Study On The Association Between Short-term Exposure To Air Pollutants And Ischemic Stroke Incidence And Projection Of Incidence Burden Attributed To PM_2.5 Of Residents In Shandong Province

BackgroundAt present,stroke has become a major global public health issue.Stroke includes both ischemic and hemorrhagic stroke.In recent years,the incidence rate of hemorrhagic stroke in China has shown a continuous downward trend,while the incidence rate of ischemic stroke has increased significantly,accounting for more than 70%of the total incident cases of stroke.Therefore,it is important to understand the current epidemiological characteristics of ischemic stroke incidence in China,and this would play an important role in improving the current prevention and control policies and clarifying the primary direction of future interventions.Although some studies,which were based on national surveys or monitoring data from limited provinces and cities,have reported the current prevalence or epidemic trend of ischemic stroke incidence in China during recent years,no study has been conducted to report the distribution and clustering characteristics of ischemic stroke incidence at the district/county level.Short-term exposure to air pollutants is one of the important factors that could cause ischemic stroke-related morbidity or mortality.However,previous studies have mainly focused on routinely monitored pollutants including fine particulate matter(PM2.5),inhalable particulate matter(PM10),nitrogen dioxide(NO2),sulfur dioxide(SO2),ozone(O3),and carbon monoxide(CO),and only a few studies have focused on the effect of particle matters with aerodynamic diameter<1 μm(PM1)on population health.In addition,using ischemic stroke-related mortality as the outcome would leave out some patients who had an ischemic stroke onset but did not die due to exposure to air pollutants,which may not fully reflect the impact of air pollutants on the population.A few studies have attempted to explore the association between short-term exposure to PM1 and the risk of hospital admission due to stroke based on admission data from a single hospital,but their conclusions were inconsistent.In addition,studies using hospital admission data from single-hospital as source data cannot include patients who visit other medical institutions in this region,which may also affect the accuracy of the results.Since 2013,the Chinese government has successively issued and implemented a series of air pollution prevention and control plans,and the air pollution level has been effectively alleviated.With the further implementation of relevant prevention and control measures such as the "Carbon Peaking Action Plan Before 2030",the air pollution level will continue to be reduced in the future.However,China will also face the problem of population ageing in the future,and the elderly is predisposed to ischemic stroke.Therefore,it is necessary to estimate the ischemic stroke incidence burden related to air pollution exposure in China under the scenario that the exposure level will be gradually reduced in the future(i.e.,the air pollution will be gradually alleviated)and the susceptible population will be gradually increased(i.e.,the number of the elderly population will be increased).At the same time,accurate assessment of the health benefits of these management measures is also an important reference for formulating and adjusting relevant future policy plans.Given the limitations of previous studies and the availability of research data,this study took Shandong Province as the study area and conducted three parts of studies.We aimed to provide a scientific basis for finding out the clustering areas of ischemic stroke incidence of residents in Shandong Province,improving the current air pollutant monitoring work,adjusting schemes for future air pollution prevention and control,and formulating strategies and measures for future population ischemic stroke intervention and prevention.Objectives1.Based on the surveillance data of ischemic stroke incidence of residents in all districts/counties in Shandong Province from 2017 to 2019,we aimed to describe the distribution of incidence among years,gender,and age groups,and study the geographical distribution and spatial clustering characteristics of ischemic stroke incidence at the district/county level.2.To quantify the association between short-term exposure to PM1,PM2.5,PM10,NO2,SO2,O3,and CO and the risk of ischemic stroke incidence among residents in Shandong Province,and to explore the discrepancies of association among different subgroups and identify potential effect modifiers at the district/county level.In addition,we aimed to evaluate the attributable incident cases and attributable fraction related to short-term exposure to each air pollutant.3.To project the attributable incident cases and attributable fraction of ischemic stroke related to short-term exposure to PM2.5 among residents in Shandong Province under different climate and air