Biostatistics-Main-生物统计学-41-54

BiostatisticsMain41-54Q41A43-year-oldAfricanAmericanmalewithastrongfamilyhistoryofhypertensionpresentstotheofficewithgeneralizedfatigueHispastmedicalhistoryisnoncontributoryHeisanactivesmokerHeiscurrentlytakingmetoproloLHisbloodpressureis162/100mmHgandhispulseis76/minTheapicalimpulseislaterallydisplaced.HisphysicalexaminationisotherwiseunremarkableEKGshowshighvoltageconsistentwithleftventricularhypertrophywithsecondaryrepolarizationchanges.Hisserumpotassiumis3.9mEq/LandestimatedGFRis36mL/min/1.73m2.HisbaselineUP/Cris0.20.ThepatientisstartedonanACEinhibitorandheshowscompliancewithfollow-upvisits.Basedonstudyresults,whichofthefollowingfactorsonfollow-upwillmostsignificantlyincreasehischanceofhyperkalemiaevent?A.Additionof3calciumchannelblockerB.AdditionofathiazidediureticC.DiscontinuationofmetoprololD.Follow-upserumpotassiumlevelof4.7meq/lascomparedlobaselineE.Follow-upUP/Crof1.1ascomparedtobaselineA41Correctanswer:DThisstudyisadoubleblindedrandomizedclinicaltrialcomparingtheIncidenceofhyperkalemiainAfrican-American(non-diabetic)patientswithCKDandhypertension(diastolicgreaterthan95mmHg).Theparentswererandomizedtoeithermetoprololamlodipine,orramiprilandfollowed,upto6years.Theprimaryendpointwastheincidenceofhyperkalemiadefinedasgreaterthan5.5mEq/lTheresultswerereportedaseventsper100parentyearsandalsoreportedashazardratios.Hazardratioistheratioofaneventrateoccurringinthetreatmentgroupcomparedtoaneventrateoccurringinthenon-treatmentgroupAratiolessthan1indicatesthatthetreatmentgrouphadasignificantlylowereventrolewhilevaluesgreaterthan1indicatethatthetreatmentgrouphadamuchhighereventrate,Inthisstudy,theriskfactorsfordevelopinghyperkalemiawerereportedashazardratiosThehighertheratio,themorelikelythatfactorisassociatedwiththeincidenceofhyperkalemia.Thefactorthathadthehighestratio(ofthechoiceslisted)isafollow-upserumpotassiumlevelbetween4-5mEq/lascomparedtolessthan4mEq/l,withahazardratioof7.25.(ChoiceA)AsshownintheresultsfigurepatientswithaGFRbetween30-40mL/m/1.73mgivenamlodipineactuallyhadalowerincidenceofhyperkalemiaper100patientyearsthanramipril.(ChoiceB)Thehazardratioforfollow-upuseofadiureticwas0.41(withastatisticallysignificantpvalue0.01).Indicatingthataddingadiureticisassociatedwithalowerincidenceofhyperkalemia.(ChoiceC)thispatientisalreadyonmetoprololandhasanormalpotassium.TheadditionofmetoprololinthestudydidnotsignificantlyincreasetheincidenceofhyperkalemiainpatientswithGFR30and≤40accordingtothefigureabove,sodiscontinuationofthedruglikelywouldnothaveasignificanteffectonthepotassium.(ChoiceE)Follow-upUP/Crof1,1ascomparedtobaselinehasahazardratioof1.84Pindicatingthatthereisanincreasedincidenceofhyperkalemia.However,thisratioIslowerthanthehazardratioforfollow-uppotassiumlevelbetween4-5mEq/Landitisnotstatisticallysignificant(p=0.16).Educationalobjective:Hazardratiosaretheratioofaneventrateoccurringinthetreatmentarmversusthenon-treatmentarm.RatiosFessthan1indicatethatthetreatmentarmbadalowereventratewhileratioshigherthanoneindicatethetreatmentarmhadahigherrateofevents.Q42Inaselectedstudyofbronchogeniccancer,itwasfoundthatinpeoplesmoking1packperday(PPD)ofcigarettes,therelativerisk(RR)was2.4.Forthosesmoking2PPG,theRRwas3,0.Forthosesmoking3PPD,theRRwas4.1.Thecorrespondingp-valuesfortheseRRswereall0.05.Whichofthefollowingstatementsistrueregardingthesefindings?A.Cigarettesmokingandbronchogeniccancerhadastrengthofassociation,butthedose-responserelationshipwasnotclearB.Smokingandbronchogeniccancerdidnothavestrengthofassociation,buthadadose-responserelationshipC.Smokingandbronchogeniccancerwereweaklyassociatedanddidnotshowacleardose-responserelationshipD.Smokingandbronchogeniccancerhadastrongassociationandadose-responserelationshipE.Smokingandbronchogeniccancerhadstrengthofassociationbutnodose-responserelationshipA42Correctanswer:DRelativerisk(RR)isusedasameasureofassociationincohortstudies.Itistheratiooftheriskintheexposedgrouptothatintheunexposedgroup.ARR1meansthatthereisapositiveassociationbetweentheriskfactorandtheoutcome,ARR1meansthatthereisanegativeassociationbetweentheriskfactorandtheoutcome.ThefartherthevalueofRRfrom1;thestrongertheassociation.Intheexample,theRRforbronchogeniccancerInsmokersis2,therebyindicatingastrongassociationbetweensmoking(riskfactor/exposure)andbronchogeniccancer(outcome).Whenexposureismeasuredonacontinuousscale(eg,numberofpacksperday[PPD]),theclassificationin2ormoreordinalcategoriesenablestherisktobeassessedasafunctionofexposure,andthedose-responseeffectcanbecalculatedfortheexposureandoutcome.Thisexampleillustratesadose-responserelationshipbetweensmokingandbronchogeniccancer(ie;theRRforbronchogeniccancerincreasesasthenumberofsmokedPPDincreases).(ChoicesA,B,C,andE)Inthestudy,therewasastrongassociationbetweencigarettesmokingandbronchogeniccancer.Therewasalsoacleardose-responserelationshipbetweensmokedPPDandbronchogeniccancer.Educationalobjective:Knowhowtointerpretthestrengthofassociationanddose-responserelationshipfromastudy.Q43Aprospectivecohortstudywasconductedtoevaluatethelong-termeffe