Neighbourhood and home environments and adolescent physical activity behaviours: Longitudinal analyses of the Olympic Regeneration in East London (ORiEL) study.

Context: Regular physical activity has been shown to prevent major non-communicable diseases and improve physical and mental health. However, a quarter of adults and the majority of adolescents do not achieve the recommended level of physical activity in England. In recent years, the home and neighbourhood environments as determinants of physical activity have received attention because of their proximal influence on everyday behaviours, especially in young people. The current literature on adolescents is however limited. This thesis aims to investigate whether features of the home and neighbourhood environments – specifically, perceptions of the neighbourhood environment, ethnic density, neighbourhood trust and social support – predict physical activity and its change over time in an ethnically diverse and deprived adolescent population. Methods: Longitudinal data from the Olympic Regeneration in East London (ORiEL) study (2012-2014) are used. Analyses are conducted on four physical activity outcomes, namely, walking to school, walking for leisure, outdoor physical activity, and pay and play physical activity. Models are estimated using generalised estimating equations, and novel methods for handling missing data with multilevel multiple imputation are applied. Results: Analyses show that adolescents’ perceptions of their neighbourhood environment (including proximity, aesthetics, street connectivity, traffic safety and personal safety) and their changes over time do not consistently predict the forms of physical activity investigated. School-level ethnic density increases the chance of walking to school in some ethnic groups and decreases it in others; whereas walking for leisure and outdoor physical activity are not consistently associated with ethnic density. Adolescents with higher perceived trust in their neighbours have higher chances of reporting outdoor physical activity, and pay and play physical activity. Finally, general social support from family, friends and significant others is shown to predict walking for leisure and its change over time. In boys only, social support from friends predicts outdoor physical activity. Discussion: This thesis advances the field methodologically and empirically by applying novel analytical approaches to important research questions. Results from this thesis contribute to our understanding of the individual, family, peer, community and neighbourhood influences on physical activity in adolescents. The predictors of physical activity identified in this thesis are mostly modifiable and therefore could be the target of policies and interventions that aim to improve physical activity.