|
Cost-effectiveness of donepezil in the treatment of mild or moderate Alzheimer's disease |
Neumann P J, Hermann R C, Kuntz K M, Araki S S, Duff S B, Leon J, Berenbaum P A, Goldman P A, Williams L W, Weinstein M 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 Donepezil was compared with no pharmacological treatment for the management of mild to moderate Alzheimer's disease.
Economic study type Cost-effectiveness analysis.
Study population The study modelled the impact of donepezil for a patient population comprising people with Alzheimer's disease categorised, using the Clinical Dementia rating (CDR) scale, into one of three stages: mild, moderate or severe. The people with Alzheimer's disease could be living in the community or in a nursing home.
Setting The study was based in the community in the USA.
Dates to which data relate The estimates for underlying disease progression were taken from a database of people with dementia who had been examined annually by clinicians between 1986 and 1995. The resources used were taken from a study published in 1993. The price year was 1997.
Source of effectiveness data Effectiveness data were derived from a review/synthesis of previously completed studies and estimates of effectiveness based on opinion.
Modelling A Markov model was used to characterise the progression of Alzheimer's disease through different disease stages and residential settings.
A Cox proportional hazards regression model was used to compare hazard ratios for drug versus placebo of transitions from the mild to the moderate stage and transitions backs to the mild stage among initially moderately affected patients.
Outcomes assessed in the review The annual transition probabilities used in the Markov model were:
stage to stage (mild through to severe and eventually death);
community to nursing home (mild, moderate, severe);
quality-of-life weights (patients: mild, moderate, severe and caregivers: mild, moderate, severe); and
effect of donepezil (mild to moderate and moderate to severe transition).
Study designs and other criteria for inclusion in the review The underlying disease progression was taken from the Consortium to Establish a Registry for Alzheimer's Disease (CERAD).
The effect of donepezil compared to placebo was taken from a 24-week randomised, double-blinded, controlled trial.
No inclusion or exclusion criteria for the studies were reported.
Sources searched to identify primary studies The sources searched to identify the primary studies were not reported.
Criteria used to ensure the validity of primary studies No criteria to ensure the validity of primary studies were reported.
Methods used to judge relevance and validity, and for extracting data The authors did not report any judgement criteria to assess the validity of the primary studies included in the review.
Number of primary studies included Four primary studies were included in the review.
Methods of combining primary studies The authors pooled the results of the 24-week trial that was used to estimate the effectiveness of donepezil. The 24-week trial evaluated the effect of 5mg and 10mg doses of donepezil. The model used a pooled estimate of the 5mg and 10mg results to increase statistical power.
Investigation of differences between primary studies The authors did not investigate the difference between the primary studies, provide an explanation of differences between individual studies, or investigate how the differences affect the estimate of the effectiveness of donepezil.
Results of the review The estimates of transition of disease progression from stage-to-stage were:
0.614 for mild to mild;
0.322 for mild to moderate;
0.042 for mild to severe;
0.021 for mild to dead;
0.043 for moderate to mild;
0.565 for moderate to moderate;
0.339 for moderate to severe;
0.053 for moderate to dead;
0.0 for severe to mild;
0.0 for severe to moderate;
0.847 for severe to severe;
0.153 for severe to dead.
The transition probabilities from living in the community to a nursing home were: 0.038 for mild disease; 0.110 for moderate disease and 0.259 for severe disease.
The quality of life weights for the patients were: 0.68 for community and 0.71 for nursing home patients with mild disease; 0.54 for community and 0.48 for nursing home patients with moderate disease; and 0.37 for community and 0.31 for nursing home patients with severe disease.
The quality of life weights for the caregivers were: 0.86 for community and 0.86 for nursing home patients with mild disease; 0.86 for community and 0.88 for nursing home patients with moderate disease; and 0.86 for community and 0.88 for nursing home patients with severe disease.
The effect of donepezil was 0.50 for mild to moderate and 2.36 for moderate to severe transition.
Methods used to derive estimates of effectiveness Expert opinion was sought from 32 clinicians who served on one of three panels (Working Group on Practice Guidelines on Alzheimer's Disease of the American Psychiatric Association, Task Force of the Dementias of the American Academy of Neurology or the Clinical Task Force of CERAD) to estimate the duration of the effect of donepezil on disease progression. Thirteen respondents answered the survey (response rate of 41%).
