We consider the problem of proving noninferiority when the comparison is based on ordered categorical data. We apply a rank test based on the Wilcoxon–Mann–Whitney effect where the asymptotic variance is estimated consistently under the alternative and a small‐sample approximation is given. We give the associated 100(1−α)% confidence interval and propose a formula for sample size determination. Finally, we illustrate the procedure and possible choices of the noninferiority margin using data from a clinical trial.
No takes yet. Share an insight, caveat, or question.
Munzel et al. (2003) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: