Open Payments recipient validation that proves every NPI before the federal file
Every recipient matched to the right NPI and licence, with the evidence
See one case, screen by screen ↓Checking that every payment or transfer of value in the federal Open Payments report is attributed to the right covered recipient — the right NPI, licence state and recipient type. Open Payments Recipient Validation does it with six agents that find and score every candidate, cite the evidence and route the uncertain matches to a transparency analyst.
How to choose AI solutions for life sciences →
Updated
Every payment or transfer of value to a US physician, non-physician practitioner or teaching hospital has to reach the federal Open Payments report against the right covered recipient: the right NPI, licence state and recipient type. The payments arrive from the expense, events, contract, research and grants systems with the name as a rep wrote it — and “Dr. J. Smith” fits three people in one city. For the records rules cannot settle, a transparency analyst searches the NPI registry and the state board by hand and pastes screenshots into a spreadsheet.
The risk sits in the calls nobody wrote down. Since program year 2021 physician assistants, nurse practitioners, clinical nurse specialists, CRNAs and certified nurse-midwives are covered too, so a sign-in that says “RN” or “Midwife” can hide a reportable payment. Speaker programs peak in the third and fourth quarters, so the backlog grows just as the year closes. And in April every recipient can review and dispute their record — a match without a recorded reason becomes rework every spring.
Estimated split: 6–12 minutes a match by hand today, against the 2.6-minute median the solution targets for a person's decision.
Six specialist agents find, compare and cite every candidate recipient; a transparency analyst confirms each uncertain match.
Northwind Bio ran a Corventa heart-failure advisory board in Columbus on Sep 12, 2026. The consulting agreement is signed “Dr. J. Smith”, and nine payment records across three systems — $14,860 — have no NPI. Here is what happens next, screen by screen, in the working solution.
Lena has 15 exceptions assigned to her; the team has 412 matches unresolved across 3,068 records, and the window opens Feb 1 — 117 days out. At the top of her list: M-2039, an advisory board contract, 9 records, $14,860, due Oct 9. The agents recommend Jennifer L. Smith, MD, NPI 1427093318, at a match score of 92%.
“Below the 0.95 line a person decides.”
The name and address normaliser kept the initial “J.” as a partial first name and set “Dr” aside. The registry matcher found 41 Smiths in Ohio and 3 with the initial J in Columbus. The licence check and the address check each point to candidate A, the covered-recipient classifier reads specialty code 207RC0000X as a physician, and the evidence writer writes the reason with numbered citations.
Score 0.92 is below the 0.95 auto-match line, so the match is routed to Lena Ortiz.
Each field of the payment record sits beside each candidate, marked agrees, variant or differs. The name and the credential cannot decide it. The practice address and the licence can: 2200 Riverside Dr and OH 35.098214 agree with candidate A and differ for James R. Smith, an internist on Knightsbridge Blvd, and Julia M. Smith, an adult-health nurse practitioner.
“Her practice is 2200 Riverside Dr, Suite 300 — the address on the contract.”
Lena clicks the licence and the evidence panel opens the state licence lookup: Jennifer Lynn Smith, MD, licence type Physician (MD), number 35.098214, active, expiring Jul 1, 2027, at 2200 Riverside Dr, Ste 300, Columbus. The consulting agreement CA-26-0918, the registry records and the HCP master record HM-140 377 are one tab away.
“Looked up live at the board or registry; the lookup is kept with the decision.”
Lena confirms the match. The NPI is written onto all 9 records in consulting contracts, speaker programs and expenses & meals. A master-data fix — link licence OH 35.098214 and NPI 1427093318 to HM-140 377 — goes to Dana Okafor. The profile is pre-validated against the federal profile rules with 0 errors and added to the submission file, program year 2026, batch 41.
“9 records across 3 systems now carry NPI 1427093318 · physician. Decided by Lena Ortiz.”
