Associations between residential greenspace exposure and mortality in 4 645 581 adults living in London, UK: a longitudinal study

BackgroundUrban greenspaces could reduce non-communicable disease (NCD) risk. The links between greenspaces and NCD-related mortality remain unclear. We aimed to estimate associations between residential greenspace quantity and access and all-cause mortality, cardiovascular disease mortality, cancer mortality, respiratory mortality, and type 2 diabetes mortality.MethodsWe linked 2011 UK Census data of London-dwelling adults (aged ≥18 years) to data from the UK death registry and the Greenspace Information for Greater London resource. We calculated percentage greenspace area, access point density (access points per km2), and distance in metres to the nearest access point for each respondent's residential neighbourhood (defined as 1000 m street network buffers) for greenspaces overall and by park type using a geographic information system. We estimated associations using Cox proportional hazards models, adjusted for a range of confounders.FindingsData were available for 4 645 581 individuals between March 27, 2011, and Dec 31, 2019. Respondents were followed up for a mean of 8·4 years (SD 1·4). All-cause mortality did not differ with overall greenspace coverage (hazard ratio [HR] 1·0004, 95% CI 0·9996–1·0012), increased with increasing access point density (1·0076, 1·0031–1·0120), and decreased slightly with increasing distance to the nearest access point (HR 0·9993, 0·9987–0·9998). A 1 percentage point (pp) increase in pocket park (areas for rest and recreation under 0·4 hectares) coverage was associated with a decrease in all-cause mortality risk (0·9441, 0·9213–0·9675), and an increase of ten pocket park access points per km2 was associated with a decreased respiratory mortality risk (0·9164, 0·8457–0·9931). Other associations were observed, but the estimated effects were small (eg, all-cause mortality risk for increases of 1 pp in regional park area were 0·9913, 0·9861–0·9966 and increases of ten small open space access points per km2 were 1·0247, 1·0151–1·0344).InterpretationIncreasing the quantity of, and access to, pocket parks might help mitigate mortality risk. More research is needed to elucidate the mechanisms that could explain these associations.FundingHealth Data Research UK (HDRUK).