|
Cost-effectiveness of memantine in community-based Alzheimer's disease patients: an adaptation in Spain |
Antonanzas F, Rive B, Badenas J M, Gomez-Lus S, Guilhaume C |
|
|
Record Status 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. Health technology The study assessed the use of memantine in patients with Alzheimer's disease (AD). This treatment was compared with standard care, which the authors defined as no pharmacological treatment.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised a hypothetical cohort of Spanish patients with AD. Based on clinical data, the initial distribution of patients in the model was: 29.5% moderately severe independent; 11.5% moderately severe dependent; 28.9% severe independent; and 30.1% severe dependent.
Setting The study setting was the community. The economic study was carried out in Spain.
Dates to which data relate The effectiveness data were derived from a review of studies published between 1975 and 2004. The resource use data were derived from a study published in 2004. The price year was 2005.
Source of effectiveness data The effectiveness data were derived from a review and synthesis of published data. More specifically, drug-specific input data for the model were derived from a Markov model developed for use in Finland (Jones et al. 2004) and the UK (Francois et al. 2004, see 'Other Publications of Related Interest' below for bibliographic details of both studies). However, Spanish-specific data (such as epidemiological data) were derived from local Spanish studies of patients with AD.
Modelling The authors adapted a Markov model that had been developed for use in Finland and the UK to a Spanish setting. The authors chose a 6-month cycle length to match that of the clinical trial. Transitions between health states were assumed to occur, on average, in the middle of a cycle. Therefore, to reflect this assumption, half-cycle corrections were performed for both the health benefits and costs. The time horizon was 2 years. Seven health states were defined: mild-moderate independent; mild-moderate dependent; moderately severe independent; moderately severe dependent; severe independent; severe dependent; and death. Severity was defined by the Mini-Mental State Examination (MMSE) as severe (MMSE 0-9), moderately severe (MMSE 10-14), or mild-moderately severe (MMSE >14). The following assumptions were made.
Memantine efficacy lasted one year. After the first year, transition probabilities for the memantine strategy became those of the standard care strategy.
Memantine had no impact on mortality, and the death probability was the same for both the standard care and memantine strategies.
Dementia and AD patients had similar mortality.
Severity transition probabilities depended on the treatment strategies and patient severity levels at the beginning of the cycle.
Dependency transition probabilities relied on the treatment, patient severity and dependency levels at the beginning of the cycle.
Outcomes assessed in the review The outcomes assessed were the severity transition probabilities for the memantine and standard care strategies, the dependency transition probabilities for the standard care strategy, and the probability of death.
Study designs and other criteria for inclusion in the review Data on severity and dependency transition probabilities were derived from a double-blind randomise controlled trial that evaluated the efficacy and safety of memantine versus placebo in 252 moderately severe to severe AD patients. Epidemiological data were obtained from three Spanish cohorts of dementia or AD patients. The other study designs were not reported.
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 Approximately 12 primary studies were included in the review.
Methods of combining primary studies Investigation of differences between primary studies No differences between the primary studies were investigated.
Results of the review For patients being treated with memantine, the probability of moving from a moderately severe health state to the other states was 20% to mild-moderate, 59% to moderately severe, and 22% to severe. For those in a severe state the probability of moving to a moderately severe state was 11% and the probability of remaining in the severe state was 89%. However, after one year of memantine treatment, transition probabilities for the memantine strategy became the same as those for the standard care strategy.
For patients receiving standard care, the probability of moving from a moderately severe health state to the other states was 19% to mild-moderate, 36% to moderately severe, and 45% to severe. For those in a severe state, the probability of moving to a moderately severe state was 3% and the probability of remaining in the severe state was 97%.
For patients receiving standard care, the probability of moving from a moderately severe independent health state to the moderately severe dependent state was 13% and the probability of remaining in the independent state was 87%.
The probability of moving from the moderately severe dependent health state to the moderately severe independent state was 23% and the probability of remaining in the dependent state was 77%.
For those in the severe independent state, the probability of moving to a severe dependent state was 18% and the probability of remaining in the independent state was 82%.
The probability of moving from the severe dependent health state to the independent state was 0%.
