OBJECTIVES: At a median follow-up of 100 months in the ATAC trial, anastrozole demonstrated significant improvements compared with tamoxifen in disease-free survival (HR: 0.85, p=0.003) and time-to recurrence (TTR) (HR: 0.76, p=0.0001). Absolute difference in TTR increased over time, 2.8% at 5 years and 4.8% at 9 years. Recurrence rates remained significantly lower on anastrozole after treatment completion. This study assesses the long-term cost-effectiveness of adjuvant treatment with anastrozole versus tamoxifen in women with early breast cancer in Sweden. METHODS: A Markov model was developed to integrate trial data with external data on costs and quality of life specific for Swedish breast cancer patients. The model adopts a life-long perspective and the primary outcome measure was QALYs. In the base-case model, TTR for tamoxifen and anastrozole diverge out to 10 years. Mortality risks for recurred and non-recurred patients are based upon trial data and national statistics. Resource utilization and utility estimates were based on Swedish data for patients in different stages of breast cancer. Utility weights were estimated with EQ-5D. All costs and benefits were discounted at 3% at 2005-year prices. RESULTS: In base-case, patients treated with anastrozole had a higher lifetime expected mean cost of SEK 16,000 compared to patients on tamoxifen, and gained on average 0.11 QALYs per patient. This corresponds to an incremental cost of approximately SEK 145,000 per QALY gained.  CONCLUSIONS: This updated cost-effectiveness analysis of the ATAC trial is conservative, neither is any assumption made of an accrued survival benefit due to a significantly lower recurrence rate, nor does the analysis consider the potential reduction of costs associated with anastrozole patent expiry. Nevertheless, the study demonstrates that up-front anastrozole treatment for five years is a cost-effective option compared to five years tamoxifen treatment in the lifetime perspective for hormone receptor-positive patients.