|
Activity Number:
|
235
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #309992 |
|
Title:
|
Bayesian Approach to Meta-Analysis in Medical Device Trials Using Mixed Effects Models
|
|
Author(s):
|
Lijuan Deng*+ and Hong Wang and Liang Li
|
|
Companies:
|
Boston Scientific Corporation and Boston Scientific Corporation and Genzyme Corporation
|
|
Address:
|
100 Boston Scientific Way, Marlborough, MA, 01752,
|
|
Keywords:
|
Bayesian approach ; meta-analysis ; Monte Carlo Markov chain ; medical device trial ; mixed effects models
|
|
Abstract:
|
A mixed effects model allowing for study-level variability is proposed to incorporate treatment effects from different medical device trials. This model accounts for variability at both patient-level and study-level. The patient-level treatment effects are assumed to have their own distributions within each study, while the study-level treatment effect follows a common underlying distribution across studies. In this Bayesian meta-analysis model, the posterior distributions and credible intervals of treatment effect are estimated using Markov Chain Monte Carlo method with WinBUGS, by adjusting for other baseline covariates such as lesion length, vessel diameter and diabetic status.
|