Her decision takes the count to 411 matches across 3,059 records. In the last 30 days the team decided 36.8 matches a day while 39.7 a day were opened. At today's pace the last match is decided on Jan 18 — 10 days after the Jan 8 year-end pre-validation. Expenses & meals holds 168 of the unresolved matches, speaker programs 131.
What closes the gap: “4 more decisions a day — about one more per reviewer each day, starting now.”
Her list is sorted by due date, with the reason each payment needs a person: a name variant, two people who fit, a credential unclear, an NPI deactivated, a payment to a practice or a foundation. Each row shows the source systems, the records and value behind it, the agents' confidence and the stage. A foundation grant and a duplicate master record also need a second person's sign-off.
M-2047 is a fellowship grant paid to Lakeview University Medical Center Foundation, a separate charity with its own tax ID. The grant agreement directs the funds in full to the hospital for two fellowship positions, and the hospital is on the program-year 2026 teaching-hospital list. The agents recommend reporting it to the teaching hospital, at 81% — and teaching hospitals are never auto-matched.
“This kind of call needs Omar Haddad's sign-off.”
Last night five source systems sent 4,812 payment records. The registry matcher found a candidate for 4,790 and looked up 298 licences; the evidence writer wrote 1,206 citations. 4,671 records were matched on their own and added to the submission file with 0 profile errors; 141 went to reviewers, 15 of them to Lena.
Matched on their own: score ≥ 0.95, no rule hit.
When a recipient disputes a payment in the review period, the recipient's profile holds every record behind it, by nature of payment and by month, with who matched it and when. Dr. Daniel Osei's 39 records, $31,800, were matched by Lena Ortiz on Oct 2 and his Georgia licence verified Oct 5. One click exports the evidence pack.
Each decision keeps its evidence for at least five years after publication.
The same payment, seen by the four people who carry it — what their reporting year looked like, and what it looks like now.
Six specialist agents find, compare and cite every candidate recipient; a transparency analyst confirms each uncertain match.
Pulls the nightly payment batch from the expense, events, accounts-payable, grants and clinical-payment systems, labelling each line's source system, nature of payment, product and amount, with the recipient kept exactly as captured.
Splits names, sets credentials aside, records nicknames and former names as variants, and standardises addresses — suites and ZIP+4.
Finds every plausible candidate in the NPI registry, state licence lookups, the HCP master and the teaching-hospital list, and scores name, credential, specialty code, address, licence and NPI separately — the licence number before the name when both exist.
Decides physician, physician assistant, nurse practitioner, clinical nurse specialist, CRNA, certified nurse-midwife, teaching hospital or not covered — from the registry specialty code and the licence type, not the credential the business typed.
Writes the short reason for each match, in 3–5 sentences, with numbered citations to the exact registry, licence and source passages, and says what the person must decide.
Writes confirmed recipients into the federal file layout and pre-validates NPI, licence and name against the federal profile file, reporting any profile error back to the exception list.
Decides each uncertain match. The agents propose; a named person decides.
Ask about any recipient, exception or source system — or say which matching rule to add. Every answer points to its source.
Why Jennifer L. Smith and not the other two?
Because the licence and the address both point to her. The contract gives Ohio licence 35.098214; the Ohio board lists that licence for Jennifer Lynn Smith, MD, and her practice is at 2200 Riverside Dr — the address on the contract. James R. Smith is an internist on Knightsbridge Blvd; Julia M. Smith is a nurse practitioner. The name alone could not decide it, so you confirm. Confirming fixes 9 records worth $14,860.
How ready are we for the February submission?
Close, but not yet on track. 117 days until the window opens on Feb 1, 2027; files must be attested by Mar 31. 412 matches are unresolved; at today's pace the last is decided on Jan 18, 10 days after year-end pre-validation — 4 more decisions a day closes the gap.
Which practitioners count as covered recipients?
