A published Markov model, with a one-year cycle length, was adapted to the Swiss setting, using local mortality, fracture incidence, and cost data from a variety of sources. The health states included healthy, healthy post-vertebral fracture, healthy post-hip fracture, healthy post-second hip fracture, vertebral fracture, hip fracture, second hip fracture, wrist fracture, and dead. A diagram of the model was provided. The time horizon of the analysis was seven years, with five years of treatment and two years of residual effect after treatment. The authors took a health care perspective.
The effectiveness of risedronate versus no intervention was based on the results of large double-blind, randomised controlled studies. The residual effect post treatment was assumed to last for two years with a linear decline from 100% to 0%. The rate of premature discontinuation of risedronate therapy was incorporated.
Monetary benefit and utility valuations:
Population-based, age-specific general utility values were taken from the Swedish general population and published utility decrements, due to fracture, were applied. The instruments used to derive these utilities were not reported.
Measure of benefit:
The primary measures of benefit were quality-adjusted life-years (QALYs) and fractures averted. A discount rate of 3% per year was applied.
The cost categories included daily in-patient costs, drug costs, diagnostics, and other services. The unit costs were collected from official prices and tariffs for Switzerland. For example, the daily in-patient costs were obtained from the medical statistics database and the socio-medical institutions database of the Swiss Federal Statistical Office. The unit costs and resource use were reported in detail. The costs were measured in Swiss francs (CHF) and were converted into Euros (EUR) at the exchange rate of one EUR equals CHF 1.6. The price year was 2005 and a discount rate of 3% per year was applied.
Analysis of uncertainty:
Univariate sensitivity analysis was conducted on all parameters. The results for 120 scenarios were reported in a table. These included patients aged 60, 65, 70, 75, and 80 years; with, without, and a combination of the risk factors of previous vertebral fracture, maternal history of hip fracture, and history of any fracture since the age of 50; and with the residual effect post treatment lasting no years, two years, and five years. A scenario analysis was also conducted on the effectiveness for patients who prematurely discontinued risedronate therapy, the probability of a new nursing home admission after a hip fracture, and the discount rate.