Fig. 1From: Type I error rates of multi-arm multi-stage clinical trials: strong control and impact of intermediate outcomesEffect of θ I on the PWER. The PWER of two 2-stage I≠D designs when the null effect on D is true and the underlying treatment effect on I varies. θ I is the true log HR on the I outcome and α j is the nominal significance level in the jth stage (j=1,2). HR hazard ratio, PWER pairwise type I error ratesBack to article page