Several medications have proved effective in reducing the fracture risk in postmenopausal women with osteoporosis. The optimal duration of use of these medications remains to be established, however. Gains in bone mineral density (BMD) persisted throughout 10 years of treatment with alendronate or 7 years with risedronate. However, proof of long-term protection against fractures was obtained only for shorter treatment periods, 4 years with alendronate and 5 years with risedronate. The persistence of treatment effects after drug discontinuation varies across medications, and further studies are needed before this point can be incorporated into treatment decisions. With raloxifene, the BMD effect observed after 3 and 4 years persisted when the drug was given for 8 years, and the fracture risk reduction was similar after 4 years and after 3 years. The long-term safety profile also was similar, with a significant decrease in the incidence of invasive estrogen-receptor-positive breast cancer and a persistent increase in the risk of deep vein thrombosis. However, a sharp drop in BMD occurred upon raloxifene discontinuation. Thus, 4 years may be appropriate for anti-resorptive drug therapy. However, the optimal treatment duration should be determined on a case-by-case basis according to the results of regular fracture-risk evaluations.