Is Your County-Level Data Hiding Critical Disparities?
The numbers look fine from a distance, but a very different picture emerges when you zoom in.
Imagine your county’s health report shows an average life expectancy of 78 years. At face value, that’s a reasonable number (close to the national average, nothing that raises an immediate alarm).
Now imagine that within that same county, residents of one neighborhood live to 87, and residents of a neighborhood 10 miles away live to 67. That’s a 20-year gap hidden inside an average that made everything look fine.
This isn’t a hypothetical, it’s exactly what the Chicago Department of Public Health found when they zoomed past county-level data into neighborhood-level analysis: residents of the Loop can expect to live more than 20 years longer than those in West Garfield Park. Same city, same county, radically different realities.
For public health departments and hospital teams making decisions about where to invest resources, which populations to prioritize, and how to structure community health improvement plans, county-level data isn’t just incomplete, it can be actively misleading.
The Averaging Problem
County-level health data is useful for many things: broad trend analysis, state and federal benchmarking, high-level resource allocation decisions. But it has a fundamental limitation — it averages across populations and geographies that may have very little in common.
A county with a dense urban core and surrounding rural communities will produce an average that accurately describes neither. A county where a wealthy suburb sits alongside a low-income neighborhood will show a middle-ground income figure that doesn’t reflect where poverty is concentrated or where investment is needed most.
The averaging problem is especially acute when discussing health outcomes. If your goal is to identify and address disparities (not just describe the average state of health in your jurisdiction) you need data that can surface variation, not smooth it over.
What Gets Missed When You Stop at the County Line
The disparities that live below the county level aren’t edge cases. They’re often where the most critical health inequities are concentrated. Here’s what typically gets masked:
Neighborhood-level mortality gaps. Like the Chicago data shows, life expectancy can vary by decades within a single county. Chronic disease burden, homicide rates, overdose deaths, and maternal mortality all cluster in ways that county averages don’t reflect.
Racial and ethnic disparities within geographies. A county with a racially diverse population may show acceptable aggregate outcomes while significant gaps persist between groups. Without disaggregated data by race and ethnicity at the sub-county level, those gaps stay invisible to planners.
Access deserts hidden by proximity. A county might have adequate hospital capacity in aggregate, but specific ZIP codes — especially rural or low-income areas — may be effectively cut off by distance, transportation, or insurance coverage.
Urban intra-city variation. In large urban counties, the differences between the north side and south side, between suburbs and the city core, can be more significant than differences between counties. County data actively obscures this.
Pockets of need in otherwise “healthy” counties. A county that performs well on most indicators may still contain census tracts with concentrated poverty, poor health outcomes, and high unmet need. Without granular data, those pockets go unaddressed.
The Geographic Levels That Actually Matter
Sub-county analysis isn’t one single thing. There are several levels of geographic granularity, each with different strengths for different kinds of decisions:
ZIP codes are often the most intuitive unit for community members and program staff. They map easily onto service areas and are useful for grant targeting and community-facing communication.
Census tracts are the gold standard for health equity analysis — small enough to capture meaningful variation, large enough to have statistically reliable data for most indicators. Most federal datasets are organized at this level.
Neighborhoods are often the most meaningful unit for community engagement and storytelling — but boundaries vary by city and data availability at this level depends on local sources.
Custom service areas defined by a hospital, health department, or community organization are increasingly important for Community Health Assessment and CHIP work, where the relevant geography is determined by the population served, not a political boundary.
The right level of analysis depends on the question you’re asking. For broad trend tracking, county data may suffice. When identifying where to deploy community health workers, where to site a new clinic, or which neighborhoods to prioritize in a CHIP, you need census tract or ZIP code data at minimum.
What Sub-County Analysis Makes Possible
When the Chicago Department of Public Health moved beyond city-wide averages to neighborhood-level life expectancy analysis — powered by the Chicago Health Atlas on Metopio’s platform — the results changed what was possible strategically.
They could see not just that a citywide gap existed between Black and non-Black residents, but which causes of death were driving it in which neighborhoods. That specificity enabled CDPH to reshape interventions with precision: the Play Streets program was restructured to target the eight community areas with the lowest chronic disease outcomes, rather than spreading resources across the city.
The Practical Barriers — and How to Overcome Them
If sub-county data is so valuable, why aren’t more organizations using it routinely? The honest answer is that it’s been hard. Three barriers come up consistently:
Data availability and reliability
Not all indicators are available at the sub-county level, and those that are may have reliability concerns for small populations with small numerators. This is a real constraint that requires thoughtful methodology — knowing when to aggregate up, when to use multi-year estimates, and when to supplement with state or local data sources that can fill federal gaps.
Time and technical capacity
Pulling, cleaning, and mapping sub-county data from multiple government sources used to require significant technical expertise and weeks of staff time. Metopio has largely eliminated this barrier — providing pre-aggregated, validated data at the census tract and ZIP code level, ready to visualize and compare across geographies without any coding required.
Communicating granular findings to non-technical audiences
Sub-county data can be powerful or paralyzing, depending on how it’s presented. Interactive maps and visualizations — accessible to community members, policymakers, and media — are essential for making granular findings actionable rather than overwhelming. Public-facing platforms that put census tract data in the hands of community advocates, not just analysts, are where the real accountability comes from.
The Bottom Line
County-level data has its place. But if your Community Health Assessment stops there, you’re almost certainly missing the people who need attention most.
The communities experiencing the worst health outcomes aren’t evenly distributed across counties — they’re concentrated in specific neighborhoods, census tracts, and ZIP codes. Finding them requires data that’s granular enough to see them.
The gap between what county data shows and what’s actually happening on the ground isn’t a data problem. It’s a resolution problem. And it’s one that’s now very solvable — for teams of any size, with any level of technical expertise.