Pharmacy supply chain intelligence
And every hop the molecule takes to get there. Built on federal sources, with the date we read each one attached to every number.
The count
Registry records are not open pharmacies. The distance between those two things is where every pharmacy count you have read goes wrong, so we show the subtraction instead of the answer.
The deactivation cross-check matters. The gap between 111,274 and the figure the trade press uses is not dead registrations, because none of these records are deactivated. It is format mix and co-location, and that is a different problem with a different fix.
What it maps
The dispensing side and the molecule side share an ontology here, which is what makes a shortage traceable to the plant that makes the drug and to the facilities that could fill the gap.
Retail, long term care, specialty, mail order, compounding, nuclear, home infusion, institutional, clinic, managed care, and unspecified. 54.9% of pharmacies hold more than one of these at once, which is why a single category label is always wrong.
Every marketed drug product by NDC, across 5,299 labeler codes, mapped to dosage form, route, and marketing category. Shortages and recalls attach here, refreshed daily from the federal feeds.
Every registered 503B outsourcing facility in the country, with registration history, last inspection, Form 483 status, and whether it compounds sterile preparations from bulk. This is who can actually fill a shortage.
| Registry code | Format | Sites | Share |
|---|
Who it is for
Member level closure telemetry and access maps by state, county, and congressional district. Sourced, dated, and refreshed, so it holds up when somebody reads it into a record.
Site formats, accreditations, and the covered entity to contract pharmacy graph, joined to ownership and geography. Public sources, entity resolved, available as an API.
Distance, density, and closure velocity, with the method written down and the source file cited for every figure. Built to be checked, not just quoted.
Early access
The dataset is loaded and the maps are running. We are talking to a small number of organizations before opening it up. If the work above is close to something you already pay for or do by hand, say so and we will come back with specifics.
One reply from a person, not a sequence. You can also write to hello@bomrx.com.
Method
Everything here comes from federal sources you can open yourself. Each figure carries the file it came from and the date we read it, so the work holds up when somebody audits it.
Licenses and statuses render with the issuing body, the status, the date we observed it, and a link to the underlying record. Facts are appended rather than overwritten, so the history stays intact and a question about last quarter has an answer.
Registry records resolve into real places by identifier first, then address, then name. Every merge keeps the score and the evidence behind it, and a human can overturn one without a migration.
The graph covers dispensing locations, drug products, and the registered plants that make them. It carries no patient, prescription, or claims data, which is a design choice and is enforced in the schema.