Sharp sensitivity bounds for mediation under unmeasured mediator-outcome confounding

TL;DR

This paper introduces a sensitivity analysis method to bound direct and indirect effects in mediation analysis, accounting for unmeasured mediator-outcome confounding without relying on parametric assumptions.

Contribution

It provides sharp bounds for mediation effects under unmeasured confounding, advancing causal inference methods in observational studies.

Findings

Develops a nonparametric sensitivity analysis technique

Provides sharp bounds for direct and indirect effects

Applicable even with unmeasured confounding

Abstract

It is often of interest to decompose a total effect of an exposure into the component that acts on the outcome through some mediator and the component that acts independently through other pathways. Said another way, we are interested in the direct and indirect effects of the exposure on the outcome. Even if the exposure is randomly assigned, it is often infeasible to randomize the mediator, leaving the mediator-outcome confounding not fully controlled. We develop a sensitivity analysis technique that can bound the direct and indirect effects without parametric assumptions about the unmeasured mediator-outcome confounding.