Using geographic information systems to evaluate and target diabetes education

Background: In order to manage diabetes, diabetes self-management education (DSME) is recommended for all those with this chronic condition. Given limited resources, it is essential to ensure diabetes education resources are targeted efficiently and effectively. To aid with this process, we analyzed county level geographical distributions of diabetes, DSME programs, and report of attending diabetes education among those with diabetes. Methods: We collected, mapped and analyzed U.S county- and individual-level data from secondary sources using ArcGIS 10 and SPSS v. 20. Web sources included age-adjusted U.S. diabetes county prevalence rates and number of adults with diabetes (County Health Rankings, 2009), rate of existing 2012 certified DSME programs per 1,000 residents with diabetes (American Diabetes Association), and percent of those with diabetes reported ever attending a diabetes education class (Behavioral Risk Factor Surveillance System, 2010). Low resource-high rate counties were defined as having below median DSME rates per 1,000 residents with diabetes and above average diabetes rates. Results: Attendance at an education class varied from 33% to 78% (p<.001). Clusters of low resource-high rate counties were found particularly in the South, while a cluster of counties in the midwest had high resources and low diabetes rates as well as the highest education attendance rates. In contrast, despite high prevalence of DSMEs, self-reported attendance was lower in many Northeast counties. Conclusions: Counties with high rates-low resources should be targeted for diabetes education programming (e.g., south) and counties with an apparent disconnect between program availability and attendance (e.g., northeast) should be further examined.