Comments on ‘Two‐sample binary phase 2 trials with low type I error and low sample size’
Comments on ‘Two‐sample binary phase 2 trials with low type I error and low sample size’
复制标题
DOI：
10.1002/sim.7359
复制
发表时间：
2017-09
影响因子：
2
通讯作者：
G. Shan
中科院分区：
文献类型：
--
作者：
G. Shan
We read with great interest the article by Litwin et al. [1], ‘Two-sample binary phase 2 trials with low type I error and low sample size’. They proposed a new randomized two-arm study with a binary outcome as the primary end point by controlling the probabilities of early termination under null and alternative hypotheses, instead of the traditional criterion with the smallest expected sample size (optimal design) or the smallest maximum sample size (minimax design) [2]. Randomized clinical trials are the gold standard in many research areas to reduce biases from a study without a control group. Suppose pC and pE are the response rates of the control group and the experimental group, respectively. Then the hypotheses for testing the efficacy of the treatment can be expressed as
