|
Abstract:
|
Natural indirect effects (NIEs) are mediated effects that can be identified when the exposure does not affect any confounders of the mediator-outcome relationship. To circumvent this assumption, so-called randomized interventional analog indirect effects (NIE^Rs), which can be identified even in the presence of exposure-induced confounding, have gained popularity in the causal mediation literature. An essential property that a putative indirect effect must possess in order to have a true mediation/indirect effect interpretation is that it must be zero whenever there is no one for whom both their exposure affects their mediator and their mediator affects their outcome. In this talk, I will demonstrate that without additional assumptions, the NIE^R does not satisfy this property. Further, I will provide examples of such additional assumptions under which this property can be recovered. Unfortunately, the NIE will also be identified under most of these additional assumptions, and so the NIE^R will provide little advantage over the NIE. Thus, while the NIE^R can tell interesting stories about joint interventions, it cannot always be relied upon to tell stories about mediation.
|
