Residential green space, air pollution, socioeconomic deprivation and cardiovascular medication sales in Belgium: A nationwide ecological study.

Green space may improve cardiovascular (CV) health, for example by promoting physical activity and by reducing air pollution, noise and heat. Socioeconomic and environmental factors may modify the health effects of green space. We examined the association between residential green space and reimbursed CV medication sales in Belgium between 2006 and 2014, adjusting for socioeconomic deprivation and air pollution. We analyzed data for 11,575 census tracts using structural equation models for the entire country and for the administrative regions. Latent variables for green space, air pollution and socioeconomic deprivation were used as predictors of CV medication sales and were estimated from the number of patches of forest, census tract relative forest cover and relative forest cover within a 600 m buffer around the census tract; annual mean concentrations of PM2.5, BC and NO2; and percentages of inhabitants that were foreign-born from lower- and mid-income countries, unemployed or had no higher education. A direct association between socioeconomic deprivation and CV medication sales [parameter estimate (95% CI): 0.26 (0.25; 0.28)] and inverse associations between CV medication sales and green space [-0.71 (-0.80; -0.61)] and air pollution [-1.62 (-1.69; -0.61)] were observed. In the regional models, the association between green space and CV medication sales was stronger in the region with relatively low green space cover (Flemish Region, standardized estimate -0.16) than in the region with high green space cover (Walloon Region, -0.10). In the highly urbanized Brussels Capital Region the association tended towards the null. In all regions, the associations between CV medication sales and socioeconomic deprivation were direct and more prominent. Our results suggest that there may be an inverse association between green space and CV medication sales, but socioeconomic deprivation was always the strongest predictor of CV medication sales.