连云港市5~15岁青少年流感样症状与空气颗粒物的关联研究

Association between influenza-like symptoms among adolescents aged 5-15 years and air particulate matter in Lianyungang City

目的探讨连云港市细颗粒物(particulate matter 2.5,PM_(2.5))、可吸入颗粒物(particulate matter 10,PM_(10))对5~15岁青少年流感样病例的短期影响,为流感样病例的预警和防治提供理论依据。方法收集2015—2018年连云港市大气污染物日均浓度、气象资料及5~15岁流感样病例日门诊量。利用Spearman等级相关分析流感样病例与空气污染物、气象因素的相关性,基于广义相加模型(generalized additive model,GAM)、分布滞后非线性模型(distributed lag nonlinear model,DLNM)探讨青少年流感样病例与PM_(2.5)、PM_(10)关联及剂量反应、滞后效应。结果在最强滞后情况下GAM单变量分析结果显示,PM_(2.5)、PM_(10)每增加一个标准差单位,流感样病例日门诊量分别增加2.4%[相对危险度(relative risk,RR)=1.024,95%置信区间(confidence interval,95%CI)=1.011~1.038]和2.2%(RR=1.022,95%CI=1.008~1.036),校正其他因素后,PM_(2.5)、PM_(10)与流感样病例日门诊量的统计学关联仍存在(PM_(2.5):RR=1.030,95%CI=1.014~1.046,P<0.01;PM_(10):RR=1.022,95%CI=1.007~1.038,P<0.01)。剂量反应关系显示,低浓度时,随PM_(2.5)、PM_(10)增加,流感样病例发生风险逐减升高,经历平台期,发病风险下降。分布滞后非线性模型显示,PM_(2.5)、PM_(10)与流感样病例日门诊量存在明显滞后效应,lag0~lag10、lag20~lag30各出现1个强滞后效应。结论PM_(2.5)、PM_(10)浓度增加可使流感样病例的发病风险升高,存在剂量反应与滞后效应。

Objective To analyze the short-term effects of PM_(2.5)and PM_(10)on influenza-like illness in adolescents aged 5-15 in Lianyungang City,so as to provide theoretical basis for early warning and prevention of influenza like cases.Methods The daily average concentration of air pollutants,meteorological data and daily outpatient visits of influenza-like illness in adolescents aged 5-15 in Lianyungang City from 2015 to 2018 were collected.Spearman rank correlation was used to analyze the correlation between influenza-like cases and air pollutants,meteorological factors.The relationship between influenza-like cases of adolescents and PM_(2.5)or PM_(10),dose response and lag effect were discussed based on generalized additive model(GAM)and distributed lag nonlinear model(DLNM).Results The univariate analysis results of generalized additive model under the strongest lag showed that the daily outpatient visits of influenza-like illness increased by 2.4%(RR=1.024,95%CI=1.011-1.038)and 2.2%(RR=1.022,95%CI=1.008-1.036)respectively for each additional standard deviation unit of PM_(2.5)and PM_(10).After adjusting for other factors,the relationship between PM_(2.5),PM_(10)and the daily outpatient visits of influenza-like illness still existed(PM_(2.5):RR=1.030,95%CI=1.014-1.046,P<0.01;PM_(10):RR=1.022,95%CI=1.007-1.038,P<0.01).Dose-effect relationship showed that at low concentration,with the increase of PM_(2.5)and PM_(10),the risk of influenza-like illness increased gradually,and decreased after the plateau period.The distributed lag nonlinear model showed that there was a significant lag effect between PM_(2.5),PM_(10)and the daily outpatient visits of influenza-like illness,and there was a strong lag effect between lag0-lag10 and lag20-lag30,respectively.Conclusion The increased concentration of PM_(2.5)and PM_(10)can increase the risk of influenza-like illness,and there is a dose effect relationship and lag effect.