Related work
Other people measuring this
This project audits one federal file. That is a narrow slice of the problem, and the people below cover parts of it we do not touch. Their work is worth reading on its own terms, not as a supplement to ours.
Every link here was opened and read, and every quotation is verbatim from the piece it cites. Listing someone is not a claim of endorsement in either direction, and nobody listed has reviewed or approved what we publish.
Defacto Health
Ron Urwongse, Co-founder
Payer network and provider directory data, with an accuracy scoring and monitoring service built on top of it.
What they cover that we do not
We audit the federal directory as published. Defacto works the payer side, which is the half of the problem the federal files do not cover, and has been tracking payer directory API behaviour since 2022. Where we measure what a file contains, they measure whether a patient could actually act on it.
“The value of compliant directory APIs for program integrity is not that directory errors are themselves evidence of fraud. What directory data provides is a map of which providers are enrolled in Medicaid.”
Meets ours at Our § 455.436 revalidation work.
“Directory accuracy is not defined simply by an attestation or by a contract. It is defined by whether a patient can find, book, and see a provider at the place and time the directory states.”
Meets ours at Our REAL Health Providers Act brief.
Meets ours at Payer orgs landed, payer endpoints did not.
Fasten Health
Jason Kulatunga, Founder
A unified medical record API that connects to health systems on a patient’s behalf, plus an open-source personal health record.
What they cover that we do not
They consume this data rather than audit it, which makes them the best evidence of what actually breaks. Their catalogue is built on the SMART User Access Brands specification, so they hit the missing brand layer in the federal directory before anyone measuring the files would notice it.
- Fasten ConnectAug 2026
Meets ours at Why we recommend Brands as the spine, not the NDH.
Where this gets discussed
Venues rather than publications. Most of what we know about what the directory is meant to do, as opposed to what it contains, came out of these rooms.
The CMS initiative the National Provider Directory sits inside, covering the interoperability framework and the patient-facing tools built on it.
Getting inPublic initiative page. CMS has run webinars and a request for information; watch the overview page for the current comment window.
CMS provider directory community
Weekly call, plus an ongoing Slack workspaceThe working group around the National Provider Directory itself: a Slack workspace and a recurring community call where the CMS directory team takes questions from implementers. Most of what this project has learned about intent, as opposed to contents, came from there.
Getting inThere is no public invite link we can point at. Ask the CMS directory team through the bulk download site, or ask anyone already participating to bring you in.
The HL7 work group that owns the National Directory of Healthcare Providers and Services implementation guide, which is the specification the CMS files are built against.
Getting inHL7 work groups are open to attend, and chat.fhir.org is where implementers actually argue about the guide. Cite the published STU1 rather than the continuous build: per its co-author the ballot and CI URLs are not stable references.
Working on this too?
If you publish research on provider or payer directory data and want it listed, open an issue. The bar is that the work is public and that a reader can check it. We will link people we disagree with, and we would rather say where we disagree than leave them out.
Open an issue on GitHub or start from the primer.