Making sense of equivalence and non-inferiority trials

Abstract

New drugs are usually compared to a placebo. Sometimes it may be unethical to give patients a placebo, so the new drug is compared with standard treatment. Trials which compare treatments may not be designed to show that one treatment is superior. These are known as non-inferiority or equivalence trials. Non-inferiority trials aim to show that the new drug is no worse than standard treatment. Equivalence trials aim to show the new treatment is no better and no worse. An equivalence boundary should be set before the trial. This is the definition of what would be the minimum important difference between the treatments. There are several traps in the interpretation of trials of non-inferiority or equivalence. The results can be influenced by many factors including the size of the equivalence boundary and whether an intention-to-treat or 'per protocol' analysis is used.