HRSA Data Warehouse - Health Center & Facility Finders
Datadory delivers hrsa data warehouse health center facility finders data: every HRSA-supported care site in the United States and its territories as an address-level geocoded row - roughly 1,400 health center awardees and look-alikes operating thousands of service delivery sites, plus Ryan White HIV/AIDS medical providers, organ transplant centers and Migrant/Seasonal Head Start locations, each carrying parent organization, county, latitude/longitude and match-precision metadata, delivered daily, weekly, or hourly.
What is the HRSA Data Warehouse - Health Center & Facility Finders dataset?
The safety net, plotted. The HRSA Data Warehouse - Health Center & Facility Finders turns the federal government's roster of HRSA-supported care sites into addressable records: every Federally Qualified Health Center delivery point, every Ryan White HIV/AIDS medical provider, every organ transplant center in the Organ Procurement and Transplantation network, and every Migrant/Seasonal Head Start location - name, parent organization, street address, ZIP, county, and a geocoded latitude/longitude pair for each.
The scale is the story. HRSA counts roughly 1,400 health center awardees and look-alikes operating thousands of individual service delivery sites, serving about 32.7 million patients - better than one in twelve Americans - with rural programs reaching over 62 million rural residents. Those sites anchor primary care in exactly the places commercial facility databases cover worst: small towns, border colonias, tribal lands, public housing.
Within the Health Care Facilities catalog this is the funded-site layer - the counterpart to hospital censuses and commercial provider lists, answering a different question: not what exists, but what the federal safety net puts where. Get a sample of this dataset cut to your states, counties or programs.
What does a sample row look like?
One site, thirteen identifying facts. Exactly as it ships:
Id : 7311
CtrNm : COMMUNITY HEALTH CENTERS BROADWAY DENTAL
ParentCtrNm : COMMUNITY HEALTH CENTERS OF THE CENTRAL COAST INC
CtrAddress : 210 N Broadway
CtrCity : Santa Maria
CtrStateAbbr : CA
CtrZipCd : 93454-4415
CountyNm : Santa Barbara County
Latitude : 34.95448901
Longitude : -120.43560803
Distance : 0.4
LocNameDesc : Data was matched at the Address level using AGOL
DwRecordCreateDt : 2026-08-21Read it as three clusters. Identity: CtrNm names the delivery site while ParentCtrNm names the grantee behind it - a dental satellite of Community Health Centers of the Central Coast Inc., which is how multi-site organizations actually behave when you count them. Place: street address through ZIP+4 and county, so the row joins cleanly to census geography, demographic overlays or your own territory map. Geocode and provenance: coordinates to seven decimal places, LocNameDesc stating openly that the match happened at the address level rather than a centroid guess, and DwRecordCreateDt stamping when the warehouse last touched the record.
The same thirteen columns repeat for every site in every program family, which is why a California dental clinic and an Alaska village clinic land in one query without special-casing.
What fields does the dataset include?
Thirteen documented columns define the core site record, each definition checked against the published dictionary during the August 2026 research pass. The design is deliberately flat: identifiers, human-readable names, postal geography, a point location, a match-quality statement and a record date. Nothing needs unpacking before it loads into a map layer or a CRM import.
Two conventions worth knowing before you build. Site and parent names arrive uppercase and unabbreviated - COMMUNITY HEALTH CENTERS BROADWAY DENTAL, not "Comm HC Brdwy Dental" - so fuzzy-matching against vendor lists is a normalization problem you control, not a mystery. And Distance is computed relative to the query point rather than stored, which makes radius studies reproducible from coordinates alone.
What does coverage look like across geography, time and granularity?
Geography - United States and its territories: all 50 states, the District of Columbia, Puerto Rico and the U.S. territories. Sites geocode to address points and carry county names, so both point maps and county-level joins work off the same rows.
Temporal - the facility roster reads as a current snapshot of the standing universe of HRSA-supported sites, with per-record creation dates attached. Program-side history lives in adjacent collections: grants dashboards spanning fiscal years 2016-2025, and Uniform Data System annual awardee reporting reaching back to 1999.
Granularity - the individual service delivery site, not the grantee rollup: one row per door that patients walk through, geocoded. Program aggregates run alongside at national, state, territory and county levels for when the site cut is more detail than the question needs.
How is the data delivered?
API, files, or your warehouse. Daily, weekly, or hourly.
Your cadence is your call regardless of how often the underlying roster moves - most teams find a scheduled warehouse load keeps their site layers and CRM exports current without anyone babysitting the pipeline. Every delivery ships the complete thirteen-field dictionary above, the sample rows, and the coverage profile mapped to the states, counties and programs you named.
Who uses this data, and for what?
