Clinically Relevant Mediation Analysis using Controlled Indirect Effect

TL;DR

This paper introduces a new mediation analysis method for multiple manipulable mediators with arbitrary causal dependencies, focusing on effects after manipulating a single mediator, which is more clinically relevant and interpretable.

Contribution

The proposed method allows for mediation analysis with multiple mediators without cross-world assumptions, enabling clinical interpretation by focusing on single mediator manipulation.

Findings

Method applied to simulated, political, and clinical datasets.

Provides guidance for manipulating mediators to optimize outcomes.

Avoids assumptions that are untestable in experiments.

Abstract

Mediation analysis allows one to use observational data to estimate the importance of each potential mediating pathway involved in the causal effect of an exposure on an outcome. However, current approaches to mediation analysis with multiple mediators either involve assumptions not verifiable by experiments, or estimate the effect when mediators are manipulated jointly which precludes the practical design of experiments due to curse of dimensionality, or are difficult to interpret when arbitrary causal dependencies are present. We propose a method for mediation analysis for multiple manipulable mediators with arbitrary causal dependencies. The proposed method is clinically relevant because the decomposition of the total effect does not involve effects under cross-world assumptions and focuses on the effects after manipulating (i.e. treating) one single mediator, which is more relevant in a clinical scenario. We illustrate the approach using simulated data, the "framing" dataset from political science, and the HIV-Brain Age dataset from a clinical retrospective cohort study. Our results provide potential guidance for clinical practitioners to make justified choices to manipulate one of the mediators to optimize the outcome.