restrictingww.00271.com
ww.00271.com 时间:2021-03-23 阅读:()
Longtermexposuretoambientairpollutionandincidenceofacutecoronaryevents:prospectivecohortstudyandmeta-analysisin11EuropeancohortsfromtheESCAPEProjectOPENACCESSGiuliaCesaroniseniorresearcher1,FrancescoForastiereresearchdirector1,MassimoStafoggiaseniorresearcher1,ZoranaJAndersenassociateprofessorinepidemiology23,ChiaraBadaloniresearchfellow1,RobBeelenseniorresearcher4,BarbaraCaraccioloresearcher56,UlfdeFaireseniorprofessorofcardiovascularepidemiology7,RaimundErbelprofessor8,KirstenTEriksenresearcher2,LauraFratiglioniprofessoringeriatricepidemiology5910,ClaudiaGalassimedicalepidemiologist11,ReginaHampelresearchfellow12,MargitHeierresearchfellow1213,FraukeHennigresearchfellow14,AgnetaHildingresearcher15,BarbaraHoffmannprofessor1416,DannyHouthuijsseniorresearcher17,Karl-HeinzJckelprofessor18,MichalKorekdoctoralstudent7,TimoLankichiefresearcher19,KarinLeanderresearcher7,PatrikKEMagnussonprofessor20,EnricaMiglioreepidemiologist11,Caes-GranOstensonprofessor15,KimOvervadprofessor2122,NancyLPedersenprofessorofgeneticepidemiology20,JuhaPekkanenJprofessor19,JohannaPenellresearcher7,GranPershagenprofessor7,AndreiPykoresearchfellow7,OleRaaschou-Nielsenheadofresearchgroup2,AndreaRanziprojectmanagerinenvironmentalepidemiology23,FulvioRicceriresearchfellow24,CarlottaSacerdotemedicalepidemiologist11,VeikkoSalomaaresearchprofessor25,WimSwartresearcher17,AnuWTurunenresearcher19,PaoloVineisprofessorinepidemiology2426,GudrunWeinmayrresearchassociate1427,KathrinWolfresearchfellow12,KeesdeHooghseniorresearchofficer26,GerardHoekassociateprofessor4,BertBrunekreefprofessor428,AnnettePetersprofessor121DepartmentofEpidemiology,LazioRegionalHealthService,00198Rome,Italy;2DanishCancerSocietyResearchCenter,2100Copenhagen,Denmark;3CenterforEpidemiologyandScreening,DepartmentofPublicHealth,UniversityofCopenhagen,1014Copenhagen,Denmark;4InstituteforRiskAssessmentSciences,UtrechtUniversity,POBox80178,3508TDUtrecht,Netherlands;5AgingResearchCenter,DepartmentofNeurobiology,CareSciencesandSociety,KarolinskaInstitutet,S-11330Stockholm,Sweden;6StressResearchInstitute,StockholmUniversity,SE-10691Stockholm,Sweden;7InstituteofEnvironmentalMedicine,KarolinskaInstitutet,Box210SE-17177Stockholm,Sweden;8WestGermanHeartCenter,UniversityHospitalofEssen,45122Essen,Germany;9StockholmGerontologyResearchCenter,11330Stockholm,Sweden;10DivisionofClinicalGeriatrics,KarolinskaUniversityHospital,SE-14186Stockholm,Sweden;11UnitofCancerEpidemiology,"CittàdellaSaluteedellaScienza"Hospital,UniversityofTurin,andCenterforCancerPreventionPiemonte,10126Turin,Italy;12HelmholtzZentrumMünchen,GermanResearchCenterforEnvironmentalHealth,InstituteofEpidemiologyII,85764Neuherberg,Germany;13CentralHospitalofAugsburg,MONICA/KORAMyocardialInfarctionRegistry,D-86156Augsburg,Germany;14IUF-LeibnizResearchInstituteforEnvironmentalMedicine,40225Düsseldorf,Germany;15DepartmentofMolecularMedicineandSurgery,KarolinskaInstitutet,KarolinskaUniversityHospital,SE-17177Stockholm,Sweden;16MedicalFaculty,UniversityofDüsseldorf,40225Düsseldorf,Germany;17NationalInsitituteforPublicHealthandtheEnvironment,POBox1,3720BABilthoven,Netherlands;18InstituteforMedicalInformatics,BiometryandEpidemiology,UniversityHospitalofEssen,D-45122Essen,Germany;19NationalInstituteforHealthandWelfare"THL",DepartmentofEnvironmentalHealth,POBox95,FI-70701Kuopio,Finland;20DepartmentofMedicalEpidemiologyandBiostatistics,KarolinskaInstitutet,SE-17177Stockholm,Sweden;21SectionforEpidemiology,DepartmentofPublicHealth,AarhusUniversity,DK-8000Aarhus,Denmark;22DepartmentofCardiology,CardiovascularResearchCenter,AalborgUniversityHospital,DK-9000Aalborg,Denmark;23EnvironmentalHealthReferenceCentre-RegionalAgencyforEnvironmentalPreventionofEmilia-Romagna,41121Modena,Italy;24MolecularandGeneticEpidemiologyUnit,HuGeF-HumanGeneticsFoundation-Turin,10126Turin,Italy;25THL-NationalNocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page1of16ResearchRESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromInstituteforHealthandWelfare,DepartmentofChronicDiseasePrevention,POB30,FI-00271Helsinki,Finland;26MRC-HPACentreforEnvironmentandHealth,DepartmentofEpidemiologyandBiostatistics,ImperialCollegeLondon,StMary'sCampus,LondonW21PG,UK;27InstituteofEpidemiologyandMedicalBiometry,UlmUniversity,89069Ulm,Germany;28JuliusCenterforHealthSciencesandPrimaryCare,UniversityMedicalCenterUtrecht,POBox85500,3508GAUtrecht,NetherlandsAbstractObjectivesTostudytheeffectoflongtermexposuretoairbornepollutantsontheincidenceofacutecoronaryeventsin11cohortsparticipatingintheEuropeanStudyofCohortsforAirPollutionEffects(ESCAPE).
DesignProspectivecohortstudiesandmeta-analysisoftheresults.
SettingCohortsinFinland,Sweden,Denmark,Germany,andItaly.
Participants100166peoplewereenrolledfrom1997to2007andfollowedforanaverageof11.
5years.
Participantswerefreefrompreviouscoronaryeventsatbaseline.
MainoutcomemeasuresModelledconcentrationsofparticulatematter5000vehicles/daymultipliedbythelengthofthoseroadsina100mbuffer.
Tovalidatethemodels,weusedtheleaveoneoutcrossvalidationmethod—thatis,systematicallysubtractingeachofthemonitoringpointsfromthemodelonebyone,andthencomparingthepredictedvalueforeachmonitoringlocationwiththemeasuredlevelatthelocationwithoutusingthismeasurementinthedevelopmentofthemodel.
3031CovariatesAllcohortshadacommonsetofpotentialconfoundersandeffectmodifiersatbaseline:maritalstatus(codedassingle,married/livingwithpartner,divorced/separated,widowed;fortheSDPPcohortonlythebinaryvariable"livingwithpartner"wasavailable;forSIDRIA-Romeallparticipantswerelivingascouplesatbaseline),education(primaryschoolorless,uptosecondaryschoolorequivalent,universitydegreeandmore),occupation(employed,unemployed,homemaker/housewife,retired),smokingstatus(current,former,never),durationofsmoking(years),smokingintensity(cigarettes/day),hypertension,anddiabetes.
Mostcohortshadinformationonadditionalcardiovascularriskfactorssuchasbodymassindex(BMI;codedas2hours/week),alcoholconsumption(never,1-3drinks/week,3-6drinks/week,>6drinks/week),andthreecohortshadinformationonuseofdrugsforhormonereplacementtherapy.
Fourof11cohortshaddataoncholesterolconcentrations.
Tenof11cohortshaddataonnoiseexposure,estimatedattheresidentialaddressesfromEuropeannoiseexposureassessmentin2007,andweusedaneightclasscategoricalvariableof5dBofexposurefrom45dBto75dBandmore(seeappendix).
Eachcohortalsousedanareabasedsocioeconomicstatusindicatorbecausesocioeconomicindicatorsatthearealevelarepredictorsofmorbidity,accesstocare,andlifestyleriskfactorsinadditiontoindividualsocioeconomiccharacteristics.
32StatisticalanalysesToevaluatetheassociationbetweenexposuretoairpollutionandincidenceofcoronaryeventsweperformedtheanalysesintwostages.
Firstly,weanalysedeachcohortusingacommonprotocolforconfounders,outcomes,andstatisticalmodelling.
WeusedCoxproportionalhazardsregressionmodels(hazardratios)withageasthetimescale.
Ascriptwithstatisticalcodewasprovidedtoallcohortspecificanalysts.
TheresultswereevaluatedcentrallyattheDepartmentofEpidemiologyinRome.
Secondly,wecarriedoutarandomeffectsmeta-analysistopoolresults.
33WecalculatedI2statisticsandPvaluesforχ2testfromCochran'sQtoquantifytheheterogeneityamongstudies.
34Wefirstestimatedhazardratiosadjustedforage,sex,andyearofenrolmentonly(model1),thenadjustedforacommonsetofindividualcovariates(maritalstatus,education,occupation,smokingstatus,smokingduration,andsmokingintensity;model2),andthenadjustedforareabasedsocioeconomicstatusindicators(model3).
WheninvestigatingtheeffectsoftrafficvariableswealsoadjustedallmodelsforbackgroundNO2levels.
Weestimatedtheeffectsofexposuretoairpollutionusingfixedincrementsofpollutants(20g/m3forNOx,10g/m3forPM10andNO2,5g/m3forPM2.
5andPMcoarse,105/mforPM2.
5absorbance).
Weperformedseveralsensitivityanalysestodealwithpotentialsourcesofbiasandtoconsidertheconfoundingroleofadditionalcardiovascularriskfactors.
Firstly,weadjustedforpossibleintermediatevariables(diabetes,hypertension)availableinallthecohorts;thenforphysicalactivity,alcoholconsumption,andBMI(availableineightcohorts);finallyweaddedtopreviouslymentionedfactorsthecholesterolconcentration(availableinfourcohorts).
Wethenanalysedtheroleoflivinginlow/highurbanisedsettings,ofnoiseexposure,andofresidentialstability(restrictingtheanalysistothoselivingatthesamebaselineaddressduringthefollow-up).
WestratifiedtheCoxmodel3forpredictorsthatdidnotmeettheproportionalityhazardassumption.
WeanalysedtheinfluentialroleofthelargestcohortbyexcludingtheDCHstudyfromtheanalysis.
Finally,weevaluatedtheroleoftheperformanceoflanduseregressionmodels,stratifyingcohortsbycrossvalidationR2>or≤60%.
Toexploretheeffectofclusteringbyareaofresidenceintheassociationbetweenexposureandmortality—thatis,residentsinthesameareausuallysharesimilarcharacteristics(socioeconomicstatus,health,accesstoservices)andhavesimilarenvironmentalandairconditions—weperformedafrailtymodeltomeasuretheroleeachareaplayed.
4Weevaluatedpotentialeffectmodificationbyaddingtomodel3aninteractiontermofexposureandoneeffectmodifierataNocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page3of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
Downloadedfromtime(sex,educationallevel,smokingstatus,BMI,hypertension,andresidenceinloworhighurbanisedsettings).
Weusedthelikelihoodratiotesttocomparethemodelswithandwithoutinteractionterms.
Forageduringfollow-up,weestimatedeffectmodificationbyaddinganinteractiontermbetweenexposureandatimedependentcategoricalvariableindicatingagegroup(75).
Toprovideinformationaboutthehealtheffectsbelowspecificthresholdvalues(20,30,and40g/m3forPM10,and15,20,and25g/m3forPM2.
