GIS analysis of diabetes related community factors within discrete, urban neighborhoods

BACKGROUND: Prior GIS-based research has analyzed diabetes-related factors in urban and rural communities. The findings suggest healthy diet, exercise, access to healthcare, and higher socioeconomic status decrease consequences of diabetes. However, limited information exists about such differences between urban neighborhoods. PURPOSE: This research uses GIS technologies to identify neighborhoods where patients who have diabetes hospitalizations live within an urban hospital's primary service area and uses GIS methods to identify differences associated with diabetes-related factors between discrete, contiguous neighborhoods. METHOD: We obtained deidentified hospital admission data from one calendar year (N=133,788) from a large urban healthcare network. Spatial references were created using zip codes. The study area was defined as the network's ‘home cluster,' consisting of five zip codes. Data from multiple sources and geographic levels were analyzed. RESULTS: The data included 4,540 patients with diabetic-related hospitalizations who were mostly female and African American. Patients with diabetic hospitalizations were not normally distributed throughout the ‘home cluster,' with differences in comorbidities and ages. DISCUSSION: The use of GIS methods can facilitate the identification of different patient groups and their relationships to neighborhoods. These results provide information to improve the understanding and use of geographic proximity and its associated factors, such as resource accessibility and the transmission of tacit knowledge, as critical factors in diabetes prevention and intervention. The results will be used to plan and then evaluate hospital-based, community-focused programs in diabetes.