高级检索

大气污染物暴露与青岛市居民脑卒中发病关系时间序列分析

葛南, 潘璐, 张欣, 李丹丹, 王寅, 尹静雅, 周慧, 于浩岩, 张秀芹, 徐春生, 方渊, 马艳, 王炳玲, 段海平

葛南, 潘璐, 张欣, 李丹丹, 王寅, 尹静雅, 周慧, 于浩岩, 张秀芹, 徐春生, 方渊, 马艳, 王炳玲, 段海平. 大气污染物暴露与青岛市居民脑卒中发病关系时间序列分析[J]. 中国公共卫生, 2024, 40(10): 1161-1168. DOI: 10.11847/zgggws1143237
引用本文: 葛南, 潘璐, 张欣, 李丹丹, 王寅, 尹静雅, 周慧, 于浩岩, 张秀芹, 徐春生, 方渊, 马艳, 王炳玲, 段海平. 大气污染物暴露与青岛市居民脑卒中发病关系时间序列分析[J]. 中国公共卫生, 2024, 40(10): 1161-1168. DOI: 10.11847/zgggws1143237
GE Nan, PAN Lu, ZHANG Xin, LI Dandan, WANG Yin, YIN Jingya, ZHOU Hui, YU Haoyan, ZHANG Xiuqin, XU Chunsheng, FANG Yuan, MA Yan, WANG Bingling, DUAN Haiping. Association between short-term exposure to air pollutants and daily stroke incidence among residents in Qingdao city: a time series analysis of disease surveillance, environmental, and meteorological monitoring data[J]. Chinese Journal of Public Health, 2024, 40(10): 1161-1168. DOI: 10.11847/zgggws1143237
Citation: GE Nan, PAN Lu, ZHANG Xin, LI Dandan, WANG Yin, YIN Jingya, ZHOU Hui, YU Haoyan, ZHANG Xiuqin, XU Chunsheng, FANG Yuan, MA Yan, WANG Bingling, DUAN Haiping. Association between short-term exposure to air pollutants and daily stroke incidence among residents in Qingdao city: a time series analysis of disease surveillance, environmental, and meteorological monitoring data[J]. Chinese Journal of Public Health, 2024, 40(10): 1161-1168. DOI: 10.11847/zgggws1143237

大气污染物暴露与青岛市居民脑卒中发病关系时间序列分析

基金项目: 青岛市科技惠民示范专项项目(23 – 2 – 8 – smjk – 18 – nsh);青岛市医疗卫生优秀人才培养项目(2022 — 2024)
详细信息
    作者简介:

    葛南(1992 – ),主管医师,硕士,研究方向:环境卫生与健康

    通讯作者:

    段海平,E-mail:duan_hp@126.com;王炳玲,E-mail:duan_hp@126.com

Association between short-term exposure to air pollutants and daily stroke incidence among residents in Qingdao city: a time series analysis of disease surveillance, environmental, and meteorological monitoring data

More Information
  • 摘要:
    目的 

    了解大气污染物暴露与山东省青岛市居民脑卒中发病的关系,为脑卒中的预防控制提供参考依据。

    方法 

    收集国家慢性病监测网络报告系统中青岛市2014 年 1 月 1 日 — 2020 年 12 月 31 日上报的脑卒中新发病例相关数据以及同期青岛生态环境监测中心大气污染物监测数据和青岛市气象台气象监测数据,应用分布滞后非线性模型(DLNM)分析大气污染物暴露与青岛市居民脑卒中发病的关系,同时控制长期趋势、星期几效应等混杂因素的影响分析单污染物模型大气污染物对脑卒中发病的效应。

    结果 

    青岛市2014 — 2020年共报告51 120例脑卒中新发病例,平均日发病数为(19.99 ± 15.53)例。暴露 – 反应关系分析结果显示,PM2.5暴露浓度 > 121.90 μg/m3、CO暴露浓度 > 1.56 μg/m3和O3暴露浓度 < 64.00 μg/m3时脑卒中发病风险均呈上升趋势,而O3暴露浓度 > 64.00 μg/m3时脑卒中发病风险呈上下波动趋势(均P < 0.05)。单污染物模型分析结果显示,PM2.5和CO暴露与脑卒中发病风险升高的易感滞后期为Lag2 d~Lag5 d,O3暴露与脑卒中发病风险升高的易感滞后期为Lag3 d~Lag6 d,PM10暴露与脑卒中发病风险升高的易感滞后期为Lag3 d~Lag4 d,SO2暴露与脑卒中发病风险升高的易感滞后期为Lag6 d;PM2.5暴露浓度每升高32.01 μg/m3、PM10暴露浓度每升高52.21 μg/m3,脑卒中Lag3 d发病的RR值分别为1.018(95%CI = 1.005~1.031)、2.027(95%CI = 1.232~3.334);O3暴露浓度每升高53.00 μg/m3、CO暴露浓度每升高0.39 mg/m3,脑卒中Lag4 d发病的RR值分别为1.155(95%CI = 1.080~1.234)、1.033(95%CI = 1.016~1.050),SO2暴露浓度每升高14.93 μg/m3,脑卒中Lag6 d发病的RR值为1.431(95%CI = 1.049~1.951)。敏感性分析结果显示,T、AP和RH的df变为4、5、6时对PM2.5、O3、CO、PM10、SO2和NO2暴露与青岛市居民脑卒中发病风险关系效应值的影响均较小,模型结果较为稳定。

    结论 

    PM2.5、PM10、O3、CO和SO2等大气污染物暴露对青岛市居民脑卒中发病具有滞后效应。

    Abstract:
    Objective 

    To investigate the association between exposure to ambient air pollutants and stroke incidence among residents of Qingdao city, Shandong province, and to provide evidence for stroke prevention and control.