5),westudiedtheeffectoffixedincrementsofPM2.
5andPM10amongpeoplewithairpollutionconcentrationsatresidencesonlybelowthesethresholds.
WeusedSTATAsoftware(versions10,11,and12)foralltheanalyses,withtheexceptionoffrailtymodelsforwhichweusedR(www.
r-project.
org).
ResultsTable1showsasummarydescriptionofthe11Europeancohorts.
Table2showsindividualcharacteristicsofparticipantsandtable3theirexposurestoairpollution.
Additionalcharacteristicsoftheparticipantsareprovidedintable4.
Theenrolmentperiodcovered15years,andtheaveragefollow-upwas11.
5years.
Theparticipatingcohortsvariedincharacteristics,availabilityofdataoncovariates,andlevelsofexposure.
Theproportionofparticipantsincludedinthestudyrangedfrom82.
4%to99.
3%oftheoriginalcohorts(overall93.
1%),afterexclusionofmissingvaluesonanyofthecovariatesinmodel3.
Therewerenodifferencesinexposurelevelsbetweenincludedandexcludedparticipants.
Among100166participantsincludedinthestudyandfollowedfor1154386personyears,therewere5157incidentcases.
TheaveragePM2.
5levelatresidencerangedfrom7.
3g/m3inSweden(SD=1.
3)to31.
0g/m3(SD=1.
7)innorthernItaly,andallotherpollutantshadsimilarpatterns.
Allexposuremodelshadgoodperformance(R2≥0.
61),andthepredictionabilityofmodelsfornitrogendioxides(basedonalargernumberofmeasurementssites)wasslightlyhigherthanthoseofparticulatemodels.
Table5showsthepooledhazardratioswith95%confidenceintervalsforincidenceofcoronaryeventsforfixedincrementsofallpollutants.
WeobservedthestrongestassociationforPM10.
Whileinmodel1(adjustedforage,sex,andcalendarperiod)allparticulatematterindicatorswerestronglyassociatedwithincidenceofcoronaryevents,whenweadjustedformaritalstatus,education,occupation,smokingstatus,smokingduration,smokingintensity,andsocioeconomicareaindicator(model3),onlyPM10showedasignificantassociation(hazardratio1.
12,95%confidenceinterval1.
01to1.
25,foreach10g/m3increase).
TherewasalsoanassociationforPM2.
5(1.
13,0.
98to1.
30,foreach5g/m3increase)andforcoarseparticles,whereasonlysmallpositiveassociationswerefoundfornitrogenoxides.
Therewasnoevidenceofaneffectoftrafficvariables.
Therewasnoevidenceofstatisticalheterogeneitybetweenthecohortspecificeffectestimates(I260)forPM10andsixcohortsforPM2.
5,wefoundstrongerassociations(hazardratio1.
18(1.
05to1.
33)for10μg/m3PM10,and1.
35(1.
04to1.
74)for5μg/m3PM2.
5).
Whenwetookindividualandarealevelcovariatesintoaccount,wefoundnoevidenceofclusteringintheneighbourhoods(datanotshown).
EffectmodificationbyageshowedstrongerPM2.
5effectsforthoseaged60-74(hazardratio1.
25,95%confidenceinterval1.
03to1.
51)andforthose>75(1.
18,0.
85to1.
64)thanamongthoseagedunder60(0.
91,0.
71to1.
15),withP=0.
11foreffectmodification.
Forallothereffectmodifiers,thePvalueofeffectmodificationwasfarfromsignificant(≥0.
22).
Table7showstheresultsofthethresholdanalysisforPM2.
5andPM10.
WhenwerestrictedtheanalysistoparticipantsfromninecohortsexposedtoPM2.
5concentrationsbelowthecurrentEuropeanlimitvalue(70comparedwith60%),wefoundincreasedeffectestimates.
Thiscouldbebecauseofreducedmisclassificationofexposure.
ThecohortswithvalidationR2≤60%,however,hadyoungparticipantsand(inparticulartheItalianones)hadhigherratesofsmokingandabetterexposureassessmentmightbenottheonlyexplanation.
ConclusionsOurstudysuggestsanassociationbetweenlongtermexposuretoinhalableparticulatematterandincidenceofcoronaryevents.
TheseassociationsremainedforexposureconcentrationsbelowthecurrentEuropeanlimits.
Theresultsofthisstudy,togetherwithotherESCAPEfindings,supportloweringofEuropeanlimitsforparticulateairpollutiontoadequatelyprotectpublichealth.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page5of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromWhatisalreadyknownonthistopicAmbientparticulatematterairpollutionisestimatedtocause3.
1milliondeathsworldwideperyear,and22%ofDALYs(disabilityadjustedlifeyears)fromischaemicheartdiseaseTheassociationbetweenlongtermexposuretoairpollutionandincidenceofcoronaryeventsremainscontroversialWhatthisstudyaddsThereisanassociationbetweenoutdoorparticulatematterandincidenceofacutecoronaryevents,evenforexposurelevelsbelowthecurrentEuropeanlimitsTheburdenofdiseaseattributabletooutdoorparticulatemattermightbeunderestimatedifonlyestimatesofmortalityareconsideredTheresultsofthisstudysupportloweringoftheEUlimitsforparticulatematterairpollutionWethankMarjanTewis,MariekeOldenwening,MarloesEeftens,UlrichQuass,Lee-JaneSallyLiufortheirhelpwithexposureassessmentanddatamanagementwithinESCAPE.
FortheFinnishpart,additionalfundingcamefromtheAcademyofFinland(projectnumber129317).
ExposureassessmentwasperformedbyTarjaYli-Tuomi,PekkaTaimisto,andArtoPennanenfromtheDepartmentofEnvironmentalHealth.
Mortality,area-levelSES,andbuildingdatawereprovidedbyStatisticsFinland.
ThefourSwedishcohorts(SNAC-K,SALT,60y,SDPP)werepartiallyfundedbytheSwedishEnvironmentalProtectionAgency,theSwedishCouncilforWorkingLifeandSocialResearchandtheSwedishHeart-LungFoundation.
TheSALTcohortwasadditionallysupportedbyNIHgrantAG-08724.
The60yearcohortwasadditionallyfundedbytheStockholmCountyCouncilandtheSwedishResearchCouncil(longitudinalresearchand0593).
TheSDPPcohortwasadditionallyfundedbytheStockholmCountyCouncil;theSwedishResearchCouncil;theSwedishDiabetesAssociation;andtheNovoNordiskScandinavia.
WethanktheHeinzNixdorfFoundation,theGermanMinistryofEducationandScience(BMBF),theGermanAerospaceCenter(DeutschesZentrumfürLuft-undRaumfahrt(DLR)),andtheGermanResearchFoundation(DFGHO3314/2-1andJO170/8-1),Bonn,Germany,fortheirgeneroussupportofHeinzNixdorfRecallstudy.
TheKORAresearchplatformandtheMONICAAugsburgstudieswereinitiatedandfinancedbytheHelmholtzZentrumMünchen,GermanResearchCenterforEnvironmentalHealth,whichisfundedbytheGermanFederalMinistryofEducationandResearchandbytheStateofBavaria.
TheSIDRIAcohortstudywaspartiallyfundedbytheItalianMinistryofHealth.
GiuseppeCostaprovideddatafromtheTurinLongitudinalStudyforthefollow-upoftheSIDRIA-Turincohort.
ExposureassessmentinTurinwasperformedbyDanielaRaffaeleandMarcoGilardetti.
NoiseassessmentinTurinwasperformedbyDanieleGrassoandJacopoFogola(RegionalAgencyfortheProtectionoftheEnvironmentofPiedmont).
WethankSimoneBucci,PatrizioPasquinelli,andEleonoraZirroforexposureassessmentinRome.
Contributors:GCcontributedtothedesign,exposureassessment,dataanalyses,anddraftedthemanuscript;FFcontributedtothedesign,providedlocalcohortdataanddraftedthemanuscript;MScontributedtothedesign,statisticalscript,anddataanalyses;ZJA,RHcontributedtothestatisticalscriptanddataanalysis;CB,FH,EM,JPen,APy,FR,andATcontributedtothedataanalyses;RB,GHcontributedtothedesign,exposureassessmentandstatisticalscript;BC,UdF,RE,LF,MH,AH,KHJ,KL,PKEM,CGO,KO,NLP,JPek,CS,andVSprovidedlocalcohortdata;CGcontributedtoexposureassessmentandprovidedlocalcohortdata;BH,ORN,GP,andPVcontributedtothedesignandprovidedlocalcohortdata;MKcontributedtotheexposureassessmentandthedataanalyses;KTE,DH,TL,AR,WS,andKdHcontributedtoexposureassessment;GWandKWcontributedtothestatisticalscript;BBandAPecontributedtothedesignanddraftedthemanuscript.
Allauthorscontributedtocriticalreadingofandcommentstothemanuscript,interpretationofdataandapprovedthefinaldraft.
GCisguarantor.
Funding:TheresearchleadingtotheseresultshasreceivedfundingfromtheEuropeanCommunity'sSeventhFrameworkProgram(FP7/2007-2011)undergrantagreementNo211250.
Thesourcesoffundinghadnoroleinstudydesign,datacollection,analyses,interpretation,anddecisiontosubmitthearticleforpublication.
Competinginterests:AllauthorshavecompletedtheICMJEuniformdisclosureformatwww.
icmje.
org/coi_disclosure.
pdfanddeclare:nosupportfromanyorganisationforthesubmittedwork;nofinancialrelationshipswithanyorganisationsthatmighthaveaninterestinthesubmittedworkinthepreviousthreeyears;nootherrelationshipsoractivitiesthatcouldappeartohaveinfluencedthesubmittedwork.
Ethicalapproval:Theoriginalcohortstudieswereapprovedbyappropriateinstitutionalreviewboardscomplyingwithallrelevantnational,state,andlocalregulations.
Transparencystatement:Theleadauthoraffirmsthatthismanuscriptisanhonest,accurate,andtransparentaccountofthestudybeingreported;thatnoimportantaspectsofthestudyhavebeenomitted;andthatanydiscrepanciesfromthestudyasplannedhavebeenexplained.
Datasharing:meta-analyticdataandstatisticalcodeareavailablefromthecorrespondingauthor.
1LimSS,VosT,FlaxmanAD,DaneiG,ShibuyaK,Adair-RohaniH,etal.
Acomparativeriskassessmentofburdenofdiseaseandinjuryattributableto67riskfactorclustersin21regions,1990-2010:asystematicanalysisfortheGlobalBurdenofDiseaseStudy2010.
Lancet2012;380:2224-60.
2BeelenR,HoekG,vandenBrandtPA,GoldbohmRA,FischerP,SchoutenLJ,etal.
Long-termeffectsoftraffic-relatedairpollutiononmortalityinaDutchcohort(NLCS-AIRstudy).
EnvironHealthPerspect2008;116:196-202.
3ChenLH,KnutsenSF,ShavlikD,BeesonWL,PetersenF,GhamsaryM,etal.
Theassociationbetweenfatalcoronaryheartdiseaseandambientparticulateairpollution:ArefemalesatgreaterriskEnvironHealthPerspect2005;113:1723-9.
4CrouseDL,PetersPA,vanDonkelaarA,GoldbergMS,VilleneuvePJ,BrionO,etal.
Riskofnonaccidentalandcardiovascularmortalityinrelationtolong-termexposuretolowconcentrationsoffineparticulatematter:aCanadiannational-levelcohortstudy.
EnvironHealthPerspect2012;120:708-14.
5GanWQ,KoehoornM,DaviesHW,DemersPA,TamburicL,BrauerM.
Long-termexposuretotraffic-relatedairpollutionandtheriskofcoronaryheartdiseasehospitalizationandmortality.
EnvironHealthPerspect2011;119:501-7.
