Examining the impact of Ozone on lung health in a vulnerable population: a Bayesian modeling strategy.

The fundamental rationale for this research is that Air pollution continues to be a major health concern not only for its impacts on morbidity but also that acute mortality may be directly associated with exposures. Mounting evidence suggests that the postnatal lung is especially sensitive, receives a disproportionate dose, and that exposure may perturb normal lung development and lead to long-standing alterations in lung function, issues sufficiently noteworthy to have attracted the attention of researchers and health care workers alike. Notable is that these findings parallel observations by research teams that demonstrate remarkable exposure-induced impacts on normal growth and development. These, other observations, and a novel programmatic paradigm afford investigators and their associates on this project unique opportunities to develop new insights regarding susceptibility to environmental insult and how exposure alters the biology of the lung and to develop tools for evaluating its affect and predict outcomes. Thus clearly, Ozone (O(3)) remains an important public health problem especially with regards to sensitive populations such as children and infants with developing lungs. Numerous recent studies document that exposures induce both short and potentially long term impacts on the childhood lung. Responses to a given exposure demonstrate marked heterogeneity with respect to age, anatomic site, species, and exposure history. Infants may be particularly at risk due to a greater inhaled dose rate. Respiratory tract surfaces are covered by an aqueous layer (epithelial lining fluid; ELF) that inhaled gases first encounter and is a complex mixture containing significant concentrations of small molecular weight antioxidants (e. g., ascorbic acid (AH(2)), glutathione, and uric acid), lipids, and proteins. In this study we use an animal model controlling for age and anatomic site and examine the effect of ozone on the antioxidant values in the ELF after varying exposures. We compare the resultant antioxidant levels by two exposure levels to filtered air which is the control level. Because of the variation in these antioxidant levels across lung tissue sections as well as over time a Bayesian modeling approach with diffuse priors applies well to this data. Also the data is transformed to a log scale allowing us to assume underlying normality of the parameters of interest. Our last objective will be to determine if there is a difference in levels of these antioxidants examining the passage of exposed air through the right caudal lobe and the geographic compartments (sections) of this tissue. The approach here is to apply a Markov Chain Monte Carlo strategy to coherently combine prior diffuse information on the normal response (level of antioxidant) with the distribution of the response probability over 3 levels of exposure (2 ozone exposure strategies plus filtered air) and derive posterior means of antioxidant level for exposure group and compare these exposure results further over regions of the lung to determine level of change in the antioxidant values over these regions. We will examine the credible regions for the parameters and the convergence properties as well. The posterior values of the parameters of the three exposures as well as the combined posterior parameters of the entire sample will be evaluated and examined. The overall goal is naturally to determine if there is significantly greater disparity in the exposure levels versus the filtered air in each of the antioxidants. The design of the study allows for section and exposure interaction results which will enhance the understanding of the mechanism of O(3) in the population of younger individuals.