get_open_payments
Returns CMS Open Payments records — the Sunshine Act database of every payment / transfer of value from drug + device manufacturers and GPOs to US physicians, non-physician practitioners, and teaching hospitals (~15M records per program year, 2019→present). LIVE passthrough to CMS's own API: results reflect CMS's current data and total_count is CMS's authoritative count for the filtered query (the results array is just the requested page). Use this when the user asks about: pharma/device money to doctors, a company's physician-payment footprint, speaker-fee / consulting / royalty programs, industry funding of research (with ClinicalTrials.gov IDs), or physician ownership stakes in manufacturers. payment_type selects the dataset (schemas differ; rows are CMS's fields verbatim): general (default) — meals, travel, consulting, speaker fees, royalties, honoraria. Fields incl. nature_of_payment_or_transfer _of_value, name_of_drug_or_biological_or_device_or_medical supply_1, covered_recipient_specialty_1. research — research payments incl. name_of_study, clinicaltrials_gov_identifier, preclinical_research_indicator. ownership — physician ownership/investment interests (total_amount_invested_usdollars, value_of_interest, terms_of_interest); recipient fields are physician*. summarize=true returns CMS's own pre-aggregated per-(company, nature) totals for the year — transaction counts + dollar totals per payment nature. The nature codes in that dataset ship without a public CMS legend; KeyVex labels the seven codes it has VERIFIED by exact count+total reconciliation against detail data (1=Consulting Fee, 2=Speaker/faculty compensation, 6=Food and Beverage, 7=Travel and Lodging, 9=Charitable Contribution, 10=Royalty or License, 14=Grant); unverified codes pass through with an empty label rather than a guess. Matching: company is a substring (matches subsidiaries: 'pfizer' catches PFIZER INC.); recipient names are EXACT (CMS stores uppercase; we uppercase for you); npi is the exact National Provider Identifier — the precise join key to get_oig_exclusions. Payments are attributed to the manufacturer AS FILED — no ticker/CIK; try the operating-company name. Program years: 2019 through the latest published year (CMS refreshes semiannually; year defaults to the latest). Pagination: limit ≤ 500 per page (CMS cap), use offset for more. Pure-publisher posture: CMS's records as filed, parsed by KeyVex (the source record is authoritative) — no derived influence scores. Disclosure ≠ wrongdoing; these are lawful, statutorily-disclosed payments.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| npi | No | Recipient National Provider Identifier (exact, 10 digits). Precise join key to get_oig_exclusions. | |
| year | No | Program year (e.g., 2023). Default: latest published year. | |
| limit | No | Records per page. Default 25, max 500 (CMS page cap). | |
| state | No | Two-letter recipient state code (e.g., 'TX'). | |
| nature | No | Substring against nature of payment (general only; e.g., 'consulting', 'speaker', 'royalty', 'food'). | |
| offset | No | Pagination offset into the filtered result set. | |
| company | No | Substring against the paying manufacturer / GPO name (e.g., 'pfizer', 'medtronic'). | |
| product | No | Substring against the associated drug / device name (general only; e.g., 'eliquis', 'ozempic'). | |
| summarize | No | true = CMS's pre-aggregated per-(company, nature) yearly totals instead of individual payments. | |
| payment_type | No | Dataset: general (default — meals/consulting/speaker fees), research, or ownership (physician stakes). | |
| recipient_last_name | No | Recipient physician / practitioner last name (exact, case-insensitive). | |
| recipient_first_name | No | Recipient first name (exact, case-insensitive). |