Online Program Home
My Program

Abstract Details

Activity Number: 232
Type: Topic Contributed
Date/Time: Monday, August 1, 2016 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #319971
Title: Meta-Mediation Analysis for Generating Mechanistic Hypotheses
Author(s): John Jackson* and Tyler J. VanderWeele
Companies: Harvard and Harvard
Keywords: mediation analysis ; natural indirect effect ; meta-analysis ; product method ; difference method

Investigators planning a mediation analysis study may want to know the possible magnitude and direction of an indirect effect from the published literature. In this paper we outline and compare methods for estimating natural indirect effects from meta-analysis of published studies of (a) total effects and direct effects on the outcome via an analogue of the difference method (b) effects of the exposure on the mediator and effects of the mediator on the outcome via an analogue of the product method. Through simulation we show that the product-method will often be less biased than the difference method, given the typical confounding control in their respective source studies.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association