A Weighted MMRM to Evaluate Overall Treatment Effectiveness

• Activity Number: 667
• Type: Contributed
• Date/Time: Thursday, August 4, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Tao Song*+ and Abdul J. Sankoh
• Companies: Vertex Pharmaceuticals and Vertex Pharmaceuticals
• Keywords: Overall treatment effectiveness ; Power ; Bias

Abstract

In controlled clinical trials, subjects are often treated over a period of time. The same outcome variable is measured repeatedly at a number of scheduled visits, and the study objective is to evaluate treatment effectiveness over time. In recent years, Mixed-effects Model for Repeated Measures (MMRM) has become one of the preferred approaches for handling missing data in clinical trials. When evaluating overall treatment effectiveness with MMRM, treatment differences at each clinical visit are estimated and equal weights are utilized to combine the visit-specific treatment differences. In this presentation, we propose the use of a weighted method to combine the visit-specific estimates. For a special case when the underline marginal covariance matrix of clinical data is compound symmetry, weighted MMRM is equivalent to MMRM. When the marginal covariance is of an unstructured pattern, we provide simulation study results that show the weighted MMRM has increased statistical power and less bias under missing not at random (MNAR) as compared to MMRM.