Residential greenness and risk of incident dementia: A prospective study of 375,342 participants.

INTRODUCTION: Incorporation of greenspace may be a novel environmental policy that might result in positive health effects; hence, this study aimed to investigate the association between residential greenness and dementia incidence. The effects of particulate air pollution on mediating dementia were also determined. METHODS: A prospective cohort study involving 375,342 UK biobank participants was conducted, in which Cox regression models were used to determine the association of greenspace exposure with the risks of all-cause dementia (ACD), Alzheimer's disease (AD) and vascular dementia (VD). Sociodemographic variables, lifestyle or dietary characteristics and apolipoprotein E4 status were controlled using two levels of adjusted models. Mediation analyses were performed to determine the mediation effects of PMs. RESULTS: The results indicated that there were 4929 ACD, 2132 AD, and 1184 VD incidents throughout the 8-year study. In the multi-adjusted model, each interquartile increment in greenspace (buffer 300m) conferred the lower risks of ACD (HR = 0.968, 95% confidence intervals [CI]: 0.938-1.000]) and VD (HR = 0.926, 95% CI: 0.867-0.989). The fourth greenspace quartile conferred also reduced risks of ACD (HR = 0.891, 95% CI: 0.804-0.989) and VD (HR = 0.778, 95% CI: 0.630-0.960) in reference to the first quartile. With regard to 1000m catchment, each interquartile increment conferred a 5.0% (95% CI: 1.8-8.1) lower risk of ACD, and the fourth greenspace quartile conferred a 10.9% (95% CI: 0.9-19.8) lower risk of ACD compared to the first quartile. The protective effect of greenness might be mediated based on the reduction of PM2.5 and PM10 (Pindirect effect<0.05). CONCLUSIONS: Increasing greenness reduces the risk of dementia. This study suggests that greenspace is an environmental strategy that helps prevent dementia.