Diabetes
Spatial analysis of healthcare access and diabetes in Pennsylvania Jessica Berman* Jessica Berman Jong Cheol Shin
Previous research has ascertained that the Healthcare Data Information Set quality measures with the lowest compliance rates and highest care gaps among a diabetic population based out of Pennsylvania (PA) were within the Eye Exams for Patients with Diabetes and Kidney Exams for Patients with Diabetes quality measures. This study examined to what extent are healthcare facilities and services pertaining to eye care (EC) and kidney care (KC) accessible to diabetics in PA. Implications of this study include: 1) increased awareness of resources to boost compliance and health outcomes and 2) increased stakeholder knowledge of geographical context leading to informed decision making and resource allocation.
This study employed ArcGIS Pro for spatial analyses and RStudio for statistical analyses on the following variables: diabetes prevalence (prev), chronic kidney disease prev, diabetic retinopathy prev, median household income, zero access to a household vehicle and medically underserved areas. Locations of hospitals, dialysis facilities, endocrinologists, nephrologists, ophthalmologists and optometrists were geocoded into XY coordinates and served as the healthcare facilities and services pertaining to EC and KC. The ArcGIS Pro tools used were: Geocode Addresses, XY Table to Point, Buffer, Thiessen Polygons and Calculate Geometry Attributes. Pearson Correlation Coefficient tests were run to determine significance and r values.
The results depict a relationship between high diabetes prev and high access to healthcare services and facilities pertaining to EC and KC: hospitals (r=-0.11, p< .001), EC (r=-0.11, p< .001) and KC (r=-0.11, p< .001). This was evidenced by Allegheny, Delaware, and Philadelphia County depicting both high diabetes prev and high densities of hospitals, EC, and KC facilities and services. The additional variables mentioned were spatially and statistically analyzed with 75% of the bivariate analyses yielding significant results.