    Methods 

    Data on new cases of stroke reported in Qingdao municipality from 2014 to 2020 were collected from the National Chronic Disease Surveillance Network Reporting System, together with air pollution monitoring data from the Qingdao Ecological Environment Monitoring Center and meteorological monitoring data from the Qingdao Meteorological Observatory for the same period. The distributed Lag nonlinear model (DLNM) was used to analyze the associations between daily mean concentrations of particulate matter with an aerodynamic diameter of less than 2.5 mum (PM2.5), particulate matter with an aerodynamic diameter of less than 10 mum (PM10), carbon monoxide (CO), ozone (O3), sulfur dioxide (SO2), and nitrogen dioxide (NO2) with daily stroke incidence, while controlling for the effects of potential confounders such as long-term trend and day of the week in Qingdao city. The effects of air pollutants on stroke incidence were analyzed using a single-pollutant model.

    Results 

    A total of 51 120 new cases of stroke were reported in the city during the period, with an average daily incidence of 19.99 ± 15.53. The exposure-response relationship analysis showed that the risk of stroke incidence increased when ambient PM2.5 concentration was higher than 121.90 μg/m3, CO concentration was higher than 1.56 μg/m3, but O3 concentration was lower than 64.00 μg/m3, while the risk of stroke incidence showed an up-and-down fluctuating trend when O3 concentration was higher than 64.00 μg/m3 (all P<0.05). The single-pollutant model analysis showed that the sensitive Lag periods for increased risk of stroke incidence were from Lag day 2 to Lag day 5 for ambient PM2.5 and CO exposure, from Lag day 3 to Lag day 6 for O3 exposure, from Lag day 3 to Lag day 4 for PM10 exposure, and from Lag day 6 for SO2 exposure. The relative risk (RR) values for daily stroke incidence at Lag day 3 were 1.018 (95% confidence interval [95%CI]: 1.005 – 1.031) and 2.027 (95%CI: 1.232 – 3.334) for each 32.01 μg/m3 increase in PM2.5 concentration and for each 52.21 μg/m3 increase in PM10 concentration, respectively. The RR values for daily stroke incidence at Lag day 4 were 1.155 (95%CI: 1.080 – 1.234) and 1.033 (95%CI: 1.016 – 1.050) for each 53.00 μg/m3 increase in O3 concentration and each 0.39 mg/m3 increase in CO concentration, respectively. The RR value for daily stroke incidence at Lag day 6 was 1.431 (95%CI: 1.049 – 1.951) for each 14.93 μg/m3 increase in SO2 concentration. Sensitivity analysis results showed that changing the degrees of freedom of daily mean temperature, barometric pressure, and relative humidity to 4, 5, and 6 had little effect on the effect of ambient PM2.5, O3, CO, PM10, SO2, and NO2 exposure on the risk of daily stroke incidence, indicating that the established models produced relatively stable results.

    Conclusion 

    Exposure to ambient PM2.5, PM10, O3, CO, and SO2 has a delayed effect on stroke incidence among Qingdao residents.

  • 脑卒中(stroke),是一种急性脑血管疾病,是由于脑部血管突然破裂或因血管阻塞导致血液不能流入大脑而引起脑组织损伤或功能障碍的一组疾病,包括缺血性脑卒中和出血性脑卒中[1 - 2]。据报道,脑卒中为世界第二大、中国第一大致死性疾病,同时也是造成残疾的主要原因,其具有高发病率、高死亡率、高复发率、高致残率和高经济负担的特点,可严重影响患者的身体健康和生活质量,已成为中国重要的公共卫生问题之一[3 - 4]。研究表明,除吸烟、高体质指数(body mass index, BMI)、高血压和高血糖外,空气污染同样为脑卒中的主要危险因素[5 - 7],无论是短期暴露或长期暴露均与脑卒中的死亡率和住院人数相关[2]。由于空气污染是一个独立于个人行为的、可改变的危险因素,因此改善空气质量可能会在降低脑卒中发病率方面发挥重要的作用[8 - 10],而了解大气污染物与脑卒中发病之间的关系则有助于评估和预测由空气质量变化引起的脑卒中疾病发作,对预防脑卒中具有重要的公共卫生意义。但国内外大多研究均以脑卒中死亡作为结局[11 - 12],然而脑卒中的发病并不全会造成死亡[13],因此采用死亡作为结局可能无法完全反映大气污染物对人群的影响,如有研究表明缺血性脑卒中的死亡发病比约为14%[14],因此若以死亡作为结局将会大幅低估受大气污染影响的人群数量。为此,本研究收集了国家慢性病监测网络报告系统中青岛市疾病预防控制中心2014 年 1 月 1 日 — 2020 年 12 月 31 日上报的脑卒中新发病例相关数据以及同期青岛生态环境监测中心大气污染物监测数据和青岛市气象台气象监测数据,应用分布滞后非线性模型分析大气污染物暴露与青岛市居民脑卒中发病的关系,同时控制长期趋势、星期几效应等混杂因素的影响分析单污染物模型大气污染物对脑卒中发病的效应,旨在了解大气污染物暴露与山东省青岛市居民脑卒中发病的关系,为脑卒中的预防控制提供参考依据。结果报告如下。

    收集国家慢性病监测网络报告系统中青岛市疾病预防控制中心2014 年 1 月 1 日 — 2020 年 12 月 31 日上报的51 120例脑卒中新发病例的相关数据,内容包括患者性别、年龄、家庭住址、疾病诊断等。按照《疾病和有关健康问题的国际统计分类(第十次修订本)》(the International Statistical Classification of Diseases and Related Health Problems 10th Revision, ICD-10)[15]编码规则,将编码为I60~I64(脑卒中)的患者作为研究对象。同期大气污染物数据来源于山东省青岛生态环境监测中心,包括细颗粒物(fine particulate matter, PM2.5)、可吸入颗粒物(inhalable particulate matter, PM10)、一氧化碳(carbon monoxide, CO)、臭氧(ozone, O3)、二氧化硫(sulfur dioxide, SO2)和二氧化氮(nitrogen dioxide, NO2)等;同期气象监测数据来源于青岛市气象台,包括日平均气温(temperature, T)、日平均气压(atmospheric pressure, AP)、日平均相对湿度(relative humidity, RH)、日累积降水量(precipitation, P)和日平均风速(wind speed, WS)等。

