An Intent-to-Treat Analysis Using a Mixed Model Approach and Data Collected After Termination of Treatment Protocol

• Activity Number: 379
• Type: Topic Contributed
• Date/Time: Thursday, August 15, 2002 : 8:30 AM to 10:20 AM
• Sponsor: IMS
• Author(s): Sati Mazumdar*+ and Patricia Houck and Kenneth Liu and Benoit Mulsant and Mary Amanda Dew and Charles Reynolds
• Affiliation(s): University of Pittsburgh and University of Pittsburgh and Merck & Company, Inc. and University of Pittsburgh and University of Pittsburgh and University of Pittsburgh
• Keywords: Pragmatic analysis ; Explanatory analysis ; Mixture Models

Abstract

Following patients for a period of time after termination of treatment protocols is a common practice in clinical trials of drug treatments. After termination of the protocol treatment, patients are usually provided with medical care as deemed appropriate. This paper presents an intent-to-treat (IT) analysis using both "off-treatment" and "non-treatment" data from a clinical trial comparing nortriptyline and paroxetine in the treatment of major depression in older patients. All patients gave their consent on entering the study for systematic assessment during their clinical follow-up. Two-piece spline models conditional on the protocol treatment dropout time are developed first, which are combined to a mixture model (Hogan and Laird. Biometrics, 1996;52: 1002-1017). The model allows the parameters of the splines to depend on the dropout time in the pragmatic analysis. Statistical inference is based on bootstrap sampling procedures. Comparable results were seen for the pragmatic analysis, random regression analysis, and the completer analysis, with no statistically significant differences between the two treatments.