Table 4:

Covariate-adjusted associations between prenatal arsenic exposure in drinking water and perinatal outcomes in the ECHO-wide Cohort Study, using multivariable mixed models.

	Birth weight Est. (95% CI)	Gestational age Est. (95% CI)	Preterm birth OR (95% CI)	SGA OR (95% CI)	LGA OR (95% CI)
Binary Model
Unexposed	Ref	Ref	Ref	Ref	Ref
Exposed	34.71 (−18.92, 88.34)	−0.02 (−0.20, 0.17)	0.95 (0.65, 1.39)	0.80 (0.54, 1.18)	1.15 (0.91, 1.47)
Count Model
Unexposed	Ref	Ref	Ref	Ref	Ref
1-3 Windows	5.40 (−57.35, 68.15)	−0.09 (−0.31, 0.13)	0.95 (0.60, 1.50)	0.88 (0.53, 1.46)	1.16 (0.86, 1.57)
4 Windows	88.84 (8.22, 169.46)	0.11 (−0.17, 0.39)	0.95 (0.56, 1.63)	0.72 (0.42, 1.23)	1.15 (0.81, 1.63)

Abbreviations: SGA: small for gestational age; LGA: large for gestational age; Est.: estimate; CI: confidence interval; OR: odds ratio.

Note: Models adjusted for maternal age at birth (scaled), year of birth (scaled), infant sex, maternal race, maternal ethnicity, pre-pregnancy BMI, any tobacco or nicotine use, maternal education, gestational diabetes, and gestational hypertension. Estimates in bold are statistically significant.