Electronic Thesis and Dissertation Repository


Doctor of Philosophy




Gilliland, Jason


A growing body of research suggests that increasing the accessibility to health-related environmental features and increasing exposure to and engagement in outdoor environments leads to positive benefits for the overall health and well-being of children. Additionally, research over the last twenty-five years has documented a decline in the time children spend outdoors. Outdoor activity in children is associated with increased levels of physical fitness, and cognitive well-being. Despite acknowledging this connection, problems occur for researchers when attempting to identify the child’s location and to measure whether a child has made use of an accessible health-related facility, or where, when and for how long a child spends time outdoors.

The purpose of this thesis is to measure children’s accessibility to, exposure to, and engagement with health-promoting features of their environment. The research on the environment-health link aims to meet two objectives: 1) to quantify the magnitude of positional discrepancies and accessibility misclassification that result from using several commonly-used address proxies; and 2) to examine how individual-level, household-level, and neighbourhood-level factors are associated with the quantity of time children spend outdoors. This will be achieved by employing the use of GPS tracking to objectively quantify the time spent outdoors using a novel machine learning algorithm, and by applying a hexagonal grid to extract built environment measures.

This study aims to identify the impact of positional discrepancies when measuring accessibility by examining misclassification of address proxies to several health-related facilities throughout the City of London and Middlesex County, Ontario, Canada. Positional errors are quantified by multiple neighbourhood types. Findings indicate that the shorter the threshold distance used to measure accessibility between subject population and health-related facility, the higher the proportion of misclassified addresses. Using address proxies based on large aggregated units, such as centroids of census tracts or dissemination areas, can result in vast positional discrepancies, and therefore should be avoided in spatial epidemiologic research.

To reduce the misclassification, and positional errors, the use of individual portable passive GPS receivers were employed to objectively track the spatial patterns, and quantify the time spent outdoors of children (aged 7 to 13 years) in London, Ontario across multiple neighbourhood types. On the whole, children spent most of their outdoor time during school hours (recess time) and the non-school time outdoors in areas immediately surrounding their home.

From these findings, policymakers, educators, and parents can support children’s health by making greater efforts to promote outdoor activities for improved health and quality of life in children. This thesis aims to advance our understanding of the environment and health-link and suggests practical steps for more well-informed decision making by combining novel classification and mapping techniques.