Estimates of effectiveness and key assumptions The mean expected duration of the effect of donepezil was 17.8 months (range: 6 months to 3 years). This estimate was taken from the responses to the following question: "For the average patient with mild to moderate Alzheimer's disease who has tolerated an appropriate dose of an acetylcholinesterase inhibitor and has had an average response, how long do you think the drug will delay progression of cognitive deterioration?"
Measure of benefits used in the economic analysis The measure of benefit used in the economic analysis was the quality-adjusted life-year (QALY). The QALY was used to reflect the desirability of living in each state. The quality of life weights were measured using the Health Utilities Index Mark II (HUI:2) in a sample of 528 caregivers of people with Alzheimer's disease stratified by disease severity (201 mild, 175 moderate and 142 severe) and setting of care (354 community and 164 nursing home). The caregivers completed questionnaires to assess quality of life of the person with Alzheimer's disease (proxy valuation) and also the caregivers' own quality of life. Benefits were discounted at a rate of 3%.
Direct costs Quantities and costs were not reported or analysed separately. The direct costs included in the analysis were medical and non-medical costs. The costs were estimated by stage of disease and care setting. The estimation of the quantities and estimation of the costs were based on published estimates of the costs of formal and informal care for patients with Alzheimer's disease. The resource use and cost data were taken from published studies. Drug costs were taken from the average wholesale price. Office visits related to obtaining donepezil and subsequent monitoring were based on Medicare payment (charge) for a visit for drug management. The resources related to a study published in 1993. The price year was 1997. Discounting was carried out at a rate of 3%. The baseline analysis reported the average costs for an 18-month time horizon from diagnosis.
Statistical analysis of costs No statistical analysis of costs was reported.
Indirect Costs The study included indirect costs, which were defined as the costs associated with unpaid care giving. Quantities and costs were not reported or analysed separately. Care-giving time was valued at the average wage rates at nursing and personal care facilities. The resources related to a study published in 1993. The price year was 1997. Discounting was carried out at a rate of 3%. The baseline analysis reported the average costs for an 18-month time horizon from diagnosis.
Currency US dollars ($). No currency conversions were reported.
Sensitivity analysis The study included a one-way sensitivity analysis of the following variables:
duration of effect (6 months, 12 months and 18 months);
drug effect (varied by one standard error);
drug cost (set at US$1,000 and US$2,000);
quality of life weights (varied by 10%);
annual discount rate (0%, 5% and 7%);
rate of disease progression (varied by 10%);
stage to nursing home transition (varied by 50%); and
rate of discontinuation (0% and 8%).
The study also looked at the effect of including direct costs only.
A threshold analysis was used to show the duration of effect at which donepezil would show economic savings.
Estimated benefits used in the economic analysis Over 18 months, the benefits to patients prescribed donepezil were estimated at 0.950 QALYs compared to 0.923 QALYs for patients not prescribed donepezil, based on the assumption that all patients initially have mild disease and are living in the community.
Over 18 months, the benefits to patients prescribed donepezil were estimated at 0.717 QALYs compared to 0.706 QALYs for patients not prescribed donepezil based on the assumption that all patients initially have moderate disease and are living in the community.
Cost results Over 18 months, the costs to society for patients prescribed donepezil were estimated at US$72,487 compared to US$72,227 for patients not prescribed donepezil based on the assumption that all patients initially have mild disease and are living in the community.
Over 18 months, the costs to society for patients prescribed donepezil were estimated at US$84,427 compared to US$83,585 for patients not prescribed donepezil based on the assumption that all patients initially have moderate disease and are living in the community.
Synthesis of costs and benefits Over 18 months, the incremental cost-effectiveness ratio for donepezil compared to no treatment was US$9,300 per QALY for patients with mild Alzheimer's disease who live initially in the community.
Over 18 months, the incremental cost-effectiveness ratio (ICER) for donepezil compared to no treatment was US$76,000 per QALY for patients with moderate Alzheimer's disease who live initially in the community.
Over a 12 month time horizon, the ICER for donepezil was US$32,000 per QALY and US$140,000 per QALY, respectively, for mild and moderate Alzheimer's disease.
