We have assessed the relationship between bone density, biochemistry a
nd life-style factors in a cross-sectional study of 434 pre- and 357 p
ostmenopausal women. Bone-mineral content (BMC) was measured in the no
n-dominant forearm using single-photon densitometry. Bone mass was cor
rected for bone size (BMC2) in a sub-set of 263 subjects. Correlation
statistics and multiple regression analyses were used to identify phys
ical and biochemical measurements that could be used to best predict B
MC or BMC2. These were combined with lifestyle factors, using multiple
logistic regression analysis, to identify women at risk of low bone d
ensity. Cut-off values were taken as the 5th percentile of the bone de
nsity distributions in pre-menopausal women. In postmenopausal women,
using BMC2 values to define ''risk'', and using age, creatinine, alkal
ine phosphatase, thyroxine treatment, and calcium supplementation as p
redictors, a sensitivity of 73% and a specificity of 95% were achieved
. Such a model may be of use as a ''first pass'' screening test for os
teoporosis.