中国环境保护重点城市空气质量指数时空变化特征

  目的   研究不同温度范围PM₁₀对成都市≥65岁老年人过早死亡累积滞后效应的影响。  方法   利用四分位间距对气温进行分层，运用分布滞后非线性Quasipoisson回归模型，分层分析2013 — 2016年不同温度范围PM₁₀对成都老年人过早死亡的累积滞后效应及其对死亡影响的持续作用时间。  结果   调整温度后，PM₁₀的短期波动对成都老年人过早死亡的影响向后延迟2～5 d，其累积滞后死亡影响持续3～5 d，PM₁₀每增加10μg/m³可使老年人非意外全死因死亡累积滞后效应增加0.2%（95 % CI = 0.0 %～0.3 %）。在高温和中等温度范围PM₁₀对老年人的过早死亡存在显著的累积滞后影响，但在低温范围PM₁₀对老年人的过早死亡影响无统计学意义（P > 0.05）；在高温范围PM₁₀每增加10μg/m³全死因（非意外）和心血管病累积滞后死亡风险分别增加3.5 %（95 % CI = 0.1 %～7.0 %）和3.3 %（95 % CI = 0.2 %～6.4 %）；在中温范围PM₁₀每增加10μg/m³全死因（非意外）、心血管病和癌症累积滞后死亡风险分别增加0.3 %（95 % CI = 0.1 %～0.5 %）、0.3 %（95% CI = 0.1 %～0.6 %）和0.5 %（95% CI = 0.2 %～0.9 %）；所有温度范围PM₁₀对呼吸疾病死亡的滞后累积效应均无统计学意义（P > 0.05）。  结论   成都PM₁₀对老年人过早死亡影响存在显著的温度差异。

Systematic review of Chinese studies of short-term exposure to air pollution and daily mortality

  Objective   To explore the difference in cumulative lag impacts of particulate matter less than 10 μm in aerodynamic diameter (PM₁₀) exposure on daily mortality in relation to various air temperature among residents aged 65 years and over in Chengdu city.  Methods   The scales of air temperature were defined by interquartile range (IQR) of annual mean temperature during 2013 – 2016. We estimated cumulative lag impacts of PM₁₀ on mortality and the duration of the cumulative lag impacts with distributed lag nonlinear quasipoisson regression model by temperature scale stratum.   Results   After controlling for daily mean temperature, a significant cumulative lag impact of PM₁₀ (excess mortality risk [ER] = 0.2 %, 95 % confidence interval [95 % CI]: 0.0 % – 0.3 %) for a 10 μg/m³ increase on non-accidental mortality was observed. The lag impacts were postponed two to five days and the lag impacts lasted for three to five days. Stratified by IQR of temperature, the associations between PM₁₀ and cumulative lag impacts of PM₁₀ on the mortality were quite different in different temperature ranges. The impacts were significant at high and medium temperature ranges. However, significant impacts were not observed at low temperature range. As for high temperature range, significant cumulative lag impacts of PM₁₀ (for a 10 μg/m³ increase) on non-accidental mortality (ER = 3.5 %, 95 % CI: 0.1 % – 7.0 %) and cardiovascular mortality (ER = 3.3 %, 95 % CI: 0.2 % – 6.4 %) were observed. As for medium temperature range, significant cumulative lag impacts of PM₁₀ (for a 10 μg/m³ increase) on non-accidental mortality (ER = 0.3 %, 95 % CI: 0.1 % – 0.5 %), cardiovascular mortality (ER = 0.3%, 95 % CI: 0.1 % – 0.6 %), and cancer mortality (ER = 0.5%, 95 % CI: 0.2 % – 0.9%) were observed. No significant cumulative lag impacts of PM₁₀ (for a 10 μg/m³ increase) on respiratory mortality were observed on whole temperature range.  Conclusion   Cumulative lag impacts of PM₁₀ on mortality of residents aged 65 years and over in Chengdu city are influenced significantly by ambient air temperature.