Pretrial Medication Bias in Randomized Antipsychotic Drug Trials

To The Editor: In two recent analyses of the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study published in the Journal, it was suggested that prestudy medication has a major impact on the results of randomized controlled trials. In the March 2007 issue, T. Scott Stroup, M.D., M.P.H., et al. (1) reported that participants who were discontinued from perphenazine in the first phase of CATIE had a relatively poor outcome if they were randomized to risperidone in phase 2. Dr. Stroup et al. explained this finding by the similarity of the respective receptor binding profiles of these antipsychotic drugs. In their article published in the December 2006 issue, Susan M. Essock, Ph.D., et al. (2) reported that participants who continued with their prestudy medication had a longer time to discontinuation than those participants whose antipsychotic was changed. Their explanation was that most patients in CATIE were only moderately ill and relatively stable at baseline. Thus, the switch of the antipsychotic might have led to a destabilization and an exacerbation of symptoms. There is a third pretrial medication scenario that has, to date, not attracted enough attention. When the first atypical antipsychotics were introduced, many patients had received haloperidol before entering the studies. Treatment resistance to haloperidol was usually an exclusion criterion, but this is difficult to define. Since—unlike the scenario in the Essock et al. analysis of the CATIE study—an acute exacerbation of symptoms was an inclusion criterion in many of these studies, it is quite possible that there was a disadvantage for the haloperidol groups because many participants received a drug to which they had not sufficiently responded in past. This situation might have artificially reduced the efficacy of haloperidol or increased the dropout rates in these groups if patients realized, from the side-effect profile, that they were again on haloperidol.

We are not aware of any published re-analyses of random controlled trials comparing atypical antipsychotics and haloperidol based on pretrial medication. We call for performing such analyses. Pretrial medication should also be considered in the design of future studies, e.g. by stratification. For example, in a recent head-to-head comparison of quetiapine and risperidone in patients with an acute exacerbation of schizophrenia, twice as many patients (3) received risperidone than quetiapine. We also wonder what the results of new randomized controlled trials comparing atypical antipsychotics with haloperidol would be in a time when, by now, most patients have received atypical antipsychotics, but hardly any patients have been treated with haloperidol. It could well turn out that haloperidol is a treatment for “atypical resistant patients.”