: A state-transition Markov model was used to estimate the clinical and economic outcomes associated with the treatment options, using effectiveness data from a randomised controlled trial. Adverse events, such as osteoporosis, endometrial cancer, thromboembolism, and deep-vein thrombosis, were incorporated into the model using separate health states. The time horizon was lifetime, with a maximum of 38 years. The cycles were six months long and half-cycle corrections were used to account for events occurring during a cycle. The authors stated that the perspective was that of the German Statutory Health Insurance payer.
The effectiveness evidence, for the first 36 months of the model, came from the Intergroup Exemestane Study (IES), which compared continuing treatment with tamoxifen against switching to exemestane. The main clinical parameters were the recurrence of breast cancer, remission, and death from breast cancer.
Monetary benefit and utility valuations:
The utility estimates, for all the relevant health states, were derived from a number of international published studies.
Measure of benefit:
The measure of benefit was the number of quality-adjusted life-years (QALYs), which were discounted at an annual rate of 5%.
The direct health care costs included those of the drugs, adverse events, and operations, chemotherapy, and terminal care for breast cancer. In-patient treatment costs were from the Diagnosis-Related Group (DRG) Browser, out-patient costs were based on the Einheitlicher Bewertungsmassstab (EBM) 2000 plus, and drugs were based on current market prices from the Red Book Service. The appropriate DRGs for the different health states were identified using medical expert opinion. Future costs were discounted at a rate of 5% per annum and the currency was Euros (EUR).
Analysis of uncertainty:
A probabilistic sensitivity analysis was performed, using 1,000 Monte Carlo simulations. The results of this analysis were presented in a cost-effectiveness acceptability curve.