Datadory notebook
Health Care Services Data Guide: 23 Datasets (2026)
This health care services data guide maps all 23 datasets Datadory catalogs for the industry — 16 primary and 7 related — anchored by CMS Medicare Care Compare's ~5,400 hospitals and the WHO Global Health Observatory's 3,093 indicators. Every one of the 16 primary sources is free, and 6 expose official APIs.
1,744 datasets. Pick your catch.
The core datasets to know
US facility quality and payments
Medicare Care Compare (Hospitals, Nursing Homes, Home Health, Dialysis) (quality score 9) is the definitive facility-level source: one row per Medicare-certified CCN across all 50 states, DC and territories, published as bulk CSVs on the CMS Provider Data Catalog (236 datasets) alongside a queryable JSON API. The nursing home file alone runs ~9 MB across 99 columns. Refreshes land quarterly in January, April, July and October, historical archives go back to 2019, and the Data Dictionary states the reporting data are commercial delivery terms with attribution appreciated.
Open Payments (Physician Payments Sunshine Act) (score 9) records payments and ownership interests reported by drug and device manufacturers: 15,498,687 rows and 91 columns in the 2024 General Payments detailed file, with program years 2019–2025 published as per-year CSVs plus a JSON data API — 74 cataloged datasets in all. No registration or integration key is required.
Federal discovery and capacity feeds
Shortage geography
Cross-country comparability
WHO Global Health Observatory Data Repository (score 9) serves 3,093 distinct indicator codes for 194 Member States — health workforce, service coverage, mortality, expenditure and risk factors — through a public OData API returning JSON or CSV. WHO Data Portal (data.who.int) (score 8) consolidates country profiles and Triple Billion dashboards (2018 baseline, 2025 projections) behind a public OData relay; its under-five mortality extract spans 1950 to the latest estimate year in an 11.2 MB CSV.
State policy and UK operations
Community corpora and prototyping
Rounding out the 7 related records are neighbors cataloged under their home industries: the NPPES NPI Registry (health care facilities), the Medical Expenditure Panel Survey (life-health insurance) and ClinicalTrials.gov (biotechnology) among them.
Health Care Services data by use case
Each workflow below maps to named sources rather than generic categories:
Who uses this data
Four kinds of teams drive most searches in this niche, and each starts from a different record:
Market researchers and consultants size health markets and build country panels, searching phrases like "health expenditure per country dataset." Three free APIs answer it: OECD Health Statistics (~86 dataflows, 38 member countries), the World Bank's 658 WDI health series for 217 economies and the WHO GHO's 3,093 indicators. Our market researchers page maps the sizing workflow.
Clinic planners and workforce recruiters hunt shortage geography with "hpsa score data by county," answered by HRSA's daily-refreshed HPSA/MUA-P marts (~35 MB, ~66 columns for primary care) and its census-tract boundary shapefiles — the inputs behind Medicare bonus-payment targeting.
How to build a Health Care Services data stack
A practical stack layers six tiers, ordered from facility-level ground truth outward:
Two design notes prevent rework. First, standardize ingestion on CSV — the universal format across all 16 sources — but budget separate parsers for NHS England's XLSX workbooks, HRSA's shapefiles and Hugging Face's Parquet. Second, keep WHO and KFF content out of any commercial redistribution or derivative product unless you clear WHO's permissions form first; their terms are the binding constraints in an otherwise unrestricted slice.
What the numbers say
Every figure below comes straight from the cataloged source records, as of August 2026:
Two of these numbers carry the most planning weight. Sixteen-of-sixteen free means a complete Health Care Services proof-of-concept can run without procurement, and the 15,498,687-row Open Payments file is large enough that per-year CSV loads — not API paging — are the only sensible ingestion pattern for payment-level analysis.
Keep reading
Continue into the connected pages:
| Number | What it counts |
|---|---|
| 23 | pooled datasets Datadory catalogs for Health Care Services (16 primary + 7 related) |
| ~5,400 · ~14,700 · ~12,500 · ~7,500 | US hospitals, nursing homes, home health agencies and dialysis facilities covered by Medicare Care Compare |
| 15,498,687 | rows (91 columns) in Open Payments' 2024 General Payments file |
| 3,093 | indicators in the WHO Global Health Observatory across 194 Member States |
Pick up where this leaves off
Every one of these ships with sample rows before you commit to anything.
Open Payments (Physician Payments Sunshine Act)
HealthData.gov Catalog
data.gov Health Dataset Collection
accrualPeriodicity · describedBy · bureauCode …+1 more
Data.gov – Medical Supplies Catalog Search
HRSA Data Warehouse - Health Workforce Shortage Areas
WHO Global Health Observatory Data Repository
Want rows instead of a pitch? Name the datasets.
API, files, or your warehouse. Daily, weekly, or hourly.
Get a sampleQuestions worth asking
How do I find physician shortage areas by county?
HRSA's Data Warehouse publishes official HPSA and MUA/P designations with scores for primary care, dental and mental health at county, census-tract and facility geography. The primary care data mart alone is a ~35 MB CSV with ~66 columns refreshed daily, and boundary shapefiles run up to ~196 MB per zip.
What data shows payments from drug companies to doctors?
CMS Open Payments discloses manufacturer payments to physicians and teaching hospitals under the Sunshine Act. The 2024 General Payments detailed file alone holds 15,498,687 rows across 91 columns, with program years 2019-2025 published as per-year CSVs plus a JSON data API — all U.S. government commercial delivery terms.
Which sources compare health systems across countries?
Three pair well. OECD Health Statistics covers ~86 health dataflows across 38 member countries via a free SDMX API, the WHO Global Health Observatory serves 3,093 indicators for 194 Member States over OData, and the World Bank's WDI adds 658 health series for 217 economies with CSV and REST access.