Individual, family and neighbourhood contributions to child excess weight and screen time among two australian cohorts

Abstract

The complexity of child overweight and obesity has been discussed in a substantial amount of literature. Yet few attempts have been made to disentangle the complexity. It is hypothesised that child weight status is influenced by multilevel factors operating within a highly complex set of social, cultural and economic conditions. Opportunities for activity or exercise and material resources, among other things, further constrain behaviour. These factors are important translators and mediators of behaviour, and thus can adversely influence health behaviours and weight status. This study sought to determine (1) what role proximal (individual and family) and more distal (neighbourhood) characteristics play in contributing to excess weight and screen time, and (2) whether screen time is a mediator of proximal and distal contributions to overweight and obesity among children in Australia. A multilevel approach examining the proximal and distal contributions using logistic regression and binary mediation statistical techniques was adopted to assess the research aims, using cross-sectional data from the Longitudinal Study of Australian Children (LSAC). LSAC is a nationally representative omnibus survey. Data from Wave 3 (conducted in 2008) was used to explore the contributions for preschoolers (4-5 year olds) and primary schoolers (8-9 year olds). More than one-fifth of preschoolers and primary schoolers were overweight or obese, while most preschoolers and primary schoolers engaged in more than the recommended threshold of two hours of screen time per day. Binary mediation analyses showed screen time to significantly mediate proximal and distal contributions to excess weight for both cohorts. Of the individual, family and neighbourhood contributions to excess weight among children aged 4-5 and 8-9 years, maternal protective parenting and excess weight (respectively) made the most substantial contributions to the likelihood of child overweight and obesity. Conversely, child preferences for active pastimes were most protective. The major contributors differed to reflect more behaviour-oriented influences when assessed in the case of screen time. Maternal excess weight featured for the younger children, while intake of energy-dense food and drinks was most influential among the older children in promoting the likelihood of excess screen time. Adequate fruit and vegetable intake was shown to be associated with reducing the likelihood of risky screen-time behaviours in children of both age groups. These relationships were confirmed when screen time was assessed as a mediator in models by which characteristics contribute to excess weight either directly or indirectly. Unexpectedly few of the neighbourhood variables were found to be significant in the analyses. Rather, individual and family level factors were mostly found to make contributions to overweight and obesity and excess screen time among preschoolers and primary schoolers. This is not to say that the hypothesised contributions of neighbourhood level characteristics should not be included in intervention. The findings presented in this thesis show that child excess weight is best understood within the environments from which it arises. Effective public health intervention strategies aimed at preventing and ameliorating child overweight and obesity in Australia would thus be best placed adopting a holistic multidimensional approach.