基于糖尿病患者队列研究评价植被覆盖对大气PM2.5健康危害的修正效应

Existing reviews and meta-analyses conclude that exposure to greenspace is beneficial for many health parameters in relatively high-income Western countries. However, despite the boom in greenspace research, no study has been conducted thus far in China. In this project, spatial analysis of MODIS (Moderate Resolution Imaging Spectrometer) satellite aerosol optical depth (AOD) will be used to estimate the concentrations of atmospheric fine particulate matter (PM2.5) and the normalized vegetation coverage index (NDVI) in personal level. Then, we will evaluate the modification of greenspace on the relationship of PM2.5 with the incidence of type 2 diabetes mellitus (T2D) in subjects from the Pearl River Delta Six Cities Cohort study which was conducted previous. A retrospective and prospective cohort study of T2D patients will be conducted to quantitatively assess the interactive effects of greenspace and ambient air pollution on the mortality and development of T2D in patients, and then determine their effect window period, the effect threshold, and early warning value according to the different susceptible population. The pollution history in components of PM2.5 will be reconstructed by dendrochemistry combined with element tracing technique and history monitoring data. During the prospective cohort study of T2D patients, we will collect the atmospheric PM2.5 samples in the selected regions in summer and winter season, respectively, and followed by analysis of characterization and major components. The attribution risk of major components in PM2.5 for T2D onset and outcome will be calculated by biomonitoring equivalent analysis under different greenspace types. Then the prior control pollutant list will be screened. Finally, a comprehensive evaluation index system of greenness, air pollution and health effects will be established, which will provide scientific basis for the government to formulate air pollution control strategies and take targeted intervention measures.

植被覆盖对大气污染健康危害的降低效应已成为国际环境卫生领域研究热点，但在我国未见报道。本项目基于我们目前开展的珠三角六城市队列研究，以二型糖尿病（T2D）患者作为研究对象，采用气溶胶光学厚度（AOD）嵌套的空间分析模型精确评估个体的归一化植被覆盖指数和大气细颗粒物PM2.5的暴露水平，评估植被覆盖对大气污染与T2D发病暴露—反应关系的修正效应。通过对T2D患者的回顾性和前瞻性队列研究，定量评估植被覆盖与大气污染对T2D患者死亡和并发症发生发展的交互效应及其效应窗口期，并根据不同易感人群确定出效应阈值和预警值；采用树木年轮化学法结合历史监测数据和现场采样，表征PM2.5及解析内载组分，基于生物监测当量分析在不同植被覆盖类型下PM2.5各内载有毒组分对T2D发病和转归的归因危险度，筛选优控污染物，构建植被覆盖与大气污染对健康影响的综合评价体系，为政府制定大气污染控制策略和采取针对性的干预措施提供科学依据。

大气污染，不仅是全球面临的重大环境问题，其对人群健康的影响及所造成的疾病负担也一直是各国政府所面对的重大公共卫生问题，因此，如何有效的改善环境降低大气污染健康危害效应是科学研究者和政府决策者所面对的挑战；作为“健康城市”的一个重要指标，植被覆盖度不仅可以改善宜居环境，还能通过降低大气污染暴露水平、促进锻炼、提高社交接触等对人群健康产生有益的影响，但目前基于中国人群开展植被覆盖对大气污染健康危害修正作用的研究不多；本项目通过大规模人群队列研究，以二型糖尿病和心血管疾病作为效应结局，采用AOD嵌套的空间分析模型构建最优的人群大气污染物（PM2.5、PM10、SO2、NO2）暴露和植被覆盖指数评估模型，定量的评估植被覆盖和大气污染对二型糖尿病和心血管疾病影响的暴露-效应关系及其交互效应，并明确颗粒物中优控组分；结果显示，大气污染暴露可显著的增加糖尿病和心血管疾病死亡和发病的风险，如PM2.5每增加10 µg/m3，居民因糖尿病死亡的风险增高了45%（HR=1.45; 95%CI: 1.41-1.49），而植被覆盖可显著的降低居民患有糖尿病的风险，NDVI指数每增加0.1个单位，居民糖尿病患病风险降低12%（OR=0.88; 95%CI: 0.82-0.94），PEBK模型显示其中有13.3%的比例归因于植被覆盖对大气细颗粒物PM2.5效应的修正作用；同时植被覆盖对大气污染健康危害的修正效应存在一定的阈值，在低污染水平下，植被覆盖的修正作用比较显著，但在高污染水平下，植被覆盖的修正效应逐渐消失；进一步通过组分解析发现氯化石蜡、多环芳烃和全氟化合物是影响糖尿病发生和转归的优控组分。这些研究结果将为大气污染的健康危害风险评估和防控策略提供科学依据。通过执行本项目，发表SCI论文30篇；培养博士研究生1人，硕士研究生14人；2名博士研究生和3名硕士研究生继续进行本项目相关研究。