Activity Number:
|
293
|
Type:
|
Invited
|
Date/Time:
|
Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Health Policy Statistics*
|
Abstract - #300168 |
Title:
|
Bayesian Methods in Cost-Effectiveness Analysis
|
Author(s):
|
Marjorie Rosenberg*+ and Dennis Fryback and Grace Flood and Bernie O'Brien and Ana Johnson-Masotti
|
Affiliation(s):
|
University of Wisconsin, Madison and University of Wisconsin and University of Wisconsin, Madison and McMaster University and Medical College of Wisconsin
|
Address:
|
975 University Avenue, Madison, Wisconsin, 53706-1323, USA
|
Keywords:
|
Economic analysis ; Clinical trials data ; Decision modeling
|
Abstract:
|
Cost-effectiveness analysis of one intervention versus another intervention to treat a medical condition requires the estimation of costs and health effects (such as mortality adjusted for quality of life) at periodic intervals over the lifetime of pertinent individuals. We will illustrate a Bayesian approach to estimate needed quantities using primary data from the Canadian Implantable Defibrillator Study (CIDS). The use of the implantable cardioverter defibrillator to reestablish normal cardiac rhythm will be compared to treatment using anti-tachyarrhythmic drugs, principally Amiodarone. Censored data is a particular problem in cost-effectiveness analyses using primary clinical trial data. Bayesian methods are useful in modeling censored data where the usual assumption of independence between failure and censored times is violated. Bayesian methods also allow for the incorporation of prior information from preceding randomized trials as was the case with the CIDS. We will present a Bayesian cost-effectiveness model that includes the correlational structure between survival and resource utilization and the uncertainty of our assumptions.
|