OBJECTIVE: Hormone replacement therapy (HRT) is the standard therapy aimed at reducing fracture risk in postmenopausal women in Australia, although newer therapies such as the bisphosphonates, calcitriol and raloxifene are used in osteoporotic women with radiographically defined fracture due to minimal trauma. There are many women, however, for whom HRT is inappropriate (eg, those with a history of reproductive cancer or demonstrated intolerance to HRT), and no publicly subsidised therapeutic options are available in Australia for such women. The aim of this study was to assess the cost-effectiveness of raloxifene in preventing osteoporotic fractures when HRT is inappropriate. METHODS: A Markov model was developed to compare raloxifene with no drug therapy in patients who are unable to use HRT. Separate analyses were performed for those who are intolerant of HRT and those unsuitable for HRT due to a history of reproductive cancer (with a consequent greater baseline risk of breast cancer). Relative efficacy assumptions in the model were taken from randomised controlled trials and the published literature. Primary outcomes included vertebral fractures, non-vertebral fractures and breast cancer in a cohort with a low bone mineral density and an average age of 65 years. The model ran for a period of 30 years, contained nine discrete states and produced cost per quality-adjusted life-year (QALY) values. Limited memory was incorporated into the model by separating each fracture health state into two states, representing a first year and then subsequent years after fracture. RESULTS: The incremental cost per QALY gained with raloxifene compared with no therapy was $33,539 in those who are intolerant of HRT and $29,780 per QALY in patients with a history of reproductive cancer. Extensive sensitivity analyses indicated the results remained robust. CONCLUSIONS: Raloxifene is a cost-effective therapy to reduce fracture risk in postmenopausal women unsuitable for HRT.