Skip to content
Research

What We Mean When We Say "Evidence"

Our labelling system, in public, so you can hold us to it.

Key takeaways

  1. We separate what research shows from what is theoretical, anecdotal, or simply unknown — on the page, not in a disclaimer.
  2. "No evidence" and "evidence of no effect" are different findings and we do not use them interchangeably.
  3. Preclinical results are labelled as preclinical. An effect in cells or animals is a hypothesis about humans, not a finding in them.

Most of the confusion in this space is not caused by people citing bad studies. It is caused by good findings being restated one confidence level too high, repeatedly, until a cell-culture result has become a claim about people.

So we label. Every substantive claim on this site carries one of five tags, and we would rather look less certain than we could get away with.

LabelMeans
EstablishedSupported by human clinical research, replicated, with the limitations stated
PreliminaryHuman data exists but is small, single-study, short, or otherwise thin
PreclinicalDemonstrated in cells or animals only. A hypothesis about humans, not a finding in them
TheoreticalA mechanism-based expectation with no direct study behind it
AnecdotalSelf-report. Real experiences, no controls, no way to separate cause from coincidence
UnknownNobody has looked, or the studies that exist do not answer the question

Two distinctions we hold to strictly. First, absence of evidence is not evidence of absence — "no studies have examined this" is a different statement from "studies found no effect", and collapsing them misleads in both directions.

Second, a mechanism is not an outcome. A plausible pathway explains why something might work. It is not evidence that it does. Most overclaiming in this industry lives precisely in that gap.

If you find a claim on this site that is labelled above what its source supports, tell us. We will correct it and say what changed.

Sources

We link to primary sources — PubMed, ClinicalTrials.gov, FDA, NIH, peer-reviewed journals, and laboratory documentation. Entries marked needs research are placeholders: the claim in the text is not yet backed by a verified citation and should be read as unconfirmed until it is.

  1. Levels of evidence and hierarchy of study design

    Peer-reviewed methodology literature

    Needs research

    Cite a standard evidence-hierarchy reference to anchor the table.

Got Peppers is owned by the same team as PepGenex and may earn a commission on links to it. That relationship never decides what we publish.

Filed undermethodologyeditorial-standardsevidence
Keep going

related reading.

Testing

What Third-Party Testing Actually Proves

It proves something specific about one sample, at one moment, by one method. That is genuinely valuable. It is also narrower than the phrase implies.