Bayesian Inference for Noninferiority Test of Matched-Pair Binary Data via Continuous Latent Model

• Activity Number: 134
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Yi Qian*+ and Deukwoo Kwon and Jeesun Jung and Jun-mo Nam
• Companies: Amgen, Inc. and University of Miami and National Institute on Alcohol Abuse and Alcoholism and National Cancer Institute
• Keywords: matched-pair ; bayesian ; latent model

Abstract

For a clinical trial with binary outcome, with positive within-pair correlation, the matched-pair design is superior in power compared to a two-sample design, and is commonly used in small-sample trials. McNemar test is well-known in this setting. Several Bayesian approaches were also developed, but the implementation requires complicated computation techniques (Ghosh et al. 2000; Shi et al. 2008, 2009). Here we propose a simplistic Bayesian approach using continuous latent model for non-inferiority test with either difference or ratio of marginal probabilities. External information can also be easily incorporated into the model. We conduct simulation study and present a real data example.