Equivalence Trials

Abstract

In the typical two-group clinical trial with outcomes of either success or failure, investigators are interested in finding out whether an innovative treatment, T, reduces the frequency of failure relative to standard therapy, S. If p_T and p_S are used to represent the respective frequencies, this question can be answered by testing the null hypothesis of equal efficacy, H₀: p_T = p_S, against the alternative hypothesis of superiority, H_A: p_T < p_S. If the null hypothesis can be rejected in favor of the alternative hypothesis, we conclude that . . .