|Long-acting risperidone compared with oral olanzapine and haloperidol depot in schizophrenia: a Belgian cost-effectiveness analysis
|De Graeve D, Smet A, Mehnert, Caleo S, Miadi-Fargier H, Mosqueda G J, Lecompte D, Peuskens J
This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn.
The study compared the following therapeutic strategies for patients with schizophrenia:
first-line treatment with long-acting risperidone, 25 mg every 14 days (ambulatory care or hospital setting);
depot haloperidol, 100 mg every 28 days (ambulatory care) or 125 mg every 28 days (hospital setting); and
oral olanzapine, 10 mg daily.
Economic study type
The study population comprised a hypothetical cohort of young schizophrenic patients treated for at least 1 year and whose disease had not been diagnosed for longer than 5 years.
The study setting was secondary care. The economic study was carried out in Belgium.
Dates to which data relate
The effectiveness data were derived from a review of studies published in 2001 and 2002. The price year was 2003.
Source of effectiveness data
The effectiveness data were derived from a review and synthesis of published studies and from estimates derived from an expert committee, which supplemented data from the literature.
A decision analytic tree model consisting of three main branches (one for each treatment strategy) was created using TreeAge DATA software. Four different health states were considered for the model:
clinical response, defined as non-relapse or non-rehospitalisation;
clinical deterioration, defined as a state characterised by positive clinical symptoms, behavioural problems, suicidal ideation or extra pyramidal symptoms;
inadequate response, similar to the clinical deterioration state, except that inadequate response could occur in the first period of each therapeutic option or after a period of clinical deterioration or inadequate response; and
long-term hospitalisation, which occurs after four therapeutic failures.
The time horizon for the model was 2 years, which was divided into six periods of 4 months each. At the end of each period, patients experiencing an inadequate response or clinical deterioration could be switched to an alternative therapy.
Outcomes assessed in the review
The outcome assessed was the relapse rate, which then allowed the authors to calculate transition probabilities at relevant time points.
Study designs and other criteria for inclusion in the review
The authors did not specify the study designs or other criteria for inclusion in the review, but the effectiveness data were based on a long-term, double-blind prospective trial and a non-randomised survey.
Sources searched to identify primary studies
Criteria used to ensure the validity of primary studies
Methods used to judge relevance and validity, and for extracting data
Number of primary studies included
Two primary studies were included in the review. The effectiveness data on haloperidol were derived from a long-term double-blind, prospective study that compared risperidone and oral haloperidol (Csernansky et al. 2002, see ,Other Publications of Related Interest. below for bibliographic details). The effectiveness data on olanzapine and risperidone were derived from a non-randomised survey that compared the two drugs (Rabinowitz et al. 2001, see ,Other Publications of Related Interest- below for bibliographic details).
Methods of combining primary studies
The primary studies were not combined.
Investigation of differences between primary studies
The authors reported that there was good congruence of results between the two studies used to derive effectiveness data.
Results of the review
The authors did not report the relapse rate estimates derived from Kaplan-Meier analysis of the two studies. However, they did report the transition probabilities (response rates) used in the model, which were based on data derived from the two studies and expert opinion.
The response rates were:
for long-acting risperidone, 91% at 4 months, 95% at 8 months, 90% at 12 months, 99% at 16 and 20 months, and 100% at 24 months;
for olanzapine, 82% at 4 months, 94% at 8 and 12 months, 99% at 16 months, 97% at 20 months, and 100% at 24 months; and
for haloperidol depot, 77% at 4 months, 84% at 8 months, 86% at 12 months, 93% at 16 months, 94% at 20 months, and 98% at 24 months.
Methods used to derive estimates of effectiveness
The effectiveness data derived from the literature were supplemented using the opinion of an expert panel. This expert panel comprised two psychiatrists and one health economist.
Estimates of effectiveness and key assumptions
The authors did not provide separate estimates of effectiveness based on those reported in the literature and those derived from experts. Therefore, the combined estimates of effectiveness have been presented in the ,Results of the Review- section.
Measure of benefits used in the economic analysis
The measure of benefits used was the proportion of patients successfully treated. This was defined as those patients who responded to initial treatment and who had none to two episodes of clinical deterioration, without needing to change treatment, over the 2-year period.
The direct medical costs of the Belgian health care system were included in the analysis. Such costs were for hospitalisation (medication, medical supervision, laboratory tests and accommodation) and ambulatory care (physician visits, injections, medications and laboratory tests). The cost of treating extrapyramidal effects was added for all strategies. The unit costs and resource use quantities were reported separately. The resources consumed by stabilised and relapsed schizophrenic patients were generally derived from an 11-member expert panel of Belgian psychiatrists using a Delphi method. Unit cost data were derived from the National Institute for Sickness and Invalidity Insurance (INAMI) and the Ministry of Health. The authors reported that in Belgium, patient subsidy impacts on expenses incurred by the INAMI according to the proportion of patients with preferential reimbursement caused by low income (VIPO). Based on a previous study, the authors assumed that 47% of schizophrenic patients were classified as VIPO. Since the costs were incurred over a 2-year period, future costs were discounted at an annual rate of 3%. The study reported the mean costs. The price year was 2003.
Statistical analysis of costs
The costs were treated as point estimates (i.e. the data were deterministic).
The indirect costs were not included.
A series of one-way sensitivity analyses were undertaken. The parameters varied were the improvement in efficacy over oral risperidone associated with the long-acting injectable formulation of risperidone, the proportion of VIPO patients, the efficacy of depot haloperidol, and dose variations for long-acting risperidone and olanzapine.
