Cost-effectiveness analysis of antipsychotics in reducing schizophrenia relapses

Schizophrenia is a severe form of mental illness that has varying presentations. As a chronic disease, it is associated with significant and long-lasting health, social and financial burdens, not only for patients but for their families, other caregivers [1] and wider society [2, 3].

According to WHO estimates, schizophrenia affects approximately 24 million people worldwide [4]. The most recent publications estimate that the median incidence of schizophrenia varies from 15.2 [5] to 20.0 per 100,000 population/year [6], although it is higher in the 15-35 year-old age group (7 per 1,000 population) [7, 8]. There are no recent epidemiological data concerning schizophrenia in Spain. Spain does not have a national registry that would make it possible to know the exact number of individuals with schizophrenia, although regional studies estimate the prevalence at 0.6-0.8% for the adult population (17 years of age and older) [9]. Available estimates from 1995 show that incidence of schizophrenia was 1.9 per 10,000 inhabitants per year for people between the ages of 15 and 54 years [10].

Conventional antipsychotic medications (chlorpromazine and haloperidol) emerged 50 years ago as the first tools on the management of schizophrenia, in concert with other important interventions, such as psychosocial and economic support and rehabilitation efforts. However, the unrivalled role of conventional antipsychotic medications has been continuously challenged by the wide range of adverse effects of these medications. Over the last 15 years, several new atypical antipsychotic medications have been introduced, including sertindole, risperidone, olanzapine, quetiapine, amisulpride, ziprasidone, aripiprazole and paliperidone [11]. In general, the new antipsychotics have shown themselves to be at least comparable in efficacy to conventional antipsychotics but with superior tolerability and a more favourable adverse effect profile, providing less extrapyramidal side effects than conventional treatment [8].

However, despite the availability of new antipsychotics, 20-30% of patients have an inadequate response to medication with 15-20% relapsing each year [2, 12]. Compliance is one of the factors associated with relapse [13]; however, some patients relapse while taking maintenance medication [14].

Relapse has wider implications for the family in general, for the provision of medical and social facilities and from a health economic perspective [15]. In terms of quality of life, it has been shown to be associated with lower quality of life than in other stable medical conditions [16]. Onset of the disease in late adolescence or early adulthood [2] together with difficulties in employment and the social stigma associated with schizophrenia could be considered the main drivers of the changes in quality of life seen in these patients [16, 17].

Additionally, schizophrenia has been shown to place a substantial economic burden on both the health care system and society worldwide due to its potentially large demands on the healthcare system [18]. The full cost of schizophrenia is high, although this is rarely appreciated by health care decision makers or other stakeholders [3].

In Spain, drug's reimbursement is a central national decision. However, there is a strong territorial decentralization of health jurisdictions in the Autonomous Communities. Therefore, the final drug financing is an Autonomous Regions' responsibility covering almost 100% of population [19]. That means that caring of schizophrenia patients is covered by the public system. In the practice, schizophrenia is diagnosed and followed-up by a psychiatrist at a hospital level, after the patient is being referred by a general practitioner. First drug prescription is usually made by the specialist, but it is required that the patient attends to a primary care level asking for drug prescription in a funded way. In the global economic crisis environment we face, the efficient use of available healthcare resources is required to maximise health benefits for people with schizophrenia and, at the same time, reduce the emotional distress and financial implications of the condition to society [18]. Use of atypical antipsychotics could increase the total cost of the disease as a result of their relatively high price. However, their effectiveness results in reduced hospital stays, thus potentially decreasing total outlay. Indeed, these more expensive treatments could be cheaper for society in the long run [3]. The efficiency, understood as combined measure of efficacy and cost, of treatments for schizophrenia is one key parameter in the decision-making process [20].

Published economic evaluations suggest that atypical (second-generation) antipsychotics are cost-effective when compared with conventional (first-generation) therapy [21]. The aim of this project was to assess the efficiency of the antipsychotics used in Spain to reduce schizophrenia relapses under the National Health System perspective.

To our knowledge, this analysis is the first economic evaluation assessing antipsychotics for prevention and treatment of schizophrenia relapses in Spain.

Atypical antipsychotics are cost-effective when compared with first generation antipsychotics (haloperidol), thus the expected higher acquisition cost of new antipsychotics are generally offset by reductions in other mental healthcare costs, particularly inpatient hospitalisation costs. This indicates that second-generation drugs may be a more efficient choice than traditional antipsychotics [21, 43].

Our study showed lower average cost-effective ratios for paliperidone ER, risperidone and haloperidol than for olanzapine, amisulpride or aripiprazole.

