Advances in Mediation Analysis

• Activity Number: 115
• Type: Topic Contributed
• Date/Time: Monday, August 4, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Eric Tchetgen and Tyler VanderWeele and Caleb Miles *+
• Companies: Harvard School of Public Health and Harvard and Harvard
• Keywords: causality ; mediation ; natural direct effects ; confounding ; interaction ; monotonicity

Abstract

Natural direct and indirect effects formalize traditional notions of mediation analysis into a rigorous causal framework. Sufficient conditions for identification of these effects were formulated by Judea Pearl under a nonparametric structural equations model. A common situation in epidemiology is that a confounder of the mediator-outcome relationship is itself affected by the exposure, in which case natural direct effects are not identified without additional assumptions, even under Pearl's nonparametric structural equations model. We show that when a binary confounder of the mediator is affected by the exposure, the natural direct effect is nonparametrically identified under the model, assuming monotonicity about the effect of the exposure on the confounder. we also show that natural direct effects are more generally identified if there is no additive mean interaction between the mediator and confounders of the mediator affected by exposure. This latter assumption is quite appealing because it does not require monotonicity of effects of the exposure. Additionally, it places no restriction on the nature of the confounders of the mediator which can be continuous or polytomous.