Long-acting antipsychotics cost-effective for treatment of schizophrenia, but evidence inconsistent

Here at Mental Elf HQ we’re expanding our skill set to include economics. Understanding the best way to value health and health care, and improving health outcomes with budget constraints in mind, are the key pastimes of economics elves. We hope to bring you the latest economic evidence in the field of mental health and to provide an additional perspective to the excellent debate and discussion facilitated by this site.

Affecting about 26 million people worldwide, schizophrenia accounts for 1.5-3% of national healthcare expenditures. High rates of relapse and hospitalisation contribute to higher costs, which have also been associated with non-compliance with medication. Long-acting injections and extended-release antipsychotic agents have been shown to improve compliance rates and symptom control, as well as reduce adverse events and relapses. It’s therefore crucial to assess the cost-effectiveness of antipsychotics for the treatment of schizophrenia, taking into account the direct and indirect costs of the disease. A new systematic review by Achilla and McCrone (2013) is the first to bring together all existing evidence from economic evaluations of the cost-effectiveness of both typical and atypical long-acting antipsychotics.

Methods

The authors reviewed all economic evaluations from PsycINFO, MEDLINE, EMBASE, the NHS Economic Evaluation Database and the Health Technology Assessment database. Letters, posters and conference abstracts were excluded. 154 records were screened, of which 28 were eligible for inclusion in the review. 13 studies were based in Europe; 5 in the US; 2 in Canada; 1 in China; and 1 in Taiwan. All studies were published in the last 15 years.

Compared with daily medication, long-acting antipsychotics can improve compliance and reduce costs

Results

Most of the studies evaluated long-acting risperidone (RLAI) compared with long-acting haloperidol (HLAI) or other atypical oral formulations. Studies that included length of hospital stays and the number of relapses as measures of effectiveness consistently showed that RLAI leads to significant benefit. However in some studies, other antipsychotic agents, such as oral olanzapine, were found to dominate RLAI.

The authors found inconsistencies in the methodologies used and the reporting of study findings. The majority of studies employed decision analytic models, including decision trees and discrete event simulations, though some had observational or trial-based designs. The studies adopted an array of different outcome measures, and time frames used in the evaluations ranged from 1 to 5 years. A number of the studies included in the review were of a poor quality. 71% of the studies met over half of the criteria in the quality checklist, though the criteria for economic studies may not be appropriate for some clinical studies.

Conclusions

The authors conclude that:

in terms of cost effectiveness, the superiority of RLAI (long-acting risperidone) over HLAI (long-acting haloperidol) and other oral antipsychotics remained a common finding in most cases.

Limitations

However, there remain many caveats to this finding. All studies in the review included the direct medical costs of schizophrenia, but none included indirect costs. This means that the societal cost of schizophrenia is underestimated. It’s also important to note that the majority of studies were funded by the Janssen-Cilag pharmaceutical company, which is the marketing authorisation holder of long-acting respiridone. It seems that assessing the quality of studies in this review was not easy. The authors suggest that more work should be done to weight items in quality criteria checklists, as simply quoting the number of items satisfied requires a subjective judgement of quality.

It’s clear that more work needs to be done to assess the cost-effectiveness of long-acting antipsychotics, taking account of both direct and indirect costs. Future studies should address the methodological limitations of previous studies, which have been nicely summarised in this review.