The Advantageous Adaptive Randomization in Clinical Trials

• Activity Number: 498
• Type: Topic Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Bob Zhong*+
• Companies: Johnson & Johnson
• Keywords: clinical trials ; randomization ; adaptive randomization ; balance

Abstract

As a statistician working in the pharmaceutical industry, a basic statistical question for a randomized study is: which randomization method is the best one for the study under consideration? To answer this question, criteria need to be established so that the "best" can be identified. In this presentation, these criteria are first discussed. Then, three familiar randomization methods, i.e., complete (simple) randomization, stratified permuted block randomization and dynamic randomization (the minimization procedure with biased coin assignment proposed by Pocock and Simon (1975)) are evaluated. We conclude that adaptive randomization is the preferred method of randomization since the adaptive randomization has more advantages over other two-randomization methods.