4. Discussion
TBS has emerged as an independent predictor of fragility fracture, with the magnitude of association being equivalent to that of bone mineral density. However, etiological determinants of TBS have not been well documented. In this family based study, we have demonstrated that approximately 45% of the between-individual variance in TBS was determined by genetic factors, after adjusting for age, gender and height. Moreover, the about one third of the covariation between TBS and lumbar spine BMD was also genetically determined. To our knowledge, this is the first demonstration of the heritability of TBS, and it is thus difficult to put the present study's finding within context. However, the index of heritability of LSBMD in this study is lower than previous twin studies'. Classical twin sties [12,13] have typically shown that the index of heritability for lumbar spine BMD ranged between 75% and 92%. In this study, we found that the heritability was much lower (51%). This difference is somewhat expected, because it is well known that twin studies tend to over-estimate the effect of genetic factors of a trait [14]. In the classical twin model, the shared environmental factors are assumed to be equal between monozygotic and dizygotic twins, and as a result, the heritability is often higher than in siblings based studies.