Editorial standards

Every page is built from a claims ledger: each factual statement is entered as a discrete claim, matched to a primary source fetched on a recorded date, and stored with the exact supporting language and an evidence grade (label statement, human RCT, human observational, systematic review, animal, in vitro, or documented absence). Statements about methylene blue's approved status come only from FDA records and the current ProvayBlue prescribing information (SPL version 28, effective 2025-05-28). Study findings are reported with the species they were observed in, and a human-trial chip is never placed on animal or laboratory data. No physician or clinician has reviewed this site, and we do not claim medical review; what we claim is that every number on the site resolves to its cited source, and the ledger that proves it is machine-readable.

Allowed sources: FDA (labels, Drugs@FDA, Drug Safety Communications), DailyMed, ClinicalTrials.gov, PubChem, NCBI Bookshelf, and PubMed-indexed peer-reviewed journals. Not allowed: supplement vendors, influencer content, wellness blogs, press releases, and secondary aggregators. Where the honest answer is an absence (no approved oral product, no chronic-use trial, no head-to-head comparison), we cite the authoritative query or review that documents the absence and say so plainly. Two 2011 FDA safety communications that FDA has since removed from its live site are cited as archived captures of the original fda.gov pages, and every medical claim they support is co-anchored to the current label.

Each page displays its published and last-updated dates. We re-review this compound's pages when any of the following changes: the ProvayBlue label (new SPL version), FDA safety communications or compounding lists, or publication of a randomized human trial that would change a conclusion. Scheduled re-checks include re-resolving every citation link. Material changes get a dated note on the affected page.

If we got something wrong, we want it gone. Use the corrections link in the site footer and include the page URL and the claim you dispute. Corrections are made on the page with a dated note; claims that fail re-verification are removed from the ledger and from every surface that rendered them. We treat a broken citation link as a defect, not a cosmetic issue.

How our numbers are made: the '38 sources' figure is the length of the numbered references list rendered from the monograph citation manifest (29 peer-reviewed, 9 government, a census computed from the same array). The '22 studies / 18 human' figures are the row count and species-column count of the study-results table. The label statistics (92.9% methemoglobin response; adverse-reaction rates) are quoted from the current ProvayBlue SPL and were not computed by us. Any internal count shown on this site must be recomputable by script from the page that displays it; if an array changes, the displayed number changes in the same update.