Given the complex composition of pollutants such as particulate matter PM in the environment, it is believed that our exposure is not limited to one chemical at a time but to several possible mixtures with varying composition and mix ratios. These mixtures could be formed because they are emitted at the same time from the same source origin as latent factors. As a result, humans are more susceptible to exposure to these mixtures as characterized by these latent factors. Factor analysis is one tool used in the literature to define mixtures. Unfortunately statistical methods of identifying mixtures are limited. Additionally, exposure to PM chemicals during pregnancy is known to be harmful to both mother and child. To address the problem a new method of identifying mixtures exposure based on extension of the traditional factor analysis combined with source apportionment methods will be used. This proposal will test the following hypothesis: Hypothesis 1: Exposure to PM mixture metals during pregnancy increased the risk of cardiovascular disease (CVD) risk factors among black and Hispanic minorities and vary by trimester. Hypothesis 2: CVD risk factors are mediators between PM metals exposure and child development, through the following Aims. 1) Quantify mixtures and estimate mother's exposure during various trimesters of pregnancy. We will use the newly developed Bayesian extended factor analysis (FA), based on the flexibility of Bayesian statistics and the decomposition of the error covariance matrix combined with SAM models to estimate metal mixtures. We will perform a spatio-temporal analysis of the mixtures using Generalized Additive Mixed Models(GAMM). We will use the inverse distance weighting method to estimate mother's inhaled quantities during each trimester of pregnancy. 2) Investigate CVD risk factors difference among black and white and interaction between air pollution and race/ethnicity by trimester. We will use logistic regression analysis and generalized spatial linear model for this aim. 3) Investigate CVD risk factors moderating between mixture metals exposure and infant mortality. We will use structural equation modeling (SEM) to investigate wich risk factors is in the pathway between exposure and child mortality. Results from the proposed studies could have implications in identifying mixtures exposures, their chemical toxicity and relative implication in CVD during pregnancy.
Exposure to Particulate Matter (PM) metals during pregnancy is known to be harmful to both mother and the new born. Cardiovascular disease (CVD) is one such harmful effect of exposure to PM metals. In the natural environment we are always exposed to mixtures of chemicals with different toxicities. Unfortunately, methods of identifying those mixtures are limited. In this K01 application, we are proposing a novel method of pollutant mixtures identification and we will investigate their association with CVD risk factors.
