Why evidence convinces some people and not others in health

The word proof sounds final. In mathematics, a demonstration can feel like an immovable rock. In real life, especially in public health, outbreaks, and urgent decisions, things are messier. The same data point can convince one person and leave another unimpressed. Under pressure, that difference is not a detail. It can shape policy, behavior, and outcomes.

In a conversation with mathematician Adam Kucharski, author of a book about proof and evidence, one idea stood out: life is full of situations that reveal large gaps between what we think is true and why we think it is true. The goal is not to eliminate uncertainty, but to navigate it without fooling ourselves.

Being right does not always convince

A classic example makes this concrete. The Monty Hall problem, based on a game show, asks you to pick one of three doors. Two hide a goat and one hides a car. After you choose, the host opens another door and reveals a goat. Many people feel it is now fifty fifty. But if you switch doors, you double your chances of winning.

The interesting part is not only the answer, but the reaction to it. Even highly capable people can feel uneasy with the result. You can accept the conclusion and still not feel satisfied. For some, walking through every possible case helps. For others, a computer simulation repeated thousands of times is what finally convinces them. The lesson is simple: the same fact can require a different explanation depending on who is listening.

Evidence under pressure and subject to revision

Kucharski connects this to day to day work during outbreaks and emerging threats. In those settings, decisions are made with incomplete information. Methods get pushed to their limits and certainty is provisional. Today you believe something because it is the best you have, but tomorrow it can change when better evidence arrives.

That is not a license to improvise. It is a reminder that evidence is not only a pile of data, it is a process. How it is collected, compared, integrated, and communicated matters. It also matters to remember that scientists, like anyone else, face incentives, time constraints, and biases.

The gap between what convinces me and what convinces you

A central part of the problem is human. Sometimes we are not arguing about data, we are arguing about identity. When someone has built a public persona around a position, it becomes very hard to write down, privately, what would change their mind. Not because it is impossible, but because the social cost is high.

A powerful exercise was mentioned that you can apply on a small scale. Write what you think will happen, or what claim you are defending, and then write what would need to occur to prove you wrong. If you cannot articulate it, you are likely defending a flag, not a hypothesis.

Multiple forms of proof instead of a silver bullet

In science and health, there is rarely a single decisive proof that ends the discussion. What works better is combining approaches.

• Use logic and internal consistency to spot contradictions.
• Use empirical evidence across settings to see whether an effect holds.
• Use plausible mechanisms to avoid magical explanations.
• Use simulations to test whether an intuition survives repeated trials.

Any one piece can be argued. Together, they build a case. And the most important part is that the case is transparent about its limits.

Talking without pushing people to the other side

The pandemic left a hard lesson. There was a loud minority with extreme claims, but there were also many people with more ordinary doubts: mistrust of companies, questions about safety in specific groups, or uncertainty because data were limited. Treating everyone as if they share the same motivation does not help.

A better approach is to ask concretely.

• What exactly worries you.
• What evidence you would accept.
• What past experiences shape your view.

That does not mean every claim deserves equal weight. It means building a bridge. If you want evidence to lead to action, the person has to feel heard and the explanation has to fit their mental model.

A practical protocol to think better

You can use these ideas in daily life without waiting for a big public debate.

1. Define the claim in one sentence. Avoid vague concepts.
2. Write your confidence level today and why.
3. List two or three tests that would change your position.
4. Actively seek evidence that could refute you, not only evidence that confirms you.
5. If time allows, simulate or test on a small scale. Repeat and compare.
6. Update in stages. Changing your mind is not a failure, it is a skill.

Conclusion

Evidence is not very useful if it does not translate into decisions that improve lives. To get there, it is not enough to accumulate data. You need a method to live with uncertainty, recognize bias, and explain why something is true in a way other people can follow. Proof is not only the answer. It is also the path that gets you there.