A Bayesian Hierarchical Model for Meta-Analysis of Rare Binary Adverse Event Data

• Activity Number: 307
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Ou Bai*+ and Xinlei Wang and Min Chen and Guanghua Xiao
• Companies: Southern Methodist University and SMU and The University of Texas Southwestern Medical Center at Dallas and UT Southwestern Medical Center
• Keywords: Bayesian hierarchical model ; ; Rare binary advervse data ; Parameter estimate ; Hypothesis test

Abstract

It is important to assess drug safety by comparing the incidence of adverse events between two treatment arms in clinical studies. Meta-analysis has been found to be useful by combining multiple studies involving the same adverse event. When the events are rare, most standard meta-analysis methods are problematic as estimation of sparse data are often biased. We propose a Bayesian hierarchical model for meta-analysis of rare binary adverse event data. We show the estimates of the treatment effect and the between-study heterogeneity parameter from the Bayesian method are less biased than the traditional moment based methods. We further develop testing procedures based on a Bayesian model selection approach.