Purpose: Oral glucocorticoids (GCs) are associated with increased fracture risk. Alendronate is a potent bone resorption inhibitor. Alfacalcidol stimulates bone formation. The study was performed to determine the cost-effectiveness of prevention of glucocorticoid-induced osteoporosis (GIOP). Methods: Cost-effectiveness analysis with a ‘life-time' time horizon and 4% discount rate for costs and effects using a Markov model to compare prevention of GIOP with alendronate and alfacalcidol to no treatment. Results of lumbar spine bone mineral density (BMD) of a randomized placebo-controlled trial (RCT) in patients starting oral GCs in a daily dosage of 7.5 mg prednisone equivalent or higher were used predict fractures and estimate incremental cost-effectiveness ratios (iCER). Also, literature based pooled estimates of the relative risk (RR) of vertebral fractures after alendronate and alfacalcidol were used to predict fractures. A threshold of € 30.000 per quality adjusted life-year (QALY) gained was assumed to be acceptable. Results. Mean age in the RCT was 60 years. In women, using BMD trial data, iCERs for alendronate and alfacalcidol were of € 94.261 and € 144.000 per QALY gained, respectively. With the pooled RRs the iCERs were € 58.111 and € 63.115, respectively. In sensitivity analysis applying pooled RRs of vertebral fractures to all types of fractures, resulted in iCERs of € 29.389 and € 26.457 per QALY gained. For alendronate and BMD as predictor of fractures the € 30.000 threshold was reached at a price of € 201 per year in women and at € 78 in men. For alfacalcidol the threshold price was € 103 in women and € 40 in men. Using pooled RRs of vertebral fractures the threshold price of alendronate was € 300 per year in women and € 106 in men. For alfacalcidol the prices were € 218 in women and € 72 in men. Conclusions. Overall, alendronate and alfacalcidol do not appear to be cost-effective in prevention of GIOP in patients with a mean age of 60 years. Only if the costs of treatment would drop considerably could prevention become cost-effective. Moreover, model assumptions regarding the association between BMD and fracture incidence had considerable (negative) impact in the cost-effectiveness.