Heterogeneous indirect effects for multiple mediators using interventional effect models

DOI10.1515/EM-2020-0023zbMATH Open1466.92197arXiv1907.08415OpenAlexW3043414626MaRDI QIDQ829924FDOQ829924

Authors: Wen Wei Loh, Beatrijs Moerkerke, Tom Loeys, Stijn Vansteelandt 

Publication date: 6 May 2021

Published in: Epidemiologic Methods (Search for Journal in Brave)

Abstract: Decomposing an exposure effect on an outcome into separate natural indirect effects through multiple mediators requires strict assumptions, such as correctly postulating the causal structure of the mediators, and no unmeasured confounding among the mediators. In contrast, interventional indirect effects for multiple mediators can be identified even when - as often - the mediators either have an unknown causal structure, or share unmeasured common causes, or both. Existing estimation methods for interventional indirect effects require calculating each distinct indirect effect in turn. This can quickly become unwieldy or unfeasible, especially when investigating indirect effect measures that may be modified by observed baseline characteristics. In this article, we introduce simplified estimation procedures for such heterogeneous interventional indirect effects using interventional effect models. Interventional effect models are a class of marginal structural models that encode the interventional indirect effects as causal model parameters, thus readily permitting effect modification by baseline covariates using (statistical) interaction terms. The mediators and outcome can be continuous or noncontinuous. We propose two estimation procedures: one using inverse weighting by the counterfactual mediator density or mass functions, and another using Monte Carlo integration. The former has the advantage of not requiring an outcome model, but is susceptible to finite sample biases due to highly variable weights. The latter has the advantage of consistent estimation under a correctly specified (parametric) outcome model, but is susceptible to biases due to extrapolation.

Full work available at URL: https://arxiv.org/abs/1907.08415

Recommendations

• Mediation analysis with multiple mediators
• Nonlinear mediation analysis with high‐dimensional mediators whose causal structure is unknown
• Generalized causal mediation analysis
• Imputation strategies for the estimation of natural direct and indirect effects
• Estimation of Controlled Direct Effects

zbMATH Keywords

direct and indirect effectseffect modificationconditional effectseffect decompositionnatural effect models

Mathematics Subject Classification ID

Applications of statistics to biology and medical sciences; meta analysis (62P10) Epidemiology (92D30)

Cites Work

• Estimating treatment effect heterogeneity in randomized program evaluation
• Imputation strategies for the estimation of natural direct and indirect effects
• Causality. Models, reasoning, and inference
• Confounding and collapsibility in causal inference
• Title not available (Why is that?)
• A note on formulae for causal mediation analysis in an odds ratio context
• Causal mediation analysis with multiple mediators
• Robust estimation of inverse probability weights for marginal structural models
• Identifying Direct and Indirect Effects in a Non-Counterfactual Framework
• Mediation Analysis with time Varying Exposures and Mediators
• Natural direct and indirect effects on the exposed: effect decomposition under weaker assump\-tions
• Hypothesis test of mediation effect in causal mediation model with high-dimensional continuous mediators
• A Multiple-Testing Procedure for High-Dimensional Mediation Hypotheses

Cited In (5)

• The identification of direct and indirect effects in studies with an unmeasured intermediate variable
• Mediation analysis with multiple mediators under unmeasured mediator-outcome confounding
• Equivalence between direct and indirect effects with different sets of intermediate variables and covariates
• Mediation analysis with multiple mediators
• Direct and Indirect Treatment Effects–Causal Chains and Mediation Analysis with Instrumental Variables

Uses Software

• medflex

This page was built for publication: Heterogeneous indirect effects for multiple mediators using interventional effect models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q829924)