资源描述：
《提高中草药随机对照试验的质量ⅰ：临床试验设计和方法学》由会员上传分享，免费在线阅读，更多相关内容在应用文档-天天文库。
1、提高中草药随机对照试验的质量Ⅰ：临床试验设计和方法学作者：卞兆祥，李幼平，刘良，吴泰相，缪江霞，关家伦，宋丽【关键词】,随机对照试验［摘要］目的：通过对中草药临床随机对照试验的设计及方法学进行质量评价，探讨如何提高中草药临床试验的质量。方法： 4.4Allocationconcealment Undoubtedly,randomizationisapoethodtoproducebalancedtreatmentgroups,entisthecriticalfactortoensurethesuccessofrandom
2、ization.Randomsequencegenerationmerelymeansthatthegroupassignmentsequenceofaparticipantly,butdoesnotensurethattheallocationsequenceplementation.Inordertoallocatetheparticipantstobalancedtreatmentgroups,thesequenceshouldbeconcealedtoinvestigators,participants,andall
3、otherstudypersonnel.Ifnot,selectionbiasmaybeintroduced,entassignmentisnolongertrulyrandomandanimbalanceinprognosticfactorsmayoccurbetentgroups.Previousstudieshaveshoentcanexaggerateclinicaleffectsupto40%,especiallyinpoorlyconductedtrials［6,7,20］,anditcanalsocausegr
4、eaterheterogenEityinresults［21］.Thusproperrandomizationshouldinvolvebothrandomsequencegenerationandpleteimplementationofthatsequencetominimizebias. Ourstudyreportedthatalthough167trialsclaimedtoberandomized,only2.4%(4/167)reportedthemethodsforallocationconcealment
5、,paredentinperiodontology［16］.Thereasonizationconcealment(and/orthereportingofit)issoloeresearchersbelievethatentisnotnecessary.Butblindingandallocationconcealmentaredifferent［22］.Blindingconcentratesonpreventingstudyparticipantsandpersonnelfromdeterminingthegroupt
6、oentbias);itsafeguardsthesequenceafterallocation.Incontrast,allocationconcealmentconcentratesonpreventingselectionandconfoundingbiases;itsafeguardstheassignmentsequencebeforeanduntilallocation.Someresearchersmaynotunderstandthet.monlyusedmethodstoconcealallocationi
7、ncludecallingacentral,coordinatingofficeforeachpatientassignmentatthetimethatthepatientpresentsforstudyinclusion;usingsequentiallynumbered,opaquesealedenvelopes;andusingnumberedbottlesorcontainers［23］. 4.5Blinding edicinedescribedtheblindingprocedure［26］.Inordert
8、ochangethissituation,moreattentionsshouldbegiventoblindinginclinicaltrials.ThisisespeciallytrueoftrialsinCHMassessmentainlysubjective. Anotherch
