Multidimensional assessment of acute and long-term effects in antidepressant drug trials.

Contrasting the findings from pharmacoepidemiological studies with the status of registration of new compounds and the legal process of reviewing the old compounds indicates a crisis of the preclinical and clinical investigations of "antidepressants". Within the last two decades, the development of more and more formalized guidelines and regulations for the performance of clinical drug trials could be observed. Certainly, this has resulted into an improved internal validity of the findings. On the other hand, there appears to be a broad consensus among practising psychiatrists that published findings of clinical trials often do not concur with the experiences of psychiatric practice indicating unsatisfactory external validity. With regard to the aim of optimizing the antidepressant therapy and to lower the costs for mental health care, it appears to be of particular importance to test the real effectiveness of an "antidepressant" as close as possible to its use in practice. (To avoid misunderstandings: This does not mean to ask for more so-called "naturalistic" study designs!) It is self-evident that also in the future controlled studies, particularly vs. placebo, cannot be dispensed with. However, more promising for a relevant assessment of old and new substances would be the use and development of suitable mathematical procedures, which might imply re-evaluations and re-classifications of already published findings; furthermore, it seems essential also to integrate the existing experiences of psychiatrists in hospitals and in private practice.(ABSTRACT TRUNCATED AT 250 WORDS)