|
Activity Number:
|
531
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #308876 |
|
Title:
|
Causal Approaches to Mediation Analyses
|
|
Author(s):
|
Thomas R. Tenhave*+ and Jennifer Faerber and Marshall Joffe
|
|
Companies:
|
University of Pennsylvania and University of Pennsylvania and University of Pennsylvania
|
|
Address:
|
607 Blockley Hall, Philadelphia, PA, 19104-6021,
|
|
Keywords:
|
|
|
Abstract:
|
In the context of two randomized psychiatry trials of behavioral interventions, we present a linear rank preserving model approach for analyzing the mediation of a randomized baseline intervention by a post-randomization factor on a follow-up outcome. Unlike standard mediation analyses, our approach does not assume that the mediating factor is randomly assigned to individuals (a form of sequential ignorability). However, there is a tradeoff with other assumptions. Consistent estimation of causal mediation effects without sequential ignorability employs weights under the G-estimation approach that are optimal in terms of semi-parametric efficiency but under sequential ignorability. In this context, we will present analyses of mediation of behavioral intervention effects on reducing depression outcomes by medication and therapy factors in the psychiatry trials.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |
