### Background

^{2)}.

^{4)}. For example, in trials involving serious outcomes such as mortality, it is unethical to use a placebo when active treatments are available. Clinical equipoise, referring to the state of true uncertainty about the relative benefits of alternative treatments under the "null" hypothesis to be tested, is an ethically necessary condition in all clinical research

^{5)}. Active controls are sometimes used to demonstrate the efficacy of a drug that may have large placebo effects. Active controls are also used to determine how experimental treatments compare to alternative treatments. Active control trials aim to demonstrate that treatments of interest have either superior effects or similar effects to the control.

^{2)}.

*in vivo*biological equivalence, and those are the ones where showing equivalence between treatments is truly of interest. When determining the effects of experimental treatments on clinical end points, it would not be sensible to investigate whether their effects are no worse than, as well as no better than, those of the control. Although noninferiority and equivalence trials have often been both referred to as "equivalence trials," they are distinct. If the intent of a study is to demonstrate that an experimental treatment is not substantially worse than a control treatment, the study is known as a noninferiority trial. However, there are some complicated issues with trials of this type that make them less reliable than typical superiority trials6-

^{9)}. Some guidelines have been provided by regulatory bodies10-

^{12)}, and the CONSORT (Consolidated Standards of Reporting Trials) Group has also published an extension of their guidelines for such trials

^{13)}.

### Demonstrating treatment effect

*P*value is <0.05 (Fig. 1). Note, that if due to a small sample size the CI is very wide, then the study is not likely able to demonstrate superiority. Thus, concluding equivalence or noninferiority on the basis of a nonsignificant test of the null hypothesis, i.e., no difference between the experimental treatment and the active control is inappropriate.

### Why noninferiority trials?

### Analysis of noninferiority trials

### Determination of noninferiority margins

### Does noninferiority imply a treatment is effective?

### Sample size considerations

### Choosing the analysis population

### Switching between superiority and noninferiority

### Conclusions

*ad hoc*determination of the margin after a trial is complete is not acceptable. Noninferiority trials are based on some directly unverifiable assumptions. In order to demonstrate assay sensitivity, it is important to evaluate the conditions of previous trials as closely as possible, including trial design and patient characteristics. Therefore, when planning a noninferiority trial, all necessary considerations should be taken to ensure that false claims of noninferiority are avoided.