Skip to main content

Table 2 Guidelines for stopping accrual to the ith research arm at intermediate analyses

From: Issues in applying multi-arm multi-stage methodology to a clinical trial in prostate cancer: the MRC STAMPEDE trial

Activity Stage Timing of analysis: control arm events Critical value
(Hazard Ratio)
I 114 1·00
II 215 0·92
III 334 0·89
  1. The intermediate analyses are timed according to the number of events reports on the control arm. Discontinuation of recruitment to a research arm will be considered if the observed hazard ratio is greater than the critical value (also known as the critical cutpoint or "hurdle")