    应用R 4.2.3 软件进行统计分析。对大气污染物和气象因素数据进行描述性分析,采用Spearman积矩相关系数描述大气污染物和气象因素之间的相关程度,并提取逐日大气污染物浓度和气象因素绘制时间序列图。与青岛市总人口数相比,每日报告的脑卒中发病数相对较少,其分布符合 Poisson分布,故本研究以Poisson分布为基础的分布滞后非线性模型(distributed Lag non-linear model,DLNM)分析PM2.5、PM10、CO、O3、SO2和NO2等大气污染物与脑卒中发病之间的滞后 – 反应关系及效应。所构建模型公式为:log[E(Yt)] = α + cb(pollutantt) + ns(T, df) + ns(AP, df) + ns(RH, df) + ns(time, df) + β × DOW + γ × holiday。式中,E(Yt)为脑卒中日病例数估计值;α为截距;cb为交叉基函数;pollutant 为大气污染物;T为日均气温,单位为℃;AP为日均气压,单位为hPa;RH为日均相对湿度,单位为%;ns为自然立方样条函数;time为时间变量,单位为d;df为自由度;βγ为系数;DOW为星期几效应;holiday为节假日效应。既往研究表明大气污染物致心脑血管事件的影响一般持续1~5 d[13,16 - 19],因此本研究将交叉基中最大滞后天数设为7 d以充分考察大气污染物的滞后效应;DOW和holiday分别代表星期几和节假日并以哑变量的形式纳入;time用来控制长期趋势和季节趋势,以ns的形式纳入,每年的df设为7;T、AP和RH以ns的形式纳入并将df设定为3,拟合其与结局的暴露 – 反应关系以控制日T、AP和RH与结局间的非线性关联。采用 Akaike 信息准则(Akaike information criterion, AIC)判断模型的拟合度。此外,本研究还在上述模型中将T、AP和RH的df 从3分别调整为4、5、6进行敏感性分析以验证模型的稳定性。检验水准为双侧检验α = 0.05。

    表  1  青岛市2014 — 2020年大气污染物和气象因素基本特征
    Table  1.  Mean, minimum, 25th percentile, median, 75th percentile, maximum, and interquartile range of daily average values of ambient air pollutants and temperature, atmospheric pressure, relative humidity, precipitation, wind speed in Qingdao city from 2014 to 2020
    项目 分类 $ \overline{x}\pm s $ Min P25 M P75 Max IQR
    大气污染物 PM2.5(μg/m3 44.01 ± 31.09 4.28 23.38 34.96 55.39 273.13 32.01
    O3(μg/m3 99.86 ± 37.39 24.12 70.17 96.38 123.17 272.04 53.00
    CO( mg/m3 0.77 ± 0.37 0.24 0.52 0.67 0.91 3.23 0.39
    PM10(μg/m3 81.63 ± 45.87 10.80 49.83 70.22 102.04 412.46 52.21
    SO2(μg/m3 17.89 ± 15.67 3.33 7.46 13.26 22.39 169.04 14.93
    NO2(μg/m3 32.17 ± 14.22 4.21 21.72 29.79 40.21 98.35 18.49
    气象因素 T(℃) 13.79 ± 9.53 – 12.06 5.19 14.81 22.33 31.00 17.14
    AP(hPa) 1009.70 ± 9.25 981.30 1001.80 1009.80 1017.20 1034.10 15.40
    RH(%) 67.85 ± 15.34 24.00 56.50 69.40 80.10 98.80 23.60
    P(mm) 1.76 ± 7.31 0.00 0.00 0.00 0.11 144.52 0.11
    WS(m/s) 2.88 ± 1.03 0.99 2.14 2.67 3.43 7.93 1.29
    下载: 导出CSV 
    | 显示表格
    图  1  青岛市2014 — 2020年逐日大气污染物浓度和气象因素的时间序列图
    Figure  1.  Daily average values of ambient air PM2.5, O3, CO, PM10, SO2, NO2 and temperature, atmospheric pressure, relative humidity, precipitation, wind speed in Qingdao city from 2014 to 2020

    青岛市2014 — 2020年共报告51 120例脑卒中新发病例,平均日发病数为(19.99 ± 15.53)例。其中,男性患者28930例(56.59%),女性患者22190例(43.41%);年龄 < 65岁患者17596例(34.42%), ≥ 65岁患者33524例(65.58%)。青岛市2014 — 2020年大气污染物和气象因素基本特征见表1。研究期间内,各大气污染物逐日浓度具有周期性分布的特点,其中PM2.5、PM10、CO、SO2和NO2浓度均呈夏季低、冬季高的特点,而O3浓度则呈夏季高、冬季低的特点;气象因素中,T和AP的周期性较为明显,而RH、P和WS的周期性不明显。逐日大气污染物和气象因素的时间序列图见图1