In the base case of the present analysis, one year of treatment with paliperidone ER or risperidone compared with haloperidol were dominant strategies. Our results are in line with economic evaluations carried out by other investigators. Risperidone against haloperidol was also a dominant strategy in Canada [44] and had a lower risk of relapse [45]. Treatment with olanzapine resulted in higher costs than risperidone [46], but it was also cost-effective when compared with haloperidol in a Mexican study [47]. In the UK, despite equivalence with respect to the costs of three alternatives, both risperidone and olanzapine were cost-effective when compared with haloperidol due to efficacy gains [48].

Given that the willingness to pay for an additional QALY in Spain is generally estimated to be within the range of €30,000 to €45,000 [25, 49] paliperidone ER and risperidone were identified as dominant therapies, olanzapine (€23,621 per QALY) could also be considered to be a cost-effective option versus haloperidol in the schizophrenia relapse treatment.

Modelling based on a decision tree to project 12-month consequences of treatment for different antipsychotic therapies has been previously used in other economics evaluations [50].

As noted in NICE analysis, the key driver of cost effectiveness for antipsychotics is their probability to reduce relapse rates. Indeed, due to lack of head to head trials between the drugs analyzed, the current analysis is built on the mixed treatment comparison built on NICE based on retrospective available data.

The main drivers of the model are probabilities. Data from a systematic review performed by NICE were used, due to the lack of head-to-head trials between the drugs analyzed, although differences in relapse definition over the 17 studies included in this meta-analysis could lead to different relapse rates and potentially influence results.

The one-year treatment period was chosen because most of the clinical trials only assessed efficacy of antipsychotics in a 52-week period, and projection of the efficacy beyond 52 weeks would be a source of bias in the analysis. On the other hand, the last guideline for schizophrenia published by NICE considered a lifetime horizon given the potential need for long-term use of APS drugs by people with schizophrenia in remission.

The primary limitation of this evaluation was the omission of other antipsychotics, such as zotepine and quetiapine. As previously justified, zotepine was not included because it is not marketed in Spain. Quetiapine was also excluded due to the lack of relevant clinical information on this drug related to schizophrenia relapse prevention.

Although antipsychotic medication is associated with a wide range of adverse events, only EPS, weight gain and metabolic effects were considered in our evaluation, as they have been identified to be those with a greater impact on cost-effectiveness ratio.

Following a conservative approach, tardive dyskinesia was not included, despite its long-term and important effects on quality of life, as this event is mainly associated with the primary comparator (haloperidol). Including this side effect could introduce an important bias in the analysis [18].

The quality-adjusted life year (QALY) is routinely used as a summary measure of health outcome for economic evaluation, which incorporates the impact on both the quantity and quality of life [51] QALYs are obtained by multiplying life years gained by an utility value. Utilities represent patient's preferences by a health state, so they are strongly related to cultural aspects. Ideal economic evaluation would be based on data derived from local population where study is performed [52]. However publications with Spanish specific utilities values in the illness of interest are scarce. Although use of values from other areas could be a potential bias' source [53], in the present study, due to the lack of published data related to Spanish population, utilities from UK where used for remission, relapse, acute EPS and weight gain.

According the results from the deterministic sensitivity analysis developed, the model seems to be quite robust, but probabilistic sensitivity analyses would have provided additional information to validate this point.

Despite these limitations, the assumptions considered in the present model appeared to be reasonable and conservative, and the results of the sensitivity analyses indicated the robustness of the findings.

The current analysis shows that relapse prevention is the key driver for cost-effectiveness of antipsychotics in Spain. This conclusion is fully aligned with the results of the NICE analysis in the UK, which acknowledges that drug acquisition costs do not determine the relative cost-effectiveness of antipsychotic medications. Indeed, antipsychotic drugs that reduce the rate and duration of hospital admissions may be cost-saving options in the long-term, despite higher acquisition costs.

With respect to other Spanish studies, our results are also aligned with a previous analysis concerning the costs of schizophrenia in Spain, which showed that drugs account for only 24% of medical costs of the disease in Spain, whereas hospitalization accounted for 73% of those costs [54].

A previous investigation in Spain has shown that number of relapses is directly related to higher treatment costs [20]. The choice of the best alternative to reduce relapses, in efficacy terms, could indeed impact in the total schizophrenia cost. According our results paliperidone ER is the most effective and the least costly strategy.

Our findings may be of interest to clinicians and others interested in outcomes and cost of mental health services among patients with schizophrenia.

In our analysis, paliperidone ER and risperidone were shown to be dominant antipsychotic therapies compared to haloperidol in Spain. However, as pointed out by NICE, it is worthwhile to highlight that schizophrenia is a highly incapacitating disease and choosing the most appropriate drug and formulation for a particular patient is crucial. The prevention of relapses is a key factor in the efficiency of the antipsychotics.