A Bayesian Hierarchical Approach for Relating PM2.5 Exposure to Cardiovascular Mortality in North Carolina

• Activity Number: 85
• Type: Contributed
• Date/Time: Monday, August 9, 2004 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics and the Environment
• Author(s): Christopher H. Holloman*+ and Steven M. Bortnick and Michele Morara and Warren J. Strauss and Catherine A. Calder
• Companies: Battelle and Battelle and Battelle and Battelle Memorial Institute and Ohio State University
• Keywords: fine particulate matter ; spatial modeling ; SHEDS-PM ; exposure simulator

Abstract

Over the past several years, studies have established that elevated PM2.5 levels and negative health effects are related; however, considerable uncertainty remains regarding the nature of the association. Since the EPA began widespread monitoring of PM2.5 levels in 1999, the epidemiological community has performed numerous observational studies modeling mortality and morbidity responses to PM2.5 levels using Poisson Generalized Additive Models (GAMs). While these models are useful for relating ambient PM2.5 levels to mortality, they give little information about the strength of the effect between exposure to PM2.5 and mortality. In order to address this question, we propose a three-stage Bayesian hierarchical model as an alternative to the classical Poisson GAM. Fitting our model in seven North Carolina counties using data covering the years 1999-2001, we find that an increase in PM2.5 exposure is linked to increased risk of cardiovascular mortality in the same day and following two days. In addition, we compare the results obtained from our model to those obtained by applying Frequentist and Bayesian versions of the classical Poisson GAM to our study population.