宁波市大气颗粒物浓度与流感样病例的时间序列研究

Impact of ambient particulate matters exposure on the risk of influenza-like illness in Ningbo,China:a time series study

目的探讨宁波市大气颗粒物中可吸入颗粒物(inhalable particulate matter,FPM_(10))及细颗粒物(fine particulate matter,PM_(2.5))的浓度对流感样病例发病的影响。方法应用分布滞后非线性模型对2014-2016年的大气颗粒物PM_(10)及PM_(2.5)的浓度与同期流感样病例的时间序列进行研究,分析两者之间的关联性,计算相应归因危险度百分比(attributable risk proportion,AR%)以评估污染物对流感样病例发病的急性效应。结果两种颗粒物浓度均与流感样病例发病数呈正相关,单污染物分析中,两种大气颗粒污染物均会造成短期内的流感样病例发病的增加,且产生的影响表现出滞后性和收获效应,其中PM_(10)产生的效应AR%值要高于PM_(2.5),滞后0～7 d的累积AR%值分别为5.13%(95%CI:2.94%～7.37%,P<0.001)和2.52%(95%CI:-0.75%～5.90%,P=0.139)。而双污染物分析中,调整其他气态污染物后,PM_(10)与流感样病例之间的关联强度发生改变,PM_(2.5)仅在纳入二氧化氮(nitrogen dioxide,NO_2)后引起滞后0～7 d的流感样病例发病的增加。结论颗粒物PM_(10)及PM_(2.5)浓度的升高会引起同期流感样病例发病的增加。

Objective To investigate the effect of ambient particulate matters with aerodynamic diameter ≤ 2. 5 μm （PM2.5） and ≤ 10 μm （PM10）on the incidence of influenza,like illness in Ningbo. Methods Time series study on PM2.5 , PM10 and the incidence of influenza-like illness from 2014 to 2016 was investigated by using the distributed lag nonlinear model （DLNM）. The corresponding attributable risk proportion（AR%) was calculated to assess the acute effect of contaminants on the onset of influenza-like illness. Results There was a positive correlation between the two kinds of particulate matters, PM10 and PM2.5, and the incidence of influenza-like cases. Analysis of single pollutant showed that both PM10 and PM2.5 increased the incidence of influenza-like illness in the short term, and the effect was characterized by hysteresis and harvesting, with the AR% of 5.13% （95% CI：2. 94%-7.37% , P 〈 0. 001) and Z 52% （95% CI：-0. 75 %-5.90%, P = 0. 139） , respectively. Furthermore, the acute health effects caused by PM10 were more severe than that caused by PM2.5. Analysis of double pollutants showed that the association between PM10 and influenza-like illness was changed af-ter adjusting for other gaseous pollutants. By contrast, PM2.5 significantly increased the incidence of influenza-like illness with a 0-7 day lag after adjusting nitrogen dioxide (NO2). Conclusion Increased concentrations of PM10 and PM2.5 may increase the incidence of influenza-like illness.