Methodology
How we grade evidence.
Every compound we cover gets a single evidence grade. It reflects the strength and relevance of the evidence in humans — not how popular, promising, or widely sold something is. A compound can be everywhere and still grade poorly.
Strong human
Backed by well-designed randomized controlled trials in people. You can have real confidence in the effect.
Preliminary human
Some human trials exist, but they're small, short, early, or mixed. Suggestive, not settled.
Animal only
Evidence comes from rodent (or other animal) studies, or from cell work. Interesting — but animal results frequently fail to translate to humans.
Theoretical
A plausible mechanism exists, but no meaningful outcome data shows it does what is claimed.
No credible evidence
Claims rest on marketing, anecdote, or nothing at all.
- Animal data is not human data. Most compounds that work in rats never pan out in people. We say so plainly.
- Mechanism is not proof. "It should work because of pathway X" is a hypothesis, not a result.
- We flag who did the research. When most studies on a compound come from a single lab or are industry-funded, that matters.
- Legal ≠ effective ≠ safe. Whether something is approved, whether it works, and whether it's safe are three different questions, and we keep them separate.
Educational content, not medical advice — a grade is never a recommendation to use a substance.