On Partial Identification of the Pure Direct Effect

TL;DR

This paper extends bounds for partial identification of the pure direct effect in causal mediation analysis to polytomous mediators and examines the sensitivity of inferences to model assumptions using real data.

Contribution

It generalizes existing bounds to polytomous mediators and explores identification under fewer assumptions, enhancing causal mediation analysis.

Findings

Bounds are extended to polytomous mediators.

Inference sensitivity depends on model assumptions.

Application to HIV treatment data illustrates practical implications.

Abstract

In causal mediation analysis, nonparametric identification of the pure (natural) direct effect typically relies on, in addition to no unobserved pre-exposure confounding, fundamental assumptions of (i) so-called "cross-world-counterfactuals" independence and (ii) no exposure- induced confounding. When the mediator is binary, bounds for partial identification have been given when neither assumption is made, or alternatively when assuming only (ii). We extend existing bounds to the case of a polytomous mediator, and provide bounds for the case assuming only (i). We apply these bounds to data from the Harvard PEPFAR program in Nigeria, where we evaluate the extent to which the effects of antiretroviral therapy on virological failure are mediated by a patient's adherence, and show that inference on this effect is somewhat sensitive to model assumptions.