Abstract:
|
Randomized trials designed to establish non-inferiority of an experimental therapy as compared to a standard one as measured by binomial proportions have been widely used. Randomization is often stratified by prognostic factors. However,there are hardly one or two hypothesis testing procedures that are mentioned in the literature. We propose a Mantel-Haenszel type test to demonstrate non-inferiority when the non-inferiority margin is not necessarily uniform in all strata. The derivation of the new test originates from Wittes and Wallenstein (1987, JASA 82). Power and sample size will be discussed. The new test will be an alternative to Yanagawa et al. (1994, Biometrics 50). This abstract reflects the views of the author and should not be construed to represent FDA's views or policies.
|