A Bayesian approach for inference from a bridging study with binary outcomes.

To address concerns that ethnic differences in any clinical trial may affect the performance of a treatment, separate clinical trials may be required for different regions. If a drug is approved in one region based on a bigger clinical trial, a bridging study with smaller size may be conducted in a new region to evaluate the similarity of the earlier clinical data to the new population. This article applies a Bayesian strategy to combine a bridging study with reference studies to assess treatment effect similarity for binary endpoints. The methods have been illustrated with a numerical example using Markov-chain Monte Carlo (MCMC) techniques.