How we fact check
Where claims come from
We check claims that are spreading online, printed on product labels, or sent to us by readers. We prioritise claims that could cost people money or affect their health.
What counts as evidence
We work down this list, starting at the top:
- Regulators and official health bodies: for example the NHS, NICE, MHRA and FSA in the UK; the FDA, NIH and FTC in the US; and EFSA in Europe.
- Clinical guidelines and official product information, such as a medicine's Summary of Product Characteristics.
- Systematic reviews and large clinical trials published in peer-reviewed journals.
- Individual studies, clearly labelled as such, with their limits explained.
We don't treat company marketing, influencer posts, forum threads or anonymous reviews as evidence.
Our three verdicts
| Verdict | What it means |
|---|---|
| ✓ Backed by evidence | Reliable sources support the claim as it's usually stated. |
| ! Partly true | There's something in it, but the claim is exaggerated, missing context, or only true for some people. |
| ✕ Not backed | Reliable evidence doesn't support the claim, or it's misleading. |
How we choose products
We decide what the evidence supports first, and only then look for products that fit. A product appears on our site because it matches the verdict, not because of the commission rate. Where a cheaper option does the same job, we say so.
For supplements, we only describe benefits using wording that regulators have approved for that nutrient. We never suggest a supplement can treat, cure or prevent a disease.
Human review
AI tools help us find and summarise research. Before anything is published, a person reads every source linked in the article and checks that it says what we say it says. Draft pages are hidden from the site until that's done.
Corrections
If we get something wrong, we fix it, update the date on the page and, for any significant change, add a note saying what changed. Report an error.