Absolute proof of efficacy can only be given by placebo controlled trials. It is, however, important to classify a drug within the spectrum of existing therapeutic alternatives and, where effective treatment is available, it may be imperative due to ethical considerations to demonstrate that one drug is as effective as another. The issue of therapeutic equivalence trials is discussed along the lines of the important items which should be defined in the protocol: a) the target parameter, which is the primary endpoint of the trial, b) the reference drug, which should be selected with respect to efficacy (superior to others), and safety (largest amount of data), c) the acceptance range, which depends on the primary endpoint, and its implication for the clinical endpoints of morbidity and mortality (the conventional acceptance range for bioequivalence trials does not apply), and d) the statistical procedures, which must take into consideration the unsuitability of the conventional power approach for confirming equivalence. In an equivalence trial, compared to those that are placebo-controlled, the proof that one drug is as effective as another relies much more upon the quality of conduct of the study according to Good Clinical Practice.