在臨床研究上，通常數個治療方法會被考慮，並且與一般標準的治療方法做比較. 在試驗性的研究當中，一個極佳的療程會被用來與標準療程做比較, 並確定是否與標準療程等價或是優於標準療程. 有關於非劣性檢定或是相等性檢定的樣本數的決定已經被研究過. 這篇文章主要提出二階段抽樣法, 亦即從數個療程當中, 找出最佳的療程, 並與標準療程做比較, 觀察他是否較標準療程佳, 或是等價於標準療程. 當非劣性檢定或相等性檢定的虛無假設被接受時, 這個試驗會被提早終止. 在固定的顯著水準以及檢定力之下, 藉由最小化期望總樣本數, 可以得到最佳的樣本數以及參數組合.

In clinical research, several experimential treaments were considered and compared to a standard teratment in the pilot study. If an experimental treatment is the best in the pilot study, then it will be confirmed whether it is better than the standard control or it is equivalent to the standard control. The general sample size determination of noninferiority test or equivalent test had been studied. This paper proposed a two-stage design which selects the best of several experimental treatments and tests whether it is equivalent to a standard control or better than a standard control. The design allows early termination with acceptance of the null hypothesis for noninferiority test or equivalent test. By minimizing expected total sample size for fixed significance level and power, optimal sample size and cut-off parameters were obtained.

• Jennison, C. (1987). Efficient group sequential tests with unpredictable group sizes. BiometrikaVol.74, P.155-165.
• Pocock, S.J. (1977). Group sequential methods in the design and analysis of clinical trials. Biometrika, Vol.64, P.191–200.
• Schuirmann D.J. (1987). A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of coverage bioavailability. Journal of Pharmacokinetics and Biopharmaceutics Vol.15,
P. 657–680.
• Simon, R. (1989). Optimal two-stage design for phase II clinical trials. Controlled Clinical Trials Vol.10, P.1-10.
• Thall, P. F. (1988). Two-stage selection and testing designs for comparative clinical trials. Biometrika Vol.75, P.303-310.
• Whitehead, J. (1986). Sample sizes for phase II and phase III clinical trials: An integrated approach. Statist. Med. Vol.5, P.459-464.