pollution management scenarios,and then they were compared with those in the baseline,aiming to understand the changing conditions of attributable incidence burden related to short-term exposure to PM2.5.Methods1.Data collectionDistrict/county-specific daily incidence of ischemic stroke data during 2017 and 2019 was obtained from the Shandong Center for Disease Control and Prevention via the noncommunicable disease surveillance system.Relevant cases were reported by medical institutions at all levels in each district/county across the province,and then were reviewed by the county-level Centers for Disease Control and Prevention.Cases were supplemented using the mortality registration data and data from other sources.China’s 2010 population census data was obtained from the National Bureau of Statistics.Daily air pollutant data covering Shandong Province were collected from the product of China High Air Pollutants.Meteorological data were collected from the European Centre for Medium-Range Weather Forecasts ERA5-Land climate reanalysis product.The gridded air pollutant and meteorological data were processed,and the daily means of air pollutant concentrations,temperature,and relative humidity were calculated as the daily exposure level of residents in each district/county.The population data in 2030 and 2060 were obtained from the Shared Socioeconomic Pathways(SSP),which was a part of the world population projections by the Intergovernmental Panel on Climate Change.PM2.5 concentration data under six scenarios in 2030 and 2060 were obtained from the projected data from the Dynamic Projection model for Emissions in China(DPEC).Under DPEC,Scenario 1 assumes China will adopt a relatively ineffective climate and clean air management pattern in the future,and the management measures in Scenario 2 to Scenario 6 will be more and more stringent.Scenario 5 assumes China will adopt relevant management measures to achieve the "carbon neutrality" and other goals,and Scenario 6 assumes China will achieve the goal of the annual PM2.5 level of 10 μg/m3 and control the range of temperature rise within 1.5℃.2.Statistical analysisThe ischemic stroke incidence rate of residents in Shandong Province was agestandardized based on the population census data,which was used to describe the distribution of incidence by year,gender,and age groups.The global autocorrelation analysis was used to judge whether there was an overall spatial correlation of ischemic stroke incidence among districts/counties.The local autocorrelation analysis was used to identify the spatial correlation of ischemic stroke incidence in local areas.The spatial clustering analysis was used to identify the clustering area of the high incidence of ischemic stroke.The two-stage model was used,and in the first stage,a distributed lag nonlinear model(DLNM)based on single-pollutant model was used to test the association between short-term exposure to PM1,PM2.5,PM10,NO2,SO2,O3 and CO and the risk of ischemic stroke incidence at the district/county level.In the second stage,the random effect meta-analysis was used to pool the effect estimate of each district/county to obtain the overall effect estimates at the province level,and then the susceptible lag period of each pollutant for the risk of ischemic stroke incidence was identified and the cumulative effect was calculated.Based on the results of the cumulative effect,the differences among subgroups were examined,and meta-regression was used to identify potential effect modifiers for the observed association at the district/county level.A series of sensitive analyses,such as changing parameters of parts of the variables in DLNM and constructing two-pollutant models,were performed.The attributable incident cases and attributable fraction were used as indicators to denote the attributable incidence burden related to short-term exposure to each pollutant.The attributable incident cases and attributable fraction related to exposure to each pollutant were calculated,and the daily limits of air pollutants recommended by World Health Organization(WHO)global air quality guidelines(AQG)and 0 μg/m3 were used as reference values,respectively.Monte Carlo simulations were used to estimate the 95%empirical confidence interval(eCI)of attributable indicators.With reference to the relevant hands-on guideline,the corresponding research assumptions were set before the projection.The baseline population data,daily incidence surveillance data,the population data in 2030 and 2060,and the daily PM2.5 concentrations of each district/county in Shandong Province under the six scenarios of DPEC were combined.Then,based on the association between short-term exposure to PM2.5 and the risk of ischemic stroke incidence from baseline data,we calculated the attributable fraction,attributable incident cases,and the amount and percentage of the changes in incident ischemic stroke cases compared with those of the baseline(i.e.,the three-year average of 2017-2019)for the total