Analytical approach:
A state-transition Markov model was developed to simulate the natural history of Alzheimer's disease and to assess the impact that treatment with donepezil would have on the costs and health outcomes associated with an ongoing risk of disease over a six-, 12-, 18-, 24-, or 30-month time horizon. The authors stated that the analysis was conducted from a societal perspective and a health service perspective.
Effectiveness data:
The main treatment effectiveness parameter was from a single RCT that studied the dose impact of 5mg or 10mg per day of donepezil compared with placebo in patients with mild or moderate Alzheimer's disease (Rogers, et al. 1998, see 'Other Publications of Related Interest' below for bibliographic details). The effects in patients receiving 5mg and those receiving 10mg were combined to increase the power of the study. The follow-up period was 24-weeks. A review of the literature was also undertaken and other clinical parameters included completion of treatment (compliance) and one-month transition probabilities between mild, moderate, severe, and death health states based on the natural history of the disease or adjusted for donepezil. These adjustments for donepezil were based on the relative risk of transitioning from mild to moderate and from moderate to mild for patients receiving donepezil compared with no treatment.
Monetary benefit and utility valuations:
The utility estimates for mild, moderate, and severe health states were calculated from a single study of 237 Alzheimer's disease patients, in the Canary Islands, Spain, who completed the European Quality of life (EQ-5D) questionnaire (Lopez-Bastida, et al. 2006, see 'Other Publications of Related Interest' below for bibliographic details). The social values of health status in the EQ-5D were taken from interviews with 300 people in a health centre in Barcelona using the time trade-off approach. These were used to calculate the utilities for EQ-5D responses.
Measure of benefit:
The benefit measure was the number of quality-adjusted life-years (QALYs) and this was discounted at an annual rate of 3%.
Cost data:
The cost categories were direct medical costs and direct non-medical costs. Direct medical costs included the cost of donepezil, other drugs, medical visits, hospital admissions, emergency visits, orthopaedic devices, and others. Direct non-medical costs included the cost of carers. The cost data was reported in Euros (EUR) and was based on the single study of 237 Alzheimer's disease patients in the Canary Islands (Lopez-Bastida, et al. 2006). The cost of donepezil was based on the proportion of patients receiving 5mg and 10mg doses in the donepezil RCT. The price year was 2006 and an annual discount rate of 3% was applied.
Analysis of uncertainty:
One-way sensitivity analyses were performed on the time horizon, discount rates, initial number of patients in the mild, moderate, and severe states, and completion of treatment. A probabilistic sensitivity analysis, using 10,000 Monte Carlo simulations, was performed to assess the uncertainty in key parameters of the model. The results of this analysis were presented in a cost-effectiveness acceptability curve.