Abstract

Diabetes is a major public health concern in the United States, with its prevalence influenced by a combination of socioeconomic, environmental, and behavioral factors. This study examines the spatial distribution of adult diabetes prevalence at the county level across the U.S., using geospatial and statistical analysis to identify high-risk areas and the underlying determinants of the disease. Utilizing data from the 2022 County Health Rankings and spatial econometric techniques, this study employs Moran’s I index, hotspot analysis, and Geographically Weighted Regression (GWR) to assess the clustering of diabetes prevalence and its association with socioeconomic variables. The findings reveal that diabetes prevalence is not randomly distributed but significantly clustered, with hotspots concentrated in the Southeastern U.S., particularly in Alabama, Georgia, and Texas. Ordinary Least Squares (OLS) regression and GWR indicate that factors such as obesity rates, child poverty, unemployment, and long commute times positively correlate with diabetes prevalence. At the same time, rurality, excessive drinking, and smoking show a negative correlation. These results underscore the importance of place-based public health interventions and policy measures to reduce diabetes disparities. Recommendations include expanding healthcare access in high-risk counties, promoting community-based diabetes prevention programs, addressing socioeconomic disparities, and enhancing the built environment to encourage healthier lifestyles.

Key words: Diabetes prevalence, spatial analysis, socioeconomic determinants, geographically weighted regression, public health policy, United States.