RootCauseAI

The method

How we decide what to tell you.

Every piece goes through the same process, and the grade you see at the top is the output of it. Here is the whole thing, so you can judge it yourself.

The scale

The three grades

A

Well established

Multiple randomised controlled trials, or a strong systematic review, pointing the same way. The effect is real and we can usually say roughly how big it is.

Example: resistance training improving strength and function as you age.

B

Reasonable

Consistent observational evidence, or small or mixed trials leaning one direction. Worth doing if it is cheap and safe, but do not expect certainty.

Example: earlier and more consistent sleep timing helping daytime alertness.

C

Unproven

Traditional use, mechanism arguments, animal studies, or a couple of small trials that disagree. We include these because you will meet them anyway, and you deserve to meet them labelled.

Example: most single-herb claims advertised on social media.

From question to page

What happens before something gets published

Start from the real question

Topics come from what people actually ask, not from what ranks well in search.

Go to the strongest evidence first

Systematic reviews and trials before blog posts. If a claim only exists in marketing copy, that fact goes in the piece.

Grade it before writing

The grade is assigned against the criteria above, before the piece is written, so the writing cannot talk itself up.

Say the size of the effect

A real but tiny benefit is described as a real but tiny benefit. This is where most health writing quietly cheats.

Flag interactions and cautions

Anything that could interact with common medication, or is not appropriate in pregnancy or for children, is marked.

Revisit when evidence moves

Grades can change. When one does, the piece says what changed and when.

Fair questions

Things people ask first

No. It is health education. It does not diagnose, does not replace your doctor, and does not tell you to change any medication. Where something needs a clinician, we say so.

No, and we never will. Food, movement and sleep genuinely improve health, and that is a different claim from curing a disease.

No. It works best alongside your doctor. If you are unwell, get seen. We will happily help you prepare better questions for that appointment.

No. Conversations are not saved. Closing the page ends it and we keep no copy.

Grades get revised and the page records what changed. That is the point of grading in the first place.

If this is how you want your health information handled, join us.