Population Density Analysis for Network Adequacy: Building Around Where Members Actually Live

The County-Center Assumption That Breaks Adequacy Models

Many network adequacy models use a county's geographic center — the geometric centroid of the county polygon — as the reference point for time-distance calculations. It is an intuitive simplification, but it is not how CMS actually measures adequacy. CMS uses member population centroids, which represent the weighted geographic center of where the plan's enrolled or projected member population actually lives within each county. For counties with uniform population distribution, the difference is minor. For counties with concentrated or non-uniform population distribution, the difference can determine whether a county passes or fails adequacy review.

The practical consequence of this distinction is that network operations teams that build adequacy models using county geometric centroids are solving the wrong problem. They may recruit and contract providers that satisfy time-distance thresholds from the county's geographic center while failing to satisfy those same thresholds from the locations where most members actually live. When CMS runs its adequacy calculation using member-weighted population centroids, gaps that didn't appear in the plan's internal model become visible — and those gaps generate deficiency findings that require corrective action.

Shifting to centroid-based adequacy analysis is one of the highest-leverage methodology improvements a network operations team can make. It doesn't require recruiting more providers — it requires recruiting the right providers in the right locations, which is a different optimization problem with a potentially different solution. Plans that have made this shift consistently report that it changes their provider recruitment targeting in ways that both improve adequacy outcomes and reduce wasted outreach effort toward providers whose location doesn't actually serve the member population.

How CMS Uses Population Centroids vs. County Centers

CMS's HPMS adequacy calculation uses member population data — either actual enrollment data for existing plans or census-based population data for new plans entering a service area — to determine the geographic points from which time-distance measurements are made. For each county in a plan's service area, CMS calculates the population-weighted centroid: the geographic point that minimizes the average distance to all members in the county, weighted by member density.

In practice, CMS uses census tract-level population data as the basis for this calculation. Each census tract in the county contributes to the centroid calculation in proportion to its population. Tracts with higher population density pull the centroid toward them; tracts with sparse population contribute less. For counties where the population is concentrated in one city or town, the centroid falls near that concentration — which may be substantially offset from the county's geographic center. For counties where the population is distributed across multiple communities, the centroid represents a weighted average that may correspond to a point between population centers rather than to any single town.

The CMS adequacy calculation then measures time and distance from this centroid to contracted providers in the plan's network. A provider must be reachable from the centroid within the applicable time-distance threshold to count toward adequacy for that county. Providers that are accessible from the county's geographic center but not from its population centroid do not count — and models that use county centers rather than population centroids will systematically overcount providers in counties where the center and centroid are significantly offset.

What Centroid-Based Analysis Reveals vs. Hides

The most important insight from centroid-based adequacy analysis is that counties are not uniform. The geographic center of a county can be misleading in both directions — it may suggest inadequacy where members are well-served, or suggest adequacy where members face real access barriers. The centroid-based approach, anchored to where members actually are, provides a more accurate picture of both.

In suburban counties, centroid-based analysis typically reveals that the effective demand for access is concentrated along specific corridors — major roads, transit lines, or developed commercial areas — rather than distributed evenly across the county's land area. Providers located in the heart of those corridors satisfy adequacy much more efficiently than providers located at the county's geographic periphery, even if both locations are technically within the county boundary. Network recruitment that targets the high-density corridors, informed by centroid analysis, contracts providers where they have the most adequacy impact per contract.

In rural counties, the centroid-based approach often reveals that the county's entire population is clustered around a single town — sometimes the county seat, sometimes a larger commercial center — and that the geographic center of the county is far from any significant population concentration. Providers located in or near the population cluster satisfy adequacy for the vast majority of county members, while providers located closer to the county's geographic center serve very few. This insight is directly relevant to provider recruitment: rural counties that appear to have broad provider gaps based on county-center analysis may be adequately served from the perspective of where members actually live.

Census Tract Population Data in Adequacy Modeling

The American Community Survey (ACS) and the decennial census provide census tract-level population data that is the standard input for centroid-based adequacy modeling. Census tracts are designed to contain roughly 4,000 people, making them a useful unit of geographic analysis that is granular enough to capture population distribution patterns without being so fine-grained that the data becomes unstable from year to year.

For adequacy modeling, the census tract population data needs to be processed into a usable format: each tract's population is associated with the tract's geographic centroid, and those tract-level centroids are then aggregated into county-level population centroids using population-weighted averaging. The resulting county centroid reflects the actual distribution of population across the county's census tracts and provides a much more accurate representation of where members are concentrated than the county's geometric center.

Network operations teams that build their adequacy models on census tract data should also be aware that census data has a lag — the most recent decennial census data is from 2020, and ACS five-year estimates provide the most current tract-level population figures but still lag real population change by several years. In rapidly growing suburban counties or counties experiencing significant demographic shifts, the census-based centroid may diverge from the actual member population centroid. Plans in dynamic markets should supplement census-based centroids with actual enrollment data — where available — to calibrate their models against real member distribution.

