Access to and Quality of Neighbourhood Public Open Space and Children's Mental Health Outcomes: Evidence from Population Linked Data across Eight Australian Capital Cities

Abstract

Neighbourhood-level interventions offer a promising opportunity to promote child mental health at a population level; however, neighbourhood effects are still regarded as a 'black box' and a better understanding of the specific design elements, such as public open space, is needed to inform actionable policy interventions.

Methods: This study leveraged data from a population linked dataset (Australian Early Development Census-Built Environment) combining information from a national census of children's developmental outcomes with individualised geospatial data. Associations between access to (within 400 m and 800 m from home), and quality of, public open space and child mental health outcomes across eight capital cities were estimated using multilevel logistic regression models, adjusting for demographic and contextual factors. Access was defined based on proximity of public open space to children's home addresses, within distance thresholds (400 m, 800 m) measured along the road network. Effect modification was tested across maternal education groups.

Results: Across the eight capital cities, inequities in access to child friendly public open spaces were observed across maternal education groups and neighbourhood disadvantage quintiles. Children with access to any type of public open space within 800 m of home had lower odds of demonstrating difficulties and higher odds of competence. Children with access to child friendly public open spaces within 800 m of home had the highest likelihood of demonstrating competence.

Conclusion: Improving access to neighbourhood public open space appears to be a promising strategy for preventing mental health difficulties and promoting competence in early childhood. Action is needed to redress socio-spatial inequities in access to child friendly public open space.

Keywords: built environment; child development; green space; inequities; mental health; public open space; social determinants.

Figure 1

Percentages of children within each maternal education subgroup who have access to public open space within 800 m and 400 m from home along the road network. Children with missing data on maternal education (n = 16,662) not shown. POS: public open space. Child friendly POS defined as having both playground and public toilet nearby; non-child friendly POS defined as those lacking either/both of these child friendly features.

Figure 2

Percentages of children within each maternal education subgroup who have access to public open space (none, non-child friendly, child friendly) within 800 m and 400 m from home along the road network). Children with missing data on neighbourhood disadvantage (n = 829) not shown. POS: public open space. Child friendly POS defined as having both playground and public toilet nearby; non-child friendly POS defined as those lacking either/both of these child friendly features.

Figure 3

Unadjusted (Model A) and adjusted (Models B and C) associations (odds ratios and 95% credible intervals) between public open space exposures and children’s odds of demonstrating externalising difficulties, internalising difficulties, and high competence. Model A: unadjusted associations. Model B: adjusted for child’s sex, language background, and English proficiency, Aboriginal and Torres Strait Islander background, additional health and education needs, and maternal education. Model C: adjusted for child- and family-level demographics (Model B) plus neighbourhood disadvantage. For all models, the reference category for public open space exposures was the no access (none) group.