Most sites in this space will tell you a compound is "promising" or rate it out of five. That is an opinion wearing a lab coat. We do something narrower and more checkable: we report exactly what human research is registered, and what it was testing.
Where the numbers come from
- Source: the ClinicalTrials.gov public API, plus FDA approval records and warning letters.
- Frequency: checked on a schedule. Every page carries the date its data was pulled — a number without a date is a number you cannot verify.
- Stored, not summarised: we keep the full record for each trial, so counts can be recomputed rather than re-typed.
What we count
Interventional human studies only. A study qualifies if it is registered, involves people, and gives them the compound.
Observational studies — which watch people who are already taking something — are listed separately and never counted toward the headline number. They cannot show that a compound causes an effect, and counting them would inflate every figure on this site.
Animal studies, cell studies, and published papers with no registry entry are not counted either. They may well exist and may well be interesting; they are just not what the number on the page measures.
What the number does and doesn't mean
A trial count measures how much has been formally studied in humans. It is not a verdict.
- Zero registered trials does not mean a compound does nothing. It means nobody has properly checked. Absence of evidence is not evidence of absence, and we will not print it as though it were.
- Fifty trials does not mean something is safe or effective for you. It means the question has been asked often.
- A registered trial is not a result. Registration means someone intended to run a study. Plenty are never completed and never published.
Why sources disagree
Search "how many human trials does BPC-157 have" and you will get several different answers from credible sources. None of them are lying — they are answering different questions, and most do not say which:
- Published papers includes unregistered and older work
- Large phased trials excludes small ones, often giving zero
- Completed trials excludes anything still running
- Registered trials at any status — what we report
Each compound page publishes all of these side by side, so you can see that our headline figure is a choice with a stated rule rather than a number we preferred.
Judgment
None. There is no grade, tier, score or rating anywhere on this site, and there is no column for one in the database. Adding one would require a physician reviewer and would turn a fact-reporting operation into a medical opinion. We have chosen not to.
Corrections
Every page has a way to report an error, and corrections are published rather than quietly edited. Every trial links to its original registry record so you can recount it yourself in about a minute.