OBJECTIVES: Tamoxifen and raloxifene has been approved by US FDA for breast cancer chemoprevention. Exemestane was also found to have promising effects on breast cancer prevention based on 3 years follow-up of MAP.3 trial. The objective of the study was to assess the cost-effectiveness of continuing therapy with tamoxifen for 5 years versus switching to exemestane or raloxifene, after 2 years of treatment with tamoxifen in postmenopausal women in the United States from third party payer perspective.  METHODS: A Markov model was developed using Treeage pro suite 2009, where cost and health outcomes were utilized for estimating quality adjusted life year (QALY). A hypothetical cohort of 1000 women with 5 year predicted breast cancer risk (≥ 1.66) based on the Gail et al. model, history of lobular carcinoma in-situ, and history of atypical hyperplasia was simulated using the Markov model. The model was developed comprising of 5 health states (Healthy, Adverse Event, Breast Cancer, Death due to Breast Cancer, and Death due to Other Causes) to calculate incremental costs per QALY gains for tamoxifen, raloxifene and exemestane. Annual transition probabilities were derived from NSABP P-1 and NSABP P-2 trials (for tamoxifen and raloxifene) and MAP.3 trial (for exemestane). Direct costs and utilities were literature-based. Costs and benefits were discounted at 3% annually. Two-way sensitivity analyses were performed by varying values for key parameters in the model.  RESULTS: Under base-case scenario, gain of QALYs per patient associated with exemestane (29 QALYS) was higher compared with tamoxifen (28 QALYS) and raloxifene (28 QALYS). The cost of gaining one QALY with exemestane ($44,723) was found to be low compared to tamoxifen ($93,053) and raloxifene ($70,940). These results were robust to the two-way sensitivity analyses performed.  CONCLUSIONS: Results suggested that switching to exemestane after 2 years of primary treatment with tamoxifen was cost-effective.