Methods to Test Treatment-by-Site Interaction Using Mediated Moderation: An Application to the TORDIA Clinical Trial

• Activity Number: 668
• Type: Contributed
• Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Kaleab Abebe*+ and Satish Iyengar and David A. Brent
• Companies: University of Pittsburgh and University of Pittsburgh and University of Pittsburgh
• Keywords: mediation ; moderation ; multisite ; trial ; logistic ; interaction

Abstract

There is little discussion about methods to explain treatment-by-site interaction in multisite clinical trials, so investigators must explain these differences post-hoc with no formal statistical tests in the literature. We extend mediated moderation techniques to logistic regression with two treatments, multiple sites, and multiple mediators. Test statistics are derived and simulation is used to estimate power for difference-in-coefficients and product-of-coefficients tests. Finally, the proposed methodology is applied to the Treatment of SSRI-Resistant Depression in Adolescents trial, which concluded that the combination of cognitive behavioral therapy and antidepressant medication had an effect on clinical response that was highly variable across six sites. The contribution of this is formal significance tests for explaining treatment-by-site interaction in multisite clinical trials.