Estimated benefits used in the economic analysis
The proportion of successfully treated patients over the 2-year period was 82.70% for long-acting risperidone, 74.80% for olanzapine, and 57.30% for haloperidol depot.
The mean costs incurred per patient over the 2-year period were EUR 16,406 for long-acting risperidone, EUR 17,074 for olanzapine, and EUR 21,779 for haloperidol depot.
Synthesis of costs and benefits
The costs and benefits were combined using a cost-effectiveness ratio (i.e. the cost per successfully treated patient). The average cost per successfully treated patient was EUR 19,839 for long-acting risperidone, EUR 22,826 for olanzapine, and EUR 38,008 for haloperidol depot. The costs and benefits were not combined using an incremental analysis, as long-acting risperidone was found to be both more effective and less expensive than olanzapine and haloperidol.
The results of the sensitivity analyses confirmed the robustness of the model to wide variations in inputs: long-acting risperidone dominated both olanzapine and haloperidol.
Long-acting risperidone represented a favourable first-line strategy for patients with schizophrenia requiring long-term maintenance treatment.
CRD COMMENTARY - Selection of comparators
The authors reported that the three interventions examined were clinically relevant options. You should decide if these interventions used in the treatment of schizophrenia are current practice in your own setting.
Validity of estimate of measure of effectiveness
The authors undertook a review of the literature, which resulted in a limited number of trials of antipsychotic treatments being identified. Using the results from two trials and opinion from an expert committee, made up of two psychiatrists and one health economist, the authors derived transition probabilities to populate their model. However, the authors did not report the results from the review of the literature and those derived from the expert committee separately. Consequently, it was unclear which results were derived from the literature and which were derived from opinion. The estimates were explored in a sensitivity analysis using appropriate ranges.
Validity of estimate of measure of benefit
The estimation of benefits was modelled using a decision analytic tree model, which was appropriate for the study question. The measure used (proportion of patients successfully treated) could be useful in this case, but it limits the comparability with other economic evaluations in other health fields. The benefits do not appear to have been discounted. This was not appropriate as the costs were discounted, and for consistency benefits would also have to be discounted.
Validity of estimate of costs
All the categories of cost relevant to the perspective adopted were included in the analysis. All major relevant costs appear to have been included in the analysis. The costs and the quantities were reported separately in the analysis, which will increase the generalisability of the authors' results. Resource use was mainly derived from an 11-member expert panel of Belgian psychiatrists using a Delphi method. The unit costs were derived using national tariffs. Limited one-way sensitivity analyses were undertaken by varying resource use and unit cost data. The costs were incurred over a 2-year period, thus the costs incurred in the second year were discounted. The price year was reported, which will aid any future inflation exercises.
The authors made appropriate comparisons of their findings with those from other studies conducted in Germany, Canada and the Netherlands which had found that long-acting risperidone was a cost-saving strategy. The issue of generalisability to other settings was partly addressed in the sensitivity analyses. The authors do not appear to have presented their results selectively, although a more detailed and thorough explanation on how transition probabilities were derived would have been desirable. The authors' conclusions would appear to reflect the scope of the analysis.
The authors reported a number of further limitations to their model. First, the use of a decision model might have simplified a complex reality. Second, the data on compliance had to be assumed. Third, the patient population used in the study might be difficult to manage. Finally, the study did not include indirect costs such as productivity losses.
Implications of the study
The authors would appear to recommend the use of long-acting risperidone as a first-line treatment option for schizophrenia in Belgium.
De Graeve D, Smet A, Mehnert, Caleo S, Miadi-Fargier H, Mosqueda G J, Lecompte D, Peuskens J. Long-acting risperidone compared with oral olanzapine and haloperidol depot in schizophrenia: a Belgian cost-effectiveness analysis. PharmacoEconomics 2005; 23(Supplement 1): 35-47
Other publications of related interest
Rabinowitz J, Lichtenberg P, Kaplan Z, et al. Rehospitalization rates of chronically ill schizophrenic patients discharged on a regimen of risperidone, olanzapine, or conventional antipsychotics. Am J Psychiatry 2001;158:266-9.
Csernansky JG, Mahmoud R, Brenner R. A comparison of risperidone and haloperidol for the prevention of relapse in patients with schizophrenia. N Engl J Med 2002;346:16-22.
Laux G, Heeg B, van Hout B, et al. Costs and effects of long-acting risperidone compared with oral atypical and conventional depot formulations in Germany. Pharmacoeconomics 2005;23 Suppl 1:49-61.
Heeg BM, van Aalst G, van den Arend JJ, et al. A discrete events model of long-term outcomes and cost of treatment with long acting risperidone in schizophrenia. Value in Health 2003;6:515-6.
Llorca PM, Miadi-Fargier H, Lancon C, et al. Cost-effectiveness of schizophrenic patient care strategies: impact of an atypical antipsychotic in a long-acting infection formulation. L'Encephale 2005;31:235-46.
Subject indexing assigned by NLM
Antipsychotic Agents /administration & Belgium; Benzodiazepines /administration & Cost of Illness; Cost-Benefit Analysis /economics; Decision Trees; Delayed-Action Preparations; Economics, Pharmaceutical; Haloperidol /administration & Humans; Risperidone /administration & Schizophrenia /drug therapy; dosage /economics /therapeutic use; dosage /economics /therapeutic use; dosage /economics /therapeutic use; dosage /economics /therapeutic use
Date bibliographic record published
Date abstract record published