Decision Analysis and Bayesian Statistics for Medical Device Studies

• Activity Number: 165
• Type: Topic Contributed
• Date/Time: Monday, July 30, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Laura Thompson*+
• Companies: FDA/CDRH
• Keywords: decision model ; pediatrics ; hierarchical model ; bayesian

Abstract

Abstract: The Pediatric Medical Device Safety and Improvement Act of 2007 provides the Center for Devices and Radiological Health at FDA the ability to use adult data to help with approval decisions regarding the effectiveness of medical devices indicated for pediatrics. Some of the considerations for using adult effectiveness data include whether the disease profile is the same in adults and pediatrics, and whether the device effect is the same. In order to help FDA make an appropriate determination, we have developed a decision model to evaluate the optimal action for a given device. The possible actions include 1) running a full-sized, stand-alone study in pediatrics, 2) not running a study in pediatrics but using adult data to make an approval decision for pediatrics, and 3) building a Bayesian hierarchical model where a small pediatric study can borrow information from previous adult studies. Since different medical devices will have different optimal decisions based on characteristics of the device and the disease treated, these characteristics sh