Predicting Overweight and Obesity Among Youth in Ontario, Canada: Evidence from Cross-sectional and Longitudinal Analyses

The percentage of Canadian children and adolescents that are overweight or obese has increased dramatically since the 1980s, with approximately 25% of youth in Canada now characterised as overweight or obese. Canadian youth are engaging in multiple risk-taking behaviours, some of which are associated with an increased risk of overweight and obesity. The school environment has the potential to influence student body mass index (BMI) through implementation of comprehensive healthy school policies, programs, and practices in the domains of physical activity and healthy eating. Guided by a socio-ecological framework, this dissertation research aimed to examine: (1) how modifiable risk behaviours cluster into unique behavioural patterns in youth, and how these behavioural patterns are related to BMI; (2) the effect of engaging in unique clusters of risky behaviours on youths’ BMI trajectories; and, (3) the effect of modifying obesity-related school policies, programs, and practices on youths’ BMI trajectories. Three manuscripts addressed these objectives using linked student- and school-level data from the COMPASS Study (COMPASS). The first manuscript represents a cross-sectional analysis (2012/13), while the second and third manuscripts were longitudinal (2012/13 to 2014/15) in nature. The first manuscript assessed the prevalence and clustering of 15 modifiable risk behaviours using latent class analysis in a sample of 18,587 youth in grades 9 to 12. Four distinct classes emerged: Traditional School Athletes, Inactive Screenagers, Health Conscious, and Moderately Active Substance Users. Youth belonging to the Traditional School Athlete, Inactive Screenager, and Moderately Active Substance User clusters were all significantly more likely to be overweight or obese, compared to the Health Conscious group. The second manuscript examined the effect of engaging in the four clusters of risky behaviours at baseline on youths’ BMI trajectories, using a linked longitudinal sample of 5,084 students in Grades 9 and 10. Using linear mixed effects models, results identified a significant difference in BMIs only at baseline in the four clusters; despite these differences, BMI increased across all clusters annually by the same amount. The third manuscript examined the effect of modifying physical activity and nutrition-related programs, policies, and practices on BMI trajectories using a linked longitudinal sample of 4,951 students in Grades 9 and 10 attending 41 COMPASS schools. Between 2012/13 and 2013/14, 26 of 41 schools implemented distinct new programs or policies, none of which used a Comprehensive School Health Approach. Results indicate that none of these school modifications were associated with improved or reduced BMI trajectories. Findings of this dissertation research have implications for future public health and school-based interventions, and highlight the need for future research in this area, particularly focused on a Comprehensive School Health approach to obesity prevention. COMPASS is uniquely positioned to evaluate similar naturally occurring school-based interventions in a cost-effective and efficient, yet scientifically robust manner by following the same students as they progress through school and are exposed to a changing school environment