Why a vivid account is not the same kind of evidence

NCCIH distinguishes testimonials and anecdotes from objective research evidence. That distinction concerns what a story can establish, not whether its teller is sincere. An account may still describe a service visit, an inconvenient schedule or a helpful question in detail.

[1]

Fictional example: the audiobook and the changed commute

Maya writes: ‘I started listening to an audiobook on the bus. I have not smoked since that week.’ The same post says that Maya changed from evening to morning shifts, stopped travelling with a colleague who smoked, and attended a cessation appointment. Maya calls the audiobook a useful companion. All names and events here are invented.

A friend reposts it as ‘An audiobook ended the cravings.’ That adds an outcome Maya did not report and singles out a cause the account does not isolate. You cannot fix the inference by choosing the appointment or the shift change as the true cause instead. The story contains several changes, but no comparison that separates their contribution.

Keep a faithful retelling: ‘Maya reports not smoking since that week and liked having something to listen to on the journey. Several circumstances changed. This account does not tell us which change caused the result.’ It neither discounts the achievement nor turns the audiobook into a tested quitting method.

Ten similar posts still do not give the missing denominator

Imagine that a page collects ten stories like Maya's. Without knowing who was invited, who chose to submit, which accounts were selected and what happened to everyone else, you cannot turn that collection into a success rate. Ten reposts of Maya's words are also still one account, not ten independent experiences.

A story ending in June does not tell you what happened afterward. Leave that unknown, rather than assuming a later setback or continued abstinence. When someone says ‘most users get the same result’, ask for group evidence with a defined outcome and follow-up, separately from the stories.

Use the story to make a better question

The useful question in this example could be ‘What happens in a cessation appointment, and can I ask about travel routines?’ You can check a service's own description. It is different from copying Maya's schedule or asking the storyteller to choose a treatment for you.

For your own support or treatment decisions, discuss the question with a qualified local health professional or cessation service. You need not post your smoking history, medicines or account screenshots beneath someone else's story. A private question can remain private.

What to keep in mind

  • Preserve what the person reported before adding an explanation.
  • Several simultaneous changes cannot be separated by this story alone.
  • A selected collection gives no rate for all users.

Common questions

Does an unsuccessful story prove a method never works?

No. It can describe a real difficulty for that person, but a single experience also cannot establish that a method has no effect in a defined population. Use the difficulty to identify a question, not to calculate a universal verdict.

Sources

The central claims on this page were checked against the sources below.

  1. NCCIH — National Institutes of Health: How Do You Know the Information Is Accurate?

    Sources checked: 2026-10-04

General information literacy. The example is entirely fictional and neither evaluates a quitting method nor predicts a personal result.