Chlorpromazine versus placebo for schizophrenia

Abstract

Background: Chlorpromazine, formulated in the 1950s, remains a benchmark treatment for those with schizophrenia.

Objectives: To evaluate the effects of chlorpromazine for schizophrenia in comparison to placebo.

Search strategy: Electronic searches of Biological Abstracts (1982-1995), The Cochrane Library (1999, Issue 2), The Cochrane Schizophrenia Group's Register (October 1999), EMBASE (1980-1995), MEDLINE (1966-1995), PsycLIT (1974-1995) were undertaken. References of all identified studies were searched for further trial citations. Pharmaceutical companies and authors of trials were contacted.

Selection criteria: Randomised controlled trials relating to people with schizophrenia, and non-affective serious/chronic mental illness irrespective of mode of diagnosis evaluating chlorpromazine (any dose) versus placebo. Primary outcomes of interest were death, violent behaviours, overall improvement, relapse and satisfaction with care.

Data collection and analysis: Citations and, where possible, abstracts were independently inspected by reviewers, papers ordered, re-inspected and quality assessed. Data were extracted by BT with CA and GA independently checking a 10% sample for reliability. Dichotomous data were analysed using random effects relative risk (RR) and the 95% confidence interval around this was estimated. Where possible the number needed to treat (NNT) or number needed to harm statistics (NNH) were calculated. Continuous data were excluded if more than 50% of people were lost to follow up, but, where possible, weighted mean difference was calculated. Sensitivity analyses have not been undertaken for this version of the review.

Main results: Over 1000 electronic records were inspected. The review currently mentions 202 papers in its Excluded Studies section and 45 studies in its Included Studies table. Six papers await assessment. Chlorpromazine reduces relapse over six months to two years (RR 0.65 CI 0.5-0.9, NNT 3 CI 2.5-4) and there is convincing evidence from trials that it promotes a global improvement in a person's symptoms and functioning (RR 0.76 CI 0.7-0.9, NNT 7 CI 5-10) although the placebo response is also considerable (nearly 40%). Fewer people allocated to chlorpromazine leave trials early (RR 0.76 CI 0.6-1.1). There are many adverse effects. Chlorpromazine is clearly sedating (RR 2.4 CI 1.7-3.3, NNH 6 CI 4-8), it increases a person's chances of experiencing acute movement disorders (RR 3.1 CI 1.3-7.6, NNH 24 CI 14-77), parkinsonism (RR 2.6 CI 1.2-5.4, NNH 10 CI 8-16) and fits (RR 2.4 CI 0.4-16). Amongst other things it clearly causes a lowering of blood pressure with accompanying dizziness (RR 1.9 CI 1. 3-2.6, NNH 12 CI 8-22) and considerable increases in weight (RR 4.4 CI 2.1-9, NNH 3 CI 2-5).

Reviewer's conclusions: This review will confirm much that clinicians and recipients of care already know but provides quantification to support clinical impression. Despite the humbling 40% improvement rate in those who were allocated to placebo, chlorpromazine's global position as the 'benchmark' treatment of those with psychoses is not threatened by this review. Chlorpromazine, in common use for nearly half a century, is a well established but imperfect treatment. Judicious use of this best available evidence should lead to better informed decisions both by carers and those with psychotic illnesses.