The “things” we adjust for in greenness epidemiology: relationships between greenness and lifestyle and environmental factors in the Swedish SCAPIS cohort

Background

Residential greenness is linked to health, but its relationships with environmental exposures and lifestyle factors—often treated as confounders or mediators—are less clear.

Objective

We investigated the associations between residential greenness and air pollution, traffic noise and lifestyle factors (smoking, alcohol consumption, leisure-time physical activity) at six study sites in the Swedish CArdioPulmonary bioImage Study (SCAPIS), using waist circumference as an illustrative outcome.

Methods

Greenness assessment was based on the average 5-year pre-recruitment values of Normalized Difference Vegetation Index (NDVI) within 250 m buffers around participants’ residences (n = 29,376; 50–65 years). We used linear regression to estimate associations between NDVI and fine particulate matter (PM2.5), respirable particulate matter (PM10), nitrogen dioxide (NO2) road traffic noise (Lden), and waist circumference; logistic regression to estimate associations between NDVI and smoking, alcohol consumption and sedentary lifestyle.

Results

NDVI (mean 0.47; range 0.08–0.79) varied across SCAPIS sites. Higher greenness was associated with lower air pollution and traffic noise and lower smoking and alcohol consumption, but not with sedentary lifestyle. Waist circumference (mean 89.4 cm in women; 99.7 cm in men) differed by site, but showed no independent association with greenness, after adjustment for urbanicity, site and socioeconomic variables. SCAPIS participants more often lived in urban, less green areas than the general population.

Significance

Greenness relates to environmental and lifestyle factors, partly in site-specific ways, underscoring the need to carefully consider what we adjust for in greenness epidemiology.