Two-stage selection and testing designs for comparative clinical trials

摘要: SUMMARY A two-stage design which selects the best of several experimental treatments and compares it to a standard control is proposed. The allows early termination with acceptance global null hypothesis. Optimal sample size cut-off parameters are obtained by minimizing expected total for fixed significance level power. many other factors. Moreover, usual error rate computations associated comparative testing do not account preliminary selection process. Consequently, overall procedure may be neither effective nor efficient identifying an treatment improvement over C. In this paper we propose new approach problem K determining whether superior control. We deal binomial setting where patient response characterized as either success or failure, ok probability Ek; k =0 corresponds For ease notation assume 01 -... OK . H0: o = =. .. OK, chosen minimize size. accompanying generalized definitions power 1 - ,3 given in ? 2. algorithm used optimization described 3. Numerical results presented 4, followed discussion relative merits proposed design.