6JerrettM,BurnettRT,MaR,PopeCA3rd,KrewskiD,NewboldKB,etal.
SpatialanalysisofairpollutionandmortalityinLosAngeles.
Epidemiology2005;16:727-36.
7KrewskiD,JerrettM,BurnettRT,MaR,HughesE,ShiY,etal.
Extendedfollow-upandspatialanalysisoftheAmericanCancerSocietystudylinkingparticulateairpollutionandmortality.
HEIResearchReport140.
HealthEffectsInstitute,2009.
8LadenF,SchwartzJ,SpeizerFE,DockeryDW.
Reductioninfineparticulateairpollutionandmortality:extendedfollow-upoftheHarvardSixCitiesstudy.
AmJRespirCritCareMed2006;173:667-72.
9MillerKA,SiscovickDS,SheppardL,ShepherdK,SullivanJH,AndersonGL,etal.
Long-termexposuretoairpollutionandincidenceofcardiovasculareventsinwomen.
NEnglJMed2007;356:447-58.
10PopeCA3rd,BurnettRT,ThurstonGD,ThunMJ,CalleEE,KrewskiD,etal.
Cardiovascularmortalityandlong-termexposuretoparticulateairpollution:epidemiologicalevidenceofgeneralpathophysiologicalpathwaysofdisease.
Circulation2004;109:71-7.
11PuettRC,HartJE,YanoskyJD,PaciorekC,SchwartzJ,SuhH,etal.
Chronicfineandcoarseparticulateexposure,mortality,andcoronaryheartdiseaseintheNurses'HealthStudy.
EnvironHealthPerspect2009;117:1697-701.
12LipsettMJ,OstroBD,ReynoldsP,GoldbergD,HertzA,JerrettM,etal.
Long-termexposuretoairpollutionandcardiorespiratorydiseaseintheCaliforniateachersstudycohort.
AmJRespirCritCareMed2011;184:828-35.
13Raaschou-NielsenO,AndersenZJ,JensenSS,KetzelM,SrensenM,HansenJ,etal.
Trafficairpollutionandmortalityfromcardiovasculardiseaseandallcauses:aDanishcohortstudy.
EnvironHealth2012;11:60.
14AtkinsonRW,CareyIM,KentAJ,vanStaaTP,AndersonHR,CookDG.
Long-termexposuretooutdoorairpollutionandincidenceofcardiovasculardiseases.
Epidemiology2013;24:44-53.
15BrookRD,RajagopalanS,PopeCA3rd,BrookJR,BhatnagarA,Diez-RouxAV,etal;AmericanHeartAssociationCouncilonEpidemiologyandPrevention,CouncilontheKidneyinCardiovascularDisease,andCouncilonNutrition,PhysicalActivityandMetabolism.
Particulatematterairpollutionandcardiovasculardisease:anupdatetothescientificstatementfromtheAmericanHeartAssociation.
Circulation2010;121:2331-78.
16ESCAPE.
Europeanstudyofcohortsforairpollutioneffects.
www.
escapeproject.
eu.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page6of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
Downloadedfrom17BeelenR,Raaschou-NielsenO,StafoggiaM,AndersenZJ,WeinmayrG,HoffmannB,etal.
Effectsoflong-termexposuretoairpollutiononnaturalcausemortality:ananalysisof22Europeancohortswithinthemulti-centerESCAPEproject.
LancetEarlyonlinepublication9December2013.
http://dx.
doi.
org/10.
1016/S0140-6736(13)62158-3.
18VartiainenE,LaatikainenT,PeltonenM,JuoleviA,MnnistS,SundvallJ,etal.
Thirty-five-yeartrendsincardiovascularriskfactorsinFinland.
IntJEpidemiol2010;39:504-18.
19MeinowB,KreholtI,LagergrenM.
AccordingtoneedPredictingtheamountofmunicipalhomehelpallocatedtoelderlyrecipientsinanurbanareaofSweden.
HealthSocCareCommunity2005;13:366-377.
20LichtensteinP,SullivanPF,CnattingiusS,GatzM,JohanssonS,CarlstrmE,etal.
TheSwedishTwinRegistryintheThirdMillennium—anupdate.
TwinResHumGenet2006;9:875-82.
21WandellPE,WajngotA,deFaireU,HelleniusML.
Increasedprevalenceofdiabetesamongimmigrantsfromnon-Europeancountriesin60-year-oldmenandwomeninSweden.
DiabetMetab2007;33:30-6.
22ErikssonAK,EkbomA,GranathF,HildingA,EfendicS,OstensonCG.
Psychologicaldistressandriskofpre-diabetesandType2diabetesinaprospectivestudyofSwedishmiddle-agedmenandwomen.
DiabetMed2008;25:834-42.
23TjnnelandA,OlsenA,BollK,StrippC,ChristensenJ,EngholmG,etal.
Studydesign,exposurevariables,andsocioeconomicdeterminantsofparticipationindiet,cancerandhealth:apopulation-basedprospectivecohortstudyof57,053menandwomeninDenmark.
ScandJPublicHealth2007;35:432-41.
24ErbelR,MhlenkampS,MoebusS,SchmermundA,LehmannN,StangA,etal.
HeinzNixdorfRecallStudyInvestigativeGroup.
Coronaryriskstratification,discrimination,andreclassificationimprovementbasedonquantificationofsubclinicalcoronaryatherosclerosis:theHeinzNixdorfRecallstudy.
JAmCollCardiol2010;56:1397-406.
25HolleR,HappichM,LwelH,WichmannHE;MONICA/KORAStudyGroup.
KORA—aresearchplatformforpopulationbasedhealthresearch.
Gesundheitswesen2005;67(suppl1):S19-25.
26CesaroniG,BadaloniC,PortaD,ForastiereF,PerucciCA.
Comparisonbetweenvariousindicesofexposuretotraffic-relatedairpollutionandtheirimpactonrespiratoryhealthinadults.
OccupEnvironMed2008;65:683-90.
27VineisP,HoekG,KrzyzanowskiM,Vigna-TagliantiF,VegliaF,AiroldiL,etal.
AirpollutionandriskoflungcancerinaprospectivestudyinEurope.
IntJCancer2006;119:169-74.
28CyrysJ,EeftensM,HeinrichJ,AmpeC,ArmengaudA,BeelenR,etal.
VariationofNO2andNOxconcentrationsbetweenandwithin36Europeanstudyareas:resultsfromtheESCAPEstudy.
AtmosphEnviron2012;62:374-90.
29EeftensM,TsaiM,AmpeC,AnwanderB,BeelenR,BellanderT,etal.
SpatialvariationofPM2.
5,PM10,PM2.
5absorbanceandPMcoarseconcentrationsbetweenandwithin20EuropeanstudyareasandtherelationshipwithNO2—resultsoftheESCAPEproject.
AtmosphEnviron2012;62:303-17.
30EeftensM,BeelenR,deHooghK,BellanderT,CesaroniG,CirachM,etal.
DevelopmentoflanduseregressionmodelsforPM(2.
5),PM(2.
5)absorbance,PM(10)andPM(coarse)in20EuropeanStudyAreas;ResultsoftheESCAPEProject.
EnvironSciTechnol2012;46:11195-205.
31BeelenR,HoekG,VienneauD,EeftensM,DimakopoulouK,PedeliX,etal.
DevelopmentofNO2andNOxlanduseregressionmodelsforestimatingairpollutionexposurein36studyareasinEurope—theESCAPEproject.
AtmosphEnvirondoi:10.
1016/j.
atmosenv.
2013.
02.
037.
32DiezRouxAV,MerkinSS,HannanP,JacobsDR,KiefeCI.
Areacharacteristics,individual-levelsocioeconomicindicators,andsmokinginyoungadults:thecoronaryarterydiseaseriskdevelopmentinyoungadultsstudy.
AmJEpidemiol2003;157:315-26.
33DerSimonianR,LairdN.
Meta-analysisinclinicaltrials.
ControlClinTrials1986;7:177-88.
34HigginsJP,ThompsonSG.
Quantifyingheterogeneityinameta-analysis.
StatMed2002;21:1539-58.
35ZanobettiA,SchwartzJ.
Theeffectofparticulateairpollutiononemergencyadmissionsformyocardialinfarction:amulticitycase-crossoveranalysis.
EnvironHealthPerspect2005;113:978-82.
36PopeCAIII,MuhlesteinJB,MayHT,RenlundDG,AndersonJL,HorneBD.
Ischemicheartdiseaseeventstriggeredbyshort-termexposuretofineparticulateairpollution.
Circulation2006;114:2443-8.
37PetersA.
Ambientparticulatematterandtheriskforcardiovasculardisease.
ProgCardiovascDis2011;53:327-33.
38BeelenR,StafoggiaM,Raaschou-NielsenM,AndersenZJ,XunWW,KatsouyanniK,etal.
Long-termexposuretoairpollutionandcardiovascularmortality:ananalysisof22EuropeancohortswithintheESCAPEproject.
Epidemiology(inpress).
39HoekG,KrishnanRM,BeelenR,PetersA,OstroB,BrunekreefB,etal.
Long-termairpollutionexposureandcardio-respiratorymortality:areview.
EnvironHealth2013;12:43.
40CesaroniG,PortaD,BadaloniC,StafoggiaM,EeftensM,MeliefsteK,etal.
Nitrogendioxidelevelsestimatedfromlanduseregressionmodelsseveralyearsapartandassociationwithmortalityinalargecohortstudy.
EnvironHealth2012;11:48.
41EeftensM,BeelenR,FischerP,BrunekreefB,MeliefsteK,HoekG.
StabilityofmeasuredandmodelledspatialcontrastsinNO2overtime.
OccupEnvironMed2011;68:765-70.
42WangR,HendersonSB,SbihiH,AllenRW,BrauerM.
Temporalstabilityoflanduseregressionmodelsfortraffic-relatedairpollution.
AtmosphEnviron2013;64:312-9.
Accepted:04December2013Citethisas:BMJ2014;348:f7412ThisisanOpenAccessarticledistributedinaccordancewiththeCreativeCommonsAttributionNonCommercial(CCBY-NC3.
0)license,whichpermitsotherstodistribute,remix,adapt,builduponthisworknon-commercially,andlicensetheirderivativeworksondifferentterms,providedtheoriginalworkisproperlycitedandtheuseisnon-commercial.
See:http://creativecommons.
org/licenses/by-nc/3.
0/.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page7of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTablesTable1|Summarydescriptionof11EuropeancohortsfromfivecountriescontributingdatatostudylongtermexposuretoambientairpollutionandincidenceofacutecoronaryeventsSIDRIAEPICTurinKORAHNRDCHSDPP60yearoldsSALTSNAC-KFINRISKVariableRomeTurin920051377230830144333569377233686608426849995Noofparticipants10289456366914908459534941464055106995399785175616256105060Personyearsatrisk87.
195.
582.
494.
099.
396.
097.
691.
092.
389.
493.
5%oforiginalcohort2051231572821353293181165204200212NoofcasesRome(Italy)Turin(Italy)Turin(Italy)Augsburg(Germany)RuhrArea(Germany)Copenhagen(Denmark)Stockholm(Sweden)Stockholm(Sweden)Stockholm(Sweden)Stockholm(Sweden)TurkuandHelsinki(Finland)Studylocation199919991993-981994-95,1999-20012000-031993-971992-981997-991998-20022001-041992,1997,2002,2007YearsofenrolmentNocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page8of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable2|Individualbaselinecharacteristicsfrom11Europeancohortsfromfivecountriescontributingdatatostudylongtermexposuretoambientairpollutionandincidenceofacutecoronaryevents.
