3.3 Unblinded SSR Methods
3.3.1 Overview
Unlike blinded sample size re-estimation methods described in the previous section, the unblinded SSR procedures utilize both the nuisance parameter and explicit treatment effect estimates in order to calculate the new sample size for the remainder of the trial. The methods vary in how they perform such adaptation but they all share a common feature: the methodology is focused on rigorously controlling the type I error. Type I error is very important to regulators in confirmatory trials (the majority of unblinded SSR trials are performed in confirmatory setting). Type I error inflation can potentially arise from multiple looks at the data (interim and final analyses) and data-driven sample size changes. ...
Get Modern Approaches to Clinical Trials Using SAS: Classical, Adaptive, and Bayesian Methods now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
