Table 4.

Logistic Regression Models for the Probability of Vitamin D Deficiency

	25-Hydroxyvitamin D
	≤20 ng/mL		≤30 ng/mL
	Coefficient	Odds ratio (95% confidence interval)	Coefficient	Odds ratio (95%confidence interval)
Multivitamin use
User	−1.63a	0.20 (0.09–0.42)a	−0.78b	0.46 (0.26–0.80)b
Nonuser	Referent	Referent	Referent	Referent
Race
Black	1.86a	6.41 (3.16–13.01)a	1.74a	5.68 (2.49–12.97)a
Asian	1.92a	6.83 (2.94–15.85)a	1.19b	3.30 (1.37–7.92)b
White	Referent	Referent	Referent	Referent
Sex
Female	−0.57d	0.57 (0.31–1.05)d	−0.64c	0.53 (0.30–0.94)c
Male	Referent	Referent	Referent	Referent
Age (y)	0.01	1.01 (0.97–1.04)	0.03c	1.03 (1.00–1.06)d
Season
Fall	−0.02	0.98 (0.47–2.04)	0.59d	1.80 (0.93–3.45)d
Winter	0.25	1.29 (0.58–2.87)	0.56d	1.76 (0.88–3.52)d
Spring	0.82d	2.28 (0.96–5.38)d	1.09b	2.97 (1.31–6.72)b
Summer	Referent	Referent	Referent	Referent
Constant	−0.96c	NA	−0.66	NA

Abbreviation: NA = not available.

^aP<0.001.

^bP<0.01.

^cP<0.05.

^dP<0.1.

To convert 25-hydroxyvitamin D to nmol/L, multiply ng/mL by 2.496.