Our evidence-grading methodology

Every compound on PeptideGrids carries one letter grade, A to D, that answers a single question: how strong is the human evidence that this compound does what it is used for? The grade is not a recommendation, a safety clearance, or a legal opinion. It is a summary of the published record as it stands today.

The grade scale

• Grade A, approved and proven. FDA approved for the use in question, supported by adequate and well-controlled human trials. Example: semaglutide for type 2 diabetes and weight management.
• Grade B, human evidence, not approved for this use. Real randomized or controlled human data exist, but the compound is not FDA approved for that indication (it may be approved for something else, or not at all).
• Grade C, preliminary or limited human evidence. Some human data exist, but they are small, uncontrolled, observational, or mixed. The signal is real but not yet reliable.
• Grade D, preclinical or anecdotal only. Evidence is limited to cell, animal, or anecdotal reports, with no credible human trials for the claimed use.
• Not evaluated. We have not finished grading the compound, or the evidence base is too sparse to place.

How we weigh the evidence

We rank study types in a fixed order: randomized controlled trials carry the most weight, then non-randomized controlled studies, then observational studies, then preclinical (cell and animal) work and case reports. A single large RCT moves a grade more than a stack of small observational papers. We also weigh sample size, replication, and whether the studied dose and route match how the compound is actually used.

How FDA status factors in

FDA approval is a ceiling check, not the whole grade. Approval for the specific use is required for an A. A compound can be FDA approved for one indication and still grade B or C for an off-label use that lacks its own trials. We track approval, non-approval, withdrawal, and compounding status separately from the evidence grade so the two are never conflated.

Drafting and review

For each compound we pull the primary literature from PubMed and Drugs@FDA, draft a sourced summary, and assign a provisional grade from the rubric. A reviewer then confirms or overrides that grade and records the reason. When new trials publish, the grade and summary are revised and the review date is updated on the page.

Sources and citations

Every published claim links to its primary source, almost always a PubMed record or an FDA document. We cite trials directly rather than secondary coverage. Chemical structures come from PubChem. We do not cite or link to compound vendors or research-chemical suppliers.

Independence

PeptideGrids has no financial relationship with any compounder, pharmacy, or supplement vendor, and is not affiliated with the FDA, NIH, or PubMed. We grade evidence; we do not sell, source, or help anyone obtain any compound. A grade is never paid placement. To flag an error, use our corrections policy.

What a grade does not mean

A high grade does not mean a compound is safe for you, legal to obtain, or appropriate for your situation, and a low grade does not mean a compound is useless, only that the human evidence is thin. This is information, not medical advice. Always consult a licensed clinician. See our full disclaimer.