Valuesarenumbers(percentages)unlessstatedotherwiseSIDRIAEPICTurinKORAHNRDCHSDPP60yearoldsSALTSNAC-KFINRISKVariableRomeTurin9200(100)5137(100)7230(100)8301(100)4433(100)35693(100)7723(100)3686(100)6084(100)2684(100)9995(100)Noofparticipants44(6)44(6)50(8)49(14)59(8)57(4)47(5)60(0)59(11)74(11)48(13)Mean(SD)age(years)4859(53)2677(52)3472(48)4275(51)2309(52)19302(54)4721(61)1967(53)3557(58)1751(65)5460(55)WomenMaritalstatus:0(0)121(2)438(6)879(11)256(6)2352(7)1272(16)*170(5)835(14)397(15)1601(16)Single9200(100)4897(95)6183(86)6292(76)3319(75)24673(69)6451(84)2628(71)4103(67)1263(47)7009(70)Married/livingwithpartner0(0)37(1)375(5)620(7)444(10)6599(18)—630(17)678(11)362(13)1051(11)Divorced/separated0(0)82(2)234(3)510(6)414(9)2069(6)—258(7)468(8)662(25)334(3)WidowedEducation:4130(45)898(17)3168(44)1024(12)499(11)10589(30)1996(26)1030(28)1332(22)705(26)3029(30)Primaryschoolorless3690(40)3668(71)3081(43)6200(65)2466(56)16943(47)3453(45)1631(44)2606(43)1128(42)5217(52)Secondaryschool1380(15)571(11)981(14)1077(13)1468(33)8161(23)2274(29)1025(28)2146(35)851(32)1749(17)UniversitydegreeandmoreOccupationalstatus:6500(71)3727(73)—4908(59)1859(42)28600(80)7088(92)1881(51)—2016(75)7092(71)Employed/selfemployed393(4)349(7)—272(3)630(14)7093(20)635(8)374(10)—668(25)610(6)Unemployed2307(25)1061(21)—1188(14)1664(38)——290(8)——351(4)Homemakerorhousewife0(0)0(0)—1933(23)280(6)——1141(31)——1942(19)RetiredSmokingstatus:3884(42)2117(41)1768(24)2162(26)1040(23)12793(36)2022(26)751(20)1224(20)394(15)2603(26)Currentsmoker2147(23)1085(21)2368(33)2517(30)1482(33)9944(28)2815(36)1429(39)2678(44)927(34)2808(28)Formersmoker3169(34)1935(38)3094(43)3622(44)1911(43)12956(36)2886(37)1506(41)2182(36)1363(51)4584(46)Neversmoker18(7)18(8)23(10)21(13)36(9)§29(10)20(10)26(13)—30(17)15(12)Mean(SD)yearsofsmokingamongeversmokers15(9)15(9)14(9)15(11)17(12)17(10)14(7)13(7)13(8)11(8)15(9)Mean(SD)Noofcigarettes/dayamongcurrentsmokers*Allexceptmarried/livingwithpartner.
29Physicalactivity(hours/week):——1660(23)3107(37)2263(51)15984(45)846(11)2510(69)1575(26)464(20)1808(18)2Alcoholconsumption(drinks/week):——457(6)3867(47)1015(23)776(2)594(8)168(5)—519(19)1320(13)Never——201(3)1212(15)1190(27)34142(98)2876(38)771(21)—634(24)4566(46)1-3——347(5)549(7)2127(49)—3873(51)1622(44)—1236(46)3139(32)3-6——6225(86)2662(32)—280(4)1123(30)—288(11)824(8)>6Diabetesmellitus:9179(100)5111(99)7118(98)7940(96)3877(87)34943(98)7597(98)3533(96)5833(96)2464(92)9546(96)No21(0)26(1)110(2)360(4)556(13)703(2)126(2)153(4)251(4)220(8)426(4)YesHypertension:9124(99)5082(99)3629(53)5098(62)2002(45)30033(84)5801(76)1770(48)4735(78)902(34)5877(59)No76(1)55(1)3255(47)3190(38)2423(55)5625(16)1837(24)1915(52)1347(22)1762(66)4042(41)Yes———229(44)231(39)——231(41)——211(78)Mean(SD)serumcholesterolNoiseexposureatbaselineaddress(dB):—62(1)44(1)466(6)721(17)126(0)—84(6)123(5)74Livinginareasoflowurbanisation:9200(100)5137(100)7230(100)3557(43)4433(100)14155(40)0(0)1450(39)2763(45)2684(100)9086(91)No0(0)0(0)0(0)4744(57)0(0)21538(60)7723(100)2236(61)3321(55)0(0)909(9)Yes*FINRISK:medianincomeratein3kmarea;SNAC-K:meanincomeinthirdsatsmallneighbourhoodlevel(SmallAreaforMarketStatistics);SALTandSDPP:meanincomeinfourcategoriesatmunicipalitylevel;60years:meanincomeinquartersatsmallneighbourhoodlevel(SmallAreaforMarketStatistics);DCH:meanincomeatmunicipalitylevel(16unitswithmedianpopulationof1500inhabitants),per/100000;HNR:unemploymentrateatneighbourhoodlevel;KORA:percentageoflowincomein5kmgrid;EPIC-Turin,SIDRIA-Turin,andSIDRIA-Rome:deprivationindex,censusblocklevel(average500inhabitants).
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page11of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable4(continued)SIDRIAEPICTurinKORAHNRDCHSDPP60yearoldsSALTSNAC-KFINRISKRomeTurinAnycategoryofdrinking.
Firstandlasttwoclassesarecollapsed.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page12of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable5|Associationbetweenexposuretopollutantsandincidenceofcoronaryevents.
ResultsexpressedperfixedincrementsPooledhazardratios(95%CI)ParticipantsCohortsFixedincrementExposureModel3Model2Model1*1.
12(1.
01to1.
25)1.
12(1.
01to1.
24)1.
15(1.
04to1.
28)1001661110PM10(g/m3)1.
06(0.
98to1.
15)1.
06(0.
98to1.
14)1.
08(1.
00to1.
17)100166115CoarsePM(g/m3)1.
13(0.
98to1.
30)1.
15(1.
00to1.
32)1.
22(1.
04to1.
44)100166115PM2.
5(g/m3)1.
10(0.
98to1.
24)1.
10(0.
98to1.
24)1.
18(1.
05to1.
32)100166111AbsorbancePM2.
5(105/m)1.
03(0.
97to1.
08)1.
03(0.
96to1.
11)1.
04(0.
96to1.
12)1001661110NO2(g/m3)1.
01(0.
98to1.
05)1.
01(0.
98to1.
05)1.
03(1.
00to1.
07)1001661120NOx(g/m3)1.
01(0.
98to1.
04)1.
01(0.
98to1.
04)1.
01(0.
98to1.
04)95733105000TrafficintensityonnearestroadadjustedforbackgroundNO2(vehicles/day)§1.
00(0.
95to1.
06)1.
00(0.
95to1.
06)1.
02(0.
96to1.
08)100166114000000Trafficloadonmajorroadsin100mbufferadjustedforbackgroundNO2(vehicles*m/day)*Adjustedforage(timevariable),yearofenrolmentandsexAsformodel1plusadjustedformaritalstatus,education,occupation,smokingstatus,smokingduration,andsmokingintensity.
Asformodel2plusadjustedforsocioeconomicarealevelvariables.
§AllcohortsexceptHNRstudy.
24Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page13of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable6|Incidenceofcoronaryevents,resultsfromsensitivityanalyses.
Figuresarepooledhazardratios(and95%CI)for10g/m3PM10and5g/m3PM2.
5,I2statistics,andPvalueforheterogeneityHazardratio(95%CI)Noofparticipants*NoofcohortsModel5g/m3PM2.
510g/m3PM101.
13(0.
98to1.
30)1.
12(1.
01to1.
25)10016611Basemodel(model3)RoleofcardiovascularriskfactorsIntermediatevariables:diabetesandhypertension:1.
11(0.
96to1.
28)1.
11(1.
00to1.
24)9952611+diabetesandhypertensionPhysicalactivity,alcoholandBMI:1.
22(1.
03to1.
44)1.
17(1.
04to1.
33)775848Basemodel,onsubsetofparticipantswithadditionalinformation1.
19(1.
01to1.
42)1.
17(1.
04to1.
33)775848+additionalinformationAllcardiovascularriskfactors(diabetes,hypertension,physicalactivity,alcohol,BMI,cholesterol)§:1.
28(0.
92to1.
79)1.
19(0.
92to1.
53)256254Basemodel,onsubsetofparticipantswithadditionalinformation1.
29(0.
92to1.
81)1.
20(0.
93to1.
56)256254+additionalinformationUrban/suburban-ruralresidencelocation:1.
13(0.
98to1.
39)1.
12(1.
01to1.
25)10016611Basemodel,onsubsetofparticipantswithadditionalinformation1.
14(0.
99to1.
31)1.
13(1.
02to1.
26)10016611+additionalinformationNoise:1.
16(0.
98to1.
37)1.
14(1.
01to1.
30)738409Basemodel,onsubsetofparticipantswithnoiseinformation1.
13(0.
95to1.
36)1.
14(1.
01to1.
31)738409+noisevariableChangeofaddress**:1.
14(0.
98to1.
33)1.
11(0.
99to1.
23)9293610Basemodel,onsubsetofcohortswithchangeofaddressinformation1.
18(0.
98to1.
42)1.
16(1.
01to1.
32)6312110Participantswhodidn'tchangeaddressduringfollow-upProportionalityhazards(PH)assumption:1.
13(0.
98to1.
30)1.
10(0.
99to1.
23)10016611VariablesthatdonotmeetPHputasstrataExclusionofDCHStudy1.
09(0.
92to1.
30)1.
12(0.
99to1.
27)6447310DataafterexclusionPerformanceoflanduseregressionmodel:1.
35(1.
04to1.
74)1.
18(1.
05to1.
33)——LeaveoneoutcrossvalidationR2>60%1.
05(0.
89to1.
25)0.
88(0.
68to1.
12)——LeaveoneoutcrossvalidationR2≤60%*FiguresdependonnumberofcohortsincludedandonmissingvaluesofincludedvariablesAdjustedforage(timevariable),yearofenrolment,sex,maritalstatus,education,occupation,smokingstatus,smokingduration,smokingintensity,andsocioeconomicarealevelindicators.
ExcludesSALT,SIDRIA-Turin,andSIDRIA-Rome.
§IncludesFINRISK,60yearolds,HNR,andKORA.
AllcohortsexceptSDPPandSIDRIA-Rome.
**AllcohortsexceptEPIC-Turin.
ForPM10:ninecohorts,80971participants(excludesFINRISKandSIDRIA-Rome).
ForPM2.
5:sixcohorts,42906participants(excludesFINRISK,DCH,EPIC-Turin,SIDRIA-Turin,andSIDRIA-Rome).
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page14of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable7|Resultsfromrandom-effectsmeta-analysesforadjustedassociationbetweenincidenceofcoronaryeventsandexposuretoPM10andPM2.
5belowvariousthresholdvalues.
Figuresarepooledhazardratios(95%CI)for10g/m3PM10and5g/m3PM2.
5,I2statistic,andPvalueforheterogeneityCohortswithexposuredataavailableforallthresholdsCohortswithexposuredataavailableforrespectivethresholdExposureandthresholdPvalueI2HR*(95%CI)NoofparticipantsNoofcohortsPvalueI2HR*(95%CI)NoofparticipantsNoofcohortsPM10(g/m3):0.
9201.
15(1.
02to1.
30)741667—————Fullrangeofexposure0.
9101.
20(1.
01to1.
41)6436370.
9101.
20(1.
01to1.
41)643637<200.
9101.
12(0.
98to1.
27)7401970.
9401.
12(0.
98to1.
27)779978<300.
9201.
15(1.
02to1.
30)7416670.
4801.
12(1.
00to1.
27)862229§<40PM2.
5(g/m3):0.
5701.
23(1.
04to1.
46)741667—————Fullrangeofexposure0.
