Long-term exposure to green and blue space and incidence of cardiovascular disease: A Danish Nurse Cohort study

BACKGROUND: Few cohort studies have examined the associations of natural surroundings (green and blue space) with cardiovascular disease (CVD) and show mixed findings. We aimed to investigate the associations between long-term exposure to green and blue space and incidence of CVD in the Danish Nurse Cohort. METHODS: We followed 19,070 female nurses living in Denmark from 1993/1999 to 2018. The shortest Euclidian distance from a residence to three types of green space (park, woodland, and heathland) and four types of blue space (lake, river, coast, and wetland), along with total count of all types of natural surroundings within a 500 meter (m), 1000 m, and 5000 m radius from a residence, were calculated using GeoDanmark data from 2005. Distance was log-transformed to correct for a right-skewed data distribution. Residential air pollution and road traffic noise data for 2005 were estimated by the Danish air pollution modeling system ('DEHM/UBM/AirGIS') and Nord2000 model, respectively. Cox proportional hazards models were used to estimate associations between green and blue space and the first-ever incidence of total CVD and certain CVD subtypes after adjusting for individual factors, air pollution, and noise. We examined effect modification by age, smoking status, occupational status, household income, and urbanicity level. Hazard ratio (HR) and 95% confidence interval (95% CI) were presented per e unit (equivalent to ∼2.72 fold) increase in distance. RESULTS: 8179 new cases of CVD were observed over 344,084 person-years. Living further from woodland was associated with higher stroke incidence (HR: 1.153; 95% CI: 1.029-1.293), while living further from heathland was associated with reduced total CVD incidence (HR: 0.975; 95% CI: 0.955-0.996). No associations were found between distance to park, woodland, lake, river, coast, or wetland and total CVD incidence. Total count of all types of green and blue spaces within a 5000 m radius was linked to a reduced risk of CVD incidence. Adjusting for air pollution or road traffic noise did not alter observed associations. Younger individuals (<50 years old) were observed to have lower HRs when living closer to all types of natural surroundings. CONCLUSION: We found that proximity to woodland was associated with a reduced risk of stroke, whilst no significant or even inverse associations were observed between proximity to other types of natural surroundings and CVD incidence. Total count of all types of natural surroundings within a 5000 m radius was negatively associated with CVD incidence, suggesting cumulative benefits of these areas.