This paper models health outcomes among adults as measured by Body Mass Index (BMI) using spatial econometric techniques that account for clustering and spillovers across neighborhoods. We model spatial spillovers among neighboring communities to determine to what extent heterogeneity and linkages across locally defined neighborhoods are important in explaining obesity data. Using survey responses tied to geographic location, demographic, behavioral, and environmental factors such as food retailer information, this study finds evidence of spatial dependence pointing to some locational impact on BMI. Our findings suggest alternative explanations for discrepancies in obesity across geographic space that currently are attributed to segregation on demographic characteristics. Preliminary indicators of spatial heteroskedasticity compel further applications of spatial econometric methods.