Bayesian variable selection approach for the simultaneous investigation of adverse event and laboratory data
View Presentation *Bradley McEvoy, CDER, FDA Keywords: Drug safety, Bayesian variable selection, high-dimensional data analysis Investigating a safety issue is a complex process, often requiring the evaluation of adverse event (AE) and laboratory data. While these data-streams provide distinct, yet supportive information for the safety issue at hand, there may be a desire to combine them. For instance, one may want to adjust the comparative risk summary for an AE (e.g., odds ratio) by accounting for whether there are differences for a laboratory parameter that serves as marker for the safety issue. Several features of the data make such an endeavor difficult, including AE data being categorical whereas laboratory measurements are continuous. In this talk these two data-streams are integrated using a Bayesian model selection framework. This work is motivated and illustrated using a drug for the bowel preparation for colonoscopy, where the safety issue of interest is electrolyte changes.
|
Key Dates
-
April 30 - May 22, 2013
Invited Abstract Submission Open -
June 4, 2013
Online Registration Opens -
August 9 - August 23, 2013
Invited Abstract Editing -
August 23, 2013
Short Course materials due from Instructors -
August 26, 2013
Housing Deadline -
September 9, 2013
Cancellation Deadline and Registration Closes @ 11:59 pm EDT -
September 16 - September 18, 2013
Marriott Wardman Park, Washington, DC