In the circulation 25-hydroxyvitamin D (25(OH)D) is bound to vitamin D-binding protein (DBP) and albumin. Only a small fraction is in the unbound, free form. According to the ' free-hormone-hypothesis ' only the free form is biologically active. Genetic differences in DBP may affect the binding to 25(OH)D and thereby the amount of free 25(OH)D. In the present study sera were obtained from 265 postmenopausal women with low bone mass density (BMD). Serum 25(OH) D, DBP and albumin were measured and the free and bio-available (free ϩ albumin-bound) 25(OH)D calculated. Based on genotyping of the polymorphisms rs7041 and rs4588, the six common DBP phenotypes were identifi ed and the free and bio-available 25(OH)D calculated according to the corresponding binding coeffi cients. Relations between measures of 25(OH)D and PTH and BMD were evaluated with linear regression adjusted for age and BMI. The calculated amount of free and bio-available 25(OH)D was 0.03% and 13.1%, respectively, of the measured total serum 25(OH)D. Adjusting for DBP phenotype affected the calculated free and bio-available 25(OH)D levels up to 37.5%. All measures of 25(OH)D correlated signifi cantly with PTH, whereas a signifi cant association with BMD was only seen for the free and bio-available 25(OH)D measures. Adjusting for the DBP phenotypes improved the associations. These relations were almost exclusively seen in subjects not using vitamin D and/or calcium supplements. In conclusion, the free and bio-available forms of 25(OH) D may be a more informative measure of vitamin D status than total 25(OH)D. Adjustment for DBP phenotype may improve this further.