Figure 5.
(a) Estimated type I error rate and 95% CI under various scenarios. (b) Estimated type I error rate and 95% CI under various design options. Adaptation rule #1: adapts percentage randomised to each dose in proportion to the current estimate of the probability that it is efficacious in MBL reduction. Adaptation rule #2: alters randomisation according to the precision of the estimated response at the ED95 (the minimum dose with near-maximal efficacy). (c) Estimated type I error rate and 95% CI, overall and under various design options.