A single number for healthcare access — measured, not guessed
Every five miles of drive time can double mortality after a heart attack. Most people find out which side of that line they live on after the ambulance has already been called. CartoChrome puts a 0–100 score on every US ZIP so that question is answered before the emergency.
US ZIP codes scored
Physician profiles
Facilities indexed
Government datasets
Conditions scored per ZIP
Peer-reviewed papers cited
What we saw
80 million Americans live inside a federally designated Health Professional Shortage Area, and almost none of them know it. The data is public. The methodology is published. Nobody had bothered to put it on a map at the scale people actually shop for houses. Source: HRSA Bureau of Health Workforce, 2024
- No way to compare two ZIP codes on healthcare — only crime, schools, and walkability
- Existing tools rank providers or facilities, never the whole system
- County-level data hides a 30-point gap between adjacent ZIPs
- Transportation, insurance, and disability — the barriers that keep providers out of reach — are ignored
What we built
A 0–100 healthcare access score for every US ZIP code, computed from an 8-component Enhanced Two-Step Floating Catchment Area model with a multiplicative SDOH penalty. Eleven score types per ZIP — overall plus ten condition-specific — rendered on a GPU-accelerated choropleth you can pan, zoom, and search.
- ZIP-level — 33,000+ scores, not 3,200 counties
- Eight access dimensions: primary care, ER, hospitals, specialists, mental health, preventive, dental, telehealth
- Six-factor SDOH penalty: insurance, economic, transportation, health literacy, disability, age
- 4M+ provider and 85K+ facility profiles in one index, cross-linked by NPI and facility ID
Why this doesn't exist anywhere else
Healthgrades has providers. CDC PLACES has outcomes. County Health Rankings has counties. Nobody has all of it at the ZIP.
| Dimension | CartoChrome | Others |
|---|---|---|
| Granularity | 33,000+ ZIP-level scores | 3,200 counties or 50 states |
| Methodology | 8-component E2SFCA with SDOH penalty | Provider counts or star ratings |
| Data breadth | 4M+ providers, 85K+ facilities, Census SDOH | Providers or facilities — rarely both |
| Visualization | Interactive choropleth at ZIP resolution | Static county heatmaps or PDF tables |
| Data freshness | 21 federal sources on an automated refresh | Annual report, proprietary licensing |
| Distribution | Free REST API + embeddable widget | Paywalled reports or CSV downloads |
Six rules we refuse to break
Every shortcut on this list was considered — and rejected — because it would have made the score less true.
Show the math
The full formula is published at /methodology/. Every coefficient, every decay function, every data source. If you disagree with a score, you can argue with the equation.
Academic methodology, not a vibe
Scoring uses E2SFCA — Enhanced Two-Step Floating Catchment Area — the peer-reviewed method for spatial accessibility cited in 36 of the 99 papers in our research library. No black-box ML, no proprietary weighting.
Refreshed automatically
NPPES pulls monthly. HRSA shortage areas quarterly. Census ACS annually. No Excel, no interns, no stale data. When CMS publishes, we recompute.
SDOH can only penalize — never inflate
Money does not create doctors. Our SDOH modifier ranges from 0.35 to 1.05, so wealth and high health literacy cannot conjure access where providers do not exist. An earlier industry approach had this backwards. We fixed it.
Every populated ZIP, not just the metros
Coverage extends to all 33,000+ US ZCTAs with resident population. Rural Mississippi gets the same algorithm as Manhattan. A county-level score hides the 20-point gap between two adjacent ZIPs.
Government data only
Twenty-one free federal sources: CMS NPPES, CMS Care Compare, HRSA HPSA, Census ACS, CDC PLACES. Zero paid licenses. Zero Data Use Agreements. Any researcher with Python can reproduce our work.
How we got here
Ninety-nine papers, twenty-one data sources, five million indexed pages. In that order.
Read 99 peer-reviewed papers on healthcare access. Picked E2SFCA as the spine of the score because it is the method most of them cite.
Wired up 21 federal data sources — CMS NPPES, HRSA HPSA, Census ACS, CDC PLACES, and 17 more — into an automated pipeline that refreshes without a human touching a spreadsheet.
Built an 8-component E2SFCA engine with component-specific decay functions (Gaussian for primary care, sigmoid for emergency), a multiplicative SDOH penalty, and a telehealth feedback modifier.
Shipped the interactive choropleth. One map. Every US ZIP, every county, every state. Overall score plus 10 condition-specific scores. Five-million-plus indexable pages.
Opened the REST API and the embeddable widget. Real estate platforms, insurers, and journalists can now pull any ZIP’s score with one authenticated HTTP call.
The research stack
Every weight, decay function, and threshold traces back to a citation.
99
Peer-reviewed papers in the research library
36
Cited directly in the scoring formula
10
Health domains represented
Domains: neighborhood access frameworks, mental health, rural care, surgery, cancer, veteran care, women's health, emergency and trauma, cardiometabolic disease, and infectious disease.
Go check your ZIP
The score is free. The methodology is public. The map opens in one click.