Decision Analysis and Bayesian Statistics for Medical Device Studies
View Presentation View Presentation
*Laura Thompson, FDA 

Keywords: decision model, pediatrics, hierarchical model, bayesian

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 effectiveness of devices indicated for pediatrics. Considerations for using adult data include whether the disease profile is the same in adults and pediatrics, and whether the device effect is the same. To help FDA make an appropriate determination, we are developing a decision model to evaluate the optimal action for a given device. Possible actions include 1)running a stand-alone study in pediatrics, 2) using only adult data to make an approval decision for pediatrics, and 3) building a Bayesian hierarchical model where a small pediatric study borrows information from adult studies. Different medical devices will have different optimal decisions based on device and disease characteristics.