Practical Bayesian Computation in Clinical Trials

• Activity Number: 172
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Computing
• Author(s): Karen Lynn Price*+ and Fang Chen and Baoguang Han
• Companies: Eli Lilly and Company and SAS Institute Inc. and Eli Lilly and Company
• Keywords: Bayesian ; computation ; SAS MCMC ; clinical trials

Abstract

In the field of medical product development, Bayesian statistical analysis is gaining acceptance for the design and analysis of clinical trials. For many problems, the Bayesian approach offers alternatives to classical approaches that are more practical and effective. Furthermore, recent advances in computational tools are alleviating the difficulty of implementing Bayesian methods by providing industry statisticians with greater flexibility and improved usability. We will demonstrate these advances in current applications to clinical trials, including network meta-analysis, methods that formally borrow historical data, and assessment of the probability that a future study will succeed. Each application is illustrated with the MCMC procedure in SAS/STAT software.