Over a 6 month time horizon, the ICER for donepezil was US$160,000 per QALY and US$440,000 per QALY, respectively, for mild and moderate Alzheimer's disease.
The threshold analysis showed that cost savings would be achieved if the effect of donepezil persisted for 24 months for mild disease but savings would never be achieved regardless of the duration of effect in moderate disease.
Authors' conclusions The authors concluded that the cost of donepezil might be partially offset by a reduction in the costs of care due to improved cognitive functioning, which resulted in a delay to more costly disease stages and settings.
CRD COMMENTARY - Selection of comparators The selection of comparators, donepezil and no treatment, was appropriate at the time this study was completed. However, decision-makers should be aware that other pharmacological treatments for Alzheimer's disease are now available and this study looks only at the additional costs and benefits associated with donepezil compared to no treatment.
Validity of estimate of measure of effectiveness The authors were not able to locate any published evidence that estimated the duration of effect for donepezil and, therefore, used a panel of experts to provide this information. The panel of experts provided a wide range of estimates for the effect of duration, which gives an indication of the level of uncertainty in the point estimate used for this variable in the baseline analysis. The effect of this uncertainty on the results of the model is explored in the sensitivity analysis. The authors did not outline criteria for studies included in the review or assess the quality of the included studies. It is therefore not possible to make an assessment of the validity of the point estimates used in the baseline analysis of the model.
Validity of estimate of measure of benefit This study used proxy valuations from the caregivers of people with Alzheimer's disease to estimate the impact on different stages of the disease on quality of life. The authors noted that there is evidence to suggest that proxy ratings tend to describe a higher level of impairment than patients' ratings of the same health state. Thus, the values for QALYs reported in this study may tend to be lower than if the patients had valued the experience of living with Alzheimer's disease. However, it is methodologically difficult to develop a valuation tool that allows patients with a disease that impairs their cognitive function to rate or value their own quality of life.
Validity of estimate of costs The cost estimates used in this study seem appropriate for the USA. Careful consideration should be given as to whether the unit costs are appropriate to the UK health care setting.
Other issues A 3% discount rate was applied to costs and benefits used in the model. This discount rate is appropriate for the US setting but not for the UK where a discount rate of 6% should be used. The sensitivity analysis showed that the choice of discount rate did not affect the findings of the model.
The authors explicitly highlighted the main limitations of this study and cautioned that decision-makers should "exercise substantial caution in interpreting the results for their own settings". In particular the authors noted that:
the information taken from the CERAD database may not be representative of the Alzheimer's disease community at large;
the evidence on the duration of the drug effect is limited;
the HUI:2 has not been validated in Alzheimer's disease; and
the transition probability to death differs by disease stage and this assumption means that the difference in QALYs between the donepezil and no treatment group is due to a mortality rather than quality of life effect, but there is no direct evidence that donepezil reduces mortality.
Implications of the study The authors suggest that this study should be viewed as a methodological demonstration that provides insights into the possible effects of donepezil and highlights the need for more data to enable informed decision-making. The authors suggest that future studies could investigate the cost-effectiveness of other interventions, including non-pharmacological interventions, for Alzheimer's disease.
Source of funding Supported by a grant from Pfizer Inc. to the Harvard Program on the Economic Evaluation of Medical Technology.
Bibliographic details Neumann P J, Hermann R C, Kuntz K M, Araki S S, Duff S B, Leon J, Berenbaum P A, Goldman P A, Williams L W, Weinstein M C. Cost-effectiveness of donepezil in the treatment of mild or moderate Alzheimer's disease. Neurology 1999; 52(6): 1138-1145 Other publications of related interest Comment in: Neurology 1999;52(6):1115-6.
Indexing Status Subject indexing assigned by NLM MeSH Alzheimer Disease /drug therapy /economics /physiopathology; Cholinesterase Inhibitors /economics /therapeutic use; Cost-Benefit Analysis; Costs and Cost Analysis; Disease Progression; Humans; Indans /economics /therapeutic use; Piperidines /economics /therapeutic use; Quality of Life AccessionNumber 21999000776 Date bibliographic record published 28/02/2002 Date abstract record published 28/02/2002 |
|
|
|