How to read a claim

You do not need a science degree to judge most claims in this area. You need four questions, asked in order, and the willingness to stop when one of them cannot be answered.

01Was it tested in humans?

This single question separates most of the noise from most of the signal. A great deal of what is claimed for research peptides comes from studies in rats, mice or cells in a dish.

Animal research is real science and it is how drug development starts. It is also a poor predictor of what happens in people. The large majority of compounds that work in animals fail in human trials, either because the effect does not appear or because harms emerge. This is the normal outcome, not a conspiracy.

The move to watch for

A study in rats reports faster tendon healing. The marketing says "shown to accelerate tendon repair". Both sentences describe the same study, but the second has quietly dropped the species. When you see a healing or performance claim, look for the word rat, mouse or in vitro in the source.

02How many people, and for how long?

Twelve people for four weeks and two thousand people for eighteen months are not the same kind of evidence. Small, short studies detect big effects and miss everything else, including most harms, which by their nature often appear late or rarely.

For context: the phase 3 trials behind the licensed GLP-1 drugs ran for over a year with thousands of participants. That is the standard a regulator requires before approving something. It is worth holding that in mind when a claim rests on a study of thirty people.

03Compared to what?

A number on its own means very little. Weight change of a given percentage over a year needs a comparison group, because people enrolled in a trial change their behaviour simply by being in one.

Placebo-controlled. The comparison is an inactive treatment. This is the standard for showing a drug does something.
Randomised. Participants were assigned by chance, which stops healthier or more motivated people clustering in one group.
Blinded. Participants, and ideally the researchers, did not know who got what. Expectation affects reported outcomes.
Open-label or uncontrolled. No comparison, or everyone knew. Useful for spotting signals, weak for proving effects.

04Who paid, and who is telling you?

Most large drug trials are funded by the manufacturer. That is how the system works, since nobody else will pay for a trial that size. It does not invalidate the results, and these trials are published in peer-reviewed journals with declared funding and registered protocols.

It does mean the funding is relevant context that is usually stripped out by the time a figure reaches social media. Every record on this site names the funder for that reason.

Then ask who is telling you. A seller writing an article about whether their product is legal is not an independent source, however many citations sit at the bottom. Check whether the person making a claim also takes payment when you act on it.

05Two patterns worth recognising

Absence of evidence presented as suppression. The argument runs: there are no human trials because the compound cannot be patented, so nobody will fund one. Patent incentives are a real problem in medicine. But "no trials have been done" and "trials were done and hidden" are completely different claims, and the first does not support the second. If nothing has been tested, nobody can honestly tell you the risk in either direction.
Mechanism presented as outcome. "It increases blood vessel formation, therefore it heals injuries faster." The first part may be established in a laboratory. The second requires an actual trial measuring actual healing in actual people. A plausible mechanism is a reason to run a trial, not a substitute for one.

06Where to check

electronic medicines compendium. If it is a licensed medicine, the SPC is the most reliable document about it.
PubMed. Search the compound name. Read the abstracts and note the species and the participant numbers.
Cochrane Library. Systematic reviews that assess the whole body of evidence rather than one study.
NICE. For whether the NHS considers something worth funding, and on what grounds.
ClinicalTrials.gov and Be Part of Research. What is actually being tested right now.

If a compound comes up empty across all of these, that is the answer. It has not been assessed, and no amount of testimonial changes that.