Treatment resistant schizophrenia: a comprehensive survey of randomised controlled trials

Background: Schizophrenia is a common serious mental health condition which has significant morbidity and financial consequences. The mainstay of treatment has been antipsychotic medication but one third of people will have a 'treatment resistant' and most disabling and costly illness. The aim of this survey was to produce a broad overview of available randomised evidence for interventions for people whose schizophrenic illness has been designated 'treatment resistant'. Method: We searched the Cochrane Schizophrenia Group's comprehensive Trials Register, selected all relevant randomised trials and, taking care not to double count, extracted the number of people randomised within each study. Finally we sought relevant reviews on the Cochrane Library and investigated how data on this subgroup of people had been presented. Results: We identified 542 relevant papers based on 268 trials (Average size 64.8 SD 61.6, range 7-526, median 56 IQR 47.3, mode 60). The most studied intervention is clozapine with 82 studies (total n = 6299) comparing it against other anti-psychotic medications. Cognitive behavioural therapy (CBT), electroconvulsive therapy (ECT), or transcranial magnetic stimulation (TMS) supplementing a standard care and risperidone supplementation of clozapine has also been extensively evaluated within trials. Many approaches, however, were clearly under researched. There were only four studies investigating combinations of non-clozapine antipsychotics. Only two psychological approaches (CBT and Family Rehabilitation Training) had more than two studies. Cochrane reviews rarely presented data specific to this important clinical sub-group. Conclusions: This survey provides a broad taxonomy of how much evaluative research has been carried out investigating interventions for people with treatment resistant schizophrenia. Over 280 trials have been undertaken but, with a few exceptions, most treatment approaches - and some in common use - have only one or two relevant but small trials. Too infrequently the leading reviews fail to highlight the paucity of evidence in this area - as these reviews are maintained this shortcoming should be addressed.