    表  2  青岛市2014 — 2020年大气污染物与气象因素相关性分析
    Table  2.  Spearman product-moment correlation coefficients among daily average values of ambient air PM2.5, O3, CO, PM10, SO2, NO2 and temperature, atmospheric pressure, relative humidity, precipitation, wind speed in Qingdao city from 2014 to 2020
    项目 PM2.5 O3 CO PM10 SO2 NO2 T AP RH P WS
    PM2.5 1.000
    O3 – 0.167 a 1.000
    CO 0.918 a – 0.295 a 1.000
    PM10 0.914 a – 0.086 a 0.827 a 1.000
    SO2 0.626 a – 0.272 a 0.721 a 0.598 a 1.000
    NO2 0.751 a – 0.254 a 0.819 a 0.758 a 0.657 a 1.000
    T – 0.395 a 0.652 a – 0.518 a – 0.351 a – 0.448 a – 0.471 a 1.000
    AP 0.304 a – 0.577 a 0.426 a 0.279 a 0.380 a 0.480 a – 0.856 a 1.000
    RH – 0.063 a 0.061 a – 0.133 a – 0.255 a – 0.282 a – 0.379 a 0.397 a – 0.457 a 1.000
    P – 0.166 a – 0.050 a – 0.153 a – 0.209 a – 0.136 a – 0.210 a 0.154 a – 0.246 a 0.311 a 1.000
    WS – 0.194 a – 0.138 a – 0.220 a – 0.131 a – 0.125 a – 0.353 a – 0.202 a 0.029 – 0.130 a 0.147 a 1.000
      注:a P < 0.05。
    下载: 导出CSV 
    | 显示表格

    相关性分析结果显示,除AP与WS间无显著相关关系外,其他6种大气污染物与气象因素间均具有相关性(均P < 0.05);其中,PM2.5、PM10、CO、SO2、NO2等5种大气污染物之间均呈正相关,与T、RH、P和WS均呈负相关,与AP均呈正相关(均P < 0.05);O3与T和RH均呈正相关,与PM2.5、PM10、CO、SO2、NO2、AP、P和WS均呈负相关(均P < 0.05)。

    图  2  青岛市2014 — 2020年大气污染物与脑卒中日报告发病数的暴露 – 反应关系
    注:图中曲线为RR值,阴影部分为其95%CI;图中横直线代表RR值为1.0。
    Figure  2.  Relative risk (red line) and its 95% confidence interval (grey area) of daily stroke incidence number associated with daily average concentration of ambient air PM2.5, O3, CO, PM10, SO2, and NO2: Poisson distributed Lag nonlinear model analysis on disease surveillance and environment/meteorological monitoring data in Qingdao city from 2014 to 2020

    暴露 – 反应关系分析结果显示,不同大气污染物与青岛市居民脑卒中日报告发病数关系的曲线形状略有差异,PM2.5、PM10和CO的效应曲线呈先平缓再升高趋势,在极高浓度时效应估计的可信区间较宽,结果不确定性大;O3的效应曲线呈先升后降之后趋于平缓再上升趋势,在极高浓度时效应估计的可信区间亦较宽,结果不确定性大;SO2的效应曲线呈整体平缓下降趋势,但由于可信区间较宽,估计的不确定性较大,但在浓度极高时的可信区间反而变窄;NO2的效应曲线呈整体平缓略有下降趋势,但由于可信区间较宽,估计的不确定性较大。PM2.5、PM10、CO、O3、SO2和NO2等 6种大气污染物与脑卒中日报告发病数的暴露 – 反应关系曲线见图2。其中,PM2.5暴露浓度 > 121.90 μg/m3、CO暴露浓度 > 1.56 mg/m3和O3暴露浓度 < 64.00 μg/m3时脑卒中发病风险均呈上升趋势,而O3暴露浓度 > 64.00 μg/m3时脑卒中发病风险呈上下波动趋势(均P < 0.05)。

    表  3  大气污染物暴露与青岛市居民脑卒中发病关系单污染物模型分析
    Table  3.  Lag day-specific relative risk and its 95% confidence interval of daily stroke incidence number associated with daily average concentration of ambient air PM2.5, O3, CO, PM10, SO2, and NO2: Poisson distributed Lag nonlinear model analysis on disease surveillance and environmental/meteorological monitoring data in Qingdao city from 2014 to 2020
    大气污
    染物
    Lag0 dLag1 dLag2 dLag3 dLag4 dLag5 dLag6 dLag7 d
    RR95%CIRR95%CIRR95%CIRR95%CIRR95%CIRR95%CIRR95%CIRR95%CI
    PM2.50.9940.973~1.0151.0040.991~1.0171.0131.002~1.0241.0181.005~1.0311.0181.005~1.0301.0121.002~1.0231.0030.991~1.0160.9920.973~1.012
    PM101.9690.798~4.8550.9310.451~1.9211.8750.960~3.6592.0271.232~3.3341.6971.022~2.8191.4000.872~2.2471.1470.775~1.6980.9360.501~1.750
    O30.9580.876~1.0471.0050.944~1.0691.0640.994~1.1391.1291.057~1.2061.1551.080~1.2341.1271.052~1.2081.0731.008~1.1421.0110.924~1.106
    CO0.9810.953~1.0111.0010.983~1.0201.0181.004~1.0331.0301.013~1.0471.0331.016~1.0501.0271.013~1.0411.0160.999~1.0321.0010.975~1.029
    SO20.9770.548~1.7440.9530.609~1.4900.9920.682~1.4421.0740.779~1.4801.1750.825~1.6731.2940.960~1.7431.4311.049~1.9511.5860.938~2.682
    NO21.0090.994~1.0251.0060.994~1.0191.0030.993~1.0130.9990.991~1.0070.9960.989~1.0040.9930.984~1.0020.990.979~1.0010.9860.973~1.000
    下载: 导出CSV 
    | 显示表格

