Methodological aspects of superiority, equivalence, and non-inferiority trials.

Abstract

Depending on the scientific hypothesis to be addressed, randomized-controlled trials (RCT) are accordingly designed. RCTs that aim to determine whether a novel, experimental therapeutic intervention (either a drug or a treatment) is superior to a placebo or control intervention, are called superiority trials. Less common are the non-inferiority RCTs, designed to assess whether a new intervention is not unacceptably worse than an already existing reference intervention. An equivalence RCT is designed to investigate whether a novel treatment is equivalently effective to another, already existing, control intervention. In equivalence and non-inferiority RCTs, the efficacy of the reference intervention (active comparator) is already established, and therefore, an untreated control group would not be ethical. In this review, using a series of examples derived from equivalence and non-inferiority/superiority RCTs, we describe the main differences and methodological aspects among these three different types of RCTs.