To determine if optimal fluoridation of public water supplies influences bone mass and fractures, we studied 2076 non-black women, all aged > or = 65 years recruited into the Study of Osteoporotic Fractures at the Pittsburgh clinic. Information on fluoride exposure was limited to community water supplies. The variable used in the analysis was years of exposure to fluoridated water in community drinking water supplies. Bone mineral density (BMD) was measured at the spine and hip using dual energy X-ray absorptiometry and at the midpoint and ultradistal radius and calcaneus using single photon absorptiometry. Prevalent and incident vertebral fractures were determined by morphometry. Incident nonspine fractures were ascertained every 4 months and confirmed by radiographic report. Exposure to residential fluoridated water had no effect on bone mass. Women exposed to fluoride for > 20 years had similar axial and appendicular bone mass to women not exposed or women exposed for < or = 20 years. There was some suggestion that women exposed to fluoride for > 20 years had a lower relative risk of nonspine fractures (relative risk, RR, = 0.73; 95% confidence interval [CI] 0.48-1.12), osteoporotic fractures, RR = 0.74 (CI 0.46-1.19), and hip fractures, RR = 0.44 (CI 0.10-1.86), compared with women not exposed, but none of these relative risks was statistically significant. There was no association with wrist or spinal fractures. Our results do not support the findings from recent ecological studies which showed an increased risk of hip fracture among individuals exposed to fluoridated public water.