    单污染物模型分析结果显示,PM2.5、PM10、O3、CO和SO2暴露与青岛市居民脑卒中的发病风险的易感滞后期略有差别,PM2.5和CO暴露与脑卒中发病风险升高的易感滞后期为Lag2 d~Lag5 d,O3暴露与脑卒中发病风险升高的易感滞后期为Lag3 d~Lag6 d,PM10暴露与脑卒中发病风险升高的易感滞后期为Lag3 d~Lag4 d,SO2暴露与脑卒中发病风险升高的易感滞后期为Lag6 d,具体见表3。其中,PM2.5暴露浓度每升高32.01 μg/m3、PM10暴露浓度每升高52.21 μg/m3,脑卒中Lag3 d发病的RR值分别为1.018(95%CI = 1.005~1.031)、2.027(95%CI = 1.232~3.334);O3暴露浓度每升高53.00 μg/m3、CO暴露浓度每升高0.39 mg/m3,脑卒中Lag4 d发病的RR值分别为1.155(95%CI = 1.080~1.234)、1.033(95%CI = 1.016~1.050),SO2暴露浓度每升高14.93 μg/m3,脑卒中Lag6 d发病的RR值为1.431(95%CI = 1.049~1.951)。

    表  4  敏感性分析
    Table  4.  Relative risk and its 95% confidence interval of daily stroke incidence number associated with daily average concentration of ambient air PM2.5, O3, CO, PM10, SO2, and NO2 by 4, 5, and 6 degrees of freedom of daily average temperature, atmospheric pressure, and relative humidity: Poisson distributed Lag nonlinear model analysis on disease surveillance and environmental/meteorological monitoring data in Qingdao city from 2014 to 2020

    大气污染物
    df
    4 5 6
    RR 95%CI RR 95%CI RR 95%CI
    PM2.5 1.019 1.006~1.032 1.019 1.006~1.032 1.019 1.006~1.032
    O3 1.152 1.078~1.231 1.152 1.078~1.232 1.150 1.076~1.230
    CO 1.035 1.018~1.052 1.036 1.019~1.053 1.035 1.018~1.052
    PM10 1.962 1.193~3.227 1.981 1.202~3.267 1.998 1.214~3.291
    SO2 1.425 1.045~1.941 1.431 1.050~1.950 1.429 1.048~1.948
    NO2 1.009 0.994~1.025 1.009 0.993~1.025 1.007 0.992~1.023
    下载: 导出CSV 
    | 显示表格

    敏感性分析结果显示,T、AP和RH的df变为4、5、6时对PM2.5、O3、CO、PM10、SO2和NO2暴露与青岛市居民脑卒中发病风险关系效应值的影响均较小,模型结果较为稳定。

    本研究中,青岛市2014 — 2020年CO、O3、SO2和NO2的平均水平均优于国家环境空气质量标准一级水平[20],而PM2.5和PM10的平均水平均优于国家环境空气质量标准二级水平[20],其中PM10、PM2.5、CO、O3和NO2平均水平低于北京市,SO2平均水平高于北京市[21],但均低于山东省的平均水平[22]。本研究结果显示,在青岛市,夏季大气污染物以O3为主,冬季大气污染物以PM10、PM2.5、SO2、NO2和CO等为主,大气污染物具有明显的季节差异,其中PM10、PM2.5、SO2、NO2和CO均呈“V”形分布,呈“夏低冬高”的季节变化特征,而O3则与此相反。这一结果与吴含[22]和贾伟杰[23]的研究一致,提示应在不同季节针对不同的大气污染物采取相应的防控措施。

    一项针对中国26个城市的病例交叉研究发现,短期颗粒物浓度升高会增加缺血性脑卒中的发病率,且PM2.5和PM10在滞后现象3d后达到其最强效应,PM2.5和PM10每增加47.5 μg/m3和76.9 μg/m3,缺血性脑卒中的住院人数分别增加1.0%和0.8%[13]。与上述研究结果一致,本研究中随着PM2.5浓度的升高,居民的脑卒中发病风险升高,当PM2.5浓度 > 121.90 μg/m3时,脑卒中发病风险呈显著上升趋势(P < 0.05),而PM10的暴露 – 反应关系却无统计学意义,虽然其每升高52.01 μg/m3时,脑卒中的发病风险也在Lag3 d~Lag4 d表现出了一定的升高。这一结果与其他研究中发现的粒径越小的颗粒物致病或致死亡效应越强的结果一致[24]。Chen L等[25]基于全国5家医院的缺血性脑卒中入院数据分析发现,Lag1 d的PM2.5和PM10暴露浓度每升高10 μg/m3,缺血性脑卒中的入院风险分别增加0.7%(95%CI = 0.0%~1.4%)和0.5%(95%CI = 0.1%~0.9%)。

    本研究结果显示,SO2暴露Lag6 d时与青岛市居民脑卒中发病存在正相关关系,这与国内外多项研究结果基本一致[26 - 27],其中对广东省广州市老年人进行的一项研究表明,SO2短期暴露与缺血性脑卒中住院风险升高有关,其浓度每升高10 μg/m3 ,Lag2 d的住院风险升高3.39%(95%CI = 1.63%~5.18%)[26]。在本研究中,O3和CO暴露亦与脑卒中发病呈正相关。吴含[22]对山东省居民的研究结果表明,O3和CO暴露浓度每升高0.42 mg/m3和73.0 μg/m3,当日缺血性脑卒中发病的RR值分别为1.008(95%CI = 1.005~1.010)和1.004(95%CI = 1.000~1.007),与本研究结果基本一致。但目前关于O3和CO对脑卒中发病影响的研究结果并不一致[28],天津市的一项研究结果显示CO对脑卒中发病的影响无统计学意义[29],而贾伟杰[23]对江西省南昌市的研究结果则显示O3对脑卒中发病的影响局限在60~69岁年龄组人群。

