Table 4.

Regression coefficients¹ and odds ratios² (OR) from unadjusted and multivariate linear and logistic regression models of maternal serum log p,p'-DDE and birth weight, gestational age and premature birth

	Birth weight (grams)			Gestational age (weeks)			Prematurity
	Regression	95% C.I.	p	Regression	95% C.I.	p	OR	95% C.I.	p
	coefficient			coefficient
Unadjusted3model
Greenland	-71.4	(-114.7; -28.1)	< 0.01	-0.2	(-0.4; 0.0)	0.02	1.13	(0.87; 1.47)	0.37
Kharkiv	-32.2	(-90.8; 26.4)	0.28	-0.2	(-0.4; 0.0)	0.09	1.43	(0.72; 2.82)	0.31
Warsaw	-131.3	(-231.5; -31.1)	0.01	-0.5	(-0.9; -0.1)	0.02	1.45	(0.73; 2.88)	0.30
Multivariate model3,4
Greenland	-56	(-99.5; -12.5)	0.01	-0.2	(-0.4; 0.0)	0.04	1.07	(0.81; 1.41)	0.64
Kharkiv	-48.9	(-109.7; 11.9)	0.12	-0.2	(-0.4; 0.0)	0.06	1.60	(0.75; 3.44)	0.23
Warsaw	-146.9	(-248.8; -45.0)	0.01	-0.6	(-1; -0.2)	< 0.01	2.44	(0.99; 6.06)	0.05
Multivariate model, with gestational age
Greenland	-39.4	(-79.0; 0.2)	0.05
Kharkiv	-30.5	(-88.7; 27.7)	0.3
Warsaw	-104.3	(-202.5; -6.1)	0.04

¹Per one unit increase of natural log p,p'-DDE

²Per twofold increase of untransformed p,p'-DDE

³Number of subjects in unadjusted/multivariate models: Greenland birth weight 519/519, gestational age 547/547, prematurity 547/547; Kharkiv birth weight 565/565, gestational age 577/577, prematurity 577/577; Warsaw birth weight 186/186, gestational age 198/198, prematurity 198/198

⁴Models for birth weight controlled for: smoking status (Greenland); age of mother (Kharkiv); age of mother, mother's BMI (Warsaw);

Models for gestational age controlled for: passive smoking (Greenland); mother's age (Kharkiv); mother's education (Warsaw);

Models for prematurity controlled for: mother's education level, mother's age, mother's BMI, marital status, passive smoking (Greenland); mother's age (Kharkiv); education level, smoking status, marital status, mother's age (Warsaw);