Vitamin D/calcium substitution is generally regarded as an effective first step treatment for glucocorticoid-induced osteoporosis (GIOP). The aim of our study was to evaluate the efficacy of the active vitamin D metabolite alfacalcidol (1alpha) compared with the native vitamin D(3) in patients with established GIOP with or without vertebral fractures. Patients on long-term corticoid therapy were given either 1 microg alfacalcidol plus 500 mg calcium per day (group A, n = 43) or 1000 IU vitamin D(3) plus 500 mg calcium (group B, n = 42). The two groups were alike in age range, sex ratio, percentages of underlying diseases, average initial bone density values (lumbar spine: mean T-score -3.28 and -3.25, respectively), and rates of vertebral and nonvertebral fractures. During the 3-year study we found a small but significant increase of lumbar spine density in group 1alpha (+2.0%, P < 0.0001) and no significant changes at the femoral neck. In the D(3) group, there were no significant changes at both sites. At the end of the study, 12 new vertebral fractures had occurred in 10 patients of the group 1alpha and 21 in 17 patients of the D(3) group. In accordance with the observed fracture rates, the alfacalcidol group showed a significant decrease in back pain (P < 0.0001) whereas no change was seen in the vitamin D group. We conclude that with the doses used in this trial, alfacalcidol is superior to vitamin D in the treatment of established GIOP.