7401.
19(1.
00to1.
42)7342070.
7401.
19(1.
00to1.
42)734207<150.
5701.
23(1.
04to1.
46)7416670.
18301.
17(0.
91to1.
50)852169§<200.
5701.
23(1.
04to1.
46)7416670.
5601.
18(1.
01to1.
39)875329§<25*Adjustedforage(timevariable),yearofenrolment,sex,maritalstatus,education,occupation,smokingstatus,smokingduration,smokingintensity.
andsocioeconomicarealevelindicatorsAllcohortsexceptHeinzNixdorfRecallStudy,EPIC-Turin,SIDRIA-TurinandSIDRIA-Rome.
AllcohortsexceptEPIC-Turin,SIDRIA-TurinandSIDRIA-Rome.
§AllcohortsexceptEPIC-TurinandSIDRIA-Turin.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page15of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromFigureHazardratiosofincidentcoronaryeventsper10g/m3PM10and5g/m3PM2.
5.
Cohortspecificandmeta-analyticresultsNocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page16of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
Downloadedfrom
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page1of16ResearchRESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromInstituteforHealthandWelfare,DepartmentofChronicDiseasePrevention,POB30,FI-00271Helsinki,Finland;26MRC-HPACentreforEnvironmentandHealth,DepartmentofEpidemiologyandBiostatistics,ImperialCollegeLondon,StMary'sCampus,LondonW21PG,UK;27InstituteofEpidemiologyandMedicalBiometry,UlmUniversity,89069Ulm,Germany;28JuliusCenterforHealthSciencesandPrimaryCare,UniversityMedicalCenterUtrecht,POBox85500,3508GAUtrecht,NetherlandsAbstractObjectivesTostudytheeffectoflongtermexposuretoairbornepollutantsontheincidenceofacutecoronaryeventsin11cohortsparticipatingintheEuropeanStudyofCohortsforAirPollutionEffects(ESCAPE).
DesignProspectivecohortstudiesandmeta-analysisoftheresults.
SettingCohortsinFinland,Sweden,Denmark,Germany,andItaly.
Participants100166peoplewereenrolledfrom1997to2007andfollowedforanaverageof11.
5years.
Participantswerefreefrompreviouscoronaryeventsatbaseline.
MainoutcomemeasuresModelledconcentrationsofparticulatematter5000vehicles/daymultipliedbythelengthofthoseroadsina100mbuffer.
Tovalidatethemodels,weusedtheleaveoneoutcrossvalidationmethod—thatis,systematicallysubtractingeachofthemonitoringpointsfromthemodelonebyone,andthencomparingthepredictedvalueforeachmonitoringlocationwiththemeasuredlevelatthelocationwithoutusingthismeasurementinthedevelopmentofthemodel.
3031CovariatesAllcohortshadacommonsetofpotentialconfoundersandeffectmodifiersatbaseline:maritalstatus(codedassingle,married/livingwithpartner,divorced/separated,widowed;fortheSDPPcohortonlythebinaryvariable"livingwithpartner"wasavailable;forSIDRIA-Romeallparticipantswerelivingascouplesatbaseline),education(primaryschoolorless,uptosecondaryschoolorequivalent,universitydegreeandmore),occupation(employed,unemployed,homemaker/housewife,retired),smokingstatus(current,former,never),durationofsmoking(years),smokingintensity(cigarettes/day),hypertension,anddiabetes.
Mostcohortshadinformationonadditionalcardiovascularriskfactorssuchasbodymassindex(BMI;codedas2hours/week),alcoholconsumption(never,1-3drinks/week,3-6drinks/week,>6drinks/week),andthreecohortshadinformationonuseofdrugsforhormonereplacementtherapy.
Fourof11cohortshaddataoncholesterolconcentrations.
Tenof11cohortshaddataonnoiseexposure,estimatedattheresidentialaddressesfromEuropeannoiseexposureassessmentin2007,andweusedaneightclasscategoricalvariableof5dBofexposurefrom45dBto75dBandmore(seeappendix).
Eachcohortalsousedanareabasedsocioeconomicstatusindicatorbecausesocioeconomicindicatorsatthearealevelarepredictorsofmorbidity,accesstocare,andlifestyleriskfactorsinadditiontoindividualsocioeconomiccharacteristics.
32StatisticalanalysesToevaluatetheassociationbetweenexposuretoairpollutionandincidenceofcoronaryeventsweperformedtheanalysesintwostages.
Firstly,weanalysedeachcohortusingacommonprotocolforconfounders,outcomes,andstatisticalmodelling.
WeusedCoxproportionalhazardsregressionmodels(hazardratios)withageasthetimescale.
Ascriptwithstatisticalcodewasprovidedtoallcohortspecificanalysts.
TheresultswereevaluatedcentrallyattheDepartmentofEpidemiologyinRome.
Secondly,wecarriedoutarandomeffectsmeta-analysistopoolresults.
33WecalculatedI2statisticsandPvaluesforχ2testfromCochran'sQtoquantifytheheterogeneityamongstudies.
34Wefirstestimatedhazardratiosadjustedforage,sex,andyearofenrolmentonly(model1),thenadjustedforacommonsetofindividualcovariates(maritalstatus,education,occupation,smokingstatus,smokingduration,andsmokingintensity;model2),andthenadjustedforareabasedsocioeconomicstatusindicators(model3).
WheninvestigatingtheeffectsoftrafficvariableswealsoadjustedallmodelsforbackgroundNO2levels.
Weestimatedtheeffectsofexposuretoairpollutionusingfixedincrementsofpollutants(20g/m3forNOx,10g/m3forPM10andNO2,5g/m3forPM2.
5andPMcoarse,105/mforPM2.
5absorbance).
Weperformedseveralsensitivityanalysestodealwithpotentialsourcesofbiasandtoconsidertheconfoundingroleofadditionalcardiovascularriskfactors.
Firstly,weadjustedforpossibleintermediatevariables(diabetes,hypertension)availableinallthecohorts;thenforphysicalactivity,alcoholconsumption,andBMI(availableineightcohorts);finallyweaddedtopreviouslymentionedfactorsthecholesterolconcentration(availableinfourcohorts).
Wethenanalysedtheroleoflivinginlow/highurbanisedsettings,ofnoiseexposure,andofresidentialstability(restrictingtheanalysistothoselivingatthesamebaselineaddressduringthefollow-up).
WestratifiedtheCoxmodel3forpredictorsthatdidnotmeettheproportionalityhazardassumption.
WeanalysedtheinfluentialroleofthelargestcohortbyexcludingtheDCHstudyfromtheanalysis.
Finally,weevaluatedtheroleoftheperformanceoflanduseregressionmodels,stratifyingcohortsbycrossvalidationR2>or≤60%.
Toexploretheeffectofclusteringbyareaofresidenceintheassociationbetweenexposureandmortality—thatis,residentsinthesameareausuallysharesimilarcharacteristics(socioeconomicstatus,health,accesstoservices)andhavesimilarenvironmentalandairconditions—weperformedafrailtymodeltomeasuretheroleeachareaplayed.
4Weevaluatedpotentialeffectmodificationbyaddingtomodel3aninteractiontermofexposureandoneeffectmodifierataNocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page3of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
Downloadedfromtime(sex,educationallevel,smokingstatus,BMI,hypertension,andresidenceinloworhighurbanisedsettings).
Weusedthelikelihoodratiotesttocomparethemodelswithandwithoutinteractionterms.
Forageduringfollow-up,weestimatedeffectmodificationbyaddinganinteractiontermbetweenexposureandatimedependentcategoricalvariableindicatingagegroup(75).
Toprovideinformationaboutthehealtheffectsbelowspecificthresholdvalues(20,30,and40g/m3forPM10,and15,20,and25g/m3forPM2.
5),westudiedtheeffectoffixedincrementsofPM2.
5andPM10amongpeoplewithairpollutionconcentrationsatresidencesonlybelowthesethresholds.
WeusedSTATAsoftware(versions10,11,and12)foralltheanalyses,withtheexceptionoffrailtymodelsforwhichweusedR(www.
r-project.
org).
ResultsTable1showsasummarydescriptionofthe11Europeancohorts.
Table2showsindividualcharacteristicsofparticipantsandtable3theirexposurestoairpollution.
Additionalcharacteristicsoftheparticipantsareprovidedintable4.
Theenrolmentperiodcovered15years,andtheaveragefollow-upwas11.
5years.
Theparticipatingcohortsvariedincharacteristics,availabilityofdataoncovariates,andlevelsofexposure.
Theproportionofparticipantsincludedinthestudyrangedfrom82.
4%to99.
3%oftheoriginalcohorts(overall93.
1%),afterexclusionofmissingvaluesonanyofthecovariatesinmodel3.
Therewerenodifferencesinexposurelevelsbetweenincludedandexcludedparticipants.
Among100166participantsincludedinthestudyandfollowedfor1154386personyears,therewere5157incidentcases.
TheaveragePM2.
5levelatresidencerangedfrom7.
3g/m3inSweden(SD=1.
3)to31.
0g/m3(SD=1.
7)innorthernItaly,andallotherpollutantshadsimilarpatterns.
Allexposuremodelshadgoodperformance(R2≥0.
61),andthepredictionabilityofmodelsfornitrogendioxides(basedonalargernumberofmeasurementssites)wasslightlyhigherthanthoseofparticulatemodels.
Table5showsthepooledhazardratioswith95%confidenceintervalsforincidenceofcoronaryeventsforfixedincrementsofallpollutants.
WeobservedthestrongestassociationforPM10.
Whileinmodel1(adjustedforage,sex,andcalendarperiod)allparticulatematterindicatorswerestronglyassociatedwithincidenceofcoronaryevents,whenweadjustedformaritalstatus,education,occupation,smokingstatus,smokingduration,smokingintensity,andsocioeconomicareaindicator(model3),onlyPM10showedasignificantassociation(hazardratio1.
12,95%confidenceinterval1.
01to1.
25,foreach10g/m3increase).
TherewasalsoanassociationforPM2.
5(1.
13,0.
98to1.
30,foreach5g/m3increase)andforcoarseparticles,whereasonlysmallpositiveassociationswerefoundfornitrogenoxides.
Therewasnoevidenceofaneffectoftrafficvariables.
Therewasnoevidenceofstatisticalheterogeneitybetweenthecohortspecificeffectestimates(I260)forPM10andsixcohortsforPM2.
5,wefoundstrongerassociations(hazardratio1.
18(1.
05to1.
33)for10μg/m3PM10,and1.
35(1.
04to1.
74)for5μg/m3PM2.
5).
Whenwetookindividualandarealevelcovariatesintoaccount,wefoundnoevidenceofclusteringintheneighbourhoods(datanotshown).
EffectmodificationbyageshowedstrongerPM2.
5effectsforthoseaged60-74(hazardratio1.
25,95%confidenceinterval1.
03to1.
51)andforthose>75(1.
18,0.
85to1.
64)thanamongthoseagedunder60(0.
91,0.
71to1.
15),withP=0.
11foreffectmodification.
Forallothereffectmodifiers,thePvalueofeffectmodificationwasfarfromsignificant(≥0.
22).
Table7showstheresultsofthethresholdanalysisforPM2.
5andPM10.
WhenwerestrictedtheanalysistoparticipantsfromninecohortsexposedtoPM2.
5concentrationsbelowthecurrentEuropeanlimitvalue(70comparedwith60%),wefoundincreasedeffectestimates.
Thiscouldbebecauseofreducedmisclassificationofexposure.
ThecohortswithvalidationR2≤60%,however,hadyoungparticipantsand(inparticulartheItalianones)hadhigherratesofsmokingandabetterexposureassessmentmightbenottheonlyexplanation.
ConclusionsOurstudysuggestsanassociationbetweenlongtermexposuretoinhalableparticulatematterandincidenceofcoronaryevents.