Besides physicians and teaching hospitals, five non-physician practitioner types have been covered since program year 2021: physician assistants, nurse practitioners, clinical nurse specialists, certified registered nurse anesthetists and certified nurse-midwives. Registered nurses and pharmacists are not covered. Two of your open exceptions are credential mix-ups: an “RN” who is a CRNA and a “Midwife” who is a CNM.
How do you handle teaching hospitals?
Teaching hospitals are matched only against the CMS list for the program year — never auto-matched. In M-2047 the payee is the hospital's foundation, a separate charity, but the grant agreement sends the funds to the hospital's fellowship, and the hospital is on the 2026 list. The agents recommend reporting it to the hospital; Omar Haddad signs off.
13 screens from the working solution, on its sample data. Pick one to see it large.
How the agents match, who signs off and where the reference data comes from are settings, on the audit trail.
| Setting | Default | Choose from |
|---|---|---|
| Auto-match line | 0.95 — the line the compliance committee approved | 0.90 · 0.95 · 0.98 |
| Send master-data fixes to the steward | On — to Dana Okafor | On · Off |
| Licence before name | On | On · Off |
| Teaching hospitals always need a person | On | On · Off |
| Confirm credential fixes on their own (RN → CRNA) | Off | On · Off |
| Sign-off: teaching hospitals and foundations | Omar Haddad | Omar Haddad · Dana Okafor · Sam Patel |
| Sign-off: not a covered recipient | Omar Haddad | Omar Haddad · Dana Okafor · Sam Patel |
| Sign-off: duplicate master records | Dana Okafor | Omar Haddad · Dana Okafor · Sam Patel |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample data · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures and regulation (42 CFR 402.105 — civil money penalties, before inflation adjustment; 42 CFR Part 403 Subpart I, §403.908 — the recipient review and dispute period; CMS Open Payments program-year 2023 data overview). People, companies, products, NPIs and addresses named on this page are fictional — characters and sample data in the working solution.
Checking that every payment or transfer of value in the federal Open Payments report is attributed to the right covered recipient — the right NPI, licence state and recipient type. Open Payments Recipient Validation does it with six agents that find and score every candidate, cite the evidence and route the uncertain matches to a transparency analyst.
The registry matcher scores every candidate from the NPI registry, state licence lookups, the HCP master and the teaching-hospital list, field by field: name, credential, specialty code, address, licence and NPI. A licence number is matched before the name. At 0.95 or above, with exact NPI, licence and name and no rule hit, the match is confirmed on its own; below it, a person decides with the candidates side by side.
Yes. The covered-recipient classifier decides physician, physician assistant, nurse practitioner, clinical nurse specialist, CRNA, certified nurse-midwife, teaching hospital or not covered from the registry specialty code and licence type — not the credential typed on a sign-in. A “not covered” decision always needs a sign-off.
They are matched only against the CMS teaching-hospital list for the program year and are never auto-matched. When a grant is paid to a hospital's foundation, the agents read where the agreement directs the funds and recommend the reportable entity; a named person signs off.
No. The agents read the source systems, the registries and the HCP master and never edit them. Credential, address and duplicate fixes found while matching go to the master-data steward as proposals, with their evidence.
Yes. Below the auto-match line a transparency analyst decides every match, the riskiest calls need a second person's sign-off, and only confirmed records enter the submission file. Every decision is on the audit trail with its evidence.
The submission clock counts the unresolved matches and records, the days to the window, the decisions made and opened each day, and the date the last match is decided at today's pace — with the extra decisions a day that clear it by year-end pre-validation.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your matching policy, your source system extracts and a sample of past payment records — and it goes live once every quality gate has passed. We will walk you through it on your own data first.
Every speaker program scored on the OIG's red flags, before and after it runs.
1-min demoThird-Party Due DiligenceEvery third party risk-tiered before the contract is signed, from its real owners.
1-min demoCompliance Investigation Case ManagementInvestigations close in a median of 18 days, every finding cited.
1-min demoWe'll run Open Payments Recipient Validation on a sample of your own past payment records.