Table 2.

Odds ratio for the risk of hip fracture

Vitamin D*	Unadjusted	MV-adj†
Number of missing pairs (total =400)	18**	y18**
Per 2.5 nM (1 ng/ml) decrease‡	1.03 (1.01, 1.05)	1.03 (1.01, 1.05)
Per 25 nM (10 ng/ml) decrease	1.30 (1.07, 1.58)	1.33 (1.06, 1.68)
p for linear trend	p = 0.009	p = 0.015
Quartiles (according to control group)
1st Quartile (<47.6 nM)	1.73 (1.13, 2.66)	1.71 (1.05, 2.79)
2nd Quartile (47.6− <60.2 nM)	1.08 (0.72, 1.63)	1.09 (0.70, 1.71)
3rd Quartile (60.2−<70.7 nM)	0.78 (0.50, 1.20)	0.82 (0.51, 1.31)
4th Quartile (70.7+ nM) (ref)	1 (ref)	1 (ref)

^*Hip fractures case and controls selection matched on age, ethnicity and blood draw date.

^**Because of deletion of cases and controls due to missing values, 18 case-control pairs were missing from our multivariable models. We also excluded these pairs from our unadjusted models to ensure the same analytic sample.

^†Multivariate adjustment includes age, body mass index, parental history of hip fracture, history of fracture, smoking, alcohol use, and total calcium intake, oral corticosteroid use and geographic region.

^‡To convert values for 25-hydroxyvitamin D to ng/ml divide by 2.5.