Bayesian ensemble learning for predicting health outcomes of multipollutant mixtures

TL;DR

This paper presents SoftBart, a Bayesian ensemble learning method that efficiently models complex, correlated multipollutant effects on health outcomes, outperforming existing methods in accuracy and variable selection.

Contribution

The paper introduces SoftBart, a novel Bayesian ensemble approach that improves efficiency, flexibility, and variable selection in modeling multipollutant health effects.

Findings

SoftBart outperforms BKMR in simulation accuracy.

It effectively identifies active variables in correlated pollutants.

Applied to Nurses' Health Study data, it provided insightful health risk estimates.

Abstract

We introduce the SoftBart approach from Bayesian ensemble learning to estimate the relationship between multipollutant mixtures and health on chronic exposures in epidemiology research. This approach offers several key advantages over existing methods: (1) it is computationally efficient and well-suited for analyzing large datasets; (2) it is flexible in estimating various correlated nonlinear functions simultaneously; and (3) it accurately identifies active variables within highly correlated multipollutant mixtures. Through simulations, we demonstrate the method's superiority by comparing its accuracy in estimating and quantifying uncertainties for both main and interaction effects with the commonly used method, BKMR. Last, we apply the method to analyze a multipollutant dataset with 10,110 participates from the Nurses' Health Study.