    此外,本研究结果还显示NO2暴露与青岛市居民脑卒中发病无显著相关性,也有一些其他关注心脑血管疾病相关结局的研究报道了类似的结果。如,Chen G等[30]利用Cox风险模型探讨缺血性卒中与空气污染关系的研究发现NO2与缺血性脑卒中的发生关系无统计学意义;山西省太原市的一项研究亦表明短期暴露于NO2与脑卒中的发病无显著相关性[31]。但山东省济南市的研究结果却显示NO2对脑卒中发病存在影响[32];另外一项纳入68篇文献的荟萃分析结果亦显示,NO2每升高10 μg/m3,脑卒中发病率的HR值为1.002(95%CI = 1.000~1.003)[27]。出现此种差异的原因可能是本研究中所采用的为脑卒中患者的报告日期而非发病时间,从脑卒中的发病到入院报告可能会间隔几小时甚至几天,从而造成大气污染物暴露短期效应的低估,从而使短期内大气污染物暴露与脑卒中发病无显著相关性。

    综上所述,PM2.5、O3、CO、PM10和SO2暴露对青岛市居民脑卒中发病存在滞后效应。为降低脑卒中发病危害,减轻家庭和社会的经济负担,有必要开展大气污染治理,合理分配卫生资源,从而降低大气污染物对居民脑卒中发病的效应影响。但本研究仍存在一定的局限性。首先,本研究以青岛生态环境监测中心大气污染物的日平均浓度作为个体暴露的估计,未分析个体水平上的大气污染物暴露,不能体现居民的真实暴露浓度,对结果可能产生误差;其次,本研究是基于群体水平而不是个体水平探讨大气污染物与居民脑卒中发病之间的关系,可能存在生态学谬误,且缺少按人群特征进行分层分析;再次,本研究未调整与居民脑卒中发病相关的其他风险因素,如烟草暴露、饮食行为、身体活动情况和患病史等;最后,本研究使用的脑卒中患者资料来源于国家慢性病监测网络报告系统,前期系统覆盖不完善导致的病例信息缺失也可能会对研究结果产生一定影响。因此,在今后的研究中可收集更多的资料按病因、疾病亚型和人群特征进行分层分析以对本研究结果加以验证。

  • 图  1   青岛市2014 — 2020年逐日大气污染物浓度和气象因素的时间序列图

    Figure  1.   Daily average values of ambient air PM2.5, O3, CO, PM10, SO2, NO2 and temperature, atmospheric pressure, relative humidity, precipitation, wind speed in Qingdao city from 2014 to 2020

    图  2   青岛市2014 — 2020年大气污染物与脑卒中日报告发病数的暴露 – 反应关系

    注:图中曲线为RR值,阴影部分为其95%CI;图中横直线代表RR值为1.0。

    Figure  2.   Relative risk (red line) and its 95% confidence interval (grey area) of daily stroke incidence number associated with daily average concentration of ambient air PM2.5, O3, CO, PM10, SO2, and NO2: Poisson distributed Lag nonlinear model analysis on disease surveillance and environment/meteorological monitoring data in Qingdao city from 2014 to 2020

    表  1   青岛市2014 — 2020年大气污染物和气象因素基本特征

    Table  1   Mean, minimum, 25th percentile, median, 75th percentile, maximum, and interquartile range of daily average values of ambient air pollutants and temperature, atmospheric pressure, relative humidity, precipitation, wind speed in Qingdao city from 2014 to 2020

    项目 分类 $ \overline{x}\pm s $ Min P25 M P75 Max IQR
    大气污染物 PM2.5(μg/m3 44.01 ± 31.09 4.28 23.38 34.96 55.39 273.13 32.01
    O3(μg/m3 99.86 ± 37.39 24.12 70.17 96.38 123.17 272.04 53.00
    CO( mg/m3 0.77 ± 0.37 0.24 0.52 0.67 0.91 3.23 0.39
    PM10(μg/m3 81.63 ± 45.87 10.80 49.83 70.22 102.04 412.46 52.21
    SO2(μg/m3 17.89 ± 15.67 3.33 7.46 13.26 22.39 169.04 14.93
    NO2(μg/m3 32.17 ± 14.22 4.21 21.72 29.79 40.21 98.35 18.49
    气象因素 T(℃) 13.79 ± 9.53 – 12.06 5.19 14.81 22.33 31.00 17.14
    AP(hPa) 1009.70 ± 9.25 981.30 1001.80 1009.80 1017.20 1034.10 15.40
    RH(%) 67.85 ± 15.34 24.00 56.50 69.40 80.10 98.80 23.60
    P(mm) 1.76 ± 7.31 0.00 0.00 0.00 0.11 144.52 0.11
    WS(m/s) 2.88 ± 1.03 0.99 2.14 2.67 3.43 7.93 1.29
    下载: 导出CSV

    表  2   青岛市2014 — 2020年大气污染物与气象因素相关性分析

    Table  2   Spearman product-moment correlation coefficients among daily average values of ambient air PM2.5, O3, CO, PM10, SO2, NO2 and temperature, atmospheric pressure, relative humidity, precipitation, wind speed in Qingdao city from 2014 to 2020

    项目 PM2.5 O3 CO PM10 SO2 NO2 T AP RH P WS
    PM2.5 1.000
    O3 – 0.167 a 1.000
    CO 0.918 a – 0.295 a 1.000
    PM10 0.914 a – 0.086 a 0.827 a 1.000
    SO2 0.626 a – 0.272 a 0.721 a 0.598 a 1.000
    NO2 0.751 a – 0.254 a 0.819 a 0.758 a 0.657 a 1.000
    T – 0.395 a 0.652 a – 0.518 a – 0.351 a – 0.448 a – 0.471 a 1.000
    AP 0.304 a – 0.577 a 0.426 a 0.279 a 0.380 a 0.480 a – 0.856 a 1.000
    RH – 0.063 a 0.061 a – 0.133 a – 0.255 a – 0.282 a – 0.379 a 0.397 a – 0.457 a 1.000
    P – 0.166 a – 0.050 a – 0.153 a – 0.209 a – 0.136 a – 0.210 a 0.154 a – 0.246 a 0.311 a 1.000
    WS – 0.194 a – 0.138 a – 0.220 a – 0.131 a – 0.125 a – 0.353 a – 0.202 a 0.029 – 0.130 a 0.147 a 1.000
      注:a P < 0.05。
    下载: 导出CSV