- Site selection and network planning - health systems, FQHCs and retail clinics screen counties for gaps by plotting funded sites against population and drive times before committing to a location.
- Referral routing - care coordination and navigation platforms resolve a patient's coordinates to the nearest HIV/AIDS medical provider or sliding-scale primary care door, with distance computed rather than guessed.
- Grant writing and needs assessment - applicants document existing supply in a target area with federal records instead of hand-counted lists, then position the proposal against the gap.
- Territory building for healthcare sales - med-device, pharma and EHR teams segment thousands of community health sites by state, county and parent organization before the first outreach email.
- Geospatial machine learning - point-layer features (nearest funded site, site density per square mile, distance to safety net) engineered straight from coordinates into access-to-care models.
- Emergency and public-health response - incident commands overlay funded-site points on flood, fire and outage footprints to see which doors go dark and where capacity shifts.
- Care-desert journalism - reporters count sites per capita and cite the federal source when a county loses its only clinic.
Which personas get the most value?
Market researchers and consultants size access markets from funded-site density crossed with demographics - see market researchers use cases. Data scientists and ML engineers get a clean point layer with match-precision metadata already attached - see data scientists use cases. Sales and growth teams turn the roster into segmented FQHC target lists keyed on parent organizations - see sales growth teams use cases. Competitive-intel and product teams watch the site universe for newly listed delivery points entering their markets - see competitive intel use cases. Journalists, academics and students get citation-grade federal sourcing for every care-access claim.
Which datasets pair with this one?
- HRSA Data Downloads - AHRF, Health Center Sites, HPSA & Transplant - the bulk-file view of the same universe plus county provider-supply tables; the head-to-head with NPPES shows when each answers the provider-location question.
- HRSA Health Center Program UDS Data 2025 - patient volumes, staffing and performance for the awardees behind these site rows.
- NPPES NPI Registry - individual-clinician identifiers to hang off the facility layer.
- American Hospital Association organizational hub - the hospital-system counterpart for teams covering both ends of the care continuum.
Field dictionary
Every field below is documented against real records. The full dictionary ships with the sample.
| field | type | definition | example |
|---|---|---|---|
Id | integer | Locator record identifier for the site. | 7311 |
CtrNm | string | Name of the health center or provider site. | COMMUNITY HEALTH CENTERS BROADWAY DENTAL |
ParentCtrNm | string | Name of the parent or grantee organization operating the site. | COMMUNITY HEALTH CENTERS OF THE CENTRAL COAST INC |
CtrAddress | string | Street address of the service delivery site. | 210 N Broadway |
CtrCity | string | City of the service delivery site. | Santa Maria |
CtrStateAbbr | string | Two-letter state abbreviation of the site. | CA |
CtrZipCd | string | ZIP code, extended to ZIP+4 where available. | 93454-4415 |
CountyNm | string | County name where the site is located. | Santa Barbara County |
Latitude | number | Geocoded latitude of the matched site address. | 34.95448901 |
Longitude | number | Geocoded longitude of the matched site address. | -120.43560803 |
Distance | number | Distance in miles between the query point and the site; computed relative to the query rather than stored. | 0.4 |
LocNameDesc | string | Statement of how the address was geographically matched, e.g. at the address level versus an area centroid. | Data was matched at the Address level using AGOL |
DwRecordCreateDt | date | Date the record was created or last refreshed in the warehouse. | 2026-08-21 |
Questions buyers ask
How many health centers are in the data?
Roughly 1,400 health center awardees and look-alikes operate thousands of individual service delivery sites, serving about 32.7 million patients nationwide. The row count reflects sites, not grantees - a multi-site organization contributes one row per delivery door.
Which provider programs does the dataset cover?
Four families: Health Center Program delivery sites including FQHCs and look-alikes, Ryan White HIV/AIDS medical providers, organ transplant centers in the OPO network, and Migrant/Seasonal Head Start center locations. Each carries the same core site schema.
How precise are the site coordinates?
Each row carries latitude and longitude plus a LocNameDesc field stating how the address was matched - for example, matched at the address level. You can filter or flag rows by match precision before they reach a map layer, so low-confidence geocodes never silently distort a distance calculation.
Does each row distinguish a site from its parent organization?
Yes. CtrNm identifies the specific delivery site and ParentCtrNm identifies the grantee operating it, so rollups by organization and counts by location come from the same table without a separate crosswalk.
Can a sample be cut to specific states, counties or programs?
Yes. Name the geography and the program family - Ryan White providers in the Southeast, transplant centers within set radii, health center sites in three counties - and the sample arrives shaped to that scope with the full field dictionary attached. Samples precede any commitment.
See the rows before you pay anything.
Name this dataset and we send real records from it — scoped to the fields you asked for.