HOW CLOSE AND HOW MUCH? LINKING HEALTH OUTCOMES TO BUILT ENVIRONMENT SPATIAL DISTRIBUTIONS

Area of Science:

• Environmental Health
• Spatial Epidemiology
• Biostatistics

Background:

• Built environment features (BEFs) influence health behaviors and outcomes.
• The proximity and density of fast food restaurants (FFRs) are potential BEFs impacting childhood obesity.

Purpose of the Study:

• To investigate the relationship between the spatial distribution of FFRs and obesity risk in schoolchildren.
• To develop and apply a novel Bayesian hierarchical modeling framework for spatial exposure analysis.

Main Methods:

• A two-stage Bayesian hierarchical model was employed.
• Stage 1: Modeled FFR distances from schools using Inhomogeneous Poisson Processes (IPP) with Nested Dirichlet Process priors.
• Stage 2: Related FFR exposure patterns to obesity risk using cluster indicators or Bayesian Kernel Machine regression.

Main Results:

• Analysis of Californian schoolchildren in 2010 revealed significant spatial patterns of FFR exposure.
• Schools with consistently more distant FFRs within a 1-mile radius were associated with lower obesity odds among 9th graders.

Conclusions:

• The spatial arrangement of the built environment, specifically FFRs, is linked to childhood obesity.
• Bayesian non-parametric and regression methods can effectively model complex spatial exposure-obesity relationships.