Skip to main content
. 2016 Jul 20;16:102. doi: 10.1186/s12887-016-0647-1

Table 2.

Multilevel regression models for variables that contribute to the change in WB_BMC, WB_BMD, FN_BMD and LS_BMD

Variables WB_BMC WB_BMD FN_BMD LS_BMD
Fixed effect Estimates ± SE P value Estimates ± SE P value Estimates ± SE P value Estimates ± SE P value
Intercept −1669.01 ± 140.83 <0.001 0.4690 ± 0.0267 <0.001 0.3354 ± 0.0508 <0.001 0.0754 ± 0.058 0.198
Osteocalcin −0.6044 ± 0.3273 0.069 −0.0003 ± 0.0001 <0.001 0.00004 ± 0.0001 0.743 −0.0003 ± 0.0001 0.021
CTX 7.6320 ± 32.3889 0.814 0.0014 ± 0.0063 0.827 −0.0307 ± 0.0118 0.011 −0.0155 ± 0.0125 0.219
Leptin −9.9625 ± 2.5970 0.001 −0.0019 ± 0.0005 0.001 −0.0031 ± 0.0010 0.002 −0.0037 ± 0.0010 0.001
Adiponectin 2.7653 ± 1.9866 0.168 0.0006 ± 0.0004 0.138 0.0010 ± 0.0007 0.173 0.0004 ± 0.0008 0.587

SE standard error, BMI body mass index, WB whole body, LS lumbar spine, FN femoral neck, BMD bone mineral density, BMC bone mineral content, CTX C-terminal telopeptide of type I collagen, MVPA moderate-to-vigorous physical activity, Model was controlled by age, BMI and Tanner stage