population and subgroups by sex and age in 2030 and 2060 under different scenarios,respectively.Monte Carlo simulations were used to estimate the 95%eCI of each attributable indicator.Results1.During 2017 and 2019,the number of ischemic stroke incident cases in Shandong Province was 1,310,620.The incident cases of ischemic stroke increased year by year in these three years,and the crude and age-standardized incidence of the overall,male,and female residents all increased.There was an upward trend in the incidence of ischemic stroke beginning from the age group of 45 to 49 years,and the incidence rates of ischemic stroke in males were higher than those of females in all age groups.The standardized incidence rate was lower in eastern Shandong,especially in coastal areas,while the standardized incidence rate in western and southern Shandong was relatively higher.The results of global autocorrelation analysis showed that the distribution of ischemic stroke incidence in Shandong Province was positively spatial autocorrelated.The local autocorrelation analysis revealed that the hot spot regions were mainly in the northwestern,central-western,and southeastern Shandong,and some districts/counties in the eastern coast and central Shandong have relatively low ischemic stroke incidence rate while their adjacent regions have relatively high ischemic stroke incidence rate.The clusters identified from the results of spatial clustering analysis were similar to the hot spot regions identified by local autocorrelation analysis.Cluster 1 included the district of Decheng,Ningjin,Linyi,Pingyuan,and Wucheng in Dezhou city;Cluster 2 included the district of Shizhong,Huaiyin,Changqing in Jinan city,Wenshang in Jining city,Ningyang and Feicheng in Tai’an city;Cluster 3 included the district of Luozhuang,Hedong,Yinan,and Tancheng in Linyi city.2.In single-lag-day of single-pollutant models,there were positive associations between short-term exposure to air pollutants and ischemic stroke incidence.The susceptible periods of PM1,PM2.5,PM10 exposure and the onset of ischemic stroke were lag0 to lag5,the susceptible periods of NO2,SO2,and CO exposure were lag0 to lag3,and the susceptible periods of O3 exposure were lag0 and lag1.The results of the cumulative effect analysis showed that the relative risks(RR)and 95%confidence interval(CI)of ischemic stroke incidence related to per interquartile range(IQR)increase in PM1,PM2.5,and PM10 concentrations at lag05 were 1.026(95%CI:1.021～1.032),1.025(95%CI:1.020～1.029),1.019(95%CI:1.014～1.024),respectively;the RR of ischemic stroke incidence related to per IQR increase in NO2,SO2,and CO concentrations at lag03 were 1.027(95%CI:1.020～1.034),1.024(95%CI:1.018～1.030),and 1.020(95%CI:1.015～1.025),respectively;the RR of ischemic stroke incidence related to per IQR increase in O3 concentration at lag01 was 1.006(95%CI:1.001-1.011).The results of subgroup analysis indicated that the associations between exposure to PM1,PM2.5,PM10,NO2,SO2,and CO and the risk of ischemic stroke incidence were relatively larger in magnitude in females than those in males,and the associations were relatively larger in magnitude in the elderly aged 65-74 and≥75 years than those aged<65 years.Exposure to O3 was positively associated with the risk of ischemic stroke incidence only in the elderly aged≥75 years.Although not all differences of the association between subgroups were statistically significant,the results revealed that females and the elderly were potentially susceptible populations.NO2,SO2,and CO exposure levels were positively associated with the risk of ischemic stroke only in the cold season,while positive association between O3 and ischemic stroke was only observed in the warm season.Results of the sensitive analysis showed that the associations between exposure to most of the air pollutants and the risk of ischemic stroke incidence were basically robust.The results of effect modifier analysis showed that the proportion of the elderly population at the district/county level had a modifying effect on the associations between exposure to PM1,PM2.5,and PM10 and the risk of ischemic stroke incidence.The associations were larger in magnitude in districts/counties with a higher proportion of the elderly population.Poverty at the district/county level had a modifying effect on the associations between exposure to all seven air pollutants and the risk of ischemic stroke incidence,and the associations were larger in magnitude in districts/counties with a higher area deprivation index.When the daily limits of air pollutants recommended by WHO AQG were used as reference values,the attributable incident cases(12,525,95%eCI:9616～15,386)and attributable fraction(2.9%,95%eCI:2.2%-3.5%)of ischemic stroke related to short-term PM2.5 exposure was the largest.When 0 μg/m3 was set as the reference value,the attributable incident cases(21,126,95%eCI:15,447～26,681)and attributable fraction(4.8%,95%eCI:3.5%～6.1%)of ischemic stroke related to