Session Ready
Exercise

Stopping rules

To avoid prolonging a study, a trial may be stopped early if there is strong evidence for efficacy at an interim analysis. Here we will derive the p-values needed to stop a trial early under the Pocock and O’Brien-Fleming spending functions.

We are assuming two-sided tests under an overall 5% significant level. The gsDesign package has been loaded for this session.

Instructions
100 XP
  • Three interim analyses and a final analysis (four in total) are planned for a study.
  • Use the gsDesign() function within the gsDesign package to generate the p-values needed under the Pocock (Fixed Nominal) Rule.
  • Find the equivalent needed under the O’Brein-Fleming Rule.