|A cost minimization study comparing vigabatrin, lamotrigine and gabapentin for the treatment of intractable partial epilepsy
|Hughes D, Cockerell O C
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.
Three antiepileptic drugs for the treatment of intractable partial epilepsy.
Economic study type
This is a cost-minimization analysis. Since no difference in efficacy between the three drugs could be established, then the aim of the study was to establish the least costly treatment.
The study population consisted of patients, over the age of 12 years, with intractable partial epilepsy. The results of the study were applied to UK epileptic patients.
The economic study was carried out in London, UK.
Dates to which data relate
The effectiveness data were taken from studies published between 1989 and 1995. The cost data were from studies, or publicly available data, published between 1991 and 1995. The price year was 1995.
Source of effectiveness data
Effectiveness data were derived from a review of previously completed studies.
Outcomes assessed in the review
The outcomes assessed in the review were the efficacy of the three drugs and their side effects, divided into minor and major side effects. Minor side effects were defined as those which did not lead to serious morbidity or mortality but which did lead to general practitioner consultation. Major side effects were defined as those which led to withdrawal from therapy.
Study designs and other criteria for inclusion in the review
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
Summary statistics from each study.
Number of primary studies included
23 primary studies were included.
Methods of combining primary studies
Investigation of differences between primary studies
Results of the review
In the absence of head to head trials, there was no evidence that the efficacy of the three drugs differed significantly. On this basis, the authors decided to conduct a cost minimization study. The side effects of the three drugs are reported in terms of incidence, the aim being to estimate the resulting level of health service utilisation. For example, 9-15% of patients on lamotrigine, 6-15% of patients on vigabatrin and 3-7% of patients on gabapentin suffer from major side effects. Full incidence results are reported in the paper, broken down by type of major and minor side effects and by drug.
Measure of benefits used in the economic analysis
Because no difference in efficacy was found between the three drugs, a cost-minimisation was undertaken. Consequently, the benefit is based on cost savings.
Quantities were reported in terms of annual frequency of health service resource utilisation for all active patients with epilepsy and percentage of patients on existing drug therapies. Costs of procedures were reported separately. The paper reported the direct costs of procedures, the direct costs of existing drug therapies, and the direct costs of the three drugs under study. The unit costs of each of the treatments and interventions associated with treatment pathways and side effects were also provided. Costs were not discounted since the study spanned only one year. The perspective adopted by the main analysis was that of the health service. Quantities were estimated as follows: the percentage of patients on existing drugs and the use of health service resources (for all epileptic patients and for the side effects caused by the three drugs) were extrapolated from the literature and were based on standard practice. The quantities for the three drugs, in terms of average daily doses, were based on the most commonly prescribed average doses, taken from the literature. The direct costs of procedures were estimated from published studies and publicly available data. The drug costs for existing drug therapies were estimated from the literature. The drug costs for lamotrigine, vigabatrin and gabapentin were taken from the Monthly Index of Medical Specialities. The price year was 1995.
Statistical analysis of costs
No statistical analyses were carried out on the costs.
The authors estimated the indirect costs due to losses of productivity. These were calculated by assuming that, where hospitalisation was required, this would result in the same number of days of lost production, and these figures were adjusted for labour participation rates for individuals with epilepsy, taken from published studies. For the latter the authors used 53%, the midpoint of two percentages reported in the studies. Quantities, in terms of the number of days of lost production, were not reported in the study. The costs were estimated using average weekly earnings taken from Department of Employment data. Costs were not discounted, since the study spanned one year of treatment. The quantity/cost boundary was that of the patient. Average weekly earnings were taken from a 1994 document and adjusted to 1995 prices, although the method used was not stated.
Sensitivity analysis was conducted on the incidence of adverse events and interventions required for standard treatment pathways. The main analysis used the midpoint of plausible ranges for these estimates, although these ranges were not reported in the study. The authors conducted a worst/best case analysis running the analysis with lowest values and highest values. Sensitivity analysis was also conducted on the maintenance dosage level of the three drugs. Lower and upper values were used.
Estimated benefits used in the economic analysis
The total direct costs, per patient, of treatment incurred from standard therapy and side effects, were 1,113.91 for lamotrigine, 1,142.57 for vigabatrin and 1,095.39 for gabapentin. The total indirect costs, per patient, of treatment incurred from standard therapy and side effects, were 23.77 for lamotrigine, 33.25 for vigabatrin and 22.74 for gabapentin. In terms of direct total costs, gabapentin was the least costly resulting in savings of 18.52 over lamotrigine and 47.18 over vigabatrin in the first year of use.
