Childhood behavior problems and unintentional injury: a longitudinal, population-based study

Although an association between behavior problems and childhood injuries has been established, the majority of studies have been cross-sectional and comorbidity has not been taken into account. The purpose of this study was to prospectively assess the relationship between behavior problems and the risk of unintentional injury in a population-based sample of Canadian children aged 4 to 11 years.

This prospective cohort study considered data from Cycles 1 and 2 of the Canadian National Longitudinal Survey of Children and Youth. The outcome was injury in Cycle 2 (2 years later). The exposure was the presence of behavior problems in Cycle 1, defined as children with hyperactivity only, aggression only, anxiety only, hyperactivity with aggression, hyperactivity with anxiety, aggression with anxiety, and hyperactivity with aggression and anxiety. All groups were compared to children with no behavior problems. Covariates included child-related, parental, and social-environmental factors.

The final weighted longitudinal sample included 2,209,886 children, of which 11.4% were injured in Cycle 2. None of the behavior groups were at significant risk of injury 2 years later. However, children who lived with a single/no parent and children who did not live with a biological parent had a significantly greater risk of injury. After controlling for confounders, children who lived with a single/no parent had more than twice the risk of having an injury.

We need to look beyond behavior problems, and possibly at family and environmental factors, to reduce the burden of injuries in the Canadian population.