On Bayes Interim-Analyses for Dynamic Safety Monitoring
Hal Li1, William Wang1
1Merck Research Laboratories, 351 N Sumneytown Pike, North Wales, PA 19454
Abstract
During clinical development and product life cycle of pharmaceutical products, safety monitoring is a dynamic, iterative process for safety evaluation, risk management and risk communication. Bayesian approach designed for learning and adaptive decision making is a natural methodology for these evaluations.
In this presentation, we discuss a proactive approach for monitoring safety profile, in response to unexpected events and events of special interest. The Bayesian interim analysis methods are discussed. We have developed a hybrid approach that combines Bayesian modeling with the frequentist approach to interim-analysis. The implications of this approach are presented, including Bayesian stopping rules as well as current FDA guidance on Bayesian analysis. We will also address the process of controlling for the possibility of false finding.
|