Why This Page Exists
The research peptide industry has a credibility problem. Marketing claims outpace evidence, animal data gets extrapolated to human outcomes without caveat, and the line between education and sales pitch is often invisible. This page documents the standards we apply when evaluating and presenting peptide research on this site — so you can decide whether our methodology is trustworthy, not just whether our conclusions sound convincing.
How We Evaluate Evidence
Evidence Hierarchy
Not all studies carry equal weight. We apply a tiered hierarchy:
- Tier 1 — Human randomized controlled trials (RCTs): The gold standard. Large sample sizes, placebo controls, and peer review. For most research peptides, these simply don’t exist. When they do (e.g., tesamorelin’s regulated clinical approval for HIV-associated lipodystrophy), we lead with that data.
- Tier 2 — Human observational and open-label studies: Useful for identifying trends but vulnerable to confounding variables. We note sample size, study duration, and funding source when available.
- Tier 3 — Animal model studies: The bulk of published peptide research. We present these findings explicitly as preclinical — animal data does not directly translate to human outcomes. A compound that accelerates tendon healing in rats may do nothing of the sort in humans.
- Tier 4 — In vitro (cell culture) studies: Demonstrate mechanisms and potential but cannot establish whole-organism effects. We reference these for mechanistic understanding, not as evidence of therapeutic outcomes.
- Tier 5 — Case reports and anecdotal evidence: We do not cite these as evidence. Individual experiences are not data.
What We Flag
When reviewing a study, we specifically look for:
- Sample size: A study with 8 rats tells you something about mechanism. It does not tell you about efficacy in any meaningful sense.
- Dosing context: Animal doses are often orders of magnitude higher (per kg) than what a human would encounter. We note when this is the case.
- Funding and conflicts of interest: Industry-funded studies are more likely to report positive results. We disclose funding sources when we can identify them.
- Replication: A finding reported by one research group is preliminary. A finding replicated by independent groups is substantially stronger.
- Publication date: A 2003 rat study is not automatically irrelevant, but the research landscape evolves. We prefer citing recent systematic reviews and meta-analyses when available.
What We Do Not Do
- We do not make health claims. We describe what published research shows. We do not state or imply that any compound treats, cures, prevents, or diagnoses any condition.
- We do not recommend dosing protocols. We report doses used in published studies as contextual information, not as recommendations.
- We do not conflate purity with efficacy. COAs prove a compound is what it claims to be. They say nothing about what it does in a biological system.
- We do not cherry-pick favorable studies. When conflicting evidence exists, we present both sides. When the evidence is overwhelmingly negative, we say so.
- We do not use before/after imagery, testimonials, or implied outcomes.
Known Biases in the Field
We acknowledge several structural biases that affect peptide research:
- Publication bias: Positive results are more likely to be published than negative ones. This inflates the apparent strength of evidence across the literature.
- Animal model limitations: Most peptide research relies on rodent models. Mice are not small humans — metabolic rates, receptor densities, and pharmacokinetics differ significantly.
- Commercial funding: Much of the published peptide research is funded by companies with commercial interests in positive outcomes. This does not invalidate the research, but it warrants scrutiny.
- Regulatory gaps: Research peptides exist in a regulatory gray zone. The lack of mandatory federal oversight of the research compound market means there are no mandatory quality or testing standards.
Our Corrections Policy
If we publish information that is later shown to be inaccurate by new research, we update the relevant article and note the correction. If you identify an error in any of our content, contact us at [email protected] and we will review and correct it.
Scientific understanding evolves. Our content should reflect the best available evidence at the time of publication, not a fixed position.
Content Production
Articles on this site are produced by Stuart Ratcliff (co-founder) and Kai (AI research assistant powered by Claude and GPT models). All content is reviewed for scientific accuracy before publication. Citations link directly to PubMed or the original journal where possible.
We use AI tools for literature search, drafting, and data organization. We do not use AI to fabricate citations, generate fake study results, or produce content that has not been reviewed by a human editor.
This editorial policy was last updated in April 2026. It is subject to revision as our standards evolve.