Ankle fractures are frequently observed in postmenopausal women although the pattern of incidence and risk factor profile suggest that ankle fracture may not be a typical osteoporotic fracture. The aims of this study were to determine the prevalence of osteopenia and vertebral fracture and to evaluate the diagnostic accuracy of dual-energy X-ray absorptiometry (DXA), anthropometry, lifestyle and reproductive factors in women who have sustained an ankle fracture. We studied 103 women aged 50-80 years (mean 63.2, 7.9 SD) with ankle fracture. These were compared with 375 women aged 50-86 years (mean 64.5, 9.1 SD) from a population-based cohort. Bone mineral density (BMD) at the lumbar spine (LS) and contralateral proximal femur (including femoral neck (FN), Ward's triangle (WT) and trochanteric region (TR)) was measured by DXA. Quantitative ultrasound (QUS) of the calcaneus and proximal digits was measured using three different devices. Radiographs of the thoracolumbar spine were taken (anteroposterior and lateral views). There were no significant differences in the prevalence of osteoporosis (T<-2.5 level) at the LS, FN and WT sites. The population-based cohort had lower TR BMD than the ankle fracture cohort. Age-and weight-adjusted Z-scores of FN BMD were significantly lower in the ankle fracture group. Age- and weight-adjusted Z-scores of QUS gave contradictory results. There were no differences in the receiver operating characteristics of DXA compared with QUS. Twenty-seven women (7%) of the population-based cohort and 10 women (10%) of the ankle fracture cohort were found to have prevalent vertebral fractures; these were not significantly different.