Table 2.

Rates of change in bone mineral density and bone turnover markers

	Symbol	Estimated parameters ± standard error (SE) for
		Femoral neck BMD	Lumbar spine BMD	sPINP	sICTP
Parameters
Baselinea	α	943.0 ± 13**	1288.0 ± 20.3**	40. 1 ± 1.3*	4.3 ± 0.12**
Rate of changeb	β	-3.5 ± 1.0**	7.4 ± 1.3**	-0.4 ± 0.18*	0.08 ± 0.02*
Variance components
Baseline	${\sigma }_{\alpha }^2$	16650 ± 2453**	38968 ± 5883**	130.8 ± 25*	0.81 ± 0.22**
Rate of change	${\sigma }_{\beta }^2$	44 ± 6*	2 ± 30	0.4 ± 0.51	0.01 ± 0.05
Covariance	${\sigma }_{\alpha \beta }^2$	43 ± 135	876 ± 283**	-1.1 ± 2.7	0.07 ± 0.03**
Random error	${\sigma }_e^2$	908 ± 131**	3234 ± 327**	52.6 ± 6.8**	0.84 ± 0.08**

sICTP, serum C-terminal telopeptide of type I collagen; sPINP, serum N-terminal propeptide of type I collagen; BMD bone mineral density.

Units of measurement: ^abaseline parameters were expressed in mg/cm² (for bone mineral density) and μg/L (for sPINP and sICTP); ^brate of change: mg/cm² per year (for bone mineral density) and μg/L per year (for sPINP and sICTP).

Note of interpretation: For femoral neck BMD, the overall average baseline measurement was 943 mg/cm², and the rate of change was -3.5 mg/cm² per year. The standard deviation (SD) of baseline BMD among individuals was $\sqrt{16650}$ = 129 mg/cm², and the SD of rates of change among individuals was $\sqrt{43}$ = 6.6 mg/cm². More importantly, the variance in BMD within individuals was 908 mg²/cm⁴. The covariance was not statistically significant, which suggests that there was no significant correlation between the baseline measurements and rates of change in BMD.

Statistical significance is indicated by *p < 0.05; ** 0.001 < p < 0.05, and *** p < 0.0001.