Confidence Intervals for a Secondary Parameter in Group Sequential Trials with Unknown Correlation

• Activity Number: 195
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: WNAR
• Author(s): Timothy Skalland*
• Keywords: Clinical Trial ; Group Sequential Methods ; Secondary Parameters ; Multiple Endpoints ; Confidence Intervals

Abstract

Group sequential trials are quite commonplace in current clinical studies. We consider a group sequential clinical trial with a primary and secondary endpoint, where the secondary endpoint is tested only if the primary endpoint is found significant (or the trial goes to maximal sample size). We extend the work of Whitehead et al. (2000) to consider situations when the correlation is unknown and must be estimated. We examine the effect of confidence interval construction for the secondary stopping mean as a result of estimating the correlation and compare Whitehead's approach to the sample mean ordering approach of Emerson and Fleming (1990).