Figure 2

Associations between maternal smoking and internalizing behaviors in children at 18 months, 36 months, and 5 years adjusting for Model A, Model B, and Model C covariates. This figure displays the individual B coefficients for associations between maternal smoking during pregnancy and internalizing behaviors at different time points for Models A, B, and C. Associations were discovered at all time points for Model A (unadjusted), and at 18 months and 36 months for Model B (main confounder model). After controlling for smoking in past pregnancies only associations at age 18 months remained significant. The effect sizes decreased as children grew older. Non-smoking mother-child pairs are the reference group. Internalizing behaviors have been log transformed and standardized. B coefficients should be interpreted as units of standard deviation difference in corrected internalizing behaviors between smokers and non-smokers. * Model A: Unadjusted model. † Model B: Main confounder model adjusted for paternal smoking, maternal alcohol consumption, maternal depressive and anxiety symptoms, maternal age, maternal education level, parity, and gestational age at birth. ‡ Model C: Main confounder model in subsample of mothers, Model B plus adjustment for smoking in previous pregnancies. For Model A, ratio of smokers to non-smokers was 4,663/57,456, 3,728/46,495 and 1,167/17,988 for analysis at time points 18 months, 36 months, and 5 years, respectively. For Model B, the ratio of smokers to non-smokers was 3,829/46,986, 3,072/38,268, and 960/14,800 for analysis at time points 18 months, 36 months, and 5 years, respectively. For Model C, the ratio of smokers to non-smokers was 2,717/30,052, 2,157/24,229, and 663/9,620 for analysis at time points 18 months, 36 months, and 5 years, respectively.