Abstract
This study investigated the relationships between vacant land and key adverse health behaviors, including smoking, insufficient sleep, and no leisure-time physical activity (No LPA), across census tracts in Chicago, Illinois. Using both global regression and geographically weighted regression (GWR), we evaluated whether neighborhood vacant land ratios (VLRs) were associated with the prevalence of these adverse health behaviors and assessed how these associations varied spatially across the city. We found significant spatial clustering in both vacant land and health behavior indicators, and the spatial clustering patterns of neighborhood vacancy and adverse health behaviors were broadly consistent. In global models, higher VLRs were associated with higher prevalence of adverse health behaviors; after accounting for spatially autocorrelated errors, the associations remained robust for smoking and insufficient sleep but were attenuated for No LPA. GWR results further revealed clear spatial non-stationarity, with stronger positive local associations concentrated in low-income neighborhoods on the south and west sides. When overlaid with Healthy Chicago Zones (HCZs), the strong vacancy-behavior associations aligned primarily with the West, Southwest, Near South, and Far South zones, highlighting these HCZs as priority areas where vacancy was most strongly linked to adverse health behaviors. Our findings support theories of neighborhood disorder and spatial inequality, emphasizing that vacant land is a potentially modifiable environmental determinant of health behaviors and calling for tailored interventions that consider local social and economic contexts to improve community health and advance health equity.