The Multiple Outcomes of Raloxifene Evaluation trial studied 7705 postmenopausal women with osteoporosis randomized to placebo, or raloxifene 60 or 120 mg/d [JAMA 282(1999): 637]. This report assesses the efficacy of raloxifene on the long-term cumulative incidence new vertebral fractures through 4 yr. New vertebral fractures was assessed from radiographs taken at baseline, yr 2-4. The primary analysis was the cumulative incidence of new vertebral fractures through 4 yr. A posthoc analysis compared the vertebral fracture risk in yr 4 alone with that observed in the first 3 yr. The 4-yr cumulative relative risks (RR) for one or more new vertebral fractures were 0.64 [95% confidence interval (CI) 0.53, 0.76] with raloxifene 60 mg/d and 0.57 (95% CI 0.48, 0.69) with raloxifene 120 mg/d. In yr 4 alone, raloxifene 60 mg/d reduced the new vertebral fracture risk by 39% [RR 0.61 (95% CI 0.43, 0.88)], which was not found to be significantly different from the RR observed in the first 3 yr in both raloxifene groups, irrespective of prevalent fracture status. The nonvertebral fracture risk was not significantly reduced [RR 0.93 (95% CI 0.81, 1.06)]. The safety profile after 4 yr was similar to that observed after 3 yr. Raloxifene 60 and 120 mg/d through 4 yr decreased the cumulative risk of new vertebral fractures in postmenopausal women with osteoporosis. The decreased vertebral fracture risk in yr 4 alone was not different from that observed in the first 3 yr.