The Power Prior in Bayesian Designs of Clinical Trials

• Activity Number: 464
• Type: Invited
• Date/Time: Wednesday, August 12, 2015 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Ming-Hui Chen*
• Companies: University of Connecticut
• Keywords: Fitting Prior ; Historical data ; Meta-Experimental Design ; Power ; Sampling Prior ; Recurrent Event Data

Abstract

Recently, Chen et al. (2011) developed a general Bayesian methodology to determine sample size for a non-inferiority trial with a focus on controlling the type I error and power. The Bayesian SSD allows for the incorporation of historical data via the power prior, which leads to a substantial reduction in the sample size compared to frequentist SSD. In this talk, we discuss the theoretical properties and computational algorithms of Bayesian SSD as well as variations in its formulation. We also review various models and design settings of Bayesian SSD using the partial borrowing power prior and partial discounting power prior. The Bayesian SSD and use of the power prior are illustrated in the design of a non-inferiority medical device clinical trial, a Bayesian meta-experimental design, and the design of a Bayesian superiority trial with recurrent events data.