Background: The selective estrogen-receptor modulators (SERMs) tamoxifen (TAM) and raloxifene (RAL), and more recently the aromatase inhibitor (AI) exemestane (EXE) have all been shown to reduce the risk of breast cancer (BC) in post menopausal (PM) women at high risk. As significant differences in adverse effect profiles and costs exist amongst these agents, determining the preferred strategy for clinical use requires consideration of both the risk-benefit profile and overall cost-effectiveness (CE) of each strategy. Prior studies have examined the CE of SERMs in this setting but similar evaluations of AIs have not yet been performed. Methods: We examined the probabilities of various BC prevention strategies, including 1) SERMs (TAM or RAL) relative to placebo and 2) AI relative to SERMs and placebo, meeting the $100,000 per quality-adjusted life year (QALY) gains threshold. A Markov model was constructed for a hypothetical cohort of PM women with varying elevated risks of developing BC and treatment related adverse events. Costs, utilities and probabilities were derived from the literature and relevant BC prevention trials including NSABP-P1, STAR and MAP 3. The analysis took a third-party payer perspective and reports costs in 2013 CDN dollars. Results: Substantial variability in the relative CE of each strategy was observed when probabilistic sensitivity analysis was performed and thus cost effectiveness acceptability frontiers (CEAFs) were used to compare the relative CE of each strategy. Both SERMs were associated with incremental gains in QALYs as well as incremental costs compared to no treatment but AI was most often the preferred BC prevention strategy at a willingness-to-pay threshold of $100,000/QALY. Conclusions: The use of EXE for BC prevention is associated with incremental QALY gains and is a cost-effective strategy in PM women at high risk of BC. The strategy of choice is dependent upon BC risk as well as the risk of adverse effects and the willingness-to-pay for a QALY gain.