short-term PM1 exposure was the largest.In addition,among the three types of particulate matter,particles with smaller size were associated with a larger attributable incident cases and attributable fraction of ischemic stroke.Among other air pollutants,the attributable incident cases(20,204,95%eCI:13,927～26,343)and attributable fraction(4.6%,95%eCI:3.2%～6.0%)of ischemic stroke related to short-term exposure to NO2 was the largest.3.Under the three pathways of SSP1,SSP2 and SSP4,the proportion of the elderly population(≥65 years)in the total population in Shandong Province would be 18.0%,16.9%,17.3%in 2030,and 42.1%,34.1%,37.4%in 2060,respectively.The proportion of incident ischemic stroke cases of the elderly population in the total population would be 74.0%,73.1%,73.0%in 2030,and 89.1%,87.1%,87.1%in 2060,respectively.In Scenario 1 to Scenario 6 of the DPEC,the total attributable incident cases and attributable fraction would be the largest in Scenario 1,and those would be the smallest in Scenario 6.Compared with the annual mean attributable fraction in 2017-2019 as the baseline,the attributable fractions in the corresponding scenarios would be smaller in the scenarios with stricter management measures.Compared with the annual mean of attributable incident cases in 2017-2019 as the baseline,the increments of the total attributable incident cases of ischemic stroke related to short-term exposure to PM2.5 of residents in Shandong Province in 2030 and 2060 would be the largest in Scenario 1.With the strengthening of the management measures,the increases in the attributable incident cases in the corresponding scenarios in 2030 and 2060 would be gradually receded.The total attributable incident cases of ischemic stroke related to short-term exposure to PM2.5 of residents in Shandong Province in 2060 would be decreased in Scenarios 5 and 6,and Scenario 6 would have the largest decrease(-35.1%,95%eCI:-48.1%～-21.4%).In addition,the attributable incident cases of the elderly would be apparently decreased in Scenario 6,with a decrease of 13.2%(95%eCI:-21.6%～-4.6%).At the district/county level,in Scenario 1,135 and 136 districts/counties in 2030 and 2060 would have increased attributable incident cases with varying degrees,respectively.In Scenario 6,there would be an increase in the attributable incident cases in 66 districts/counties in 2030,and in only 12 districts/counties in 2060.Conclusions1.There was an upward trend in the ischemic stroke incidence of residents in Shandong Province from 2017 to 2019.The distribution of ischemic stroke incidence in Shandong Province was spatially clustered at the district/county level,and the clustering areas were mainly in the northwestern,central-western,and southeastern Shandong.In the future,further population surveys should be carried out in these areas to explore the specific causes of the high-incidence of ischemic stroke clustering areas.Then targeted prevention and intervention should be carried out based on relevant clues to reduce the ischemic stroke incidence of residents in these areas.2.There were positive associations between short-term exposure to PM1,PM2.5,PM10,NO2,SO2,O3,and CO and risk of ischemic stroke incidence of residents in Shandong Province,but the susceptible period varied across these air pollutants.Females and the elderly were potentially susceptible populations.The associations were larger in magnitude in districts/counties with a higher proportion of the elderly population or with a higher area deprivation index.The attributable incident cases and attributable fraction of ischemic stroke related to short-term exposure to PM1 were the largest among all air pollutants.In the future,areas with a high proportion of the elderly population or poor areas should be focused on when carrying out interventions on improving coping ability with air pollution of the population.Relevant departments should also strengthen the monitoring of ambient PMi,and reduce the health hazards caused by air pollution exposure.3.In a long period in the future,the proportion of the elderly population in Shandong Province will continue to increase.If China insists on implementing effective clean air policies in the future,the attributed incidence burden of ischemic stroke related to short-term exposure to PM2.5 caused by the ageing population would be reduced.If the "carbon neutrality" goal can be achieved on schedule,the attributable incident cases of ischemic stroke related to short-term exposure to PM2.5 in the total population would be lower in 2060 than that in the baseline.If energy transition can be strengthened at the same time to control the extent of temperature rise,the health benefits would be more pronounced.Relevant departments should make arrangements in advance,plan population and socioeconomic development rationally,and strengthen the health protection of the elderly to fight against PM2.5 exposure,to scientifically cope with the health challenges brought about by an ageing population.