OBJECTIVE: Recent studies demonstrate the cost-effectiveness of inhaled corticosteroids (ICS) in adults. However, they fail to account for the impact of ICS-induced bone mineral density (BMD) loss and resultant hip fractures. METHODS: We used a previously published Markov model to compare quick relievers (e.g., b-agonists) alone vs. the addition of ICS therapy. State-space dimensions included patient age, clinical history, and lung function (FEV1% predicted). Risk functions were estimated using symptom, exacerbation, and hospitalization rates obtained from literature reviews and analyses of primary, cross-sectional data. Societal costs were derived from published economic studies of inpatient and outpatient asthma. We used published data retrieved from 3 RCTs and 3 qualified observational studies to estimate yearly change in hip BMD ranging from 0.000933g/cm2 to -0.00058g/cm2 per 100mcg standard ICS dosage. ICS side-effects were modeled as an annual reduction in age-predicted BMD. Risk functions to link BMD to hip fracture rates and QOL weights for hip fracture were derived from published estimates. RESULTS: In the absence of BMD decline, we estimated an incremental cost-effectiveness ratio of $15,000 per quality-adjusted life-year (QALY) for ICS therapy. In sensitivity analysis, we considered 3 alternative BMD decline scenarios (none; mean; high) and 4 alternative models of the temporal impact of ICS on BMD (10-, 20-, 30-year, and lifetime). Incremental cost-effectiveness ratios under various combinations of these two variables ranged from $15,000 to over $130,000/QALY. The ICS therapy strategy was dominated (i.e., it produced both higher costs and lower total QALYs) under the "high" BMD decline scenario with temporal risk greater than "30-years." CONCLUSION: The economic attractiveness of inhaled steroids is highly dependent upon what is assumed about their impact on BMD and the resultant fracture rate. More research is needed on these variables before definitive cost-effectiveness conclusions can be drawn.