The healthcare cost of managing osteoporotic fractures is projected to rise because of the change in population demographics. To reduce the fracture epidemic, strategies are needed to identify those at risk early to allow preventative intervention to be implemented. The aim of this study was to investigate if low-cost community-based assessments, such as calcaneal ultrasound and falls risk assessments, can discriminate a group of elderly women at risk of osteoporotic fracture from those at lower risk. Over the period of 2002-2003, 104 community-dwelling women (mean age 71.3, standard deviation 5.8) were recruited via various modes including advertisements in community newspaper and community centres. These women underwent dual-energy X-ray absorptiometry bone mineral density (BMD) and calcaneal quantitative ultrasound (QUS) measurements; spinal radiography; and performance-based assessment of strength, mobility and balance. The women were classified into a 'high risk' (osteoporotic) group, based on low BMD (T-score of </= -2.5) and/or a history of a fragility fracture, or a 'low risk' (non-osteoporotic) group. Multiple stepwise logistic regression identified decreased speed of sound (SOS), a QUS variable, and poor mobility [slower performance in the timed 'up and go' (TUG) task], as significant discriminators, adjusted for age [SOS: adjusted odds ratio (OR) = 2.96, 95% confidence interval (CI) = 1.64-5.12, P < 0.001; TUG: adjusted OR = 1.69, 95% CI = 1.04-2.73, P = 0.031]. This result suggests that there may be a potential role for primary healthcare practitioners to use calcaneal QUS and functional mobility assessment (TUG) to screen for post-menopausal women at risk of osteoporotic fracture in community settings.