在第二階段臨床試驗中，Simon (1989) 二階段設計是最經典和廣為人知的單臂設計之一。Simon 兩階段設計解決的問題是檢測某種試驗藥物的反應機率是否低於某一低水準，其檢測點必须是二元的。而Tsou 等人(2008) 對於某一試驗藥物有效性的平均水準是否低於某一低水準的單臂設計，在將二元點改為連續點方面，為Simon 兩階段設計的一個修正。Shi 和Yin (2016)在很多方面修改了Simon 兩階段設計：一是改變第一階段的決策法則（與Simon (1989) 檢測的假設相同），二是在第二階段將單臂試驗改為雙臂試驗；三是在第二階段使用貝葉斯方法做決策。我們的設計在將二元點改為連續點方面，可視為Shi 和Yin (2016) 的延伸，即將Shi 和Yin (2016) 以及Simon (1989) 中某種藥物的反應機率改為Tsou (2008) 等人的某種藥物的有效性的平均水準。

In Phase II clinical trials, Simon’s (1989) two-stage design is one of the most classical and well-known designs. The question Simon’s two-stage design tackles
with is to test whether the response probability of an experimental drug is less than some uninteresting level, which demands that the endpoints are binary. Besides, Simon’s design is a single-arm design. Tsou et al. (2008) is to test whether the mean level of efficacy of an experimental drug is less than some uninteresting level,
which makes itself a revision of Simon’s two-stage design in the aspect of changing binary endpoints to continuous endpoints and also a single-arm design. Shi and Yin
(2016) modifies Simon’s two-stage design in several aspects: the first is to change the decision rule in the first stage with the same hypothesis tested as Simon (1989), the second is to change single-arm trial to double-arm trial in the second stage, and the third is to use Bayesian method to make a decision in the second stage. Our design is an extension of Shi and Yin (2016) since we change the binary endpoints, i.e., the response probability in Shi and Yin (2016) and Simon (1989), to continuous endpoints in Tsou et al. (2008), which is the mean level of efficacy of a drug.

Hoff P D. A first course in Bayesian statistical methods. Springer Science & Business Media, 2009.
Shi H, Yin G. Bayesian Two-Stage Design for Phase II Clinical Trials with Switching Hypothesis Tests. Bayesian Analysis, 2016.
Simon R. Optimal two-stage designs for phase II clinical trials. Controlled clinical trials, 1989, 10(1): 1-10.
Tsou H H, Hsiao C F, Chow S C, Liu J. P. A two-stage design for drug screening trials based on continuous endpoints. Drug Information Journal, 2008, 42(3): 253-
262.