Page 1
Stat 13, Intro. to Statistical Methods for the Life and Health Sciences.
1. Experiments and aspirin example. 2. Random sampling, random assignment, and blocking. 3. Blinding. 4. Portacaval shunt example. 5. Coverage, non-response bias, adherer bias, and clofibrate example. 6. More about confounding factors.
Finish reading chapter 4. http://www.stat.ucla.edu/~frederic/13/F16 .HW2 is due Oct 18 and is problems 2.3.15, 3.3.18, and 4.1.23.FreetutoringintheStatsClub,MondaysandTuesdaysfrom5:00-6:00pminFranz2258A
1
Page 2
1.Experimentsandaspirinexample.
• Inanexperiment,theresearcherssettheleveloftheexplanatoryvariableforeachsubject.
• Theselevelsmaycorrespondtoatreatmentandcontrol.
• Welldesignedexperimentscancontrolforconfoundingvariablesbymakingthetreatmentandcontrolgroupsverysimilarexceptforwhattheexperimentermanipulates.
Page 3
Aspirinexample.
Physicians’HealthStudyI (studyaspirin’saffectonreducingheartattacks.
• Startedin1982with22,071malephysicians.
• Thephysicianswererandomlyassignedintooneoftwogroups.
• Halftooka325mgaspirineveryotherdayandhalftookaplacebo.
Page 4
Results• Intendedtogountil1995,theaspirinstudywas
stoppedin1988afterfindingsignificantresults.• 189(1.7%)heartattacksoccurredintheplacebo
groupand104(0.9%)intheaspiringroup.(45%reductioninheartattacksfortheaspiringroup.)
• Whataboutconfoundingvariables?Couldtheaspiringroupbedifferentthantheplacebogroupinsomeotherways?– Didtheyhaveabetterdiet?– Didtheyexercisemore?– Weretheygeneticallylesslikelytohaveheartattacks?– Weretheyyounger?
Page 5
TheBigIdea• Confoundingvariablesareoftencircumventedin
experimentsduetotherandomassignmentofsubjectstotreatmentgroups.
• Randomlyassigningpeopletogroupstendstobalanceoutallothervariablesbetweenthegroups.
• Soconfoundingvariables,includingonestheresearchersdidn'tanticipate,shouldberoughlyequalizedbetweenthetwogroupsandthereforeshouldnotbeconfounding.
• Thus,causeandeffectconclusionsaresometimespossibleinexperimentsthroughrandomassignment.(Itmustbeawellrunexperiment.)
Page 6
2.Randomsamplingandrandomassignment.
• Withobservationalstudies,randomsamplingisoftendone.Thispossiblyallowsustomakeinferencesfromthesampletothepopulationwherethesamplewasdrawn.
• Withexperiments,randomassignmentisdone.Thispossibleallowsustoconcludecausation.
Page 7
• ThePhysician’sHealthStudy usedrandomassignment.Diditalsouserandomsampling?
• No,hardlyanyexperimentsuserandomsampling,butgettheirsubjectsinotherways.
• ThePhysician’sHealthStudysentoutinvitationlettersandquestionnairestoall261,248malephysiciansbetween40and84yearsofagewholivedintheUnitedStates.
• Ofthe59,285whowerewillingtoparticipateinthetrial,26,062weretoldtheycouldnotbecauseofsomemedicalconditionorcurrentmedicaltreatment.
Page 8
• Sotowhatgroupcanwegeneralizetheresultsthattakingaspirincanreduceheartattacks?– Justphysiciansinthestudy?– Allmalephysiciansbetween40-84yearsold?– Allmalephysicians?– Allmalesbetween40-84yearsolds?– Allmales?– Everyonebetween40-84yearsold?– Everyone?
Page 9
ArticleBaselineDemographicsAfterRandomAssignment
Parameter Placebo(n=129)
Uceris(n=128)
Meanage,years(range) 39.9(12–68) 37.6(13–66)Men 77(59.7) 70(54.7)Women 52(40.3) 58(45.3)Meandiseaseduration(yrs) 6.3 5.5Duration≤1 year,n(%) 23(17.8) 28(21.9)Duration>5 years,n(%) 51(39.5) 44(34.4)Proctosigmoiditis 64(49.6) 58(45.3)Left-sidedcolitis 44(34.1) 37(28.9)MeanbaselineUCDAIscore 6.2 6.5MeanbaselineEIscore 6.6 6.5Priormesalazine use 75(58.1) 66(51.6)Priorsulfasalazineuse 28(21.7) 33(25.8)
Sandborn WJ,TravisS,MoroL,JonesR,Gautille T,Bagin R,HuangM,YeungP,BallardED2nd Once-dailybudesonideMMX®extended-releasetabletsinduceremissioninpatientswithmildtomoderateulcerativecolitis:resultsfromtheCOREIstudy. Gastroenterology 2012Nov;143(5):1218-26
Page 10
BlockingandRandomAssignment
• Thegoalinrandomassignmentistomakethetwogroupsassimilaraspossibleinallwaysotherthanthetreatment.