High-Density Suburban Corridors That Straddle County Lines

One of the most common adequacy modeling challenges involves high-density suburban corridors that straddle county lines. In metropolitan areas, population density often follows development patterns — major roads, transit corridors, employment centers — that cross county boundaries without regard for the administrative divisions that structure CMS adequacy calculations. A densely populated suburban corridor may have the majority of its population on one side of a county line, with the most accessible providers on the other side.

CMS adequacy calculations are county-specific — each county is evaluated independently against its applicable time-distance threshold. A provider located across the county line may be closer to the majority of members in the adjacent county than any provider within the county itself, but if the provider is not contracted in the plan's service area for the adjacent county, it cannot count toward that county's adequacy. Plans operating in multi-county metropolitan areas need to be particularly attentive to the geography of population distribution relative to county lines, because the county-line constraint interacts with population centroid location in ways that create adequacy gaps that are invisible in county-center analysis.

The practical strategy for cross-county corridor situations is to map the population centroids for all counties in the metropolitan service area simultaneously, identify providers within or near each county that are accessible from the county's population centroid, and prioritize contracting with providers that are geographically positioned to satisfy multiple counties' adequacy requirements simultaneously. Providers located near county boundaries in high-density corridors can often count toward adequacy for two or more counties, making them particularly high-value targets for network recruitment in metropolitan markets.

Rural Counties Where Population Is Concentrated Near One Town

The rural county scenario is in many ways the mirror image of the suburban corridor challenge. Rural counties are geographically large — some covering thousands of square miles — but their populations are often concentrated in a single town or a small number of communities. The county's geographic center may be in an uninhabited rural area far from any significant population concentration, while the majority of the county's residents live in or near a single population center that may be offset from the geographic center by tens of miles.

For adequacy modeling, this means that the CMS population centroid for a rural county falls near the dominant population center, not near the county's geographic center. A provider located in or near the dominant town satisfies adequacy for most of the county's population, even if the provider is not centrally located from a geographic standpoint. Conversely, a provider located at the county's geographic center — which might seem like an optimal central location from a mapping perspective — may actually be less accessible to the majority of county residents than a provider in the dominant town.

This insight has direct implications for rural network recruitment strategy. In single-center rural counties, the highest-adequacy-impact recruitment target is a provider in or near the dominant population center, not a provider at the county's geographic center. Plans that have been recruiting rural providers based on geographic centrality rather than population centrality may have been prioritizing the wrong locations — and may have contracted providers that contribute less to CMS-measured adequacy than their geographic position suggests.

Practical Steps to Run Centroid-Based Adequacy Analysis

Building a centroid-based adequacy model requires three data inputs: census tract population data with geographic centroids, county boundary files that enable tract-to-county assignment, and contracted provider location data with geographic coordinates. Each of these data sets is publicly available — census tract data from the Census Bureau, county boundaries from the Census TIGER/Line shapefiles, and provider location data from NPPES — and can be combined in a GIS environment or purpose-built adequacy analysis platform.

The centroid calculation process involves, for each county in the service area: identifying all census tracts that fall within the county boundary, calculating the population-weighted average latitude and longitude across those tracts, validating that the resulting centroid falls within the county boundary (edge cases at county boundaries sometimes produce centroids that fall slightly outside), and storing the centroid coordinates as the adequacy measurement point for that county.

Once county centroids are established, the adequacy calculation follows the standard CMS methodology: for each county and each provider specialty category, calculate the driving time and distance from the county centroid to each contracted provider of the applicable specialty, count providers that fall within the applicable time-distance threshold, and compare the count to the CMS minimum ratio for the county's geographic classification. The output is a county-by-specialty adequacy matrix that reflects CMS's actual measurement approach rather than a geometric approximation.

How This Shapes Provider Recruitment Targeting

The most immediate operational impact of centroid-based adequacy analysis is on provider recruitment targeting. When adequacy gaps are identified using population centroids rather than county centers, the gap locations are often different — sometimes dramatically so — from where county-center analysis would have directed recruitment resources. This changes the list of high-priority outreach targets and, in some cases, changes which counties are flagged as priority gaps at all.

Centroid-based targeting allows network operations teams to evaluate potential recruitment targets not just by whether they are in the gap county, but by how much adequacy impact their location would have. A provider located near the county's population centroid — close to where most members are — contributes more to adequacy than a provider at the county's geographic periphery, even if both are within the county boundary. Recruitment resources directed toward high-centroid-proximity providers produce more adequacy improvement per contracted provider than resources directed toward peripherally located providers.

Blueprint Network Hub incorporates population centroid data into its adequacy scoring engine, providing network operations teams with centroid-based adequacy scores for each county in their service area. The provider recruitment dashboard displays adequacy impact scores for potential recruitment targets based on the provider's proximity to the county's population centroid — allowing recruiters to prioritize outreach toward providers who will move the adequacy needle most efficiently. For plans building out new service areas or closing persistent adequacy gaps, this capability transforms the recruitment prioritization process from a geographic approximation into a data-driven targeting exercise grounded in CMS's actual measurement methodology.