Fig. 2

Estimated sample sizes by baseline mortality and absolute mortality reduction. This figure examines the total sample size needed to identify an absolute mortality reduction of 3 to 15 % assuming three control group mortality rates (30, 20, and 10 %). The assumptions in this figure is that power is 80 % for a two-sided test and that 1:1 randomization will be employed (for example, a total N of 3000 on the y-axis implies a n = 1500 in each treatment arm). Source: author calculations (MOH)