The probability of death for all AD patients was 7.3% for a 6-month period.
Measure of benefits used in the economic analysis The measure of benefits used was the time of independence during a 2-year period (i.e. the average time patients spent in an independent state). The health benefits were discounted at an annual rate of 6%.
Direct costs The direct costs included in the analysis were those to the health care system and care givers. These included the emergency room and physician visits, social and health services, and caregiver costs for medication. The costs were derived from a Spanish cohort, while resource use was assessed using the Resource Utilisation in Dementia (RUD) questionnaire. The source of the unit costs was not reported. All future costs were discounted at an annual rate of 6%. The authors reported the average and incremental costs. All costs were updated to 2005 prices using the Spanish Consumer Price Index.
Statistical analysis of costs The authors provided mean values. They also provided median values, standard deviations (SDs) and percentiles, all derived from the Monte Carlo simulation.
Indirect Costs The indirect costs included in the analysis were those associated with informal care-giving time and lost productivity. The costs were derived from a Spanish cohort. In this study, care-giving time was assessed using the RUD questionnaire and valued using the hourly wage of a nursing assistant for basic daily activities (e.g. personal care, dressing and washing) and the minimum hourly wage of a domestic help for instrumental activities (e.g. general cleaning and cooking). The authors reported that they were unclear how lost productivity was calculated. All future costs were discounted at an annual rate of 6%. The authors reported the average and incremental costs. All costs were updated to 2005 prices using the Spanish Consumer Price Index.
Sensitivity analysis The authors carried out a Monte Carlo simulation using 10,000 iterations. Each of the iterations involved drawing a value for each of the a priori distributions defined for each parameter in the model. The results of this analysis allowed computation of the distribution of outcomes per treatment strategy, incremental outcomes, and acceptability curves.
The authors also undertook sensitivity analysis under six scenarios:
Scenario 1: the treatment effect of memantine was assumed to last only 6 months.
Scenario 2: memantine was assumed to have no effect on severity.
Scenario 3: the minimal effect of memantine on dependency was applied by considering the lower bound of the odds ratio confidence limits.
Scenario 4: no transition to the mild-moderate severity stage was assumed to occur.
Scenario 5: no dependency weights were applied to the costs.
Scenario 6: no discount rate was applied to the outcomes of the model.
Estimated benefits used in the economic analysis Over 2 years, the average time of independence was 0.715 (SD=0.084) years (median 0.717; 95% confidence interval, CI: 0.543 to 0.876) for patients treated with memantine and 0.513 (SD=0.058) years (median 0.510; 95% CI: 0.408 to 0.636) for patients receiving standard care.
Therefore, the incremental time of independence for patients treated with memantine compared with those receiving standard care was 0.202 (SD=0.063) years (median 0.203; 95% CI: 0.074 to 0.321).
Cost results Over 2 years, the average societal cost was EUR 24,055 (SD=10,070) (median 22,453; 95% CI: 8,928 to 48,225) for patients treated with memantine and EUR 24,721 (SD=11,514) (median 22,938; 95% CI: 7,709 to 52,566) for patients receiving standard care.
Therefore, the savings generated from treating patients with memantine compared with standard care were EUR 667 (SD=2,611) (median 405; 95% CI: 6,467 to additional costs of 4,120).
Synthesis of costs and benefits The costs and benefits were not combined as memantine was found to be dominant over standard care (i.e. it was both more effective and less costly). The results of the Monte Carlo simulation showed that, when considering a willingness-to-pay of EUR 30,000 per year of independence gained, the memantine strategy had a 98.1% probability of being cost-effective. At a lower threshold of EUR 10,000, the cost-effectiveness probability was 87.7%.
The results of the one-way sensitivity analyses showed that the results were robust to changes in parameter values. The authors found that for a willingness-to-pay of EUR 30,000 per year of independence gained, the minimum cost-effectiveness probability of memantine across all analyses was 88.1%.
Authors' conclusions The study supports memantine as a cost-effective treatment strategy for patients with moderately severe to severe Alzheimer's disease (AD) in Spain.
