SUMMARY
A Markov state transition model was developed to predict 10-year cumulative incidence-based costs of postmenopausal breast cancer. Using published clinical data, the model simulates, during 10 periods of 1 year each, the disease progression through nine health states: node-negative and node-positive early cancer; local relapse; metastasis (each with its follow-up states); and death. The health insurance perspective was taken and a 3% annual discount rate on future costs was applied. The cost per health state was obtained from a chart review of 118 patients with different disease states. The average cumulative 10-year cost per patient was €31,774 (95% confidence interval €30,536 - 33,012), of which 30% was due to hospitalisation and 28% to systemic treatment and radiotherapy. The model provided a valid simulation of long-term breast cancer outcome. This model may serve as a valid baseline for economic assessment of new interventions in prevention or early treatment.