Sample Size Re-Estimation Using Re-Sampling to Correct for Initial Trial Assumptions

• Activity Number: 591
• Type: Topic Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Jeff Maca*+
• Companies: Quintiles
• Keywords: Sample size re-esitmation ; Adaptive designs ; re-sampling

Abstract

In designing a clinical study, one important aspect is to correctly determine the sample size needed to have the desired power for the study. However, it is often the case that there is no precise information available prior to the start of the study which would allow for the sample size calculations to be performed with confidence. The process can be further complicated if the models used in the final primary analysis have many unknown parameters, such as covariates and correlations, or if there is complicated multiplicity schemes are in place. One approach to this problem would be to gather information about the design parameters during the study, and then update the final sample size estimates based on estimation of these parameters at an interim analysis. An alternative approach proposed to use re-sampling to best estimate the needed sample size to obtain the desired power for the study. For all methods, it must be ensured that proper statistical methodology is in place to ensure the validity of the final tests used in the primary analysis.