Comparisons between psychotropic drugs: must the risk of side effects dictate our practices?

Recently, SOHO and CATIE's studies in the field of schizophrenic disorders asserted that molecules apparently showing the most side-effects are not only the most effective but also lead to the least changes in treatments. Can we generalise this assertion to other domains of pharmacological treatments, such as in mood and anxiety disorders? And, more generally, do we possess information about comparison between different types of molecules in these different fields.

Review of the literature (medline-psycinfo-psycarticles) addressing these three psychiatric disorders, and comparing efficacy of treatments, or cost-effectiveness studies.

Although there is a plethora of publications about the efficacy of given molecules vs. placebo, studies comparing molecules are scarce, and studies on cost-effectiveness in natural environments are even more scarce. Independently of the type of disorders under study, the last few years' efforts to completely limit side effects seem to have resulted in a loss of efficacy. Moreover, the previously held hypothesis suggesting that the fewer the side effects, the lesser the need to change treatment has not been confirmed. The duration of a treatment is more dependent upon its efficacy than on other variables.

Clinicians cannot determine the absence of noxiousness of a molecule as their primary criterion of choice. In contrast, they should carefully balance side effects and efficacy. In Psychiatry, there is a lack of studies about cost-effectiveness in natural environments.