Response to Malik et al. and Brewer Letters
To the Editor: We thank Dr. Malik et al. and Dr. Brewer for their thoughtful comments. Dr. Brewer inquires about the fate of the study in which we reported a post hoc analysis of a large industry-sponsored trial that compared aripiprazole and haloperidol for the treatment of schizophrenia. Our analysis found that the apparent superiority of aripiprazole among patients early in the course of their illness was likely due to substantial side effects in the haloperidol-treated group, perhaps as a result of excessive dosing of haloperidol in the parent trial. We submitted our paper to several journals before it was recently accepted for publication (1). The difficulty of publishing negative results is a well-established phenomenon in clinical trial research in general (2) and for psychiatric trials in particular (3), and it is a primary source of publication bias. Nonpublication of negative results may be a result of authorial or organizational reluctance to submit negative findings, or a relative undervaluing of such studies by reviewers or editors.
We commend Dr. Malik et al. for the creative ways in which they teach residents in their program about the ethical challenges of interacting with the pharmaceutical industry.
1. : Aripiprazole vs haloperidol in early-stage schizophrenia. J Psychiatr Res (in press)Google Scholar
2. : Publication bias in clinical trials due to statistical significance or direction of trials results. Cochrane Database Syst Rev 2009; 1:MR000006Medline, Google Scholar
3. : Selective publication of antidepressant trials and its influence on apparent efficacy. N Engl J Med 2008; 358:252–260Crossref, Medline, Google Scholar