How we grade evidence

Not all research is equal. A laboratory finding is a clue, an observation in people is a pattern, and a well designed randomised trial is a test. We show the type of evidence behind every claim, so you can see how strong it is for yourself.

Choose Detailed for the fuller picture, or Scientific for mechanisms and study data.

We use five levels, from weakest to strongest.

1. Traditional use. Documented in classical texts or long practice. Valuable as a starting point, but not tested.
2. Laboratory evidence. Studies in cells or animals. They explain how something might work, but often do not translate to people.
3. Observational evidence. Patterns seen in large groups of people. They can show links, but not prove cause.
4. Clinical evidence. Randomised controlled trials, where people are assigned by chance to a treatment or a comparison. This is how cause and effect is tested.
5. Systematic reviews. Several trials gathered and appraised together, usually the most reliable summary.

What else we look at

How many people were studied, for how long, who funded the work and whether the results have been reproduced. A large, independent, repeated finding is worth more than a small, single, sponsored one.

Why we do this

Because honesty builds trust. Some of our ingredients have strong evidence, others have promising early data and some rest mainly on tradition. You deserve to know which is which.

Our hierarchy follows the principles of the Oxford Centre for Evidence-Based Medicine levels of evidence and the GRADE approach, which also considers risk of bias, inconsistency, indirectness, imprecision and publication bias.