In a population-based, retrospective study, we investigated the effect of a previous fracture on the risk of a later fracture. A questionnaire was mailed to 46,353 postmenopausal women aged 50-80 years and 29,802 (64%) responded. Questions were asked about fractures after age 25 and in what year they had occurred. The fractures were grouped according to whether they occurred within 5 and 10 years before answering the questionnaire. The occurrence of a fracture preceding these time intervals increased the odds ratio (OR) of having sustained a hip fracture to 1.6 (95% CI 1.1-2.3) for a previous ankle fracture and to 3.5 (95% CI 2.4-5.0) for a previous humerus fracture. Corresponding figures for having sustained a spine fracture were 1.5 (95% CI 1.1-2.1) for a previous ankle fracture and 4.5 (95% CI 3.4-5.9) for a previous spine fracture. It seems possible to select women for evaluation and intervention against osteoporosis by using information on previous fragility fractures.