Likelihood-Based Methods for Exposure-Adjusted Incidence Data Analysis in Clinical Trials

• Activity Number: 674
• Type: Topic Contributed
• Date/Time: Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Jing Xu*+
• Companies: Millennium Pharmaceuticals, Inc.
• Keywords: Exposure Adjusted Event Rates ; Parametric Approach ; Annualized Event Rate ; Person-year rate ; Incident Density Rate

Abstract

In clinical trials, likelihood based methods, model based or non-model based, can be used for incidence density data analysis. In this talk, we will first review the most frequently used likelihood based approaches in epidemiology studies and cases they have been applied in phase III clinical trials to summarize clinical adverse experience or efficacy data. Limitations of these methods will be discussed. We will further discuss appropriate alternative model based approaches for phase III trial exposure adjusted incidence data analysis. Comparison of these approaches based on simulation results will be presented.