Introduction: Diabetes remains a major chronic disease in the U.S., disproportionately affecting communities facing poverty, pollution, and weak health infrastructure. While clinical risk factors are well-known, fewer studies have examined how environmental and structural conditions vary across space and time to shape diabetes prevalence. This study investigates the role of socio-environmental exposures—PM2.5, crime, greenness (NDVI), poverty, and food access—in county-level diabetes trends. Hypothesis: We hypothesized that socio-environmental conditions—including air pollution, crime, food access, and greenness—jointly and unevenly influence diabetes prevalence, and that vegetation cover may buffer the negative effects of poverty. Methods: Using CDC’s U.S. Diabetes Surveillance data (2012–2020), we built a spatio-temporal dataset for all U.S. counties, integrating NDVI from NASA’s MODIS satellite, County Health Rankings indicators (crime, obesity, inactivity, food access), and Census poverty estimates. Linear mixed-effects models assessed temporal trends with county and year as random intercepts. GWR was used to explore spatial variation in predictor effects. Associations between NDVI, crime, and diabetes were evaluated using grouped exposure levels. Results: Diabetes prevalence was higher in counties with elevated poverty, air pollution, crime, and limited food access. Counties with more vegetation showed weaker poverty–diabetes associations, suggesting a buffering effect. Spatial analyses revealed persistent hotspots in the Southeast and emerging clusters in the Midwest. GWR showed that air pollution and food access were stronger predictors in urban areas, while crime and poverty had greater influence in rural regions. Conclusion: This study shows that socio-environmental exposures significantly influence diabetes prevalence, with effects varying across geography and time. Greenness appears to mitigate structural disadvantages, offering insights for place-based prevention. Results support targeting social and environmental inequities in chronic disease policy and highlight the value of spatio-temporal models in guiding public health interventions.
Suresh Nath Neupane (Mon,) studied this question.