Which reference values should we use when measuring calcaneus with ultrasound? - The objective of this study was to compare Quantitative Ultrasound Index t-scores (QUI t-scores) obtained from two different reference data (European referenca data and US caucasian female reference data) in order to determine differences and similarites of their ability to detect age-related changes and their correlations to dual energy X-ray absorptiometry (DXA) bone mineral density (BMD) of the spine and the hip. 166 women whose BMD of the spine and the femoral neck was measured by DXA had their calmal, osteopenic and osteoporotic. They were stratifield to four age bands (<50, 50-69, 60-69, &>70) to study age-related changes. pearson correlation coefficients were calculated to eşamine the relationship between two different QUI t-scores and the relationship between these different QUI t-scores at various age-bands -were compared by ANOVa.Turkey HSD multiple comparison test was used to compare t-zscore differences between groups. Kappa scores were calculated to examine the aggrement betxeen t-scores two different QUI t-scores abtoanide badet on different reference data (r= 0.995). Mean t-scores based on US caucasian female reference data xas 0.7+1.6 and 0.9+1.1 based on European reference data. The difference between themeans of the t-scores was 0.17 (95 % Cl: 0.09, 0.25) and this difference was statisticallysiginificant (P<0.001). T-scores based on two different reference data were the highest in younger age groups and the lowest in the elderly. Difference between different t-scores at different age groups was found statistically significant. T-scores based on both reference data were the highest in normal subjects and the lowest in those with osteoporosis, the difference between two different t-scores being significant in the normal and the osteoporotics. The corelations between t-scores based on European reference data and lumbar spine (r=0.58) and the femoral neck BMD (r=0.57) were greater than those between t-scores base on US caucazsian female reference data and BMD (r= 0.55 and r)0.51). Although the agrement between t-scores in risk determination was sinnificant (Kappa = 0.796), it showed discrepencies in some subjects. These results indicate that t-scores that are calculated based on different populations display a somewhat different trend both in discriminating age-related changes in acustic parameters of the calcaneus and in determining BMD changes. In conclusion, this investigation reveals the iportance of the establishment of our normative data concerning ultrasound measurements of the calcaneus.