Osteoporosis is a major public health problem in postmenopausal women. Identifying patients at risk of developing an osteoporosis related fracture remains one of the challenges of the next decades. Although it is widely recognized that the gold standard procedure for osteoporosis diagnosis is bone densitometry, economic issues or availability of the densitometers may prevent its use under a mass screening scenario. A risk assessment instrument, the Simple Calculated Osteoporosis Risk Estimation (SCORE), has been developed to be used for selecting women likely to have low bone mass density (BMD) for referring for bone densitometry. Low BMD was defined as 2 standard deviations or more below the mean of young healthy adult. The promises shown by the original validation of SCORE have to be further tested on different populations. A couple of published studies did not reach the same level of satisfaction. OBJECTIVE: The aim of our study is to evaluate this prescreening tool in a large sample of Caucasian women, at different levels of BMD and at various sites of measurement. METHOD: We gathered medical data on 4035 patients aged 45 years or more, either consulting spontaneously or referred for a BMD measurement to an outpatient osteoporosis center located in Liege, Belgium. BMD measures, using DEXA technology, were listed at the hip (both total and neck) and the lumbar spine (L2-L4). RESULTS: More than 47 % of our sample presented with low BMD at, at least one of the sites considered. At the recommended cutoff point of 6, SCORE had a sensitivity of 91.5% to detect a low BMD at any site, a specificity of 26.5%, a positive predictive value of 52.8% and a negative predictive value of 77.7%. For the hip the associated indicators of SCORE ability to identify low BMD reached respectively 97.4%, 21.1%; 20.1% and 97.5%. For the femoral neck the parameters yielded were respectively 95.9%, 23.8%, 34.4% and 93.4%. At the lumbar spine SCORE performed less well. Area under the Receiver Operating Characteristic curve to detect low BMD at the total hip and the femoral neck were 0.760 and 0.738 respectively. CONCLUSIONS: In view of these results we can conclude that, notwithstanding the high values of sensitivity achieved (which are even higher in our experience than those reported in the princeps paper), SCORE specificity is too low to be used as a diagnostic tool for screening patients at high risk to later develop osteoporosis.