Prenatal environmental tobacco smoke exposure and early childhood body mass index

Abstract

Maternal smoking during pregnancy is associated with increased risk of childhood overweight body mass index (BMI). Less is known about the association between prenatal secondhand tobacco smoke (SHS) exposure and childhood BMI. We followed 292 mother-child dyads from early pregnancy to 3 years of age. Prenatal tobacco smoke exposure during pregnancy was quantified using self-report and serum cotinine biomarkers. We used linear mixed models to estimate the association between tobacco smoke exposure and BMI at birth, 4 weeks, and 1, 2 and 3 years. During pregnancy, 15% of women reported SHS exposure and 12% reported active smoking, but 51% of women had cotinine levels consistent with SHS exposure and 10% had cotinine concentrations indicative of active smoking. After adjustment for confounders, children born to active smokers (self-report or serum cotinine) had higher BMI at 2 and 3 years of age, compared with unexposed children. Children born to women with prenatal serum cotinine concentrations indicative of SHS exposure had higher BMI at 2 (mean difference [MD] 0.3 [95% confidence interval -0.1, 0.7]) and 3 (MD 0.4 [0, 0.8]) years compared with unexposed children. Using self-reported prenatal exposure resulted in non-differential exposure misclassification of SHS exposures that attenuated the association between SHS exposure and BMI compared with serum cotinine concentrations. These findings suggest active and secondhand prenatal tobacco smoke exposure may be related to an important public health problem in childhood and later life. In addition, accurate quantification of prenatal secondhand tobacco smoke exposures is essential to obtaining valid estimates.

Figure 1

Mean weight and height z-score from birth to 3 years of age by prenatal serum cotinine category.* † *-Adjusted for child age (in years), child age squared, child age cubed, maternal age (<25, 25–34, and >34 years), education (<12, 12, and >12 years), race (non-Hispanic white, non-Hispanic black, and other), marital status (married and non-married), depression at baseline home visit (moderately or severely depressed and mildly or minimally depressed), breast feeding duration since last visit (years), parity (0, 1, and >1), income ($10,000 increments), and maternal BMI at 16 weeks gestation. †-Bars represent 95% confidence intervals