The effect of prenatal maternal cigarette smoking on children's BMI z-score with SGA as a mediator.

BACKGROUND: The goal of this study was to assess the effect of prenatal maternal cigarette smoking on children's BMI z-score trajectories, and to evaluate whether small-for-gestational-age (SGA) acts as a potential mediator between prenatal maternal cigarette smoking and child's BMI z-score at 4 years of age.

METHODS: Group-based trajectory modeling (GBTM) methods were employed to describe and classify developmental BMI z-score trajectories (the outcome of interest) in children from 9 months to 4 years of age (n = 5221) in the Early Childhood Longitudinal Study, Birth Cohort (ECLS-B) study (2001-2005). Further analysis examined whether the identified BMI z-score trajectories varied with the exposure, prenatal maternal cigarette smoking. Mediation analyses were utilized to examine whether being SGA (binary measure) acted as a potential mediator in the relationship between prenatal maternal cigarette smoking and BMI z-score among 4-year-old children.

RESULTS: Using GBTM, two BMI z-score trajectory groups were identified: normal BMI z-score (57.8%); and high BMI z-score (42.2%). Children of mothers who smoked cigarettes during pregnancy were 2.1 times (RR 95% CI: 1.1-4.0, P value = 0.023) more at risk of being in the high BMI z-score trajectory group. Prenatal cigarette smoking was positively related to SGA at birth, but SGA was inversely related to BMI z-score at 4 years. The direct effect (0.19, 95% CI: 0.18, 0.19; P value < 0.001) of maternal cigarette smoking status during pregnancy on BMI z-score among 4-year-old children was stronger and in the opposite direction of the indirect effect (-0.04, 95% CI: -0.04, -0.04; P value < 0.001) mediated through SGA.

CONCLUSIONS: In this study, prenatal maternal cigarette smoking was positively associated with the high BMI z-score group, as well with SGA. The effects of prenatal smoking on BMI z-score at 4 years appears to act through pathways other than SGA.