Recent Developments of Bayesian Meta-Analysis for Safety Evaluation in Randomized Clinical Trials

• Activity Number: 493
• Type: Topic Contributed
• Date/Time: Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Karen Price*+
• Companies: Eli Lilly and Company
• Keywords: Bayesian ; meta-analysis ; safety ; clinical trials

Abstract

The use of Bayesian meta-analysis methods in the context of safety evaluation provides a number of advantages over the use of traditional approaches alone. The Bayesian approach allows the synthesis of evidence from multiple sources with structures that incorporate multiple treatments, which is common when combining multiple RCTs. One can also incorporate other sources of evidence via prior distributions. All results are expressed in terms of probability statements which greatly enhance the decision maker's ability to evaluate potential safety signals. For example, it is straightforward to obtain the posterior probability of an odds ratio for a safety event of interest, comparing treatment to control, being greater than a number of different thresholds of interest (e.g. 1.4 and 1.8). Additional advantages include the ability to easily model trial heterogeneity and fully incorporate and assess uncertainty, so vital for safety evaluation.

In this presentation, we provide an overview of Bayesian meta-analysis, with an emphasis on the advantages of using Bayesian meta-analysis for safety evaluation and provide a case example.