A1366
Title: Estimation of heterogeneous causal mediation effects
Authors: Yi Zhao - Indiana University (United States) [presenting]
Abstract: Mediation analysis is widely used to elucidate the mechanisms through which an exposure influences an outcome of interest. However, standard causal mediation methods may yield inconsistent conclusions, potentially due to unmodeled heterogeneity in mediation effects across individuals. A new framework is proposed that incorporates covariate-treatment and mediator-treatment interactions within a linear structural equation modeling system. Causal assumptions are discussed and heterogeneous natural direct and indirect effects are parameterized as functions of patient characteristics. A modified covariate approach is introduced to relax hierarchical constraints in models with interactions, and generalized lasso regularization is employed to achieve parsimony in high-dimensional settings. Asymptotic properties of the proposed estimators are established and finite-sample performance is demonstrated through extensive simulation studies. Application to data from the Alzheimer's Disease Neuroimaging Initiative reveals substantial heterogeneity in the mediation effect of cerebrospinal fluid volume change on the association between APOE-$\varepsilon4$ carrier status and cognitive decline, identifying a subgroup of individuals who may be particularly susceptible to cognitive impairment through this Disease pathway.
