|
Activity Number:
|
628
- Complex Data Analysis with Mental Health Applications
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2018 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Mental Health Statistics Section
|
|
Abstract #330317
|
|
|
Title:
|
Mediation Analysis with Item Response Theory Model for Ordinal Variables: An Application to a Hypothetical Clinical Trial
|
|
Author(s):
|
Yun Zhang* and Xiang Li and Pilar Lim
|
|
Companies:
|
Janssen Research & Development, LLC and Statistics and Decision Sciences, Janssen Research & Development, LLC and Janssen Research & Development, LLC
|
|
Keywords:
|
ordinal mediation analysis;
Item response theory;
counterfactual approach
|
|
Abstract:
|
Mediation analysis is a useful tool to understand the mechanisms for how the intervention affects outcome. The expansion of mediation analysis to ordinal mediators and/or ordinal outcomes are challenging. In this paper, we propose mediation analysis techniques to estimate direct and indirect effect by combining Item Response Theory (IRT) models and the standard mediation analysis approaches. The use of IRT models allows us to extend the well-studied mediation analysis approaches for continuous variables to ordinal variables by transforming the original ordinal scale to continuous latent score. We conducted simulation studies to evaluate and compare the performance of the proposed method to other available approaches for ordinal outcomes. We will discuss the application of the method to a hypothetical clinical trial where an endpoint was measured using a 7-item rating scale.
|
Authors who are presenting talks have a * after their name.