• Sometimethereareknownconfoundersandyoucanblockon(controlfor)thesevariables.
• Forexample,ifoursubjectsconsistof60%femalesand40%males,wecanforceeachgrouptobe60%femaleand40%male,usingamatchedpairdesign.
• Blockingmakessensewhenthereareknownconfoundersyouwanttocontrolfor.Butrandomlyassigningsubjectstogroupsmakesthemassimilaraspossibleevenintermsofunknownconfounders.
Page 11
3.Blinding.
Eveninexperiments,thetreatmentandcontrolgroupscanbedifferentinwaysotherthantheexplanatoryvariable.Thisisespeciallytruewhentheresponsevariableissomewhatsubjective.Painisanexample.Onestudyfoundthat1/4ofpatientssufferingfrompost-operativepain,whengivenaplacebo(justapillofsugarandwater)claimedtheyexperienced"significantpromptpainrelief".Thisshowsthatpeoplecannotjudgetheirownlevelsofpainverywell,andmaybeinfluencedbythebeliefthattheyhavetakenaneffectivetreatment.Thusinanexperimentwithsucharesponsevariable,researchersshouldensurethatwhoeverisrecordingtheresponsevariable(typicallythesubjecthim/herself)doesnotknowwhetherthesubjectreceivedthetreatmentorthecontrol.Thisiscalledblinding.
Page 12
3.Blinding.Peoplemightnotbeabletojudgetheirownlevelsofpainverywell,andmaybeinfluencedbythebeliefthattheyhavetakenaneffectivetreatment.Thusinanexperimentwithsucharesponsevariable,researchersshouldensurethesubjectdoesnotknowwhetherheorshereceivedthetreatmentorthecontrol.Thisiscalledblinding.Inadouble-blind experiment,neitherthesubjectnortheresearcherrecordingtheresponsevariableknowstheleveloftheexplanatoryvariableforeachsubject(i.e.treatmentorcontrol).
Page 13
4.Portacaval shuntexample.Thefollowingexampleshowstheimportanceofdoingarandomizedcontrolledexperiment.Theportacaval shuntisamedicalprocedureaimedatcurbingbleedingtodeathinpatientswithcirrhosisoftheliver.Thefollowingtablesummarizes51studiesontheportacaval shunt.Thepoorlydesignedstudieswereveryenthusiasticaboutthesurgery,whilethecarefullydesignedstudiesprovethatthesurgeryislargelyineffective.
DegreeofenthusiasmDesign HighModerateNoneNocontrols 24 7 1Controls,butnotrandomized 10 3 2Randomizedcontrolled 0 1 3
Page 14
4.Portacaval shuntexample.Whydidthepoorlydesignedstudiescometothewrongconclusion?Alikelyexplanationisthatinthestudieswherepatientswerenotrandomlyassignedtothetreatmentorcontrolgroup,byandlargethehealthierpatientsweregiventhesurgery.Thisalonecouldexplainwhythetreatmentgroupoutlivedthecontrolgroupinthesestudies.
DegreeofenthusiasmDesign HighModerateNoneNocontrols 24 7 1Controls,butnotrandomized 10 3 2Randomizedcontrolled 0 1 3
Page 15
5.Moreproblemswithstudies,andClofibrate example.Surveysareobservational.• Coverageisacommonissue.Coverageistheextenttowhich
thepeopleyousampledfromrepresenttheoverallpopulation.Asurveyatafancyresearchhospitalinawealthyneighborhoodmayyieldpatientswithhigherincomes,highereducation,etc.
• Non-responsebiasisanothercommonproblem.Poorcoveragemeansthepeoplegettingthesurveydonotrepresentthegeneralpopulation.Non-responsebiasmeansthatoutofthepeopleyougavethesurveyto,thepeopleactuallyfillingitoutandsubmittingitaredifferentfromthepeoplewhodidnot.
• Sameexactissuesinwebsurveys.
Page 16
5.Moreproblemswithstudies,andClofibrate example.Non-responsebiasissimilartoadhererbias,inexperiments.Adrugcalledclofibrate wastestedon3,892middle-agedmenwithhearttrouble.Itwassupposedtopreventheartattacks.1,103assignedatrandomtotakeclofibrate,2,789toplacebo(lactose)group.Subjectswerefollowedfor5years.Isthisanexperimentoranobservationalstudy?
Clofibrate patientswhodiedduringfollowupadherers 15%non-adherers 25%total 20%
Page 17
5.Moreproblemswithstudies,andClofibrate example.Non-responsebiasissimilartoadhererbias,inexperiments.Adrugcalledclofibrate wastestedon3,892middle-agedmenwithhearttrouble.Itwassupposedtopreventheartattacks.1,103assignedatrandomtotakeclofibrate,2,789toplacebo(lactose)group.Subjectswerefollowedfor5years.Isthisanexperimentoranobservationalstudy?
