A Romanian instrument to facilitate bone density measurement indication in postmenopausal women

The aim of the study is to find and valid a clinical instrument which identifies the women that need a bone density measurement because of their high risk for osteoporosis.

A number of 356 women were enrolled in this study. They filled in a study formular and their bone density was measured by DEXA exam (dual energy X-ray absorbtiometry) in the lumbar spine and proximal femur. Statistical correlations between the dependent variable minimal T-score and the other variables (clinical risk factors) were established. The most significant osteoporosis risk factor were included in a classification tree for the dependent variable minimal T-score with 5 splits, 5 independent splitting variables and 6 terminal knots.

The most significant osteoporosis predictors included in our classification tree are: current weight, decreasing of height, sedentary life, alcohol consumption, number of deliveries. In predicting osteoporosis, the classification tree has high values both for sensitivity and for specificity (88.2%, respectively 74.11%).The osteoporosis prevalence is very high (74.34%) within the group predicted as having osteoporosis (53.7% of all women). The osteoporosis prevalence was 11% within the group predicted as having osteopenia.

The classification tree facilitates a bone density measurement strategy and establishes the criteria for initiation of the treatment in high risk osteoporosis cases. It allows postponing the DEXA-exam for low risk women and selective measurement of the bone density for intermediar osteoporosis risk women.