PM2.5污染对中国人口死亡率的影响——基于346个城市面板数据的实证分析

中国生态文明建设和“健康中国”战略强调切实治理影响人口健康的环境问题,建设健康人居环境。论文基于2000年和2010年中国人口普查资料以及2005年和2015年各省级行政单元1%人口抽样调查等数据资料,论文使用探索性空间分析方法刻画中国城市人口死亡率的时空变化特征,并采用空间回归方法,揭示城市PM_2.5的平均浓度对人口死亡率的影响及其空间溢出效应,以及社会经济因素对PM_2.5—人口死亡率关联的调节效应。结果表明：① 中国城市人口死亡率的空间分布特征呈现明显的异质性,高死亡率地区早期集聚分布于西南地区,2005年后在西南地区、华北地区、华东地区和华中地区呈现逐渐集聚分布态势。低死亡率地区长期集中分布于西北地区、东北地区、长三角地区、珠三角地区和京津两市。② 人口死亡率的分布存在空间关联性,高—高类型地区早期集中分布于西南地区,后期向东扩展;低—低类型地区主要分布于北疆、内蒙古西部和广东省及其周边地区。③ 城市PM_2.5浓度对人口死亡率具有显著的正向影响,并且对邻近地区的人口死亡率具有显著的空间溢出效应。④ 中国城市PM_2.5浓度对人口死亡率的影响存在学历差异和城乡差异,地区高学历人群集聚可降低PM_2.5的健康风险,城镇化发展进程缓慢则会加重PM_2.5的健康风险。研究旨在为防范空气污染暴露导致的健康风险、建设健康人居环境提供科学依据。

Effects of PM2.5 concentration on mortality in China: A study based on city-level panel data

Health hazards and risks caused by air pollution have become a public topic. Ecological civilization construction and "Healthy China" strategy emphasize the alleviation of environmental stressors and the construction of healthy living environment. Using city-level data of population census and the 1% provincial sample demographic survey from 2000 to 2015, this study examined the spatial-temporal patterns of mortality and the effects of PM_2.5 concentration on mortality based on spatial regression models. It further examined the moderation effects of regional socioeconomic conditions on the PM_2.5 concentration-mortality association. Analytical results are as follows: 1) There is a considerable regional variation in mortality rate in China. Cities with high mortality rates initially were maincdly concentrated in Southwest China, and they become increasingly concentrated in Southwest China, North China, East China, and Central China after 2005. Cities with low mortality rates have long been concentrated in Northwest China, Northeast China, the Yangtze River Delta, the Pearl River Delta, and Beijing and Tianjin. 2) There is a significant spatial correlation in mortality rates. From 2000 to 2015, mortality rates became increasingly concentrated in particular regions. High-high type areas were concentrated in the southwest in the early stage and then expanded to the east. Low-low type areas are mainly distributed in the north of Xinjiang, the west of Inner Mongolia, and Guangdong Province and its surrounding areas. 3) The concentration of PM_2.5 has a positive correlation with mortality rates and a significant spatial spillover effect on mortality rates in neighboring areas. 4) The impact of PM_2.5 on mortality is subject to the influences of educational differences and urban-rural divides: regions with great concentration of highly educated people are less vulnerable to PM_2.5 health impact, and areas with a low level of urbanization are more subject to health risks of PM_2.5. The results show that regional prevention and control of air pollution is important, and in order to reduce health risks, more attention should be paid to the development of high-quality urbanization, optimization of economic structure, and promotion of residents' health literacy. Our findings can provide a scientific reference for the environmental risk assessment of air pollution and the construction of healthy living environments.