Medivalent

No. Medivalent is a reference tool. It tells you what brand a medicine is sold under in another country and what the active ingredient is — that’s information, not advice. Always confirm with a pharmacist before substituting any medication.

Every result is the output of a deterministic SQL join over public government drug registries — RxNorm (US), NHS dm+d (UK), ANSM BDPM (FR), AEMPS CIMA (ES), CBG-MEB (NL), Health Canada DPD, and others. We canonicalize the active ingredient name to the WHO INN so the cross-country join works regardless of how each country spells it (acetaminophen ↔ paracetamol, albuterol ↔ salbutamol, paracétamol ↔ paracetamolo, etc.).

Tell us — open an issue at github.com/medivalent/medivalent/issues. We’ll trace the bad row to its registry source. If the registry itself was wrong (does happen), we file a correction with them. If our pipeline mis-canonicalized, we fix the canonicalization. Either way, the audit trail starts with the source link on the row.

Every product row carries a source_version reflecting the registry’s release date when we ingested it. Most registries publish weekly to monthly. Our ingest workflow refreshes whenever triggered (manually via GitHub Actions for now). We’ll move to scheduled refreshes once the cadence per source stabilizes.

Because the source registry didn’t carry an OTC vs prescription classification, and we won’t guess. RxNorm (the US data source) notably doesn’t include this field. Where openFDA’s NDC monograph index has a verdict, we use it. Otherwise the badge says unclassified rather than silently rounding to OTC.

On the /browse directory, when a country has no OTC/Rx classification at all (currently the US — see above), we relax the filter to include all products of the molecule rather than show an empty page. The molecule-level otc_certain: falseflag tells the UI this happened. It’s a stopgap; we’re running per-substance OTC enrichment from openFDA to drop the relax.

Coverage varies by country. Some registries publish complete bulk downloads (UK dm+d, Canada DPD, Spain AEMPS, France ANSM, Netherlands CBG-MEB, Switzerland Swissmedic, Brazil ANVISA, Ireland HPRA — those are 80k+ products combined). Others are auth-walled or PDF-only and we maintain hand-curated stubs of the most-recognized brands (Italy AIFA, Germany BfArM, Saudi SFDA, UAE MOHAP, Mexico COFEPRIS, Singapore HSA, New Zealand Medsafe). Per-country status is on /sources.

No. Medivalent has no commercial relationships with any retailer. The chips on a result page (Albert Heijn, Boots, Walgreens, etc.) are simple search-URL constructions that take you to that retailer’s own search results for the brand. We don’t earn commission, we don’t track clicks, we don’t have inventory data. They’re there because finding the medicine is the user’s actual end goal, not learning what it’s called.

Where we surface verbatim manufacturer warnings, contraindications, or excipients, that text comes from FDA-approved labels via openFDA — which is US-only and English. EU registries provide structured product data (brand, strength, form, marketing holder) but their full label text lives in PDFs we don’t mirror. For NL and FR products we do link to the official PDF leaflet on the regulator’s own site (CBG-MEB and ANSM respectively).

Not today. The site is a normal web app served from Cloudflare’s edge; results require a query to our API. A native iOS app with offline cache is on the long-term roadmap.