A Decision Theoretic Approach to Sample Size Determination for Clinical Trials

In this talk a Bayesian decision theoretic approach to the sample size determination problem for clinical trials is presented. In contrast to the most frequently used methods, which are based on the required size and power of the trial for a specified treatment effect, this approach treats the problem as a decision problem and employs a utility function to find the optimal sample size of the trial. The expected net benefit function which is the expected benefit of subsequent use of a new treatment, under consideration, minus the cost of conducting the trial is maximized. Three decision makers namely; the drug company which decides on the size of the trial, patients or their medical advisors who decide whether or not to use the new treatment and regulatory authorities who decide whether or not to grant a license to a new treatment will be considered.