Analytical approach:
The economic evaluation used a 20-year Markov model to simulate the costs and benefits of disease progression and survival, for a cohort of Swedish patients with cognitive impairment or dementia. The model had five states: minor cognitive impairment; mild, moderate, and severe Alzheimer's disease; and death. The authors stated that a societal perspective was taken.
Effectiveness data:
Patient characteristics were from Swedish prevalence statistics. The conversion rate was the primary measure of effectiveness. This was the rate at which patients with minor cognitive impairment developed mild Alzheimer's disease. There were no effectiveness data for the treatment. It was assumed that it halted the conversion from minor cognitive impairment, for half the patients, and that overall survival was improved. Patients who started the model in the minor cognitive impairment state had a probability of transition to mild Alzheimer's disease that was based on a systematic review of minor cognitive impairment. The probability of death in the minor cognitive impairment state was based on Swedish epidemiological data. The transition probabilities between other states were from the Kungsholmen Project. The states and progression were defined using the Mini Mental State Examination.
Monetary benefit and utility valuations:
Utility scores were from a Swedish study of utility for patients with minor cognitive impairment and dementia. These scores were assigned to each model state.
Measure of benefit:
The primary measure of benefit was quality-adjusted life-years (QALYs). The benefits were discounted at 3% annually.
Cost data:
The costs of disease-modifying treatment were assumed. The resource use data, for each model state, were from the Nordanstig-Kungsholmen Project. The unit costs were from a Swedish National Board of Health and Welfare report. State-specific costs included long-term institutional care, hospital care, hospital clinic visits, day care, home services, and informal care. Informal care assumed a mix of employed and retired informal carers. The costs were in 2005 Swedish kronor (SEK). They were discounted at 3% annually.
Analysis of uncertainty:
One-way and probabilistic sensitivity analyses were conducted. The discount rates were varied between zero and 5%; the conversion risk was increased from 10% to 25%; the treatment drug costs were assumed to be similar to those of standard care (SEK 10,000) or much greater (SEK 300,000); the costs of informal care were raised to reflect professional care; and no survival benefit with treatment was assumed. The probabilistic sensitivity analysis varied mortality and the rate of conversion from minor cognitive impairment to Alzheimer's disease, for 1,000 simulations.