Abstract：
Short-term exposure to PM2.5 or O3 can increase mortality risk; however, limited studies have evaluated their interaction. A multicity time series study was conducted to investigate the synergistic effect of PM2.5 and O3 on mortality in China, using mortality data and high-resolution pollutant predictions from 272 cities in 2013-2015. Generalized additive models were applied to estimate associations of PM2.5 and O3 with mortality. Modification and interaction effects were explored by stratified analyses and synergistic indexes. Deaths attributable to PM2.5 and O3 were evaluated with or without modification of the other pollutant. The risk of total nonaccidental mortality increased by 0.70% for each 10 μg/m3 increase in PM2.5 when O3 levels were high, compared to 0.12% at low O3 levels. The effect of O3 on total nonaccidental mortality at high PM2.5 levels (1.26%) was also significantly higher than that at low PM2.5 levels (0.59%). Similar patterns were observed for cardiovascular or respiratory diseases. The relative excess risk of interaction and synergy index of PM2.5 and O3 on nonaccidental mortality were 0.69% and 1.31 with statistical significance, respectively. Nonaccidental deaths attributable to short-term exposure of PM2.5 or O3 when considering modification of the other pollutant were 28% and 31% higher than those without considering modification, respectively. Our results found synergistic effects of short-term coexposure to PM2.5 and O3 on mortality and suggested underestimations of attributable risks without considering their synergistic effects.
摘要：
短期接触PM2.5或O3会增加死亡风险；然而，有限的研究评估了它们的相互作用。进行了多元时间序列研究，以研究PM2.5和O3对中国死亡率的协同作用。使用2013-2015年272个城市的死亡率数据和高分辨率污染物预测。应用广义累加模型来估计PM2.5和O3与死亡率的关联。通过分层分析和协同指数探讨了修饰和交互效应。在有或没有改变其他污染物的情况下，评估了可归因于PM2.5和O3的死亡。当O3水平较高时，PM2.5每增加10μg/m3，总的非意外死亡风险增加0.70%，相比之下，在低O3水平下为0.12%。在高PM2.5水平（1.26％）下，O3对总非意外死亡率的影响也显着高于低PM2.5水平（0.59％）。对于心血管或呼吸系统疾病观察到类似的模式。PM2.5和O3的交互作用和协同作用指数对非意外死亡率的相对超额风险分别为0.69%和1.31,具有统计学意义。分别。考虑修改其他污染物时，由于短期暴露于PM2.5或O3而导致的非意外死亡比不考虑修改的高28％和31％。分别。我们的结果发现了短期共同暴露于PM2.5和O3对死亡率的协同作用，并建议在不考虑其协同作用的情况下低估了可归因风险。
