Some Aspects of the Bayesian Three-Arm Noninferiority Trial Design

• Activity Number: 639
• Type: Topic Contributed
• Date/Time: Thursday, August 7, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Learning and Data Mining
• Author(s): Samiran Ghosh*+ and Santu Ghosh
• Companies: Wayne State University and Wayne State University School of Medicine
• Keywords: Bayesian ; Continuous Outcome ; Non-Inferiority ; Three-arm trial

Abstract

With the availability of existing treatment options in many therapeutic areas, comparative effectiveness research (CER) is gaining higher importance for better and informed health care decisions. Randomized controlled clinical trials (RCTs) are an indispensable source of information about efficacy as well as effectiveness of treatments in almost any disease area including CER. The effect of including an active comparator arm in a RCT is immense. This gives rise to superiority and non-inferiority trials and in this talk we will focus our attention on the latter one. The development of statistical techniques for non-inferiority (NI) trial has a long history albeit predominantly via classical/frequentist approach. However, availability of historical placebo-controlled trial is useful and if integrated in the current NI design, can provide better precision for CER and Bayesian paradigm provides a natural path to integrate historical as well as current trial data via sequential learning in the NI setup. We will discuss both fraction margin and fixed margin based Bayesian approach for three-arm trial with continuous outcome.