BACKGROUND: Schizophrenia is often a severe and disabling psychiatric disorder. Antipsychotics remain the mainstay of psychotropic treatment for people with psychosis. In limited resource and humanitarian contexts, it is key to have several options for beneficial, low-cost antipsychotics, which require minimal monitoring. We wanted to compare oral haloperidol, as one of the most available antipsychotics in these settings, with a second-generation antipsychotic, olanzapine.
OBJECTIVES: To assess the clinical benefits and harms of haloperidol compared to olanzapine for people with schizophrenia and schizophrenia-spectrum disorders.
SEARCH METHODS: We searched the Cochrane Schizophrenia study-based register of trials, which is based on monthly searches of CENTRAL, CINAHL, ClinicalTrials.gov, Embase, ISRCTN, MEDLINE, PsycINFO, PubMed and WHO ICTRP. We screened the references of all included studies. We contacted relevant authors of trials for additional information where clarification was required or where data were incomplete. The register was last searched on 14 January 2023.
SELECTION CRITERIA: Randomised clinical trials comparing haloperidol with olanzapine for people with schizophrenia and schizophrenia-spectrum disorders. Our main outcomes of interest were clinically important change in global state, relapse, clinically important change in mental state, extrapyramidal side effects, weight increase, clinically important change in quality of life and leaving the study early due to adverse effects.
DATA COLLECTION AND ANALYSIS: We independently evaluated and extracted data. For dichotomous outcomes, we calculated risk ratios (RR) and their 95% confidence intervals (CI) and the number needed to treat for an additional beneficial or harmful outcome (NNTB or NNTH) with 95% CI. For continuous data, we estimated mean differences (MD) or standardised mean differences (SMD) with 95% CIs. For all included studies, we assessed risk of bias (RoB 1) and we used the GRADE approach to create a summary of findings table.
MAIN RESULTS: We included 68 studies randomising 9132 participants. We are very uncertain whether there is a difference between haloperidol and olanzapine in clinically important change in global state (RR 0.84, 95% CI 0.69 to 1.02; 6 studies, 3078 participants; very low-certainty evidence). We are very uncertain whether there is a difference between haloperidol and olanzapine in relapse (RR 1.42, 95% CI 1.00 to 2.02; 7 studies, 1499 participants; very low-certainty evidence). Haloperidol may reduce the incidence of clinically important change in overall mental state compared to olanzapine (RR 0.70, 95% CI 0.60 to 0.81; 13 studies, 1210 participants; low-certainty evidence). For every eight people treated with haloperidol instead of olanzapine, one fewer person would experience this improvement. The evidence suggests that haloperidol may result in a large increase in extrapyramidal side effects compared to olanzapine (RR 3.38, 95% CI 2.28 to 5.02; 14 studies, 3290 participants; low-certainty evidence). For every three people treated with haloperidol instead of olanzapine, one additional person would experience extrapyramidal side effects. For weight gain, the evidence suggests that there may be a large reduction in the risk with haloperidol compared to olanzapine (RR 0.47, 95% CI 0.35 to 0.61; 18 studies, 4302 participants; low-certainty evidence). For every 10 people treated with haloperidol instead of olanzapine, one fewer person would experience weight increase. A single study suggests that haloperidol may reduce the incidence of clinically important change in quality of life compared to olanzapine (RR 0.72, 95% CI 0.57 to 0.91; 828 participants; low-certainty evidence). For every nine people treated with haloperidol instead of olanzapine, one fewer person would experience clinically important improvement in quality of life. Haloperidol may result in an increase in the incidence of leaving the study early due to adverse effects compared to olanzapine (RR 1.99, 95% CI 1.60 to 2.47; 21 studies, 5047 participants; low-certainty evidence). For every 22 people treated with haloperidol instead of olanzapine, one fewer person would experience this outcome. Thirty otherwise relevant studies and several endpoints from 14 included studies could not be evaluated due to inconsistencies and poor transparency of several parameters. Furthermore, even within studies that were included, it was often not possible to use data for the same reasons. Risk of bias differed substantially for different outcomes and the certainty of the evidence ranged from very low to low. The most common risks of bias leading to downgrading of the evidence were blinding (performance bias) and selective reporting (reporting bias).
AUTHORS' CONCLUSIONS: Overall, the certainty of the evidence was low to very low for the main outcomes in this review, making it difficult to draw reliable conclusions. We are very uncertain whether there is a difference between haloperidol and olanzapine in terms of clinically important global state and relapse. Olanzapine may result in a slightly greater overall clinically important change in mental state and in a clinically important change in quality of life. Different side effect profiles were noted: haloperidol may result in a large increase in extrapyramidal side effects and olanzapine in a large increase in weight gain. The drug of choice needs to take into account side effect profiles and the preferences of the individual. These findings and the recent inclusion of olanzapine alongside haloperidol in the WHO Model List of Essential Medicines should increase the likelihood of it becoming more easily available in low- and middle- income countries, thereby improving choice and providing a greater ability to respond to side effects for people with lived experience of schizophrenia. There is a need for additional research using appropriate and equivalent dosages of these drugs. Some of this research needs to be done in low- and middle-income settings and should actively seek to account for factors relevant to these. Research on antipsychotics needs to be person-centred and prioritise factors that are of interest to people with lived experience of schizophrenia.
