BAYES DISCRETE SEQUENTIAL BOUNDARIES FOR CLINICAL TRIALS

Abstract

In clinical trials, the use of sequential stopping rules or stopping boundaries has been advocated from both frequentist and Bayesian perspectives. Elsewhere, we have shown that a large family of untruncated stopping boundaries that are of interest from a frequentist point of view are also Bayes optimal boundaries. In this paper we examine a specific class of truncated rules, focusing especially on those related to work of [1]. We show that it is possible to choose a cost and loss structure such that stopping boundaries of the type discussed by Lan and DeMets are also optimal Bayesian boundaries.

Keywords:

• Bayes sequential decision procedure
• Brownian motion
• Clinical trials
• Group sequential design
• Spending function

ACKNOWLEDGMENTS

Thanks are due to a referee and an associate editor for their detailed, constructive and thoughtful comments and suggestions which we believe have led to a significant improvement to this paper.