El-Sayeh et al (Reference El-Sayeh, Morganti and Adams2006) raise some important issues regarding the design and reporting of clinical trials. However, we feel that the conclusion that ‘aripiprazole has been licensed despite the fact that few reliable data on this drug are publicly available’ merits further clarification. Aripiprazole was first approved in November 2002 in the USA, and in 2004 in Europe, based on the submission of a substantial body of evidence to the regulatory authorities on more than 4000 patients. However, Bristol-Myers Squibb and Otsuka Pharmaceuticals are committed to reporting trial results as completely as possible, and publication of pivotal studies has taken place subsequent to approval.

All the aripiprazole clinical studies were conducted in accordance with regulatory requirements and using accepted standards (Reference Marder, McQuade and StockMarder et al, 2003; Reference Naber and LambertNaber & Lambert, 2004). Such studies have inherent restrictions, and we recognise that patients enrolled may not always reflect those seen in everyday care. We understand the value of all study types - randomised controlled trials, naturalistic, retrospective, observational - in helping to determine the benefit-risk profile, and have recently completed a series of studies with more naturalistic designs and with large sample sizes, to explore the benefits in a wide range of patients (Reference Tandon, Marcus and StockTandon et al, 2006; Reference Kerwin, Millet and HermanKerwin et al, 2007, details of the other study can be obtained from http://www.clinicaltrials.gov, trial number NCT00237939). These complete studies support the profile of aripiprazole established in the clinical studies reviewed by El-Sayeh et al in their systematic analysis.

With respect to the suggestion that deaths occurring during the aripiprazole studies have not been widely reported, it is our practice to report any deaths or adverse events applicable to a study and we have done so consistently in our publications. Deaths unfortunately do occur during studies, just as they do in real-world situations.

We are committed to continued openness and disclosure of clinical study results, and as such will continue to work closely with El-Sayeh et al.