Do green and blue spaces in the residential neighbourhood have an effect on multimorbidity? A comparative, observational study of 48,589 UK Biobank participants

• Few studies have comparatively assessed the impact of different urban green and blue spaces on multimorbidity and chronic health. • Urban inland blue spaces reduced the odds of having complex multimorbidity. • Total amount of greenery increased the odds of having multimorbidity. • Parks, street trees, gardens and proximity to coast had no effect on multimorbidity. • It is not exposure to green but exposure to blue space that reduces the risk of multimorbidity in urban-dwelling middle-aged and older adults. Availability of green and blue space in the urban residential neighbourhood can reduce the risk of poor mental and physical health, however, little is still known about different types of urban green and blue spaces and their differential impact on individuals that have multiple physical and mental chronic health conditions. We conducted a cross-sectional study of 48,589 UK Biobank participants to analyse the relationship between exposure to seven types of urban green and blue spaces (parks, street trees, domestic gardens, total green space, inland blue space, proximity to coast and total green and blue space) with five multimorbidity outcomes: simple (2 long-term health conditions (LTCs)), complex (3 LTCs or 4+LTCs), cardio-metabolic, respiratory, and mental multimorbidity. Amount (% in 1500 m circular buffers) and proximity (Euclidean distance) of green and blue spaces in the residential neighbourhood were computed individually for each UK Biobank participant using remote sense data from European Urban Atlas. Analyses were adjusted for the Bonferroni correction for multiple testing to reduce the risk of false positive results. Sensitivity analyses were conducted by buffer size (300 m and 3000 m). Individuals that have a higher proportion of inland blue spaces in their residential neighbourhood had lower odds of multimorbidity. For every percent increase in the amount of inland blue space in 3000 m buffer, the odds of having complex multimorbidity (3LTCs) decreased by 3 % (OR:0·97; 95 % CI:0·95–0·98; p-value:0·0002) after applying the Bonferroni correction. In contrast, individuals with a higher amount of total green space within a 1500 m buffer had a higher risk of having 4+ LTCs (OR:1·01; 95 % CI:1·00–1·01; p-value:0.00005). Other types of green and blue spaces had no effect on our outcomes. Urban inland blue spaces, such as rivers and canals, are integral parts of life in UK but they have often been overlooked in observational health research. Future policy should aim to incorporate blue spaces in the design of urban regeneration and public health interventions.