CRD COMMENTARY - Selection of comparators A justification was given for using no pharmacological treatment as the comparator. Memantine was the only licensed pharmacological treatment indicated for moderately severe and severe AD patients. You should decide if the comparator represents current practice in your own setting.
Validity of estimate of measure of effectiveness The authors did not report that a systematic review of the literature had been undertaken to identify all relevant research and minimise biases. The authors reported that their model was adapted from UK and Finnish settings by using epidemiological data specific to Spain. The authors provided appropriate details of the studies included in the review. Further, appropriate sensitivity analyses, both one-way and probabilistic, were used to test the uncertainty in the parameter values and their impact on the model's results.
Validity of estimate of measure of benefit The estimation of benefits was modelled using a Markov model. However, the measure of benefit used (i.e. time of independence) is difficult to compare with the benefits from other treatments or interventions in other disease areas, making it difficult to prioritise resources. The use of health measures such as quality-adjusted life-years would have been more desirable as they are easier to generalise and compare across different disease areas.
Validity of estimate of costs All the costs relevant to the societal perspective adopted were included in the analysis. No major relevant costs appear to have been omitted. However, as the authors acknowledged, it was unclear how some costs (e.g. productivity costs) were estimated. The costs and resource use were not reported separately, which will limit the generalisability of the authors' results. Resource use and costs were derived from published studies. Appropriate sensitivity analyses were conducted to assess the uncertainty in the parameter values and their impact on the model's results. Although discounting was not strictly necessary, as the costs were accrued over a 2-year period, all future costs were discounted. The price year was appropriately reported, which will aid any future inflation exercises.
Other issues The authors reported that results from the model using Finnish and UK data demonstrated that treatment with memantine improved the patients' quality of life, prolonged independence, delayed institutionalisation, and reduced total societal costs in comparison with standard care. The generalisability to other settings was addressed in the sensitivity analysis. The authors do not appear to have presented their results selectively and their conclusions reflected the scope of the analysis. The authors acknowledged a number of further limitations to their study. First, no costs per dependency level were found in the literature for a Spanish setting. Second, the estimated cost weights were similar to those computed on the basis of a UK cohort. Finally, some health states used in the model were measured using scales that are inherently subject to possible error.
Implications of the study The authors did not make any explicit recommendations. However, they would appear to recommend the use of memantine for patients with moderately severe to severe AD in Spain.
Bibliographic details Antonanzas F, Rive B, Badenas J M, Gomez-Lus S, Guilhaume C. Cost-effectiveness of memantine in community-based Alzheimer's disease patients: an adaptation in Spain. European Journal of Health Economics 2006; 7: 137-144 Other publications of related interest Because readers are likely to encounter and assess individual publications, NHS EED abstracts reflect the original publication as it is written, as a stand-alone paper. Where NHS EED abstractors are able to identify positively that a publication is significantly linked to or informed by other publications, these will be referenced in the text of the abstract and their bibliographic details recorded here for information
Jones RW, McCrone P, Guilhaume C. Cost-effectiveness of memantine in moderately severe to severe Alzheimer's disease: an analysis based on a probabilistic Markov model from a UK perspective. Drugs Aging 2004;21:607-20.
Francois C, Sintonen H, Sulkava R, Rive B. Cost-effectiveness of memantine in moderately severe to severe Alzheimer's disease: a Markov model in Finland. Clin Drug Invest 2004;24:373-84.
Neumann PJ, Hermann RC, Kuntz KM, et al. Cost-effectiveness of donepezil in the treatment of mild or moderate Alzheimer's disease. Neurology 1999;52:1138-45.
Ward A, Caro JJ, Getsios D, et al. Assessment of health economics in Alzheimer's disease (AHEAD): treatment with galantamine in the UK. Int J Geriatr Psychiatr 2003;18:740-7.
Indexing Status Subject indexing assigned by NLM MeSH Alzheimer Disease /drug therapy /economics; Cost-Benefit Analysis; Dopamine Agents /economics /therapeutic use; Humans; Markov Chains; Memantine /economics /therapeutic use; Randomized Controlled Trials as Topic; Spain AccessionNumber 22006008263 Date bibliographic record published 28/02/2007 Date abstract record published 28/02/2007 |
|
|
|