TY - JOUR
T1 - A hierarchical testing procedure for three arm non-inferiority trials
AU - Ghosh, Santu
AU - Guo, Wenge
AU - Ghosh, Samiran
N1 - Funding Information:
We express our sincere thanks to the Associate Editor and two anonymous reviewers for their useful comments and suggestions on an earlier versions of this manuscript which led to this improved one.
Publisher Copyright:
© 2022 Elsevier B.V.
PY - 2022/10
Y1 - 2022/10
N2 - 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.
AB - 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.
KW - Extended skew-normal distribution
KW - FWER
KW - Non-inferiority trials
UR - http://www.scopus.com/inward/record.url?scp=85130374131&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85130374131&partnerID=8YFLogxK
U2 - 10.1016/j.csda.2022.107521
DO - 10.1016/j.csda.2022.107521
M3 - Article
AN - SCOPUS:85130374131
SN - 0167-9473
VL - 174
JO - Computational Statistics and Data Analysis
JF - Computational Statistics and Data Analysis
M1 - 107521
ER -