Every claim, checked against its source.

Last updated: September 4, 2026

Fact-checking and medical review are different jobs

Medical review is a clinician’s judgment about whether an article’s health guidance is sound and safe. Fact-checking is narrower and more literal: does the study cited actually exist, does it say what the article says it says, is the number right, is the quotation accurate, is the mechanism described correctly. An article can pass one and fail the other, which is why both happen.

What gets checked

Before a health article publishes, an editor checks that:

• Every cited source is real and reachable. Study identifiers such as PubMed IDs and DOIs are verified against the source, not copied from another article. A citation that cannot be found is removed, and the claim it supported is removed or rewritten.
• Each source supports the sentence it is attached to. We read the source, not just its abstract, wherever the full text is available. A study on mice does not support a claim about people; a study of one dose does not support a claim about another.
• Statistics, dosages, and ranges match the source. Numbers are checked against the original figure, with units and context intact.
• The strength of the evidence is described accurately. A single small trial is called that. A systematic review is called that. “Studies show” has to be backed by studies that show it.
• Health claims have a reference. Our publishing tools flag a health claim without a supporting reference before an editor sees the draft; the editor resolves each flag by adding a source, softening the claim, or cutting it.
• Traditional-use and anecdotal claims are labeled. Saying a remedy has been used for a purpose for centuries is a historical claim and is checked as one. It is never presented as clinical evidence that the remedy works.

What counts as a good source

In rough order of preference, we rely on:

1. Systematic reviews and meta-analyses of human studies.
2. Randomized controlled trials in people.
3. Observational studies in people, such as cohort and case-control studies.
4. Clinical guidelines and position statements from recognized medical and scientific bodies, and reference works such as pharmacology and nutrition texts.
5. Animal, cell, and laboratory studies, used to explain a possible mechanism and labeled as preliminary.

We prefer sources indexed in PubMed or published by government health agencies, universities, and professional societies. We do not cite press releases, product marketing, or other health websites as evidence for a health claim. Where an article draws on traditional medical systems, we say so and describe the tradition rather than dressing it up as a trial.

How you can check our work

Every health article carries a numbered reference list, and inline citation marks in the text jump to the entry they rely on, so you can see which source stands behind which sentence and follow it to the original. We would rather you check than take our word for it.

After publication

Fact-checking does not end when an article goes live. Research is revised, studies are retracted, guidelines change, and readers write in. When any of those things affects a claim we have published, the article is corrected, its update date is refreshed, and a material change is recorded in the update log at the end of the article. Articles migrated from the earlier version of DrAxe.com are being re-checked to this standard on a rolling basis, starting with the most-read health topics. How corrections are handled is set out in the editorial policy.

Report an error

If you find a claim you believe is wrong, a citation that does not say what we say it says, or a number that does not match its source, email help@draxe.com with the article’s address, the sentence in question, and, if you have it, the source you are relying on. Every report is read and checked. If you are right, we will fix it and tell you.