A0545
Title: Mediation in causal survival analysis under competing risks using longitudinal modified treatment policies
Authors: Ivan Diaz - NYU Langone Health (United States) [presenting]
Abstract: Longitudinal modified treatment policies (LMTP) have been recently developed as a novel method to define and estimate causal parameters that depend on the natural value of treatment. LMTPs represent an important advancement in causal inference for longitudinal studies as they allow the non-parametric definition and estimation of the joint effect of multiple categorical, numerical, or continuous exposures measured at several time points. The LMTP methodology is extended to mediation problems with time-varying mediators. Identification results and non-parametric locally efficient estimators that use flexible data-adaptive regression techniques are presented to alleviate model misspecification bias while retaining important asymptotic properties such as root-n-consistency. An application in the estimation of the (positive) effect of intubation on survival amongst hospitalized COVID-19 patients is presented, and it is decomposed into its (negative) effect through acute kidney injury and its (positive) effect through other paths.
