In regulatory applications active controlled clinical trials may or may not have a placebo arm. Such a trial may be expected to demonstrate that a test treatment is not inferior to the active control. In the absence of a placebo arm, the effect of the test treatment can be established by showing non-inferiority or superiority over the control. Statistical considerations pertaining to non-inferiority testing versus superiority testing are often quite different. This presentation will discuss the following difficult subjects: non-inferiority margin determination, primary analysis method, necessary sensitivity analyses, strength of evidence, multiplicity issues management.