Assessment of PM2.5-related health effects: A comparative study using multiple methods and multi-source data in China

Xiaoyun Hou,Qinghai Guo,Yan Hong,Qiaowei Yang,Xinkui Wang,Siyang Zhou,Haiqiang Liu
DOI: https://doi.org/10.1016/j.envpol.2022.119381
IF: 8.9
2022-04-01
Environmental Pollution
Abstract:In China, PM<sub>2.5</sub> pollution has caused extensive death and economic loss. Thus, an accurate assessment of the spatial distribution of these losses is crucial for delineating priority areas for air pollution control in China. In this study, we assessed the PM<sub>2.5</sub> exposure-related health effects according to the integrated exposure risk function and non-linear power law (NLP) function in 338 prefecture-level cities in China by utilizing online monitoring data and the PM<sub>2.5</sub> Hindcast Database (PHD). Our results revealed no significant difference between the monitoring data and PHD (p value = 0.66 &gt; 0.05). The number of deaths caused by PM<sub>2.5</sub>-related Stroke (cerebrovascular disease), ischemic heart disease, chronic obstructive pulmonary disease, and lung cancer at the national level estimated through the NLP function was 0.27 million (95% CI: 0.06-0.50), 0.23 million (95% CI: 0.08-0.38), 0.31 million (95% CI: 0.04-0.57), and 0.31 million (95% CI: 0.16-0.40), respectively. The total economic cost at the national level in 2016 was approximately US$80.25 billion (95% CI: 24.46-132.25). Based on a comparison of Z statistics, we propose that the evaluation results obtained using the NLP function and monitoring data are accurate. Additionally, according to scenario simulations, Beijing, Chongqing, Tianjin, and other cities should be priority areas for PM<sub>2.5</sub> pollution control to achieve considerable health benefits. Our statistics can help improve the accuracy of PM<sub>2.5</sub>-related health effect assessments in China.
environmental sciences
What problem does this paper attempt to address?