CDC PLACES: Local Data for Better Health
Centers for Disease Control and Prevention (CDC)
Modeled rates of diabetes, depression, and smoking for every census tract and ZIP code in the US.
Small-area estimates of chronic disease, health risk behaviors, preventive care, disability, and social needs for counties, places, census tracts, and ZIP code tabulation areas. CDC builds them by modeling BRFSS survey responses with census population data. It is the only free source of tract-level health measures for the whole country.
Access
- Cost
- FreeNo charge; downloadable from data.cdc.gov with a Socrata API.
- License
- Public domainPublic domain (U.S. government work)
- Sign-up
- No sign-up
- How you get it
- API and Bulk downloadCSV, GeoJSON, shapefile
What's in it
- One row is
- One row per geographic unit and measure, with a modeled prevalence and confidence interval
- Covers
- All US counties, places, census tracts, and ZCTAs
- Years
- 2016-present (as 500 Cities until 2019)
- Updated
- Annual
- Size
- Roughly 40 measures for about 70,000 census tracts and 30,000+ ZCTAs
Quality
Methodologically documented and widely used, but every value is a model output, so tract differences mostly reflect demographics rather than measured local behavior.
- Estimates are modeled from demographics, not measured locally
- Do not use for trends because methods and inputs change
- Most recent release was August 2024 at time of check
Connects to
- Behavioral Risk Factor Surveillance System (BRFSS): The survey responses PLACES models from
- County Health Rankings & Roadmaps: Uses several PLACES county measures
Questions it could answer
- Which census tracts have the highest modeled rates of both diabetes and lack of insurance?
- How does modeled depression prevalence vary across a city's neighborhoods?