DOSE, DROP-OUT AND SYMPTOMATOLOGY - IMPOR TANT METHODOLOGICAL PROBLEMS IN TESTING NEW ANTIDEPRESSANTS

During the last decade, several new antidepressants have been introduc ed. This paper discusses methodological problems in randomized, clinic al trials in which these new components are compared with tricyclic an tidepressants. A correct randomization is not enough to produce conclu sive results. A sufficient dose of the new drug as well as the referen ce substance is essential and ways of fulfilling this need are discuss ed. Since drop-out rate during clinical trials is often high and may l ead to inconclusive results, drop-out as an end-point has to be consid ered as a measure of effect. Finally, preliminary results of a compari son of hospitalized and outpatients are presented. Hospitalized patien ts display higher levels of severity and endogenous features, which, i n part, may explain the contradictory results between outpatient and i npatient studies.