It has been suggested that quantitative ultrasound (QUS) could be used as a selective population pre-screen, to maximise the cost effectiveness of referral for dual energy X-ray absorptiometry (DXA) assessment of bone mineral density (BMD). We set out to examine how such an approach might perform in the assessment of women who were referred by general practitioners for DXA via the open access service in Cardiff. In 115 women aged 40-80 (mean 69) years we used DXA to measure BMD at lumbar spine and hip, and QUS to measure broadband ultrasound attenuation (BUA) in the heel. A bottom-up approach was used to estimate the costs of DXA and QUS. We examined the cost effectiveness of using QUS as a pre-screen, only referring subjects for the more expensive DXA assessment if BUA were less than a pre-determined threshold. The unit costs of pencil-beam DXA and QUS were approximately 44 UK pounds and 16 UK pounds respectively. We identified a BUA threshold of 60 dB/MHz as the most cost effective, and calculated a sensitivity of 81% and specificity of 89% in identifying those subjects whom DXA assessment subsequently identified as having osteoporosis. At the BUA threshold of 60 dB/MHz, pre-screening saved 969 UK pounds at the expense of missing ten women with osteoporosis as diagnosed by DXA. Therefore the cost per additional woman with osteoporosis identified using DXA alone was only 97 UK pounds. QUS assessment does not appear to have a significant cost effective benefit as a pre-screen for DXA in the studied population. A QUS pre-screen would be cost effective only if this investigation could be performed at a substantially lower cost.