Community built environment optimization and avoidable PM2.5-attributable mortality in China: A nationwide health
Jian Song1, Xunrong Cheng1, Rong Song1
1Department of Epidemiology and Health Statistics, School of Public Health, Anhui Medical University, No.81 Meishan road, Shushan District, Hefei, Anhui 230031, China; Inflammation and Immune Mediated Diseases Laboratory of Anhui Province, China.
Background:
The design of built environment with health considerations can mitigate population exposure to PM2.5 air pollution. However, quantifiable evidence regarding the health benefits of community-level built environment interventions in reducing the PM2.5-related disease burden in China, while accounting for spatial variations, remains limited.
Methods:
We first conducted a quantitative health impact assessment to characterize spatiotemporal patterns and disparities in PM2.5-attributable mortality at the community scale in China (2013-2019). Employing a three-stage analytical framework integrating machine learning and spatial statistics, we identified critical built environment determinants of ambient PM2.5 concentrations and estimated preventable PM2.5-attributable deaths under built environment modification scenarios, accounting for geographic heterogeneity. Finally, we projected PM2.5-related disease burdens under different Shared Socioeconomic Pathways (SSPs) for China in 2050, highlighting the importance of optimizing the community-built environment.
Results:
Despite a national decline in PM2.5-attributable mortality between 2013 and 2019, substantial spatial disparities persisted. In explaining the baseline spatial variation of PM2.5 concentrations, population density was the most important feature, followed by NDVI (Normalized Difference Vegetation Index) and building density. Under standardized counterfactual scenarios (adjustment to the 75th percentile), enhancing community greenness (NDVI), an intervention applicable to 43.4% of communities, was projected to avert approximately 47,838.0 (95%CI: 46,906.1-48,768.9) annual deaths nationally. In comparison, interventions on population density and building density, which demonstrated more spatially constrained applicability, could avert 18,763.8 (95% CI: 18,397.7-19,130.3) and 12,626.6 (95% CI: 12,380.7-12,872.6) deaths, respectively. The health benefits of all interventions exhibited profound spatial heterogeneity, for instance, increasing NDVI was associated with mortality reductions ranging from 0.6% in the Beijing-Tianjin-Hebei region to 5.5% in the Fenwei Plain. Future projections for 2050 indicated that even under an ambitious sustainability pathway (SSP1-2.6), the estimated mortality reduction falls short of the Sustainable Development Goal target.
Conclusions:
This study suggests that optimizing the community-built environment may serve as a valuable complement to existing strategies for reducing PM2.5-attributable mortality. Its benefits are highly region-specific, necessitating spatially-tailored strategies to effectively mitigate disease burden.
