In many clinical experiments multiple measurements are required for evaluation of the results. Several methods have been illustrated in the literature to deal with this problem. Some of these solve the problem of multiple testing bias just by looking at the significance of the most pronounced difference. On the other hand, some other methods are designed to take into account all the information available from an experiment. The purpose of this paper is to review these methods and compare the results they gave when applied to a clinical trial in which drug efficacy was assessed by 14 outcome measures.