1. Why Peptide Data grades evidence

Every compound profile on this site carries one of four evidence grades: Strong, Moderate, Limited, or Anecdotal. The grade is not a marketing label and it is not a legal status. It is a summary judgement about what kind of research exists on a compound, how well that research was conducted, and how directly it applies to humans.

The research-peptide literature is unusual in that a compound can accumulate hundreds of citations — almost all of them in cell culture or animal models — while having no controlled human data at all. A bibliography count is therefore a poor proxy for evidence. Grading exists to make that distinction explicit, so a reader can see at a glance whether a claim rests on large randomised trials or on a single small study from decades ago.

Consistency is the point. The same four tiers are applied to every compound, including licensed medicines and obscure research materials, and the reasoning is published alongside the grade. The site carries no vendor advertising, and no grade is influenced by commercial relationships.

2. The four grades

Strong. Multiple well-conducted human studies — ideally randomised controlled trials, with at least one systematic review or meta-analysis — that agree in direction and were conducted in populations that resemble the question being asked. Evidence at this level is typically also reflected in a licensed medicine and a regulator-assessed product information document.

Moderate. Consistent human data, but with material gaps: few trials, small samples, short follow-up, a single population, or conflicting results that replication has not resolved. Findings are plausible and probably real, but the size and durability of any effect remain uncertain.

Limited. Human data exist but are thin, dated, unreplicated, or methodologically weak — or the compound has a substantial preclinical literature that has never been translated into adequate human studies. Many research peptides sit here. The grade is a statement about the evidence, not a verdict on the compound.

Anecdotal only. No meaningful controlled human data. The evidence base is forum reports, vendor material, or uncontrolled observation, sometimes alongside in vitro or animal work. Nothing at this level supports a claim about effects in people.

3. How a grade is assigned

Five questions drive the decision.

3.1 What is the highest-quality study design available?

Study design sets the ceiling. A systematic review of randomised trials sits above a single randomised trial, which sits above a cohort study, which sits above a case series, which sits above a case report or an in vitro experiment. This ordering is conventional and is set out in the Oxford Centre for Evidence-Based Medicine's levels of evidence and in the GRADE approach to rating certainty of evidence.

3.2 Are the findings replicated, and by whom?

A single positive trial is a hypothesis; two independent trials pointing the same way are evidence. Independent replication is weighted heavily, and findings that originate almost entirely from one laboratory, one institution, or one region are treated as weaker than the raw study count suggests.

3.3 How direct is the evidence?

An animal model of a condition is not the condition in a human. Where human data are absent or minimal, no volume of preclinical work lifts the grade above Limited, because the central question — whether the finding survives translation — has not been tested. Reviews of preclinical validity have repeatedly shown how poorly animal findings predict human results.

3.4 How robust is each study?

Sample size, randomisation, blinding, pre-registered outcomes, control of confounders, attrition, and completeness of reporting all matter. Small, unblinded, single-arm studies using surrogate endpoints generate effect sizes that rarely survive replication. Where a study's methods cannot be assessed — because the report is incomplete or the paper is not accessible — that uncertainty is recorded rather than ignored.

3.5 Is the record stable?

A grade describes the literature as it stands today. Retractions, corrections, and unresolved concerns about a body of work are treated as downgrades, because a mechanism claim that has been withdrawn cannot support a grade. Where a compound's core mechanism literature has been retracted, that is recorded on the profile and reflected in the grade.

4. What a grade is not

  • It is not a statement that a compound works, or that it is safe. It is a statement about the quantity and quality of research.
  • It is not a legal status. Evidence grade and UK legal status are recorded separately. A compound can be well-evidenced and a prescription-only medicine, or poorly evidenced and legally supplied for research. Legality in the UK turns on the Medicines Act 1968, the Human Medicines Regulations 2012, and — where applicable — the Misuse of Drugs Act 1971, not on how good the science is.
  • It is not a dose, a protocol, or a recommendation. Peptide Data does not provide consumption, self-administration, or dosing advice for any compound, and nothing in the grading system should be read as such. All compounds described here are research materials.
  • It is not permanent. Grades move in both directions as the literature changes.

5. How to read a grade alongside a profile

A grade is the headline; the profile is the argument. Each compound page sets out the mechanism, the research record by study type, the protocols commonly discussed in the literature, storage and reconstitution considerations, the blood-work markers discussed in research contexts, UK legal status, and full references. Read the grade to know how much weight the evidence can carry; read the profile to see which studies produced it.

6. Corrections and how to challenge a grade

Grading is only as good as its sourcing, and we would rather be corrected than consistent. If a profile misstates a study, misses a trial, or applies a grade the record does not support, every reference is listed so the claim can be checked directly. Corrections are handled editorially, and where a grade changes, the reason is recorded.

Methodology note: Peptide Data publishes in a research-use-only context. Compounds discussed on this site are research materials; nothing on this page is medical advice, and none of it should be read as guidance for human use.

This article is AI-researched and editorially reviewed. It is provided for research and educational purposes only and is not medical advice. Research peptides are not licensed for human consumption in the UK.