Editorial standards

The core monograph, evidence table, and claim manifest are built from a claims ledger matched to primary sources and graded by study design, species, tested substance, and research provenance. No named physician or clinician has reviewed this site, and we do not claim medical review. The generated guide archive and interactive tools have not completed the standard required for high-risk medical publication; they are visibly labeled, set noindex, and excluded from sitemaps and AI citation files until individually reviewed and approved.

We cite primary sources: peer-reviewed literature via Europe PMC and PubMed Central, and government records from FDA, ClinicalTrials.gov and PubChem. Of this site's 45 references, 86.7 percent are peer-reviewed journal articles or government records; the remainder are one clearly labeled preprint, cited only as corroboration of an absence, and this site's own archived search results, which are shipped so anyone can recompute them. Every reference URL is checked for a 200 response at build time. We do not cite vendor pages, supplement blogs or press releases as evidence. When a document does not contain what it is widely said to contain, we say that rather than paraphrasing it into shape: FDA's July 2026 briefing package has an FDA Evaluation heading for epitalon with no text beneath it, so we quote the proposal and attribute no reasons the document does not state.

Pages carry a visible last-updated date. The regulatory sections are rebuilt whenever FDA acts: the July 2026 advisory committee proposal is a recommendation and the rulemaking is pending, so that section will change and we will date the change. The provenance census is re-run against the same archived query so that a changed number is always a changed literature and never a changed method. If an independent laboratory publishes an in vivo replication, or any human study of epitalon rather than the extract, that is the update this page most wants to make.

If you find an error, tell us and we will fix it and say what changed. Corrections are made on the page itself with the date, not silently. We are particularly interested in hearing from researchers who can point us to studies outside the originating network, or to any human pharmacokinetic data, since both are recorded here as absences.