Socioeconomic deprivation modifies green space and mental health associations: A within person study

Living in an area with good availability and accessibility of residential green spaces such as parks, woodlands, and residential gardens can improve mental health and reduce the global disease burden. Unlike for physical health, it is not well understood if mental health and green space associations might be modified by local area deprivation. Existing evidence for this association comes from cross-sectional studies, widely considered vulnerable to confounding and bias. Individual time-invariant mental health status, personality, and other factors may result in positive effect modification on green space and mental health associations in more deprived areas. We use fixed-effects models that remove time-invariant confounding by calculating differences within-persons to eliminate this bias and add to the existing evidence. We modelled changes in mental health status, green space, and deprivation (relative to the within-person mean) within 54,666 individuals with a combined total of 300,710 mental health scores from one of the world's largest panel surveys: "Understanding Society" in the UK. We found a positive effect of increasing residential green space on mental health and this was positively modified and intensified by area deprivation before and after adjusting for confounding. Our results support providing green space to protect against the negative impact of socioeconomic deprivation on health, particularly for those moving from a less deprived to a more deprived area.