Ohio State University - Department of Statistics
Characterizing variation in human exposure to toxic substances over large populations often requires an understanding of the geographic variation in environmental levels of toxicants. This knowledge is essential when the primary routes of exposure are through interactions with environmental media, as opposed to more individual-specific exposure routes (e.g., occupational exposure). In this study, we focus on modeling the spatial variation in the concentration of arsenic, a toxic heavy metal, in air, soil, and water across the state of Arizona. We then synthesize this background information with individual-specific arsenic exposure measurements from EPA's National Human Exposure Assessment Survey (NHEXAS) in a Bayesian hierarchical pathways model. Our approach uses mixture model components to account for the spatial misalignment between NHEXAS individuals' county of residence and the various types of background information about arsenic levels in environmental media. We discuss the implications of performing this spatial data assimilation in assessing the relative importance of various exposure pathways and, more generally, the issue of assessing model fit in large multilevel statistical models. This is joint work with Peter Craigmile, Jian Zhang, Hongfei Li, Rajib Paul, and Noel Cressie.