CONCLUSIONS This longitudinal twin study documented that genetic factors explain 44-56% of the between-individual variance in bone loss at femoral neck, lumbar spine, and forearm in postmenopausal Caucasian women, providing a rationale for identifying the specific genes involved. BACKGROUND Although there is a significant genetic effect on peak BMD, until recently, no substantive studies on heritability of bone loss in human were available. The aim of the study was to estimate the heritability of the bone loss at multiple sites in postmenopausal Caucasian women. METHODS Postmenopausal female monozygotic (MZ) and dizygotic (DZ) twins aged 40 or above at baseline were selected from the TwinsUK registry and followed up for an average of 8 years (range 5-14 years). All twins were noncurrent hormone replacement therapy users and not on any osteoporosis treatment. They had dual-energy X-ray absorptiometry (DXA) scans of their hip, lumbar spine, and forearm several times (range 2-9) during the follow-up period. Individual bone losses at femoral neck, lumbar spine, and forearm were estimated by linear regression modeling. Structural equation modeling was utilized to estimate the heritability of the bone loss. RESULTS A total of 712 postmenopausal Caucasian female twins (152 MZ and 204 DZ pairs) were included. MZ twins were older and had slightly lower BMD at all sites than DZ twins. DZ twins had slightly higher bone loss at lumbar spine, but similar at femoral neck and forearm compared to MZ twins. Intraclass correlation coefficients (ICC) for the bone loss at all sites were significantly higher in MZ than DZ twin pairs (p = 0.0045, 0.0003, and 0.0007 for femoral neck, lumbar spine, and forearm, respectively), indicating a significant genetic influence on bone loss at these sites. After adjustment for age at baseline and weight change during the follow-up, the heritability estimate was 47% (95% CI 27-63%) for bone loss at femoral neck, 44% (95% CI 27-58%) for lumbar spine, and 56% (95% CI 44-65%) for forearm. CONCLUSIONS Our data suggest that up to 56% of the between-individual variance in bone loss is due to genes, providing a rationale to identify specific genetic factors for bone loss.