The first RCT was carried out 60 years ago ( UK MRC trial of streptomycin in tuberculosis, 1948).Modern drug treatments for mental disorders also started emerging in the last 60 years.First RCT in psychiatry was published in July 1955 (Reserpine in the treatment of anxious and depressed patients). Jose Miyar and Clive Adams examines the content and quality of research conducted in schizophrenia over the last five decades in this article.
The Cochrane Schizophrenia group’s data base is the most comprehensive one of all published and unpublished trials in this area. The register in 2008 held 12,211 reports of 10,000 trials.77% of this were published in journals and 10% in conferences. The survey show that schizophrenia research is under represented in leading journals. Only 28% of these trials were available on Pubmed . 93% of these trials focused on treatment. The number of trials per year relevant to schizophrenia is rising from about 35 in the 1950s and 1960s, 100 in 1970s and 1980s, 200 in 1990s to 650 per year so far in the 2000s. There is a tendency to have multiple reports from one trial, one trial was published 122 times!. Multiple articles often repeat findings in different format though in some cases multiple publications are very much necessary to communicate all relevant information that is generated. Authors consider that “multiple publications serve only to confuse—often giving the impression that there are more data than there really are. This can be used as a marketing tool to highlight favourable findings of a treatment and hide disadvantageous facts amidst the confusion”.
It is interesting to note that an increasing proportion of research is now published in Mandarin (18%).In 1998 there were only 216 trials from China , but by 2008 22% of all trials come from china. The average number of trial participants rose from 65 people in 1998 to 132 in 2008. The authors consider that an outcome such as ‘‘clinically important improvement in mental state’’ to show a 20% difference between groups a study would have to have around 150 participants in each arm (a = .05, power 85%).However, the median and mode estimates of study size of 60 participants indicate that schizophrenia trials remain tiny in most cases.
Drugs remain the most common intervention with a marked increase in the study of risperidone. Risperidone has become the benchmark control medication (instead of haloperidol). The study show that there is a proliferation of ratings scales. Every fifth trial is introducing a new rating scale. 40% of these scales may not be clearly validated and are therefore likely to produce exaggerated estimates of treatment effect. This makes the comparison across trials more difficult and interpretation of results more complicated.
Sample size of schizophrenia trials remains small. Multiple publications are becoming more common, and new outcomes scales are proliferating. There is need for agreed standardized sets of outcomes.More research is now coming from previously underrepresented countries like China.
An interesting development is the COMET Initiative which brings together researchers interested in the development and application of agreed standardised sets of outcomes, known as a ‘core outcome set.’ These sets are envisaged to represent the minimum that should be measured and reported in all clinical trials for a specific condition.
Clinicians and patients can hope that research community would take these observations seriously and there would be much more good quality, larger trials that address the important issues that matters.
Summary of the article:
Content and quality of 10 000 controlled trials in schizophrenia over 60 years. Miyar J, Adams CE.Schizophr Bull. 2013 Jan;39(1):226-9. doi: 10.1093/schbul/sbr140.