Establishing Sample Size in Clinical Trials Having Multiple Hypotheses

• Activity Number: 256
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Alan Barry Davis*+ and Mary M. Poole and Young Kim
• Companies: Pharmanet Development Group, Inc. and Pharmanet Development Group, Inc. and Pharmanet Development Group, Inc.
• Keywords: power ; three-arm clinical trial ; equivalence ; superiority

Abstract

Many clinical trials have as their objective the rejection of one or more null hypothesis. A clinical trial having k>2 groups can require significance of multiple pre-planned comparisons in order to meet a specific objective. The probability of reaching this objective, or Power, is the probability that a hypothetical alternative to the null hypothesis falls within the region of rejection determined by the joint comparisons. As an example, a three-group parallel trial is considered, in which each subject is to be evaluated as a treatment success or failure. Subjects in group one will receive a new, or test, product; group two subjects will receive a standard reference product, and subjects in group three will receive a placebo.

This poster will present the power of rejecting the individual and joint comparisons as a function of sample size and treatment-group success rates.