Table 4.

Multivariable logistic regression analysis of bone mineral density

	Men			Premenopausal women			Postmenopausal women
	Beta	SE	P value	Beta	SE	P value	Beta	SE	P value
Lumbar spine
Model 1	−0.2899	0.2570	0.2604	−0.1820	0.1338	0.1753	−0.0778	0.1560	0.6186
Model 2	−0.0194	0.2472	0.9375	−0.1244	0.1320	0.3471	0.1261	0.1449	0.3847
Model 3	0.0777	0.2517	0.7578	−0.1400	0.1365	0.3063	0.1055	0.1482	0.4772
Total femur
Model 1	−1.1271	0.3862	0.0038	0.0260	0.1426	0.8553	−0.2160	0.1653	0.1924
Model 2	−1.0141	0.3913	0.0101	0.0034	0.1454	0.8118	−0.0549	0.1586	0.7296
Model 3	−0.8842	0.3892	0.0240	0.0266	0.1511	0.8609	−0.0038	0.1690	0.8201
Femoral neck
Model 1	−0.8639	0.4739	0.0695	0.1211	0.1328	0.3625	−0.0713	0.1646	0.6650
Model 2	−0.6742	0.4917	0.1716	0.1004	0.1335	0.4528	0.0031	0.1565	0.8409
Model 3	−0.5278	0.4737	0.2663	0.0847	0.1392	0.5433	−0.0006	0.1648	0.9969

Model 1 is adjusted for age and body mass index.

Model 2 is adjusted for age, body mass index, smoking status, ethanol intake, regular exercise, educational level and income.

Model 3 is adjusted for age, body mass index, smoking status, ethanol intake, regular exercise, educational level, income, 25-hydroxyvitamin D, stress level and tinnitus.

Results are presented as an estimated beta with corresponding SE and P value.