Quantitative ultrasound is a promising technique to evaluate fracture risk. To test the hypothesis that reduction of one ultrasound measurement, broadband ultrasound attenuation (BUA), is associated with an increased risk of vertebral fracture in postmenopausal women, we performed a cross-sectional analysis of 442 women aged 55-80 years enrolling in a clinical trial of alendronate. Prior to randomization, we obtained bone mineral density (BMD) measurements of the hip, spine, and whole body, radiographs of lateral lumbar and thoracic spine, and duplicate measurements of calcaneal BUA with a Walker Sonix UBA575. Vertebral fractures were defined by morphometry and semiquantitative reading by a radiologist. After adjustment for potential confounders, the relative risk of vertebral fracture was 1.8 (95% CI 1.4-2.3) for each standard deviation reduction in BUA; for each standard deviation reduction in BMD, the relative risk was 1.7 (95% CI 1.3-2.1) at the femoral neck, and 2.2 (95% CI 1.7-2.9) at the spine. Adjustment for hip, spine, or whole body BMD did not significantly alter the relation between BUA and vertebral fracture. For example, after adjusting for spine BMD, the relative risk of vertebral fracture was 1.5 (95% CI 1.1-2.0) for each standard deviation reduction in BUA. We conclude that low BUA is associated with the presence of vertebral fractures in postmenopausal women. The relation between BUA and vertebral fracture is similar to that observed for BMD and vertebral fractures. Furthermore, the relation between BUA and vertebral fracture persists after adjustment for BMD.(ABSTRACT TRUNCATED AT 250 WORDS)