Noninferiority trial ($d>-\Delta$)

A trial testing whether an intervention is no worse than a comparator by more than a prespecified clinically justified margin. For a higher-is-better outcome and difference $d$, a lower confidence bound above $-\Delta$ supports the stated margin. Failure of a superiority test is not sufficient; fidelity, attrition and intention-to-treat and sensitivity analyses affect interpretation.

The maximum clinically acceptable disadvantage specified before seeing results. Its justification and the direction of a favourable outcome must be explicit.