Online Program Home
My Program

Abstract Details

Activity Number: 244 - Missing Data; Causal Inference
Type: Contributed
Date/Time: Monday, July 30, 2018 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #328540 Presentation
Title: Assessing Indirect Effect in a Mediation Model with a Censored Mediator
Author(s): Jian Wang* and Sanjay Shete
Companies: The University of Texas MD Anderson Cancer Center and The University of Texas MD Anderson Cancer Center
Keywords: mediation model; censored mediator; indirect effect
Abstract:

A mediation model is a statistical approach assessing the direct and indirect effects of an initial variable on an outcome by including a mediator. In practice, investigators can observe censored data. Currently, most approaches for mediation model with censored data focus on censored outcomes but not censored mediators. In this study, we proposed an approach to assess the indirect effect in a mediation model where the mediator is a censored variable, using the accelerated failure time model and a multiple imputation approach. Using simulations, we established the bias in estimating coefficients of different paths in the mediation model and indirect effects when using the existing approaches (i.e. naïve approach, complete-case analysis, Tobit mediation model). We conducted simulation studies to investigate the performance of the proposed approach and compare it to that of the existing approaches. The proposed approach accurately estimates the coefficients of different paths and the indirect effects. We applied the proposed and existing approaches to investigate the indirect effect of age at menopause on the association between SNPs and fasting glucose levels.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program