Itisanexperiment.DoesClofibrate work?Clofibrate patientswhodiedduringfollowup
adherers 15%non-adherers 25%total 20%
Page 18
Clofibrate patientswhodiedduringfollowupadherers 15%non-adherers 25%total 20%
--------------------------------------------------------------------------Placebo
adherers 15%nonadherers 28%total 21%
Thosewhotookclofibrate didmuchbetterthanthosewhodidn'tkeeptakingclofibrate.Doesthismeanclofibrate works?
Page 19
Clofibrate patientswhodiedduringfollowupadherers 15%non-adherers 25%total 20%
--------------------------------------------------------------------------Placebo
adherers 15%nonadherers 28%total 21%
Thosewhoadheredtoplaceboalsodidmuchbetterthanthosewhostoppedadhering.
Page 20
Clofibrate patientswhodiedduringfollowupadherers 15%non-adherers 25%total 20%
--------------------------------------------------------------------------Placebo
adherers 15%nonadherers 28%total 21%
Allinalltherewaslittledifferencebetweenthetwogroups.
Page 21
Clofibrate patientswhodiedduringfollowupadherers 15%non-adherers 25%total 20%
--------------------------------------------------------------------------Placebo
adherers 15%nonadherers 28%total 21%
Adherersdidbetterthannon-adherers,notbecauseofclofibrate,butbecausetheywerehealthieringeneral.Why?
Page 22
Clofibrate patientswhodiedduringfollowupadherers 15%non-adherers 25%total 20%
--------------------------------------------------------------------------Placebo
adherers 15%nonadherers 28%total 21%
Adherersdidbetterthannon-adherers,notbecauseofclofibrate,butbecausetheywerehealthieringeneral.Why?• adherersarethetypetoengageinhealthierbehavior.• sickpatientsarelesslikelytoadhere.
Page 23
6.Moreaboutconfoundingfactors.• Byaconfoundingfactor,wemeananalternativeexplanation
thatcouldexplaintheapparentrelationshipbetweenthetwovariables,eveniftheyarenotcausallyrelated.Typicallythisisdonebyfindinganotherdifferencebetweenthetreatmentandcontrolgroup.Forinstance,differentstudieshaveexaminedsmokersandnon-smokersandhavefoundthatsmokershavehigherratesoflivercancer.Oneexplanationwouldbethatsmokingcauseslivercancer.Butisthereanyother,alternativeexplanation?
• Onealternativewouldbethatthesmokerstendtodrinkmorealcohol,anditisthealcohol,notthesmoking,thatcauseslivercancer.
Page 24
6.Moreaboutconfoundingfactors.• Anotherplausibleexplanationisthatthesmokersareprobably
olderonaveragethanthenon-smokers,andolderpeoplearemoreatriskforallsortsofcancerthanyoungerpeople.
• Anothermightbethatsmokersengageinotherunhealthyactivitiesmorethannon-smokers.
• Notethatifonesaidthat“smokingmakesyouwanttodrinkalcoholwhichcauseslivercancer,”thatwouldnotbeavalidconfoundingfactor,sinceinthatexplanation,smokingeffectiveiscausallyrelatedtolivercancerrisk.
Page 25
6.Moreaboutconfoundingfactors.• Aconfoundingfactormustbeplausiblylinkedtoboththe
explanatoryandresponsevariables.Soforinstancesaying“perhapsahigherproportionofthesmokersaremen”wouldnotbeaveryconvincingconfoundingfactor,unlessyouhavesomereasontothinkgenderisstronglylinkedtolivercancer.
• Anotherexample:left-handednessandageatdeath.PsychologistsDianeHalpernandStanleyCoren lookedat1,000deathrecordsofthosewhodiedinSouthernCaliforniainthelate1980s andearly1990sandcontactedrelativestoseeifthedeceasedwererighthanded orlefthanded.Theyfoundthattheaverageagesatdeathofthelefthanded was66,andfortherighthanded itwas75.Theirresultswerepublishedinprestigiousscientificjournals,NatureandtheNewEnglandJournalofMedicine.
Page 26
6.Moreaboutconfoundingfactors.Allsortsofcausalconclusionsweremadeabouthowthisshowsthatthestressofbeinglefthanded inourrighthanded worldleadstoprematuredeath.
Page 27
6.Moreaboutconfoundingfactors.• Isthisanobservationalstudyoranexperiment?
Page 28
6.Moreaboutconfoundingfactors.• Isthisanobservationalstudyoranexperiment?Itisanobservationalstudy.• Arethereplausibleconfoundingfactorsyoucanthinkof?
Page 29
6.Moreaboutconfoundingfactors.• Aconfoundingfactoristheageofthetwopopulationsin
general.Leftiesinthe1980swereonaverageyoungerthanrighties.Manyoldleftieswereconvertedtorightiesatinfancy,intheearly20thcentury,butthispracticehassubsided.Thusinthe1980sand1990s,therewererelativelyfewoldleftiesbutmanyyoungleftiesintheoverallpopulation.Thisaloneexplainsthediscrepancy.