To assess the utility of quantitative ultrasound (QUS) of the heel for osteoporosis screening, we studied a group of 170 early postmenopausal women using both QUS of the heel and dual-energy X-ray absorptiometry (DXA) at the spine, hip, forearm, and whole body. On the basis of the linear regression results between QUS and DXA, a 95% bone mineral density (BMD) estimate confidence range was defined. Correlation coefficients between the QUS measurements and DXA ranged from 0.26 to 0.63. The confidence ranges for the estimated BMD based on a QUS measurement of the heel were large, such that an estimation of skeletal BMD at any of the DXA sites measured was not possible. For example, an estimate of the normative anteroposterior spine BMD (i.e. the T-score or the Z-score) based on a calcaneal ultrasound reading would have an error of +/- 1.9 standard deviations. Results for predicting the normative BMD of the other DXA regions were similar, with expected errors ranging from +/- 1.4 to +/- 2.0 standard deviations. We therefore conclude that QUS is not suited for the screening of early postmenopausal women for low axial or peripheral BMD. However, QUS may have a role as an independent predictor of fracture by measuring skeletal properties in addition to bone density.