Aromatase inhibitors (AIs) have been evaluated clinically in a wide range of breast cancer treatment settings. Although these agents appear to have clinical superiority, they are more expensive than the therapies (primarily tamoxifen) they have been compared with, thus economic evaluation is required to consider their incremental value to the payer. This paper reviews published economic evaluations of AIs as first-line therapy for advanced cancer, and as adjuvant therapy for early breast cancer. The evaluations in the advanced setting demonstrate a range of different modelling techniques and consider the payer's perspective in three healthcare systems. There is broad similarity in the application of standard cohort Markov modelling techniques to evaluate AIs in the early breast cancer setting, covering four separate health system perspectives. AIs appear cost effective compared with current practice. An analysis in the advanced setting suggests that letrozole may be the more cost-effective AI. No direct comparisons of alternative AIs in the adjuvant setting are reported, although indirect comparisons may be feasible. Future evaluations of treatment strategies over the entire course of the disease may also be needed.