TheseassociationsremainedforexposureconcentrationsbelowthecurrentEuropeanlimits.
Theresultsofthisstudy,togetherwithotherESCAPEfindings,supportloweringofEuropeanlimitsforparticulateairpollutiontoadequatelyprotectpublichealth.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page5of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromWhatisalreadyknownonthistopicAmbientparticulatematterairpollutionisestimatedtocause3.
1milliondeathsworldwideperyear,and22%ofDALYs(disabilityadjustedlifeyears)fromischaemicheartdiseaseTheassociationbetweenlongtermexposuretoairpollutionandincidenceofcoronaryeventsremainscontroversialWhatthisstudyaddsThereisanassociationbetweenoutdoorparticulatematterandincidenceofacutecoronaryevents,evenforexposurelevelsbelowthecurrentEuropeanlimitsTheburdenofdiseaseattributabletooutdoorparticulatemattermightbeunderestimatedifonlyestimatesofmortalityareconsideredTheresultsofthisstudysupportloweringoftheEUlimitsforparticulatematterairpollutionWethankMarjanTewis,MariekeOldenwening,MarloesEeftens,UlrichQuass,Lee-JaneSallyLiufortheirhelpwithexposureassessmentanddatamanagementwithinESCAPE.
FortheFinnishpart,additionalfundingcamefromtheAcademyofFinland(projectnumber129317).
ExposureassessmentwasperformedbyTarjaYli-Tuomi,PekkaTaimisto,andArtoPennanenfromtheDepartmentofEnvironmentalHealth.
Mortality,area-levelSES,andbuildingdatawereprovidedbyStatisticsFinland.
ThefourSwedishcohorts(SNAC-K,SALT,60y,SDPP)werepartiallyfundedbytheSwedishEnvironmentalProtectionAgency,theSwedishCouncilforWorkingLifeandSocialResearchandtheSwedishHeart-LungFoundation.
TheSALTcohortwasadditionallysupportedbyNIHgrantAG-08724.
The60yearcohortwasadditionallyfundedbytheStockholmCountyCouncilandtheSwedishResearchCouncil(longitudinalresearchand0593).
TheSDPPcohortwasadditionallyfundedbytheStockholmCountyCouncil;theSwedishResearchCouncil;theSwedishDiabetesAssociation;andtheNovoNordiskScandinavia.
WethanktheHeinzNixdorfFoundation,theGermanMinistryofEducationandScience(BMBF),theGermanAerospaceCenter(DeutschesZentrumfürLuft-undRaumfahrt(DLR)),andtheGermanResearchFoundation(DFGHO3314/2-1andJO170/8-1),Bonn,Germany,fortheirgeneroussupportofHeinzNixdorfRecallstudy.
TheKORAresearchplatformandtheMONICAAugsburgstudieswereinitiatedandfinancedbytheHelmholtzZentrumMünchen,GermanResearchCenterforEnvironmentalHealth,whichisfundedbytheGermanFederalMinistryofEducationandResearchandbytheStateofBavaria.
TheSIDRIAcohortstudywaspartiallyfundedbytheItalianMinistryofHealth.
GiuseppeCostaprovideddatafromtheTurinLongitudinalStudyforthefollow-upoftheSIDRIA-Turincohort.
ExposureassessmentinTurinwasperformedbyDanielaRaffaeleandMarcoGilardetti.
NoiseassessmentinTurinwasperformedbyDanieleGrassoandJacopoFogola(RegionalAgencyfortheProtectionoftheEnvironmentofPiedmont).
WethankSimoneBucci,PatrizioPasquinelli,andEleonoraZirroforexposureassessmentinRome.
Contributors:GCcontributedtothedesign,exposureassessment,dataanalyses,anddraftedthemanuscript;FFcontributedtothedesign,providedlocalcohortdataanddraftedthemanuscript;MScontributedtothedesign,statisticalscript,anddataanalyses;ZJA,RHcontributedtothestatisticalscriptanddataanalysis;CB,FH,EM,JPen,APy,FR,andATcontributedtothedataanalyses;RB,GHcontributedtothedesign,exposureassessmentandstatisticalscript;BC,UdF,RE,LF,MH,AH,KHJ,KL,PKEM,CGO,KO,NLP,JPek,CS,andVSprovidedlocalcohortdata;CGcontributedtoexposureassessmentandprovidedlocalcohortdata;BH,ORN,GP,andPVcontributedtothedesignandprovidedlocalcohortdata;MKcontributedtotheexposureassessmentandthedataanalyses;KTE,DH,TL,AR,WS,andKdHcontributedtoexposureassessment;GWandKWcontributedtothestatisticalscript;BBandAPecontributedtothedesignanddraftedthemanuscript.
Allauthorscontributedtocriticalreadingofandcommentstothemanuscript,interpretationofdataandapprovedthefinaldraft.
GCisguarantor.
Funding:TheresearchleadingtotheseresultshasreceivedfundingfromtheEuropeanCommunity'sSeventhFrameworkProgram(FP7/2007-2011)undergrantagreementNo211250.
Thesourcesoffundinghadnoroleinstudydesign,datacollection,analyses,interpretation,anddecisiontosubmitthearticleforpublication.
Competinginterests:AllauthorshavecompletedtheICMJEuniformdisclosureformatwww.
icmje.
org/coi_disclosure.
pdfanddeclare:nosupportfromanyorganisationforthesubmittedwork;nofinancialrelationshipswithanyorganisationsthatmighthaveaninterestinthesubmittedworkinthepreviousthreeyears;nootherrelationshipsoractivitiesthatcouldappeartohaveinfluencedthesubmittedwork.
Ethicalapproval:Theoriginalcohortstudieswereapprovedbyappropriateinstitutionalreviewboardscomplyingwithallrelevantnational,state,andlocalregulations.
Transparencystatement:Theleadauthoraffirmsthatthismanuscriptisanhonest,accurate,andtransparentaccountofthestudybeingreported;thatnoimportantaspectsofthestudyhavebeenomitted;andthatanydiscrepanciesfromthestudyasplannedhavebeenexplained.
Datasharing:meta-analyticdataandstatisticalcodeareavailablefromthecorrespondingauthor.
1LimSS,VosT,FlaxmanAD,DaneiG,ShibuyaK,Adair-RohaniH,etal.
Acomparativeriskassessmentofburdenofdiseaseandinjuryattributableto67riskfactorclustersin21regions,1990-2010:asystematicanalysisfortheGlobalBurdenofDiseaseStudy2010.
Lancet2012;380:2224-60.
2BeelenR,HoekG,vandenBrandtPA,GoldbohmRA,FischerP,SchoutenLJ,etal.
Long-termeffectsoftraffic-relatedairpollutiononmortalityinaDutchcohort(NLCS-AIRstudy).
EnvironHealthPerspect2008;116:196-202.
3ChenLH,KnutsenSF,ShavlikD,BeesonWL,PetersenF,GhamsaryM,etal.
Theassociationbetweenfatalcoronaryheartdiseaseandambientparticulateairpollution:ArefemalesatgreaterriskEnvironHealthPerspect2005;113:1723-9.
4CrouseDL,PetersPA,vanDonkelaarA,GoldbergMS,VilleneuvePJ,BrionO,etal.
Riskofnonaccidentalandcardiovascularmortalityinrelationtolong-termexposuretolowconcentrationsoffineparticulatematter:aCanadiannational-levelcohortstudy.
EnvironHealthPerspect2012;120:708-14.
5GanWQ,KoehoornM,DaviesHW,DemersPA,TamburicL,BrauerM.
Long-termexposuretotraffic-relatedairpollutionandtheriskofcoronaryheartdiseasehospitalizationandmortality.
EnvironHealthPerspect2011;119:501-7.
6JerrettM,BurnettRT,MaR,PopeCA3rd,KrewskiD,NewboldKB,etal.
SpatialanalysisofairpollutionandmortalityinLosAngeles.
Epidemiology2005;16:727-36.
7KrewskiD,JerrettM,BurnettRT,MaR,HughesE,ShiY,etal.
Extendedfollow-upandspatialanalysisoftheAmericanCancerSocietystudylinkingparticulateairpollutionandmortality.
HEIResearchReport140.
HealthEffectsInstitute,2009.
8LadenF,SchwartzJ,SpeizerFE,DockeryDW.
Reductioninfineparticulateairpollutionandmortality:extendedfollow-upoftheHarvardSixCitiesstudy.
AmJRespirCritCareMed2006;173:667-72.
9MillerKA,SiscovickDS,SheppardL,ShepherdK,SullivanJH,AndersonGL,etal.
Long-termexposuretoairpollutionandincidenceofcardiovasculareventsinwomen.
NEnglJMed2007;356:447-58.
10PopeCA3rd,BurnettRT,ThurstonGD,ThunMJ,CalleEE,KrewskiD,etal.
Cardiovascularmortalityandlong-termexposuretoparticulateairpollution:epidemiologicalevidenceofgeneralpathophysiologicalpathwaysofdisease.
Circulation2004;109:71-7.
11PuettRC,HartJE,YanoskyJD,PaciorekC,SchwartzJ,SuhH,etal.
Chronicfineandcoarseparticulateexposure,mortality,andcoronaryheartdiseaseintheNurses'HealthStudy.
EnvironHealthPerspect2009;117:1697-701.
12LipsettMJ,OstroBD,ReynoldsP,GoldbergD,HertzA,JerrettM,etal.
Long-termexposuretoairpollutionandcardiorespiratorydiseaseintheCaliforniateachersstudycohort.
AmJRespirCritCareMed2011;184:828-35.
13Raaschou-NielsenO,AndersenZJ,JensenSS,KetzelM,SrensenM,HansenJ,etal.
Trafficairpollutionandmortalityfromcardiovasculardiseaseandallcauses:aDanishcohortstudy.
EnvironHealth2012;11:60.
14AtkinsonRW,CareyIM,KentAJ,vanStaaTP,AndersonHR,CookDG.
Long-termexposuretooutdoorairpollutionandincidenceofcardiovasculardiseases.
Epidemiology2013;24:44-53.
15BrookRD,RajagopalanS,PopeCA3rd,BrookJR,BhatnagarA,Diez-RouxAV,etal;AmericanHeartAssociationCouncilonEpidemiologyandPrevention,CouncilontheKidneyinCardiovascularDisease,andCouncilonNutrition,PhysicalActivityandMetabolism.
Particulatematterairpollutionandcardiovasculardisease:anupdatetothescientificstatementfromtheAmericanHeartAssociation.
Circulation2010;121:2331-78.
16ESCAPE.
Europeanstudyofcohortsforairpollutioneffects.
www.
escapeproject.
eu.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page6of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
Downloadedfrom17BeelenR,Raaschou-NielsenO,StafoggiaM,AndersenZJ,WeinmayrG,HoffmannB,etal.
Effectsoflong-termexposuretoairpollutiononnaturalcausemortality:ananalysisof22Europeancohortswithinthemulti-centerESCAPEproject.
LancetEarlyonlinepublication9December2013.
http://dx.
doi.
org/10.
1016/S0140-6736(13)62158-3.
18VartiainenE,LaatikainenT,PeltonenM,JuoleviA,MnnistS,SundvallJ,etal.
Thirty-five-yeartrendsincardiovascularriskfactorsinFinland.
IntJEpidemiol2010;39:504-18.
19MeinowB,KreholtI,LagergrenM.
AccordingtoneedPredictingtheamountofmunicipalhomehelpallocatedtoelderlyrecipientsinanurbanareaofSweden.
HealthSocCareCommunity2005;13:366-377.
20LichtensteinP,SullivanPF,CnattingiusS,GatzM,JohanssonS,CarlstrmE,etal.
TheSwedishTwinRegistryintheThirdMillennium—anupdate.
TwinResHumGenet2006;9:875-82.
