Premium
On the three‐arm non‐inferiority trial including a placebo with a prespecified margin
Author(s) -
Hida Eisuke,
Tango Toshiro
Publication year - 2010
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.4099
Subject(s) - margin (machine learning) , assay sensitivity , placebo , medicine , sample size determination , randomized controlled trial , statistics , clinical trial , randomization , fraction (chemistry) , mathematics , computer science , surgery , machine learning , chemistry , alternative medicine , organic chemistry , pathology
Three‐arm trials including the experimental treatment, an active reference treatment and a placebo are recommended in the guidelines of the ICH and EMEA/CPMP as a useful approach to the assessment of assay sensitivity. Generally, the acceptable non‐inferiority margin Δ has been defined as the maximum clinically irrelevant difference between treatments in many two‐arm non‐inferiority trials. However, many recent articles discussing three‐arm trials have considered a design with unknown Δ which is the prespecified fraction f of unknown effect size of the reference drug, where the prespecified fraction f is treated as if it were a revised margin. Therefore, these methods cannot be applied to the case where the acceptable non‐inferiority margin must be a prespecified difference between treatments. In this paper, we propose a statistical test procedure for three‐arm non‐inferiority trials with the margin Δ defined as a prespecified difference between treatments under the situation that the primary endpoints are normally distributed with a common, but unknown, variance. In addition, we derive the optimal allocation that minimizes the total sample size. The proposed method is illustrated with data on a randomized controlled trial on major depressive disorder. Copyright © 2010 John Wiley & Sons, Ltd.