Introduction Childhood obesity is a major public health concern in Canada as nearly 17% of children between 2 and 11 are overweight and more than 7% are obese. The objective of this study is to examine whether neighbourhood-level predictors affect BMI trajectories among young children.
Methods We conducted a secondary data analysis of the National Longitudinal Survey of Children and Youth. A cohort of over 6000 2- and 3-year-old children were followed between 1994 and 2004 in the sequence of bi-annual interviews. Multi-cohort latent growth curve modelling techniques for hierarchical data were employed to assess an independent effect of neighbourhood characteristics on BMI trajectories, after controlling for a number of child- and family-level covariates. Neighbourhood conditions were assessed by indicators related to the physical environment (built and physical) in which the child lives, as well as factors related to socio-economic status of its inhabitants.
Results Overall, between the ages of 2 and 12, the estimated BMI trajectory followed the expected U-shaped pattern. The parameter estimates of this trajectory varied significantly, both across-children and across-neighbourhoods. In the unadjusted model, the between-neighbourhood variance constituted approximately 20% of the total variance in these estimates. The results from the final model suggest that a statistically significant portion of the between-neighbourhood variance was accounted by the proposed neighbourhood-level predictors.
Conclusion Neighbourhood-level predictors were identified as significant predictors of the variance in BMI trajectories, suggesting that the neighbourhood characteristics play an important role in shaping BMI trajectories among young children in Canada.