Incorporation of Prior Information in a Clinical Trial

• Activity Number: 297
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Manuela Buzoianu*+
• Companies: FDA
• Keywords: medical device ; clinical trial ; Bayesian statistics ; informative prior ; elicitation

Abstract

The use of informative prior is not common in the Bayesian clinical trials submitted to FDA. In the context of medical devices, when there is value recognized in the prior clinically relevant data or similar studies, Bayesian methodology may be adopted in designing a new trial in which such initial data is used as an informative prior. A reason is that the new trial might have smaller sample size or duration by borrowing strength from previous studies. However, limited new data collected in the new trail could be dominated by the prior. To avoid overly influencing the trial results, a discounting weight can be applied to the prior data presumably based on elicitation from "experts". In this talk I will discuss some aspects of informative borrowing, specifically the influence of the informative prior on the analysis of a Bayesian trial.