21WandellPE,WajngotA,deFaireU,HelleniusML.
Increasedprevalenceofdiabetesamongimmigrantsfromnon-Europeancountriesin60-year-oldmenandwomeninSweden.
DiabetMetab2007;33:30-6.
22ErikssonAK,EkbomA,GranathF,HildingA,EfendicS,OstensonCG.
Psychologicaldistressandriskofpre-diabetesandType2diabetesinaprospectivestudyofSwedishmiddle-agedmenandwomen.
DiabetMed2008;25:834-42.
23TjnnelandA,OlsenA,BollK,StrippC,ChristensenJ,EngholmG,etal.
Studydesign,exposurevariables,andsocioeconomicdeterminantsofparticipationindiet,cancerandhealth:apopulation-basedprospectivecohortstudyof57,053menandwomeninDenmark.
ScandJPublicHealth2007;35:432-41.
24ErbelR,MhlenkampS,MoebusS,SchmermundA,LehmannN,StangA,etal.
HeinzNixdorfRecallStudyInvestigativeGroup.
Coronaryriskstratification,discrimination,andreclassificationimprovementbasedonquantificationofsubclinicalcoronaryatherosclerosis:theHeinzNixdorfRecallstudy.
JAmCollCardiol2010;56:1397-406.
25HolleR,HappichM,LwelH,WichmannHE;MONICA/KORAStudyGroup.
KORA—aresearchplatformforpopulationbasedhealthresearch.
Gesundheitswesen2005;67(suppl1):S19-25.
26CesaroniG,BadaloniC,PortaD,ForastiereF,PerucciCA.
Comparisonbetweenvariousindicesofexposuretotraffic-relatedairpollutionandtheirimpactonrespiratoryhealthinadults.
OccupEnvironMed2008;65:683-90.
27VineisP,HoekG,KrzyzanowskiM,Vigna-TagliantiF,VegliaF,AiroldiL,etal.
AirpollutionandriskoflungcancerinaprospectivestudyinEurope.
IntJCancer2006;119:169-74.
28CyrysJ,EeftensM,HeinrichJ,AmpeC,ArmengaudA,BeelenR,etal.
VariationofNO2andNOxconcentrationsbetweenandwithin36Europeanstudyareas:resultsfromtheESCAPEstudy.
AtmosphEnviron2012;62:374-90.
29EeftensM,TsaiM,AmpeC,AnwanderB,BeelenR,BellanderT,etal.
SpatialvariationofPM2.
5,PM10,PM2.
5absorbanceandPMcoarseconcentrationsbetweenandwithin20EuropeanstudyareasandtherelationshipwithNO2—resultsoftheESCAPEproject.
AtmosphEnviron2012;62:303-17.
30EeftensM,BeelenR,deHooghK,BellanderT,CesaroniG,CirachM,etal.
DevelopmentoflanduseregressionmodelsforPM(2.
5),PM(2.
5)absorbance,PM(10)andPM(coarse)in20EuropeanStudyAreas;ResultsoftheESCAPEProject.
EnvironSciTechnol2012;46:11195-205.
31BeelenR,HoekG,VienneauD,EeftensM,DimakopoulouK,PedeliX,etal.
DevelopmentofNO2andNOxlanduseregressionmodelsforestimatingairpollutionexposurein36studyareasinEurope—theESCAPEproject.
AtmosphEnvirondoi:10.
1016/j.
atmosenv.
2013.
02.
037.
32DiezRouxAV,MerkinSS,HannanP,JacobsDR,KiefeCI.
Areacharacteristics,individual-levelsocioeconomicindicators,andsmokinginyoungadults:thecoronaryarterydiseaseriskdevelopmentinyoungadultsstudy.
AmJEpidemiol2003;157:315-26.
33DerSimonianR,LairdN.
Meta-analysisinclinicaltrials.
ControlClinTrials1986;7:177-88.
34HigginsJP,ThompsonSG.
Quantifyingheterogeneityinameta-analysis.
StatMed2002;21:1539-58.
35ZanobettiA,SchwartzJ.
Theeffectofparticulateairpollutiononemergencyadmissionsformyocardialinfarction:amulticitycase-crossoveranalysis.
EnvironHealthPerspect2005;113:978-82.
36PopeCAIII,MuhlesteinJB,MayHT,RenlundDG,AndersonJL,HorneBD.
Ischemicheartdiseaseeventstriggeredbyshort-termexposuretofineparticulateairpollution.
Circulation2006;114:2443-8.
37PetersA.
Ambientparticulatematterandtheriskforcardiovasculardisease.
ProgCardiovascDis2011;53:327-33.
38BeelenR,StafoggiaM,Raaschou-NielsenM,AndersenZJ,XunWW,KatsouyanniK,etal.
Long-termexposuretoairpollutionandcardiovascularmortality:ananalysisof22EuropeancohortswithintheESCAPEproject.
Epidemiology(inpress).
39HoekG,KrishnanRM,BeelenR,PetersA,OstroB,BrunekreefB,etal.
Long-termairpollutionexposureandcardio-respiratorymortality:areview.
EnvironHealth2013;12:43.
40CesaroniG,PortaD,BadaloniC,StafoggiaM,EeftensM,MeliefsteK,etal.
Nitrogendioxidelevelsestimatedfromlanduseregressionmodelsseveralyearsapartandassociationwithmortalityinalargecohortstudy.
EnvironHealth2012;11:48.
41EeftensM,BeelenR,FischerP,BrunekreefB,MeliefsteK,HoekG.
StabilityofmeasuredandmodelledspatialcontrastsinNO2overtime.
OccupEnvironMed2011;68:765-70.
42WangR,HendersonSB,SbihiH,AllenRW,BrauerM.
Temporalstabilityoflanduseregressionmodelsfortraffic-relatedairpollution.
AtmosphEnviron2013;64:312-9.
Accepted:04December2013Citethisas:BMJ2014;348:f7412ThisisanOpenAccessarticledistributedinaccordancewiththeCreativeCommonsAttributionNonCommercial(CCBY-NC3.
0)license,whichpermitsotherstodistribute,remix,adapt,builduponthisworknon-commercially,andlicensetheirderivativeworksondifferentterms,providedtheoriginalworkisproperlycitedandtheuseisnon-commercial.
See:http://creativecommons.
org/licenses/by-nc/3.
0/.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page7of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTablesTable1|Summarydescriptionof11EuropeancohortsfromfivecountriescontributingdatatostudylongtermexposuretoambientairpollutionandincidenceofacutecoronaryeventsSIDRIAEPICTurinKORAHNRDCHSDPP60yearoldsSALTSNAC-KFINRISKVariableRomeTurin920051377230830144333569377233686608426849995Noofparticipants10289456366914908459534941464055106995399785175616256105060Personyearsatrisk87.
195.
582.
494.
099.
396.
097.
691.
092.
389.
493.
5%oforiginalcohort2051231572821353293181165204200212NoofcasesRome(Italy)Turin(Italy)Turin(Italy)Augsburg(Germany)RuhrArea(Germany)Copenhagen(Denmark)Stockholm(Sweden)Stockholm(Sweden)Stockholm(Sweden)Stockholm(Sweden)TurkuandHelsinki(Finland)Studylocation199919991993-981994-95,1999-20012000-031993-971992-981997-991998-20022001-041992,1997,2002,2007YearsofenrolmentNocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page8of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable2|Individualbaselinecharacteristicsfrom11Europeancohortsfromfivecountriescontributingdatatostudylongtermexposuretoambientairpollutionandincidenceofacutecoronaryevents.
Valuesarenumbers(percentages)unlessstatedotherwiseSIDRIAEPICTurinKORAHNRDCHSDPP60yearoldsSALTSNAC-KFINRISKVariableRomeTurin9200(100)5137(100)7230(100)8301(100)4433(100)35693(100)7723(100)3686(100)6084(100)2684(100)9995(100)Noofparticipants44(6)44(6)50(8)49(14)59(8)57(4)47(5)60(0)59(11)74(11)48(13)Mean(SD)age(years)4859(53)2677(52)3472(48)4275(51)2309(52)19302(54)4721(61)1967(53)3557(58)1751(65)5460(55)WomenMaritalstatus:0(0)121(2)438(6)879(11)256(6)2352(7)1272(16)*170(5)835(14)397(15)1601(16)Single9200(100)4897(95)6183(86)6292(76)3319(75)24673(69)6451(84)2628(71)4103(67)1263(47)7009(70)Married/livingwithpartner0(0)37(1)375(5)620(7)444(10)6599(18)—630(17)678(11)362(13)1051(11)Divorced/separated0(0)82(2)234(3)510(6)414(9)2069(6)—258(7)468(8)662(25)334(3)WidowedEducation:4130(45)898(17)3168(44)1024(12)499(11)10589(30)1996(26)1030(28)1332(22)705(26)3029(30)Primaryschoolorless3690(40)3668(71)3081(43)6200(65)2466(56)16943(47)3453(45)1631(44)2606(43)1128(42)5217(52)Secondaryschool1380(15)571(11)981(14)1077(13)1468(33)8161(23)2274(29)1025(28)2146(35)851(32)1749(17)UniversitydegreeandmoreOccupationalstatus:6500(71)3727(73)—4908(59)1859(42)28600(80)7088(92)1881(51)—2016(75)7092(71)Employed/selfemployed393(4)349(7)—272(3)630(14)7093(20)635(8)374(10)—668(25)610(6)Unemployed2307(25)1061(21)—1188(14)1664(38)——290(8)——351(4)Homemakerorhousewife0(0)0(0)—1933(23)280(6)——1141(31)——1942(19)RetiredSmokingstatus:3884(42)2117(41)1768(24)2162(26)1040(23)12793(36)2022(26)751(20)1224(20)394(15)2603(26)Currentsmoker2147(23)1085(21)2368(33)2517(30)1482(33)9944(28)2815(36)1429(39)2678(44)927(34)2808(28)Formersmoker3169(34)1935(38)3094(43)3622(44)1911(43)12956(36)2886(37)1506(41)2182(36)1363(51)4584(46)Neversmoker18(7)18(8)23(10)21(13)36(9)§29(10)20(10)26(13)—30(17)15(12)Mean(SD)yearsofsmokingamongeversmokers15(9)15(9)14(9)15(11)17(12)17(10)14(7)13(7)13(8)11(8)15(9)Mean(SD)Noofcigarettes/dayamongcurrentsmokers*Allexceptmarried/livingwithpartner.
29Physicalactivity(hours/week):——1660(23)3107(37)2263(51)15984(45)846(11)2510(69)1575(26)464(20)1808(18)2Alcoholconsumption(drinks/week):——457(6)3867(47)1015(23)776(2)594(8)168(5)—519(19)1320(13)Never——201(3)1212(15)1190(27)34142(98)2876(38)771(21)—634(24)4566(46)1-3——347(5)549(7)2127(49)—3873(51)1622(44)—1236(46)3139(32)3-6——6225(86)2662(32)—280(4)1123(30)—288(11)824(8)>6Diabetesmellitus:9179(100)5111(99)7118(98)7940(96)3877(87)34943(98)7597(98)3533(96)5833(96)2464(92)9546(96)No21(0)26(1)110(2)360(4)556(13)703(2)126(2)153(4)251(4)220(8)426(4)YesHypertension:9124(99)5082(99)3629(53)5098(62)2002(45)30033(84)5801(76)1770(48)4735(78)902(34)5877(59)No76(1)55(1)3255(47)3190(38)2423(55)5625(16)1837(24)1915(52)1347(22)1762(66)4042(41)Yes———229(44)231(39)——231(41)——211(78)Mean(SD)serumcholesterolNoiseexposureatbaselineaddress(dB):—62(1)44(1)466(6)721(17)126(0)—84(6)123(5)74Livinginareasoflowurbanisation:9200(100)5137(100)7230(100)3557(43)4433(100)14155(40)0(0)1450(39)2763(45)2684(100)9086(91)No0(0)0(0)0(0)4744(57)0(0)21538(60)7723(100)2236(61)3321(55)0(0)909(9)Yes*FINRISK:medianincomeratein3kmarea;SNAC-K:meanincomeinthirdsatsmallneighbourhoodlevel(SmallAreaforMarketStatistics);SALTandSDPP:meanincomeinfourcategoriesatmunicipalitylevel;60years:meanincomeinquartersatsmallneighbourhoodlevel(SmallAreaforMarketStatistics);DCH:meanincomeatmunicipalitylevel(16unitswithmedianpopulationof1500inhabitants),per/100000;HNR:unemploymentrateatneighbourhoodlevel;KORA:percentageoflowincomein5kmgrid;EPIC-Turin,SIDRIA-Turin,andSIDRIA-Rome:deprivationindex,censusblocklevel(average500inhabitants).
