Schizophrenia: Do we really need placebo-controlled studies?

Objective: To investigate whether placebo control is necessary to prove eff icacy in short-term studies in schizophrenia. Design: This study compares t he efficacy results of placebo-controlled studies versus positive controlle d studies, that is controlled studies without a placebo control, in the sho rt-term treatment of chronic schizophrenia. Results: Concerning mean improv ement on the BPRS, the placebo arms showed in two cases a worsening, in one case almost no change, and in the remaining studies (6) the improvement wa s between 1 and 5%. The percentage mean improvement in the haloperidol arms of the placebo-controlled studies was comparable to the percentage mean im provement in the corresponding arms of the non-placebo-controlled studies. The highest percentage responders in the placebo-groups was 43% and the low est was 6%. Moreover the responder rates in the atypical antipsychotic and haloperidol arms of the non-placebo-controlled studies were, in two of the three studies, in the same order of magnitude as the responder rates of the placebo arms in the placebo-controlled studies. The overall dropout rates in the placebo arms was between 48% and 80% and were higher than the drop o ut rates in the atypical neuroleptic arms and haloperidol arms of the place bo-controlled studies. The dropout rates due to an insufficient response in the atypical neuroleptic arms and haloperidol arms of the non-placebo-cont rolled studies were lower when compared to corresponding treatment arms of the placebo-controlled studies. Conclusion: In contrast to the mean improve ment on the BPRS, responder rates in the placebo arms varied considerably f rom study to study. Responder rates in the atypical antipsychotic and halop eridol arms of the non-placebo-controlled studies were, in two of the three studies, of the same order of magnitude as the responder rates of the plac ebo arms in the placebo-controlled studies. These results indicate that pla cebo control is necessary. Moreover as responders are a more clinically rel evant outcome measure when compared to mean improvement on a rating scale, placebo-controlled studies are still needed. However, consensus on responde r definition should be agreed upon. For the moment, alternatives to placebo -controlled studies are inadequate in demonstrating efficacy in studies wit h schizophrenic patients. (C) 1998 Published by Elsevier Science B.V./ECNP. All rights reserved.