    表  3   大气污染物暴露与青岛市居民脑卒中发病关系单污染物模型分析

    Table  3   Lag day-specific relative risk and its 95% confidence interval of daily stroke incidence number associated with daily average concentration of ambient air PM2.5, O3, CO, PM10, SO2, and NO2: Poisson distributed Lag nonlinear model analysis on disease surveillance and environmental/meteorological monitoring data in Qingdao city from 2014 to 2020

    大气污
    染物
    Lag0 dLag1 dLag2 dLag3 dLag4 dLag5 dLag6 dLag7 d
    RR95%CIRR95%CIRR95%CIRR95%CIRR95%CIRR95%CIRR95%CIRR95%CI
    PM2.50.9940.973~1.0151.0040.991~1.0171.0131.002~1.0241.0181.005~1.0311.0181.005~1.0301.0121.002~1.0231.0030.991~1.0160.9920.973~1.012
    PM101.9690.798~4.8550.9310.451~1.9211.8750.960~3.6592.0271.232~3.3341.6971.022~2.8191.4000.872~2.2471.1470.775~1.6980.9360.501~1.750
    O30.9580.876~1.0471.0050.944~1.0691.0640.994~1.1391.1291.057~1.2061.1551.080~1.2341.1271.052~1.2081.0731.008~1.1421.0110.924~1.106
    CO0.9810.953~1.0111.0010.983~1.0201.0181.004~1.0331.0301.013~1.0471.0331.016~1.0501.0271.013~1.0411.0160.999~1.0321.0010.975~1.029
    SO20.9770.548~1.7440.9530.609~1.4900.9920.682~1.4421.0740.779~1.4801.1750.825~1.6731.2940.960~1.7431.4311.049~1.9511.5860.938~2.682
    NO21.0090.994~1.0251.0060.994~1.0191.0030.993~1.0130.9990.991~1.0070.9960.989~1.0040.9930.984~1.0020.990.979~1.0010.9860.973~1.000
    下载: 导出CSV

    表  4   敏感性分析

    Table  4   Relative risk and its 95% confidence interval of daily stroke incidence number associated with daily average concentration of ambient air PM2.5, O3, CO, PM10, SO2, and NO2 by 4, 5, and 6 degrees of freedom of daily average temperature, atmospheric pressure, and relative humidity: Poisson distributed Lag nonlinear model analysis on disease surveillance and environmental/meteorological monitoring data in Qingdao city from 2014 to 2020


    大气污染物
    df
    4 5 6
    RR 95%CI RR 95%CI RR 95%CI
    PM2.5 1.019 1.006~1.032 1.019 1.006~1.032 1.019 1.006~1.032
    O3 1.152 1.078~1.231 1.152 1.078~1.232 1.150 1.076~1.230
    CO 1.035 1.018~1.052 1.036 1.019~1.053 1.035 1.018~1.052
    PM10 1.962 1.193~3.227 1.981 1.202~3.267 1.998 1.214~3.291
    SO2 1.425 1.045~1.941 1.431 1.050~1.950 1.429 1.048~1.948
    NO2 1.009 0.994~1.025 1.009 0.993~1.025 1.007 0.992~1.023
    下载: 导出CSV
  • [1] 马林, 巢宝华, 曹雷, 等. 2007 — 2017年中国脑卒中流行趋势及特征分析[J]. 中华脑血管病杂志: 电子版, 2020, 14(5): 253 – 258.
    [2] 吕泰标, 黄瑜亮, 杨逸成, 等. 短期与长期空气污染暴露对脑卒中影响[J]. 中国公共卫生, 2021, 37(4): 760 – 763.
    [3]

    Tu WJ, Wang LD. China stroke surveillance report 2021[J]. Military Medical Research, 2023, 10(1): 33. DOI: 10.1186/s40779-023-00463-x

    [4]

    Feigin VL, Stark BA, Johnson CO, et al. Global, regional, and national burden of stroke and its risk factors, 1990 – 2019: a systematic analysis for the Global Burden of Disease Study 2019[J]. The Lancet Neurology, 2021, 20(10): 795 – 820. DOI: 10.1016/S1474-4422(21)00252-0

    [5]

    Ma Q, Li R, Wang L, et al. Temporal trend and attributable risk factors of stroke burden in China, 1990 – 2019: an analysis for the Global Burden of Disease Study 2019[J]. The Lancet Public Health, 2021, 6(12): e897 – e906. DOI: 10.1016/S2468-2667(21)00228-0

    [6] 王鑫钰, 苏文丽, 黄亚楠, 等. BMI与脑卒中发病风险的剂量 – 反应Meta分析[J]. 中国循证医学杂志, 2022, 22(11): 1330 – 1338.
    [7] 周盛年, 孙晓晗, 周雪颖, 等. 山东省脑卒中流行病学及其危险因素分析[J]. 中华神经科杂志, 2019, 52(9): 716 – 716.
    [8]

    Yuan S, Wang J, Jiang Q, et al. Long-term exposure to PM2.5 and stroke: a systematic review and meta-analysis of cohort studies[J]. Environmental research, 2019, 177: 108587.

    [9]

    Li M, Edgell RC, Wei J, et al. Air pollution and stroke hospitalization in the Beibu Gulf Region of China: a case-crossover analysis[J]. Ecotoxicology and Environmental Safety, 2023, 255: 114814. DOI: 10.1016/j.ecoenv.2023.114814

    [10]

    Tian F, Cai M, Li H, et al. Air pollution associated with incident stroke, poststroke cardiovascular events, and death: a trajectory analysis of a prospective cohort[J]. Neurology, 2022, 99(22): e2474 – e2484.

