Skip to main content
. 2017 Dec 17;2017:9032141. doi: 10.1155/2017/9032141

Table 2.

Results of the univariate and multivariate logistic regression models for the association of subject characteristics with vitamin D hypovitaminosis with the cutoff value of 17 ng/mL.

Characteristic p value Modalities for OR calculation OR (95% CI)
Univariate logistic regression
Age 0.41 Continuous 0.99 (0.96–1.02)
Duration of menopause 0.46 Continuous 0.99 (0.97–1.02)
Parity 0.75 Continuous 1.01 (0.94–1.09)
BMI 0.05 <30 kg/m2 versus ≥30 kg/m2 0.61 (0.36–1.01)
Habitat/housing 0.32 Urban versus rural 1.31 (0.76–2.26)
Education level 0.81 Education versus no education 0.82 (0.50–1.33)
Skin phototype 0.82 Dark versus light 1.13 (0.69–1.84)
Veiling 0.09 Yes or no 2.56 (0.84–7.84)
T-score spine 0.95 Continuous 0.99 (0.83–1.18)
DMO spine 0.93 Abnormal versus normal 0.98 (0.58–1.65)
T-score hip 0.19 Continuous 1.12 (0.93–1.36)
DMO hip 0.67 Abnormal versus normal 0.80 (0.49–1.31)
Multivariate logistic regression
BMI 0.02 <30 kg/m2 versus ≥30 kg/m2 1.826 (1.081–3.083)
Veiling 0.0353 Yes or no 3.526 (1.090–11.400)
Season <0.0001 Summer versus other pooled seasons 4.159 (2.456–7.043)

p value was calculated for each variable according to the Mann–Whitney test or chi-square test.