The Neighbourhood Physical Environment and the 24-Hour Movement Behaviour Composition Among Children
INTRODUCTION: The combination of time spent in physical activity, sedentary behaviour, and sleep within a day is known as the 24-hour movement behaviour composition. Recent research using compositional data analysis (CoDA), a branch of statistical analysis appropriate for compositional data, such as the 24-hour movement behaviour composition, has identified that the 24-hour movement behaviour composition is an important determinant of health. The potential for the physical features of a person’s neighbourhood environment to promote healthy movement behaviours has also been identified. However, research has not yet considered which neighbourhood environment features are associated with all components of the 24-hour movement behaviour composition. The objective of this thesis was to use CoDA to identify neighbourhood physical environment features that are associated with the 24-hour movement behaviour composition among children. METHODS: 458 children aged 10-13 years from Kingston, ON, were studied. Neighbourhood physical environment features were measured in 1 km buffers from the participants’ homes using geographic information system (GIS) software and databases. Twenty-three neighbourhood environment characteristics were measured, and from these, six indices were created, including walkability, traffic safety, dedicated play and activity spaces, non-dedicated play and activity spaces, noise, and artificial light. Moderate-to-vigorous physical activity, light physical activity, sedentary time, and sleep duration were measured over seven days using accelerometers and sleep logs. CoDA was used to establish differences in the movement behaviours between groups defined by the neighbourhood physical environment features. RESULTS: A total of 29 associations between neighbourhood environment features and movement behaviours were examined and only three exposures had statistically significant differences in movement behaviour compositions. All three significant associations were for non-dedicated play and activity spaces. These significant results included subtle differences, did not follow dose-response patterns, and were not always in the expected direction. CONCLUSIONS: Neighbourhood physical environment features were not meaningful correlates of 24-hour movement behaviours.