    [11]

    Verhoeven JI, Allach Y, Vaartjes ICH, et al. Ambient air pollution and the risk of ischaemic and haemorrhagic stroke[J]. The Lancet Planetary Health, 2021, 5(8): e542 – e552. DOI: 10.1016/S2542-5196(21)00145-5

    [12] 秦绪成, 李伟伟, 马昭君, 等. 2014 — 2020年连云港市脑卒中死亡特征和趋势分析[J]. 现代预防医学, 2023, 50(2): 216 – 221.
    [13]

    Liu H, Tian Y, Xiang X, et al. Ambient particulate matter concentrations and hospital admissions in 26 of China′s largest cities: a case–crossover study[J]. Epidemiology, 2018, 29(5): 649 – 657. DOI: 10.1097/EDE.0000000000000869

    [14] 魏常松, 江国虹, 王德征, 等. 2007 — 2015年天津市居民脑卒中发病和死亡流行特征[J]. 中华预防医学杂志, 2018, 52(12): 1300 – 1304.
    [15] 卫生部卫生统计信息中心. 国际疾病分类 (ICD – 10) 应用指导手册[M]. 北京: 协和医科大学出版社, 2001.
    [16]

    Gu J, Shi Y, Chen N, et al. Ambient fine particulate matter and hospital admissions for ischemic and hemorrhagic strokes and transient ischemic attack in 248 Chinese cities[J]. Science of the Total Environment, 2020, 715: 136896. DOI: 10.1016/j.scitotenv.2020.136896

    [17]

    Ban J, Wang Q, Ma R, et al. Associations between short-term exposure to PM2.5 and stroke incidence and mortality in China: a case-crossover study and estimation of the burden[J]. Environmental Pollution, 2021, 268: 115743.

    [18]

    Zhang Y, Zhang L, Wei J, et al. Size-specific particulate air pollution and hospitalization for cardiovascular diseases: a case-crossover study in Shenzhen, China[J]. Atmospheric Environment, 2021, 251: 118271. DOI: 10.1016/j.atmosenv.2021.118271

    [19]

    Zhang Y, Fang J, Mao F, et al. Age- and season-specific effects of ambient particles (PM1, PM2.5, and PM10) on daily emergency department visits among two Chinese metropolitan populations[J]. Chemosphere, 2020, 246: 125723. DOI: 10.1016/j.chemosphere.2019.125723

    [20] 中华人民共和国国家质量监督检验检疫总局, 中国国家标准化管理委员会. GB 3095 — 2012 环境空气质量标准[S]. 北京: 中国环境科学出版社, 2016.
    [21] 赵泽濛, 郭默宁, 谭鹏, 等. 空气污染短期暴露与北京市高血压患者缺血性脑卒中入院关联分析[J]. 中国公共卫生, 2022, 38(10): 1258 – 1263.
    [22] 吴含. 山东省居民大气污染物短期暴露与缺血性脑卒中发病的关联及PM2.5归因发病负担预估研究[D]. 济南: 山东大学, 2022.
    [23] 贾伟杰. 南昌市空气污染物暴露与缺血性脑卒中入院风险的关联研究[D]. 南昌: 南昌大学, 2023.
    [24]

    Wu H, Zhang B, Wei J, et al. Short-term effects of exposure to ambient PM1, PM2.5, and PM10 on ischemic and hemorrhagic stroke incidence in Shandong Province, China[J]. Environmental Research, 2022, 212: 113350.

    [25]

    Chen L, Zhang Y, Zhang W, et al. Short-term effect of PM1 on hospital admission for ischemic stroke: a multi-city case-crossover study in China[J]. Environmental Pollution, 2020, 260: 113776. DOI: 10.1016/j.envpol.2019.113776

    [26] 许瑞君, 叶韵韶, 林巧绚, 等. 大气污染物短期暴露对老年人缺血性脑卒中住院的影响研究[J]. 公共卫生与预防医学, 2022, 33(1): 6 – 11.
    [27]

    Niu Z, Liu F, Yu H, et al. Association between exposure to ambient air pollution and hospital admission, incidence, and mortality of stroke: an updated systematic review and meta-analysis of more than 23 million participants[J]. Environmental health and preventive medicine, 2021, 26: 1 – 14. DOI: 10.1186/s12199-020-00923-z

    [28] 杨逸成, 黎嘉雯, 孙博, 等. 空气污染与心脑血管疾病的研究进展[J]. 中国医学科学院学报, 2022, 44(2): 318 – 323.
    [29] 徐佳辉, 江婷婷, 王士玉, 等. 天津市大气污染物对脑卒中发病影响的时间序列分析[J]. 中国慢性病预防与控制, 2023, 31(3): 188 – 193.
    [30]

    Chen G, Wang A, Li S, et al. Long-term exposure to air pollution and survival after ischemic stroke: the China national stroke registry cohort[J]. Stroke, 2019, 50(3): 563 – 570. DOI: 10.1161/STROKEAHA.118.023264

    [31] 李泽星. 大气污染与卒中发病之间的短期关联: 中国太原的时间序列研究[D]. 太原: 山西医科大学, 2022.
    [32]

    Chen C, Wang X, Lv C, et al. The effect of air pollution on hospitalization of individuals with respiratory and cardiovascular diseases in Jinan, China[J]. Medicine, 2019, 98(22): e15634. DOI: 10.1097/MD.0000000000015634

图(2)  /  表(4)
计量
  • 文章访问数:  190
  • HTML全文浏览量:  17
  • PDF下载量:  64
  • 被引次数: 0
出版历程
  • 收稿日期:  2023-10-06
  • 修回日期:  2024-02-27
  • 录用日期:  2024-07-01
  • 刊出日期:  2024-10-09

目录

/

返回文章
返回