Children's mobility and environmental exposures in urban landscapes: A cross-sectional study of 10–11 year old Scottish children

Research into how the environment affects health and related behaviour is typically limited in at least two ways: it represents the environment to which people are exposed using fixed areal units, and, it focuses on one or two environmental characteristics only. This study developed a methodology for describing children's mobility and the complexity of their environmental exposure across a 1934 km(2) study area, including urban, suburban and rural zones. It conceptualised and modelled this area as a landscape, comprised of spatially discrete amenities, infrastructure features, differing land covers/use and broader environmental contexts. The model used a 25 m(2) grid system (∼3 million cells). For each cell, there was detailed built-environment information. We joined data for 100 10/11-year-old children who had worn GPS trackers to provide individual-level mobility information for one week during 2015/16 to our model. Using negative binomial regression, we explored which landscape features were associated with a child visiting that space and time spent there. We examined whether relationships between the features across our study area and children's use of the space differed by their sociodemographic characteristics. We found that children often used specific amenities outside their home neighbourhood, even if they were also available close to home. They spent more time in cells containing roads/transportation stops, food/drink retail (Incidence rate ratio (IRR):4.02, 95%CI 2.33 to 6.94), places of worship (IRR:5.98, 95%CI 3.33 to 10.72) and libraries (IRR:7.40, 95%CI 2.13 to 25.68), independently of proximity to home. This has importance for the optimal location of place-based health interventions. If we want to target children, we need to understand that using fixed neighbourhood boundaries may not be the best way to do it. The variations we found in time spent in certain areas by sex and socio-economic position also raise the possibility that interventions which ignore these differences may exacerbate inequalities.