Sources & Methodology

We prioritize sources in the following order:

• FDA regulatory documents — approval labels, safety communications, REMS programs, shortage lists, and drug databases (e.g., Drugs@FDA, Orange Book).
• ClinicalTrials.gov — study registrations and posted results.
• Peer-reviewed primary clinical trials — published in PubMed-indexed journals.
• Systematic reviews and meta-analyses — including Cochrane reviews.
• Government health agencies — NIH, CDC, NLM, EMA, WHO.
• Professional medical guidelines — from recognized specialty societies.
• Manufacturer labeling — only for factual product details (e.g., storage, formulation); never as the sole basis for benefit claims.

• Anonymous peptide blogs or influencer content.
• Vendor marketing materials or product pages.
• Social media posts, Reddit threads, or forum anecdotes.
• Affiliate content presented as clinical evidence.
• AI-generated citations that cannot be located and verified in primary sources.
• Any study that cannot be independently accessed, read, and verified.

We use evidence labels — not numerical scores alone — to communicate research strength. A high evidence score requires well-designed human trials, consistent results, and replication, not merely many publications.

• Strong clinical evidence — multiple high-quality randomized controlled trials (RCTs) with consistent results, often supported by systematic reviews.
• Moderate clinical evidence — well-designed RCTs or observational studies with generally consistent results.
• Preliminary human evidence — early-phase trials, small sample sizes, or single studies requiring replication.
• Preclinical evidence — animal or in-vitro studies only; no confirmed human data.
• Mechanistic hypothesis — a plausible biological mechanism is proposed but not yet tested clinically.
• Insufficient evidence — too little research to draw conclusions.
• Conflicting evidence — studies report contradictory results.

When assessing a clinical study, we consider:

• Study design — RCTs rank above observational studies, which rank above case reports.
• Sample size — larger, adequately powered studies carry more weight.
• Population — whether participants reflect the intended user population.
• Endpoints — whether outcomes are clinically meaningful or surrogate markers.
• Duration — short-term efficacy may not predict long-term outcomes.
• Funding and conflicts — industry-funded studies are noted and assessed for potential bias.
• Replication — a single study, no matter how well designed, rarely establishes a clinical fact.

We never present an indirect comparison of results from separate trials as a head-to-head clinical conclusion. Cross-trial observations can be illustrative but are affected by differences in study design, population, dosing, and endpoints. We label all comparisons clearly as:

• Head-to-head evidence — direct comparison within a single trial.
• Cross-trial observations — indirect comparison across different studies (limited reliability).
• Preclinical evidence — animal or in-vitro data only.
• Early-stage human evidence — Phase 1 or small Phase 2 trials.
• Unverified claims — marketing or anecdotal claims without scientific support (flagged as such).

Every significant clinical claim includes a source citation with: the source type, publication or update date, date accessed where applicable, evidence classification, human or preclinical designation, study population, and key limitations. References are listed at the bottom of each compound profile and guide.