2018—2019年成都市多种污染物联合作用对呼吸系统疾病门诊就诊人数的影响研究  

Combined effects of multiple air pollutants on number of outpatient visits for respiratory diseases in Chengdu from 2018 to 2019

作　　者：张莹 尹春苗 罗欢 王玉霞[4] ZHANG Ying;YIN Chun-miao;LUO Huan;WANG Yu-xia(School of Atmospheric Sciences,Plateau Atmosphere and Environment Sichuan Provincial Key Laboratory,Research Institute of Meteorology,Environment and Health,Chengdu University of Information Engineering,Chengdu,Sichuan 610225,China;不详)

机构地区：[1]成都信息工程大学大气科学学院,高原大气与环境四川省重点实验室气象环境与健康研究院,四川成都610225 [2]中卫市气象局 [3]成都市双流区气象局 [4]电子科技大学附属医院四川省人民医院

出　　处：《环境与健康杂志》2024年第12期1035-1042,F0003,共9页Journal of Environment and Health

基　　金：四川省自然科学基金(2024NSFSC0775);四川省科技厅项目(2020YJ0425);第二次青藏高原科学考察与研究(STEP)计划(2019QZKK0103);成都市双流区空气污染与健康气象服务研究项目(SCHBZX-2023-CG022);成都信息工程大学科研项目(KYTZ201723)。

摘　　要：目的分析2018—2019年成都市多种污染物联合作用对呼吸系统疾病门诊就诊人数的影响。方法收集2018—2019年四川省人民医院逐日呼吸系统疾病门诊就诊总人数和上呼吸道感染人数资料、同期气象要素和6种大气污染物(O3、PM_(2.5)、SO_(2)、NO_(2)、CO和PM_(2.5-10))资料,采用贝叶斯核机器回归(Bayesian Kernel Machine Regression,BKMR)模型,分析多种污染物在单滞后(1-2 d)和累积滞后(01-04 d)的情况下共同作用对呼吸系统疾病发病人数的影响效应。根据暴露变量的后验纳入概率(PIP),最终选出混合物中对呼吸系统疾病发病影响最明显的4种主要空气污染物PM_(2.5)、SO_(2)、NO_(2)和PM_(2.5-10)(PIP>0.1)纳入模型,进一步开展4种主要污染物联合作用的健康风险效应研究。结果4种大气污染物联合作用时,NO_(2)和PM_(2.5-10)对呼吸系统疾病的健康风险效应起主导作用。以lag1为例,NO_(2)浓度升高10μg/m3对上呼吸道感染的影响风险由第50百分位时的2.35%(95%CI:0.78%~5.30%)增大为第75百分位时的5.17%(95%CI:2.22%~10.83%),对总呼吸系统疾病影响风险由第50百分位时的2.29%(95%CI:0.80%~4.99%)增大为第75百分位时的19.50%(95%CI:9.42%~39.29%)。4种污染物与疾病发病之间的关联均呈非线性关系。不同滞后时间下4种大气污染物联合作用对疾病的交互影响效应存在明显差异,NO_(2)-PM_(2.5)、NO_(2)-SO_(2)在lag1、lag01和lag02时,对总呼吸系统疾病的影响存在交互效应,而PM_(2.5-10)-SO_(2)对总呼吸系统疾病的影响只在lag2时呈现出交互效应;NO_(2)-SO_(2)对上呼吸道感染的影响只在lag01时呈现出交互效应。结论成都市多污染物联合作用对呼吸系统疾病发病影响明显,应关注NO_(2)和PM_(2.5-10)对该地区呼吸系统疾病的健康风险效应,并进一步加强多污染物联合作用对人群健康的影响效应研究。Objective To understand the combined effect of multiple air pollutants on the number of outpatients with respiratory diseases in Chengdu from 2018 to 2019.Methods The data of daily outpatient visits for respiratory diseases and upper respiratory tract infections,meteorological factors and air pollutants(O_(3),PM_(2.5),SO_(2),NO_(2),CO and PM_(2.5-10))in Sichuan Provincial People’s Hospital from 2018 to 2019 were collected.Bayesian Kernel Machine Regression(BKMR)model was used to analyze the combined effect of multiple pollutants on the number of respiratory diseases under single lag(1-2)and cumulative lag(01-04).According to the posterior inclusion probability(PIP)of exposure variables,four major air pollutants(PM_(2.5),SO_(2),NO_(2)and PM_(2.5-10))(PIP>0.1)with the most significant impact on the incidence of respiratory diseases in the mixture were selected for further study on the health risk effects of the combined effects of the four major air pollutants.Results When the four air pollutants were combined,NO_(2)and PM_(2.5-10)played a dominant role in the health risk effect on respiratory diseases.For lag1,with a 10μg/m~3increase in NO_(2)concentration,the risk of upper respiratory tract infection increased from 2.35%(95%CI:0.78%-5.30%)at the 50 th percentile to 5.17%(95%CI:2.22%-10.83%)at the 75 th percentile.The risk of total respiratory diseases increased from 2.29%(95%CI:0.80%-4.99%)at the 50 th percentile to 19.50%(95%CI:9.42%-39.29%)at the 75 th percentile.Non-linear relationships between the four air pollutants and the incidence of diseases.There were significant differences in the interaction effects of the combined effects of the four air pollutants on diseases at different lag times.Specifically,there were interaction effects of NO_(2)-PM_(2.5)and NO_(2)-SO_(2)on total respiratory diseases at lag1,lag01 and lag02.Interaction effects of PM_(2.5-10)-SO_(2)on total respiratory diseases were observed only in lag2.The effects of NO_(2)-SO_(2)on upper respiratory tract infection showed an interaction ef

关 键 词：呼吸系统疾病 上呼吸道感染 大气污染物 贝叶斯核函数机器回归 协同效应 

分 类 号：R122.2[医药卫生—环境卫生学] R181.3[医药卫生—公共卫生与预防医学]