A New GIS-driven Approach to Optimize Service Area Boundaries for ACOs

My Folder

Within many healthcare systems, a patient population is usually defined by who has already been seen for care. These systems often use the patient registries in their EMRs to define patient populations and base their analysis on current diagnoses or self-identified demographics. But there is a better way using GIS location technology and supporting data sources that may be especially useful for ACOs and other organizations concerned with managing population health—optimized service area boundaries.

Using GIS Location Technology in Healthcare

A more precise approach would be to first objectively define the geography within which the health system operates, and second to define the subsequent population belonging to that overall service area. This is known as network coverage optimization, and it provides a more robust way to define boundaries and identify populations.

Once the geography-based care population is well defined, value-based strategies, such as disease cohort underwriting or at-risk contracting, become more feasible. Within a rapidly changing healthcare market, it makes sense to leverage location analytics for more robust strategic assessments for healthcare systems, physician groups, health exchanges, and other payer groups. Using geography provides a picture of all the possible patients available to a health system rather than just those who have been seen. In other words, healthcare organization of all kinds should use GIS location technology to clearly define patient populations and determine sound strategies and decisions.

Determining Adequate Service Coverage

One simple approach to gauge adequate service coverage across population density is to visualize access times for various healthcare facilities as shown in Figure 1.

drive time to access healthcare facilityFigure 1: An example showing drive time access to healthcare facilities within a given service area.

At Health Catalyst, we’ve developed a more comprehensive approach than this using objective and independent methods for defining geography, such as Dartmouth Atlas hospital referral regions, a hierarchy of facility levels with appropriate drive time isochrones, and medical specialties-based central place theory. These complementary methods were employed to generate the minimum bounding geometry of the overall reach of a given healthcare system as shown in Figure 2.

minimum bounding geometryFigure 2: An example of using minimum bounding geometry to define the reach of a health system.

Further, to better measure the overall viability of this health system, a network coverage score was calculated using zip code and population-based statistics (also shown in Figure 2). This score (shown as a percentage) represents a system’s enrolled patients as compared to the total population. This type of objective scoring could also be used to compare health systems in various locations. No protected patient health information was necessary and all calculations were aggregated to the zip code level.

Lastly, consumer data from sources such as Census.gov, Esri Tapestry market segmentation, and Medicare.gov were used to characterize the population defined by the final resulting boundary as shown in Figure 3.

leveraging socioeconomic and demographic data to undersatnd health system boundariesFigure 3: Leveraging socioeconomic and demographic data will enhance the understanding of health system boundaries and covered population.

Optimized Service Area Boundaries Result in a Better Understanding for Population Health Management Strategy

So, how does this really work?

The combined minimum bounding geometry (using drive time access to facilities, central place theory borders, and hospital referral regions) was generated for a Minneapolis healthcare system. A total of 871 zip codes containing some 5,669,493 people were identified as bounded by the service area thus defined. However, in this geographic service area, only 1,266,157 people were attributable to that specific health system. Thus, among the many opportunities this tool elucidates, one is how to best attract some percentage of the remaining 4.4 million people as new customers for the health system.

ACOs Benefit from GIS Technology

Using GIS-powered analytics as described above to define their patient populations, ACOs derive a better understanding of their enrolled patients and eligible payer groups. This strategic population analysis coupled with novel visualizations can yield better decisions in population health management, leading to improving quality and lowered cost, both imperative for ACOs to thrive.

In summary, GIS location technology can be used to readily identify objective service area boundaries relevant to a specific healthcare system. This automated method of defining the service area also identified the exact population for which health coverage is currently provided. Population characteristics from various external source data are also leveraged.

This health network optimization tool was built in collaboration with GISi (a platinum Esri partner) and is available by invitation at http://cloud.gisinc.com/hc/.

Powerpoint Slides

Would you like to use or share these concepts? Download this presentation highlighting the key main points.

Click Here to Download the Slides

Loading next article...