The problem of therapeutic equivalence with paired qualitative data: an example from a clinical trial using haemophiliacs with an inhibitor to factor VIII

The problem of demonstrating the equivalence of two treatments or drugs occurs fairly frequently in medicine, but is quite often misformulated. The common error is to assume equivalence is demonstrated if the usual null hypothesis of no difference is not rejected, which ignores the potential of a substantial beta-error. We describe here two procedures for testing equivalence when the data are paired and dichotomous, and provide sample size formulas. We give an example based on a clinical trial of haemophiliacs with inhibitor to Factor VIII.