Non-inferiority trials are becoming very popular for comparative effectiveness research. Non-inferiority trials establish that the effect of an experimental treatment is not worse than that of a reference treatment by more than a specified margin. A three-arm non-inferiority trial that includes the placebo, experimental treatment, and a reference treatment is considered. It has been criticized that the conventional approach for three-arm non-inferiority trials loses power for the non-inferiority hypothesis test unless the power of the assay sensitivity test is close to one. In order to overcome this situation, a novel hierarchical testing procedure with two stages for three-arm non-inferiority trials is developed. The family-wise error rate (FWER) is investigated analytically and numerically of the proposed test procedure. Numerical studies indicate that the suggested method controls FWER and has more power than the traditional approach particularly when the power of that assay sensitivity test is not close to one. Through these empirical studies, it is shown that the proposed method can be successfully applied in practice.
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- Computational Mathematics
- Computational Theory and Mathematics
- Applied Mathematics
- Extended skew-normal distribution
- Non-inferiority trials