Bayesian G-Computation for Estimating Impacts of Interventions on Exposure Mixtures: Demonstration With Metals From Coal-Fired Power Plants and Birth Weight

Abstract

The importance of studying the health impacts of exposure mixtures is increasingly being recognized, but such research presents many methodological and interpretation difficulties. We used Bayesian g-computation to estimate effects of a simulated public health action on exposure mixtures and birth weights in Milwaukee, Wisconsin, in 2011-2013. We linked data from birth records with census-tract-level air toxics data from the Environmental Protection Agency's National Air Toxics Assessment model. We estimated the difference between observed and expected birth weights that theoretically would have followed a hypothetical intervention to reduce exposure to 6 airborne metals by decommissioning 3 coal-fired power plants in Milwaukee County prior to 2010. Using Bayesian g-computation, we estimated a 68-g (95% credible interval: 25, 135) increase in birth weight following this hypothetical intervention. This example demonstrates the utility of our approach for using observational data to evaluate and contrast possible public health actions. Additionally, Bayesian g-computation offers a flexible strategy for estimating the effects of highly correlated exposures, addressing statistical issues such as variance inflation, and addressing conceptual issues such as the lack of interpretability of independent effects.

Keywords: Bayesian methods; air pollution; air toxics; causal inference; g-computation; metals.

Figure 1

Estimated (National Air Toxics Assessment (19, 20)) ambient arsenic concentrations in Milwaukee County, Wisconsin, by US Census tract, 2011. Black dots represent the locations of the 3 coal-fired power plants in operation in Milwaukee County during the study period.

Figure 2

Bivariate scatter plots, univariate kernel density plots, and Spearman correlation matrix for a hypothetical intervention to reduce exposure to 6 airborne metals by decommissioning 3 coal-fired power plants, Milwaukee County, Wisconsin, 2011. Hg, mercury; Se, selenium; Be, beryllium; As, arsenic; Cr, chromium; Ni, nickel.

Figure 3

Posterior mean difference (points) and 95% credible intervals (bars) for contrasts between population interventions in which all individuals have arsenic, beryllium, chromium, mercury, nickel, and selenium exposures simultaneously set to a percentile of their observed values, where the referent intervention is to set all exposures to the population median values (medians are given in Table 1). A) Results from the primary model with Bayesian model averaging (model A in Table 3); B) results from the hierarchical Bayesian model with no selection (model J in Table 3). Scatter plot smoothing lines (gray) are shown for visual reference.