23 health outcome measures — obesity, diabetes, hypertension, mental health, insurance coverage, and more — for every US ZIP code. Part of the EnrichZip dataset — all 33,000+ US ZIP codes, instant download.
How to use this data
Healthcare Site Selectors
Identify underserved markets by combining health need (high chronic disease rates) with access gaps (low provider density). A ZIP with 18% diabetes prevalence and few endocrinologists is a clear opportunity for a specialty practice or urgent care clinic.
Health Insurance Actuaries
Price risk more precisely with actual health outcome data at the ZIP level. CDC PLACES gives you chronic disease prevalence, behavioral risk factors, and preventive care utilization — far richer inputs than demographic proxies alone.
Public Health Researchers
Analyze geographic health disparities, model social determinants of health, and benchmark community health outcomes without building your own data pipeline from CDC's raw census tract files.
AI Health Applications
Feed structured, clean health outcome data directly to AI models to power patient matching, risk stratification, or care gap identification by geography. Our first customers include AI platforms connecting patients with providers.
Columns Included
Source: Centers for Disease Control and Prevention, PLACES Program. All columns are pre-joined and ready to use — no API key or GIS software required.
| Column Name | Description |
|---|---|
cdc_obesity_pct |
% adults with obesity |
cdc_diabetes_pct |
% adults with diagnosed diabetes |
cdc_hypertension_pct |
% adults with high blood pressure |
cdc_heart_disease_pct |
% adults with coronary heart disease |
cdc_stroke_pct |
% adults who have had a stroke |
cdc_asthma_pct |
% adults with current asthma |
cdc_copd_pct |
% adults with COPD |
cdc_cancer_pct |
% adults with cancer (excluding skin cancer) |
cdc_depression_pct |
% adults with depression |
cdc_mental_health_pct |
% adults with poor mental health 14+ days/month |
cdc_physical_health_pct |
% adults with poor physical health 14+ days/month |
cdc_smoking_pct |
% adults who currently smoke |
cdc_binge_drinking_pct |
% adults who binge drink |
cdc_no_exercise_pct |
% adults with no leisure-time physical activity |
cdc_no_insurance_pct |
% adults without health insurance |
cdc_no_checkup_pct |
% adults with no routine checkup in past year |
cdc_no_dental_pct |
% adults with no dental visit in past year |
cdc_sleep_disorder_pct |
% adults with insufficient sleep |
cdc_disability_pct |
% adults with any disability |
cdc_food_insecurity_pct |
% adults experiencing food insecurity |
cdc_housing_insecurity_pct |
% adults experiencing housing insecurity |
cdc_loneliness_pct |
% adults reporting loneliness or social isolation |
cdc_high_cholesterol_pct |
% adults with high cholesterol |
About This Data
Health outcome data by ZIP code is one of the most sought-after datasets in healthcare analytics, but it's also one of the hardest to work with in its raw form. CDC PLACES publishes estimates at the census tract level — over 85,000 tracts across the US — which requires significant GIS work to aggregate to ZIP codes and join with other data sources.
EnrichZip pre-processes this for you. We map all 85,000 census tracts to their corresponding ZIP codes, weight by population, and produce a single clean row per ZIP covering 23 health measures. You get the analytical value of PLACES without the data engineering overhead.
The 23 measures span chronic disease prevalence (diabetes, hypertension, heart disease, COPD, cancer), behavioral risk factors (smoking, obesity, physical inactivity, binge drinking), preventive care utilization (checkups, dental visits, health insurance coverage), and social determinants (food insecurity, housing insecurity, loneliness). Together they give a comprehensive picture of a community's health burden and care needs.
For healthcare organizations, the combination of high chronic disease prevalence and low preventive care utilization is a particularly powerful signal. A ZIP where 20% of adults have diabetes but 35% haven't had a checkup in the past year represents both unmet need and a potential patient acquisition opportunity for a primary care or specialty practice.
AI-powered healthcare applications are an increasingly important buyer of this data. Patient matching platforms, care gap identification tools, and population health management systems all benefit from clean, structured health outcome data at the geographic level. Because EnrichZip data is already clean and documented, it can be fed directly into AI models without preprocessing — dramatically accelerating time to insight.
FAQ
CDC PLACES (formerly 500 Cities) is a CDC program that uses small area estimation methods to produce local health outcome estimates at the census tract level for all US localities. The estimates are derived from Behavioral Risk Factor Surveillance System (BRFSS) survey data, combined with census demographics and predictive models. EnrichZip maps these tract-level estimates to ZIP codes using Census crosswalk files, producing a single health outcome row per ZIP code.
CDC PLACES estimates are modeled, not measured — they're statistical predictions based on BRFSS survey responses calibrated to census demographics. For large ZIP codes they're highly reliable; for very small or rural ZIPs with few survey respondents, estimates have higher uncertainty. CDC publishes confidence intervals at the tract level; our ZIP-level aggregates are weighted by population. Treat the data as a population-level signal, not a precise clinical measure.
Our dataset uses CDC PLACES 2023 release data, which incorporates BRFSS surveys from 2020-2021. CDC releases updated PLACES data annually. We rebuild our dataset when new PLACES data becomes available.
Yes — this is one of the primary reasons customers buy our data. The clean, structured format (one row per ZIP, consistent column names, documented definitions) makes it immediately usable with AI tools like Claude or GPT. You can paste a ZIP-level health profile into an AI and immediately ask questions like 'which of these markets has the highest need for mental health services?' without any data cleaning or preprocessing.