the selection was steering them. The same mechanism that decides what you scroll past decides what you find when you go looking for answers.

No conspiracy. Manufactured doubt — and honest worry.

Two things are true at once, and the page holds both. First, some health misinformation is manufactured doubt: contrarian claims deliberately produced and amplified to make a settled question look unresolved — the same playbook the tobacco and fossil-fuel industries used, documented on the Science page. Second, most of the people sharing health misinformation are not malicious. They are worried — about a child, a diagnosis, a side effect — and are seeking answers in the place that answers everything else. The US Surgeon General's 2021 advisory, which called health misinformation an "urgent threat," is careful to draw exactly this line between deliberate deception and honest questioning.

The cost

The harm is not abstract. Measles cases rose roughly 30% worldwide in 2019, and countries that had nearly eliminated the disease have seen it return as vaccination coverage slips. Health misinformation also leads people to refuse effective treatment for cancer and heart disease, and to harass the health workers trying to communicate accurate guidance. The mechanism is the same one running through this whole site: a system that rewards what is engaging, not what is true, distorts how people weigh risk — and with health, the cost of a distorted risk picture can be measured in lives.

How we know — what is settled, what is genuinely uncertain, and the honest line we hold +

Getting this topic right requires keeping three categories apart. Settled (refuted, not contested): vaccines are safe and effective, and the MMR–autism link is false — established by the retraction of the original fraudulent paper and by very large subsequent studies. Treating that as an open question is itself misinformation. Genuinely uncertain and openly monitored: like all medicines, vaccines have real side effects, the vast majority minor and rare ones tracked by safety-surveillance systems. Saying so is not a concession to anti-vaccine claims — it is the honesty that distinguishes science from doubt-manufacturing, and pretending vaccines have zero risk feeds the "they're hiding something" narrative. The strategy: manufactured doubt exploits the gap between those two by inflating the rare and monitored into the catastrophic and concealed.

On concentration. The Center for Countering Digital Hate reported in 2021 that a small number of accounts ("the Disinformation Dozen") produced a large share of anti-vaccine content on the major platforms. We mention it because it illustrates how amplification can concentrate, but the specific figure was disputed by Meta and the analysis is from an advocacy group, so we treat it as suggestive rather than settled — the spine of this page rests on the peer-reviewed and official sources above. The antidote is the one the rest of the site argues for: trusted, correction-transparent sources (a doctor, the systematic reviews on the Tools page) and learning to recognise the technique of manufactured doubt, rather than trying to chase down every individual false claim.

What you can do →
Where to get help →
The Manosphere →
What is Truth? — Science →