Table 2. Bivariable and multivariable logistic regression analysis of the determinants of preeclampsia.

	Bivariable analysis					Multivariable modela
	B	OR	95% C.I.		P	B	OR	95% C.I.		P
			Lower	Upper				Lower	Upper
Delta 25(OH)D ≥30 (nmol/L)	-1.334	0.264	0.113	0.613	0.002	-1.510	0.221	0.084	0.581	0.002
Multifetal gestation	1.497	4.467	1.502	13.281	0.007	2.428	11.332	2.343	54.821	0.003
Obesity T1	1.514	4.544	2.758	7.486	<0.001	1.221	3.391	1.726	6.663	<0.001
Nulliparity	1.308	3.698	2.259	6.053	<0.001	1.432	4.188	2.279	7.697	<0.001
DBP T1	0.086	1.090	1.058	1.122	<0.001	0.089	1.094	1.055	1.134	<0.001
Preexisting medical condition T1	0.959	2.609	1.008	6.748	0.048	1.234	3.435	1.031	11.447	0.045
Excessive GWG	0.887	2.429	1.539	3.834	<0.001	0.517	1.677	0.951	2.958	0.074
Assisted reproduction	0.444	1.559	0.987	2.463	0.057	0.240	1.271	0.642	2.515	0.491
Age ≥40 years T1	-0.546	0.579	0.140	2.401	0.579	0.119	1.126	0.240	5.275	0.880
	Bivariable analysis					Alternative multivariable models
	B	OR	95% C.I.		P	B	OR	95% C.I.		P
			Upper	Lower				Upper	Lower
25(OH)D T1 (nmol/L)	0.004	1.004	0.995	1.014	0.354	0.004	1.004	0.991	1.016	0.571
25(OH)D T3 (nmol/L)	-0.005	0.995	0.988	1.002	0.173	-0.010	0.990	0.981	1.000	0.043
Delta 25(OH)D (nmol/L)	-0.009	0.991	0.982	0.999	0.028	-0.013	0.987	0.977	0.998	0.021

25(OH)D = 25-hydroxyvitamin D, T1 = first trimester, GWG = gestational weight gain, DBP = diastolic blood pressure, T3 = third trimester

Dichotomous: multifetal gestation, age ≥40 years, obesity, nulliparity, excessive GWG, preexisting medical condition, assisted reproduction and delta 25(OH)D ≥30 (4th quartile vs. quartiles 1–3).

^a.Adjusted for 25(OH)D at T1, month of conception, gestational age at sampling, baseline tobacco use, Northern European birth country and employment status at baseline