Disease Mapping with Disparate Spatial Data

Ecological regression studies are widely used to explore relationships between disease rates and levels of exposure to environmental risk factors. The raw data for such studies, such as disease case counts, environmental pollution concentration measurements and the reference population at risk, are measured at disparate levels of spatial aggregation but are commonly accumulated to a single common geographical scale to facilitate statistical analysis. In this traditional approach heterogeneous exposure distributions within the aggregate areas may lead to biased inference, while individual attributes such as age, gender and smoking habits must either be summarized to provide area level covariate values or used to stratify the analysis.

This presentation offers a spatial regression analysis of the effect of traffic pollution on respiratory disorders in children. The analysis features data measured at disparate, non-nested scales, including spatially varying covariates, latent spatially varying risk factors, and case-specific individual attributes.

The problem of disparate discretizations is overcome by relating all spatially-varying quantities to a continuous latent underlying random field. Case-specific individual attributes are accommodated by treating cases as a marked point process. Inference in these hierarchical Poisson/L'evy models is based on simulated samples drawn from Bayesian posterior distributions, using Markov chain Monte Carlo methods.