ASSESSMENT OF THE RISK OF POSTMENOPAUSAL OSTEOPOROSIS USING CLINICAL FACTORS

OBJECTIVE We wished to assess the predictive value of the main clinical risk factors for osteoporosis over a low vertebral bone mineral density. DESIGN A cross-sectional study was made of a cohort of peri and post-menopausal women (mean age, 54 years). PATIENTS One thousand, five hundred and sixty-five normal white women were selected from among the women referred to our menopause clinic for screening and prevention of osteoporosis. MEASUREMENTS Each woman had replied to a detailed standardized questionnaire including the main clinical risk factors and had her bone density measured using dual photon absorptiometry. RESULTS The predictive value for a low vertebral bone mineral density (2 SD below the normal young adult value) was assessed for 15 historical and anthropometric variables. Among these, age, age at menarche, weight, height, menopause and its duration, were independent predictors of a low bone mineral density, in a multiple logistic regression analysis. Odds ratios were calculated for each of these variables, weight, menopause and its duration being the three most influential variables. At best this model makes it possible to correctly classify 73% of women with a low bone mineral density and 66% of those with a normal bone mineral density. If this model is used for screening, it could possibly save 25% of bone densitometry examinations. CONCLUSIONS Direct bone densitometry remains indispensable to assess osteoporosis risk, since risk factors alone are not sufficient for accurate delineation of either low or normal bone mineral density.