Effects of Breastfeeding and Maternal Smoking During Pregnancy On Body Mass in dex (Bmi) Trajectories Among Children in Canada

Abstract

Background: Research has suggested that breastfeeding and maternal smoking, among other factors, can impact child overweight and obesity. Breastfeeding has been identified as having a protective effect. Smoking during pregnancy has been associated with higher odds of childhood overweight. Objectives: To examine how smoking during pregnancy and breastfeeding, two key modifiable factors associated with maternal behaviour, affect body mass index (BMI) trajectories among young children between the ages of two and 11. Methods: We conducted secondary data analysis of the National Longitudinal Survey of Children and Youth data collected from 1994 to 2002. Analysis was based on a sample of over 3,500 two-year old children who were followed until the age of 11. Multi-cohort latent growth modeling techniques for data with individually varying times of observations were employed. To assess an independent effect of smoking and breastfeeding on BMI trajectories, the following child- and family-level covariates were included: child's weight and health status at birth, pregnancy length, maternal diabetes, as well as mother's age and socio-economic status. We also examined gender and ethnic differences in the BMI trajectories by comparing Aboriginal and non-Aboriginal children, by gender. Results: Overall, between the ages of two and 11, the estimated BMI trajectory followed the expected U-shaped pattern. BMI trajectories varied significantly across children in terms of the baseline scores and the linear growth parameter. In the unadjusted models, smoking was associated with higher child BMI score at baseline, while children who were breastfed had, on average, lower BMI score at baseline. Neither of the two risk factors had a statistically significant effect on growth parameters associated with the BMI change between the ages of two and 11. Adjusting for the effects of background variables, the association between breastfeeding and BMI at age two was still statistically significant. The effect of smoking, however, became non-significant. Finally, there were some gender differences and differences between Aboriginal and non-Aboriginal children both in terms of the shape of BMI trajectories and effects of background variables on these trajectories. Conclusion: Both maternal smoking during pregnancy and breastfeeding were significant predictors of childhood BMI scores. After controlling for a number of covariates, only breastfeeding had a significant effect on BMI growth trajectories for children between the ages of two and 11. These findings suggest that the promotion of breastfeeding could be an important obesity prevention strategy for all children, regardless of their background.