These costs were then applied to the number of new patients in one year in the UK epileptic population who could be prescribed one of the three drugs. This was estimated, from a published study, to total 9,000 patients. The total annual direct cost savings from prescribing gabapentin were 166,680 per annum over lamotrigine, and 424,620 per annum over vigabatrin. The total indirect costs per patient for one year were 23.77 for lamotrigine, 33.25 for vigabatrin and 22.74 for gabapentin.
The sensitivity analysis using lowest and highest values for incidence rates resulted in direct cost savings associated with the use of gabapentin over lamotrigine of between 20.70 and 16.32 and over vigabatrin of between 20.23 and 74.12. The ranking of direct costs was therefore not modified by varying the incidence rates. The sensitivity analysis using lowest and highest maintenance dosage of the drugs did affect the ranking of direct costs. The range of costs for lamotrigine was 805.97 to 1517.82, for vigabatrin 910.82 to 1786.94, and for gabapentin 1063.96 to 1511.04. Lamotrigine appeared to be less costly at the lower end of the dosage levels and gabapentin at the higher end. No sensitivity analysis was conducted on indirect costs.
Synthesis of costs and benefits
A synthesis of costs and benefits was not conducted since this study was a cost-minimization analysis.
The authors concluded that there may be potential cost savings to the NHS if new patients with intractable partial epilepsy were initially treated with gabapentin rather than the other available drug therapies. Using gabapentin would also lead to indirect cost savings due to the relatively low incidence of major side effects requiring hospitalisation.
CRD COMMENTARY - Selection of comparators
The selection of the three drugs was justified by the fact that they were new antiepileptic drugs used to treat patients with intractable epilepsy and had emerged as the main adjunctive treatments for patients who had failed combinations of the more established antiepileptic drugs. The rationale for the selection of comparators was clear.
Validity of estimate of measure of benefit
The main assumption of the study was that there was no difference in efficacy between the three drugs. The study did not report the methods used to search the literature and this would have been useful, given the importance of this assumption. Furthermore, the occurrence of side effects due to the drugs was not included in a measure of benefits, which could have allowed a full cost-effectiveness analysis. Again, the methods used for searching the literature for the incidence of side effects should have been stated. Another issue, raised by the authors in the discussion, was the maintenance dosage levels of the drugs. The sensitivity analysis showed that the results were sensitive to this parameter. If dosage levels had an impact on efficacy and/or on effectiveness in terms of side effects and in terms of compliance, then a cost-effectiveness analysis with a measure of benefits including side effects would have been more appropriate.
Validity of estimate of costs
The study thoroughly explored both the standard and side-effects-induced use and cost of health care resources. However the study only uses point estimates for the costs and a sensitivity analysis of the direct costs would have been useful. The sensitivity analysis used for the study took lowest and highest values of incidence rates and maintenance dosage levels. This type of sensitivity analysis is a little limited in that it is unlikely that all the lowest or highest values for the parameters, will occur together. A sensitivity analysis of the parameters used for indirect costs was lacking. The labour participation rate used was 53%, the midpoint of 35% and 72% taken from two published studies. Given the width of this range some sensitivity analysis should have been conducted. Furthermore, productivity losses were assumed to occur only for those in paid employment and not for people engaged in other types of activities, such as voluntary or domestic work. Since the population study included patients over 12, a more detailed analysis could also have included non-attendance at school. Although the authors did state that the indirect costs used were conservative, a sensitivity analysis would have been useful.
The lack of reporting of the methods used to review the literature for the efficacy of the drugs and the incidence rates of side effects was not entirely compensated for by a thorough sensitivity analysis of the point estimates. The authors reported that this was one of the first studies to explore the cost-effectiveness of anti-epilepsy drugs. Sub-populations of epileptic patients were not considered which could affect the generalisability of the study to different patient domains. In particular a break down by patient age would have been useful, since patients over the age of 12 were included in the study. A discussion of the incidence of side-effects in children and adults for example, could have informed this issue.
Implications of the study
A study of the cost-effectiveness of the treatment with lamotrigine, vigabatrin and gabapentin with measures of benefits including the side effects, and taking into account compliance rates, would build on the results of this published study. Ideally, such a study would use actual patient data.
Hughes D, Cockerell O C. A cost minimization study comparing vigabatrin, lamotrigine and gabapentin for the treatment of intractable partial epilepsy. Seizure 1996; 5(2): 89-95
Other publications of related interest
Comment in: Seizure 1996;5(3):247-8.
Subject indexing assigned by NLM
Acetates /administration & Amines; Anticonvulsants /administration & Cost Control; Cyclohexanecarboxylic Acids; Epilepsies, Partial /drug therapy; Humans; Triazines /administration & Vigabatrin; derivatives /economics /therapeutic use; dosage /analogs & dosage /economics /therapeutic use; dosage /economics /therapeutic use; dosage /economics /therapeutic use; gamma-Aminobutyric Acid /administration &
Date bibliographic record published
Date abstract record published