Many studies have recently been carried out to compare the effectiveness of various antipsychotic agents.

We wished to determine what could be learnt from these studies about the effectiveness of antipsychotics, and whether this should influence choice of treatment.

We carried out a literature search using PUBMED. We directed special attention to large studies comparing medications in schizophrenia. These studies included CUTLASS, CATIE, SOHO, CAFE, and EUFEST (Kahn 2008, Haro 2006, Jones 2006, Lieberman 2005, Perkins 2008), as well as studies by Tihonen 92006, 2009) The studies were critically reviewed.

The different studies measure different aspects of care, at different phases of the illness. This illustrates the need to study schizophrenia through a stage model. Furthermore, different outcome measures are proposed. We believe that discontinuation data, on which some studiesdepend, is a weak measure of effectiveness. Important doubts arise regarding the choice of patients (those requiring switching of antipsychotic) in CUTLASS, and also regarding potential observer bias in this study.

There is much information to be gathered from the results of such studies. However interpretation is made more difficult by problems within the study designs themselves. On balance it does appear that different antipsychotics have different effectiveness. The main difficulty in putting these results into practice is concern among clinicians and patients regarding side effects.