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page11of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable4(continued)SIDRIAEPICTurinKORAHNRDCHSDPP60yearoldsSALTSNAC-KFINRISKRomeTurinAnycategoryofdrinking.
Firstandlasttwoclassesarecollapsed.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page12of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable5|Associationbetweenexposuretopollutantsandincidenceofcoronaryevents.
ResultsexpressedperfixedincrementsPooledhazardratios(95%CI)ParticipantsCohortsFixedincrementExposureModel3Model2Model1*1.
12(1.
01to1.
25)1.
12(1.
01to1.
24)1.
15(1.
04to1.
28)1001661110PM10(g/m3)1.
06(0.
98to1.
15)1.
06(0.
98to1.
14)1.
08(1.
00to1.
17)100166115CoarsePM(g/m3)1.
13(0.
98to1.
30)1.
15(1.
00to1.
32)1.
22(1.
04to1.
44)100166115PM2.
5(g/m3)1.
10(0.
98to1.
24)1.
10(0.
98to1.
24)1.
18(1.
05to1.
32)100166111AbsorbancePM2.
5(105/m)1.
03(0.
97to1.
08)1.
03(0.
96to1.
11)1.
04(0.
96to1.
12)1001661110NO2(g/m3)1.
01(0.
98to1.
05)1.
01(0.
98to1.
05)1.
03(1.
00to1.
07)1001661120NOx(g/m3)1.
01(0.
98to1.
04)1.
01(0.
98to1.
04)1.
01(0.
98to1.
04)95733105000TrafficintensityonnearestroadadjustedforbackgroundNO2(vehicles/day)§1.
00(0.
95to1.
06)1.
00(0.
95to1.
06)1.
02(0.
96to1.
08)100166114000000Trafficloadonmajorroadsin100mbufferadjustedforbackgroundNO2(vehicles*m/day)*Adjustedforage(timevariable),yearofenrolmentandsexAsformodel1plusadjustedformaritalstatus,education,occupation,smokingstatus,smokingduration,andsmokingintensity.
Asformodel2plusadjustedforsocioeconomicarealevelvariables.
§AllcohortsexceptHNRstudy.
24Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page13of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable6|Incidenceofcoronaryevents,resultsfromsensitivityanalyses.
Figuresarepooledhazardratios(and95%CI)for10g/m3PM10and5g/m3PM2.
5,I2statistics,andPvalueforheterogeneityHazardratio(95%CI)Noofparticipants*NoofcohortsModel5g/m3PM2.
510g/m3PM101.
13(0.
98to1.
30)1.
12(1.
01to1.
25)10016611Basemodel(model3)RoleofcardiovascularriskfactorsIntermediatevariables:diabetesandhypertension:1.
11(0.
96to1.
28)1.
11(1.
00to1.
24)9952611+diabetesandhypertensionPhysicalactivity,alcoholandBMI:1.
22(1.
03to1.
44)1.
17(1.
04to1.
33)775848Basemodel,onsubsetofparticipantswithadditionalinformation1.
19(1.
01to1.
42)1.
17(1.
04to1.
33)775848+additionalinformationAllcardiovascularriskfactors(diabetes,hypertension,physicalactivity,alcohol,BMI,cholesterol)§:1.
28(0.
92to1.
79)1.
19(0.
92to1.
53)256254Basemodel,onsubsetofparticipantswithadditionalinformation1.
29(0.
92to1.
81)1.
20(0.
93to1.
56)256254+additionalinformationUrban/suburban-ruralresidencelocation:1.
13(0.
98to1.
39)1.
12(1.
01to1.
25)10016611Basemodel,onsubsetofparticipantswithadditionalinformation1.
14(0.
99to1.
31)1.
13(1.
02to1.
26)10016611+additionalinformationNoise:1.
16(0.
98to1.
37)1.
14(1.
01to1.
30)738409Basemodel,onsubsetofparticipantswithnoiseinformation1.
13(0.
95to1.
36)1.
14(1.
01to1.
31)738409+noisevariableChangeofaddress**:1.
14(0.
98to1.
33)1.
11(0.
99to1.
23)9293610Basemodel,onsubsetofcohortswithchangeofaddressinformation1.
18(0.
98to1.
42)1.
16(1.
01to1.
32)6312110Participantswhodidn'tchangeaddressduringfollow-upProportionalityhazards(PH)assumption:1.
13(0.
98to1.
30)1.
10(0.
99to1.
23)10016611VariablesthatdonotmeetPHputasstrataExclusionofDCHStudy1.
09(0.
92to1.
30)1.
12(0.
99to1.
27)6447310DataafterexclusionPerformanceoflanduseregressionmodel:1.
35(1.
04to1.
74)1.
18(1.
05to1.
33)——LeaveoneoutcrossvalidationR2>60%1.
05(0.
89to1.
25)0.
88(0.
68to1.
12)——LeaveoneoutcrossvalidationR2≤60%*FiguresdependonnumberofcohortsincludedandonmissingvaluesofincludedvariablesAdjustedforage(timevariable),yearofenrolment,sex,maritalstatus,education,occupation,smokingstatus,smokingduration,smokingintensity,andsocioeconomicarealevelindicators.
ExcludesSALT,SIDRIA-Turin,andSIDRIA-Rome.
§IncludesFINRISK,60yearolds,HNR,andKORA.
AllcohortsexceptSDPPandSIDRIA-Rome.
**AllcohortsexceptEPIC-Turin.
ForPM10:ninecohorts,80971participants(excludesFINRISKandSIDRIA-Rome).
ForPM2.
5:sixcohorts,42906participants(excludesFINRISK,DCH,EPIC-Turin,SIDRIA-Turin,andSIDRIA-Rome).
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page14of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromTable7|Resultsfromrandom-effectsmeta-analysesforadjustedassociationbetweenincidenceofcoronaryeventsandexposuretoPM10andPM2.
5belowvariousthresholdvalues.
Figuresarepooledhazardratios(95%CI)for10g/m3PM10and5g/m3PM2.
5,I2statistic,andPvalueforheterogeneityCohortswithexposuredataavailableforallthresholdsCohortswithexposuredataavailableforrespectivethresholdExposureandthresholdPvalueI2HR*(95%CI)NoofparticipantsNoofcohortsPvalueI2HR*(95%CI)NoofparticipantsNoofcohortsPM10(g/m3):0.
9201.
15(1.
02to1.
30)741667—————Fullrangeofexposure0.
9101.
20(1.
01to1.
41)6436370.
9101.
20(1.
01to1.
41)643637<200.
9101.
12(0.
98to1.
27)7401970.
9401.
12(0.
98to1.
27)779978<300.
9201.
15(1.
02to1.
30)7416670.
4801.
12(1.
00to1.
27)862229§<40PM2.
5(g/m3):0.
5701.
23(1.
04to1.
46)741667—————Fullrangeofexposure0.
7401.
19(1.
00to1.
42)7342070.
7401.
19(1.
00to1.
42)734207<150.
5701.
23(1.
04to1.
46)7416670.
18301.
17(0.
91to1.
50)852169§<200.
5701.
23(1.
04to1.
46)7416670.
5601.
18(1.
01to1.
39)875329§<25*Adjustedforage(timevariable),yearofenrolment,sex,maritalstatus,education,occupation,smokingstatus,smokingduration,smokingintensity.
andsocioeconomicarealevelindicatorsAllcohortsexceptHeinzNixdorfRecallStudy,EPIC-Turin,SIDRIA-TurinandSIDRIA-Rome.
AllcohortsexceptEPIC-Turin,SIDRIA-TurinandSIDRIA-Rome.
§AllcohortsexceptEPIC-TurinandSIDRIA-Turin.
Nocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page15of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
DownloadedfromFigureHazardratiosofincidentcoronaryeventsper10g/m3PM10and5g/m3PM2.
5.
Cohortspecificandmeta-analyticresultsNocommercialreuse:Seerightsandreprintshttp://www.
bmj.
com/permissionsSubscribe:http://www.
bmj.
com/subscribeBMJ2014;348:f7412doi:10.
1136/bmj.
f7412(Published21January2014)Page16of16RESEARCHon21March2021byguest.
Protectedbycopyright.
http://www.
bmj.
com/BMJ:firstpublishedas10.
1136/bmj.
f7412on21January2014.
Downloadedfrom
- restrictingww.00271.com相关文档
- plateletww.00271.com
- CHEK2ww.00271.com
- movementsww.00271.com
- openww.00271.com
- Currentww.00271.com
- Mermaidww.00271.com
轻云互联,香港云服务器折后22元/月 美国云服务器 1核 512M内存 15M带宽 折后19.36元/月
轻云互联成立于2018年的国人商家，广州轻云互联网络科技有限公司旗下品牌，主要从事VPS、虚拟主机等云计算产品业务，适合建站、新手上车的值得选择，香港三网直连（电信CN2GIA联通移动CN2直连）；美国圣何塞（回程三网CN2GIA）线路，所有产品均采用KVM虚拟技术架构，高效售后保障，稳定多年，高性能可用，网络优质，为您的业务保驾护航。官方网站：点击进入广州轻云网络科技有限公司活动规则：1.用户购...
易探云2核2G5M仅330元/年起，国内挂机宝云服务器,独立ip
易探云怎么样?易探云是国内一家云计算服务商家，致力香港服务器、国内外服务器租用及托管等互联网业务，目前主要地区为运作香港BGP、香港CN2、广东、北京、深圳等地区。目前，易探云推出深圳或北京地区的适合挂机和建站的云服务器，国内挂机宝云服务器（可选深圳或北京地区）,独立ip;2核2G5M挂机云服务器仅330元/年起！点击进入：易探云官方网站地址易探云国内挂机宝云服务器推荐：1、国内入门型挂机云服务器...
ShockHosting日本机房VPS测试点评
这个月11号ShockHosting发了个新上日本东京机房的邮件，并且表示其他机房可以申请转移到日本，刚好赵容手里有个美国的也没数据就发工单申请新开了一个，这里做个简单的测试，方便大家参考。ShockHosting成立于2013年，目前提供的VPS主机可以选择11个数据中心，包括美国洛杉矶、芝加哥、达拉斯、杰克逊维尔、新泽西、澳大利亚、新加坡、日本、荷兰和英国等。官方网站：https://shoc...
ww.00271.com为你推荐
-
今日油条油条晚上炸好定型明天可再复炸吗？今日油条油条的由来及历史甲骨文不满赔偿公司倒闭员工不满一年怎么赔偿陈嘉垣反黑阿欣是谁演的 扮演者介绍www.522av.com在白虎网站bhwz.com看电影要安装什么播放器？8090lu.com《8090》节目有不有高清的在线观看网站啊？www.zjs.com.cn怎么查询我的平安信用卡寄送情况4400av.com在www.dadady.com 达达电影看片子很快的啊杨丽晓博客明星的最新博文555sss.com拜求：http://www.jjj555.com/这个网站是用的什么程序