A0856
Title: Heterogeneous causal mediation analysis with Bayesian additive regression trees
Authors: Chen Liu - University of Pittsburgh (United States) [presenting]
Xu Qin - University of Pittsburgh (United States)
Jiebiao Wang - University of Pittsburgh (United States)
Abstract: Causal mediation analysis can help explain how an exposure affects an outcome. The mediation effects are often heterogeneous based on individual characteristics, but most existing methods ignore this heterogeneity and estimate the population average effects. To address this gap, a heterogeneous causal mediation analysis method is developed using Bayesian regression tree ensembles. Distinct from traditional methods, the approach captures complex non-linear interactions and heterogeneous effects in mediation processes more flexibly, offering a refined understanding of the heterogeneity of causal mechanisms. By sampling from the posterior trees of mediator and outcome models, rigorous credible intervals are obtained for causal mediation effects. Partial dependent plots are also used to illustrate which moderators play more important roles and how each effect changes with a moderator. Utilizing simulated datasets, the superiority of the approach is demonstrated in the accurate estimation and inference of heterogeneous mediation effects, especially in scenarios characterized by non-linear relationships and interaction effects. The proposed method is applied to estimate heterogeneous mediation effects in genetic mechanisms of Alzheimer's disease.
