One of the most innovative areas of combining datasets in healthcare is the melding of health-related administrative and clinical data with other sources such as census and demographic data via geographic information systems, or GIS technology.
While the use of GIS in healthcare is not entirely new, more granular data and the ability to combine different sources, such as photo diary data with geo-mapping platforms, is extending the ability of both public health officials and health system planners to better tailor their services to their patients and communities.
Successfully exploiting this sort of mashup capability will also be paramount in order for health delivery systems to maximize their investments in population health management and accountable care models, early adopters of GIS data in healthcare say.
Also See: Is Geomedicine Reaching a Tipping Point?
Michael Dulin, M.D., chief clinical officer for analytics and outcomes research at Charlotte, N.C.-based Carolinas HealthCare System, began pondering the demographics of primary care delivery and using demographic and GIS data to make it more efficient a decade ago, and has been in the forefront of demonstrating how the pieces fit together. As the principles of preventive care inspired by the Affordable Care Act begin to manifest themselves, Dulin believes more organizations will have to take advantage of what's out there in terms of demographic data as well as who in the community can best put it to use.
"I think it's fairly early in the game, but obviously this is very much in the value world of healthcare delivery and it is one of the foundational tenets of the Affordable Care Act how do we get people early into preventative care services, keep them healthy, and keep them active in the community? That was certainly the driver behind the work here, this idea that we need to be in this value-based world."
Dulin's work goes back to the middle of the last decade, when he first began to wonder why patients he saw in CHS clinics did not seem to match the demographics of the surrounding community.
"It made me wonder what was happening, how were people accessing healthcare?" Dulin said. "And I thought GIS would be a good way for us to start to look at how communities utilize healthcare and see if we can intervene to enhance appropriate access to primary care services."
Dulin and his colleagues pioneered the GIS-enhanced approach to analyzing community health data with a project called Multiattribute Primary Care Targeting Strategy, or MAPCATS. Described in an article in the Journal of the American Board of Family Medicine in 2010, MAPCATS used a combination of GIS technologies, including census-based demographic data from San Diego-based Claritas, and ArcGIS geo-coding software from Redlands, Calif.-based ESRI Products. Clinically, it analyzed attributes of emergency department visits to measure which ones might have been able to be replaced by a primary care visit instead, and also measured insurance status of each patient within the study and patient utilization of the community's primary care safety net.
The results of the study, Dulin and his team found, confirmed that MAPCATS provided a means for using information about patients patterns of healthcare utilization and community-level data to identify areas where access to primary care services was limited and could be improved.
Since developing MAPCATS, Dulin has increased the number of variables well beyond ED utilization and safety net visits to include data such as distance to the nearest grocery store, distance to nearest park or green space, and distance to a primary care clinic, data elements collected by the University of North Carolina at Charlotte Metropolitan Studies department. The work has advanced out of the research lab and into corporate planning, Dulin said CHS administrators now have access to dynamic interactive GIS-enabled portals to aid in planning.
From the beginning, Dulin has involved community members in designing and evaluating GIS-enabled research at CHS. For MAPCATS and the subsequent work that expanded on it, the researchers formed a community advisory board of ambulatory providers, community members, and research team members that was involved from development of the research question to selection of attributes to oversight of research processes and review of results.
For more recent work, the CHS team dug even deeper into both the community and technology to obtain more granular information. Using the PhotoVoice visual diary platform, the researchers teamed planning students students from UNCC with high school students to document the physical features in neighborhoods and how usage patterns of them affected the communities within them. The PhotoVoice work was part of a larger study on the social determinants of health, and the researchers found the additional data was invaluable. For example, they found variations in healthcare utilization and the prominence of social variables directing the GIS model (such as immigration and acculturation, education, safety, and socioeconomic status) were all identified as significant issues within the identified neighborhoods through PhotoVoice.
Perhaps even more significantly, the researchers found PhotoVoice results provided information that the geospatial models could not address or illuminate.
"For example, several large apartment complexes were identified during the PhotoVoice study as being home to many of the newest immigrants living in the community," they found. "The invisibility of these areas in the study up to this point reinforced the notion that, although the community-level data gleaned through census-based, government- and hospital-derived sources are extensive and reliable, they can become quickly outdated or underestimate certain populations. This is especially true for disenfranchised or at-risk populations."
As a result, Dulin and his colleagues concluded field-based assessments and qualitative analyses are especially crucial for understanding rapidly growing immigrant populations and transitioning communities, and, in this particular study, also built community engagement and expertise.
"Part of the idea was to both recognize people who lived in the neighborhood as experts on the neighborhood and what influenced their health there, but also to link them with a mentor and it was very successful in a number of ways," Dulin said. "A number of those kids went on to college at UNCC and one went on to do research with us, So the idea is helping to build community capacity and at the same time, overcoming some of the barriers that exist at the community level."
Moving forward, Dulin said GIS data could be further integrated with aggregated patient data to help build healthier neighborhoods. One example of that is the Carolinas Tracker, a platform recently launched by CHS that will allow the system's clinicians access to patient fitness and activity data.
"The key to that is, as a physician, I really don't want a million lines of Fitbit data to show up in your EHR, but that data needs to flow into our data warehouse where we can run analytics on it, and we can then do all sorts of things with it," he said. "For example, are people who live near an area with more green space and parks more active? Is there something we can do with the public health department to try to create those green spaces in areas where there are people who are high risk, who have low levels of activity?"
Ultimately, he said, GIS technology is a vital component of piecing together the puzzle that community- and population-based healthcare will have to be, one that will help both delivery systems and public health officials reach out and help people engage with the community-based resources right around them to help keep them healthy and out of the hospital.
"I'm a real believer in the community-based approach. The social network, your neighbors, the environment around you, they're all things that are very influential in terms of your health," he said. "Within the walls of the hospital system, we really influence a very small proportion of health. How can we become a proactive hospital to influence that social context to improve health and outcomes?"
Register or login for access to this item and much more
All Health Data Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access