Bayesian Design of Noninferiority Clinical Trials via Conditional Borrowing of Historical Data with Applications

• Activity Number: 522
• Type: Topic Contributed
• Date/Time: Wednesday, August 7, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Ming-Hui Chen*+
• Companies: University of Connecticut
• Keywords: Decision Rule ; Fitting Prior ; Markov chain Monte Carlo ; Sampling Prior ; Simulation ; Type I Error

Abstract

We develop a general decision-rule based Bayesian method for the design of non-inferiority clinical trials with a focus on controlling type I error and power. The idea of conditional borrowing is based on comparability between the future data and the historical data. The fitting priors are constructed to allow for incorporation or no incorporation of historical data while the sampling priors are constructed for computing Bayesian Type I error and power. Various properties of the proposed Bayesian methodology are examined and simulation-based computational algorithms are developed. The proposed methodology is applied to the design of a non-inferiority medical device clinical trial with historical data from previous trials to demonstrate superiority of the Bayesian method in sample size reduction.