Every Claim We’ve Checked

All 138 claims we have checked, grouped by subject — supplements, training, sweeteners, diets, sleep and the rest — each with its confidence rating.

Every claim on this site has been checked against the primary research and given one of four confidence ratings. There are 138 of them so far, and this page is all of them, grouped by subject.

If you would rather browse by how confident we are, the Established, Supported, Preliminary and Unsupported pages do that instead.

What the four ratings mean
  • established (19)
    Settled. Large, consistent evidence and the direction is not contested.
  • supported (34)
    Real, with good evidence behind it — but smaller or less certain than Established.
  • preliminary (35)
    Something is here and nobody can tell you what yet. Unresolved on purpose.
  • unsupported (50)
    Checked, and the evidence does not back the claim.

Sweeteners (7)

Sugar substitutes, and the studies people quote about them.

Supplements (34)

What the tub claims, and what the trials found.

Sleep, stress and recovery (15)

The parts of fat loss that happen when you are not training.

Meaning and connection (2)

Purpose, belonging, and the places people are said to live longest.

Applied from the outside (13)

Saunas, ice baths, sweat suits, LED panels and compresses — things done to your body rather than by it.

Diets and eating patterns (10)

Whole approaches to eating, checked one claim at a time.

Hormones (8)

Blood levels, what they mean, and what moves them.

Training (22)

Whether a given kind of exercise does what it is sold as doing.

Body composition (6)

What your body is made of, and what the scale can and cannot see.

Food and drink (21)

Individual foods, and the claims attached to them.

Guides

Round-ups built from the verdicts above. They carry no confidence rating of their own, because a guide is not a single claim — each entry inside links to the verdict that justifies it.

Our own data

Not a round-up. This is a measurement the desk made itself, and it is the only page here that needed both halves of a job most publishers only do one side of.

  • What health advice actually reaches people — we measured how far twenty claims travelled, then read the studies behind every one. Two thirds of the attention sat on claims that did not hold up — and the handful that survived had something in common we did not expect.
  • Who paid for the study — we record who funded the research on every verdict. On most of them, we could not find out. The surprise is not industry money.
  • Two kinds of “no evidence” — nobody ran the study, or somebody ran it and found nothing. Both get reported the same way, and only one of them means the claim is wrong.

Nothing here is medical advice. This is journalism — we read the research and report what it says. For anything about your own health, talk to a clinician.