To examine the association between osteoarthritis (OA), as defined by radiographic evidence and self report, and osteoporotic fractures, falls, and bone loss in a cohort of elderly white women.


A cohort of 5,552 elderly women from the Study of Osteoporotic Fractures was followed up prospectively for a mean of 7.4 years. Self-reported, physician-diagnosed OA was recorded at interview, and radiologic OA of the hip and hand were defined from pelvis and hand radiographs obtained at baseline by validated techniques. Prevalent and incident vertebral fractures were detected by vertebral morphometry, and data on incident fractures and falls were collected by postcard surveys; fractures were confirmed by radiography. Bone mineral density (BMD) was measured on 2 occasions at the hip, lumbar spine, and calcaneus, and rates of bone loss were calculated.


Women with radiographic hip OA had a reduced risk of recurrent falls in the first year (relative risk [RR] 0.7, 95% confidence interval [95% CI] 0.5–0.95). However, those with self-reported OA had an increased risk of falls (RR 1.4, 95% CI 1.2–1.5). Radiographic hip OA was associated with reduced bone loss in the femoral neck compared with controls (mean ± SD −0.29 ± 0.09%/year versus −0.51 ± 0.03%/year; P = 0.018). However, radiographic hip OA showed nonsignificant trends toward increased bone loss at the calcaneus and lumbar spine. There was no significant association between self-reported OA or radiographic hand OA with bone loss. No definition of OA was associated with incident nonvertebral fracture, hip fracture, or vertebral fracture.


Despite having increased BMD compared with controls, subjects with OA did not have a significantly reduced risk of osteoporotic fracture, although there was a trend toward a reduced risk of femoral neck fractures in subjects with severe radiographic OA. The failure of the observed increase in BMD to translate into a reduced fracture risk may be due, in part, to the number and type of falls sustained by subjects with OA. Patients with OA should not be considered to be at a lower risk of fracture than the general population. Physicians should be aware that a high BMD in patients with OA may be falsely reassuring.