BACKGROUND: Postmenopausal women with a prior fracture have an increased risk for future fracture. Whether a history of non-vertebral fracture defines a group of women with low bone mass but without osteoporosis for whom alendronate would prevent new non-vertebral fracture is not known. SUBJECTS AND METHODS: Secondary analysis of data from the Fracture Intervention Trial (FIT). Of 2,785 postmenopausal women with a T-score at the femoral neck between -1 and -2.5 and without prevalent radiographic vertebral deformity, 880 (31.6%) reported experiencing a fracture after 45 years of age. Women were randomized to placebo or alendronate (5 mg/day years for the first 2 years and 10 mg/day thereafter) and were followed for an average of 4.2 +/- 0.5 years. Incident non-vertebral fractures were confirmed by x-rays and radiology reports. RESULTS: In the placebo arm, a self-report of prior fracture identified women with a 1.5-fold (hazard ratio [RH] 1.46, 95% C.I. 1.04-2.04) increased risk for incident non-vertebral fracture. However, there was no evidence that the effect of alendronate differed across subgroups of women with (RH 1.26 for alendronate vs placebo, 95% C.I. 0.89-1.79) and without prior fracture (RH 1.02 for alendronate vs placebo, 95% C.I. 0.76-1.38; P = 0.37 for interaction). CONCLUSION: Assessing a clinical risk factor, prior non-vertebral fracture, did not identify women with low bone mass for whom alendronate reduced future non-vertebral fracture risk.