哈佛博士流行病学培训讲义.docx
- 文档编号:13405945
- 上传时间:2023-06-13
- 格式:DOCX
- 页数:24
- 大小:26.29KB
哈佛博士流行病学培训讲义.docx
《哈佛博士流行病学培训讲义.docx》由会员分享,可在线阅读,更多相关《哈佛博士流行病学培训讲义.docx(24页珍藏版)》请在冰点文库上搜索。
哈佛博士流行病学培训讲义
Instructor:
谢晶,JingXie
jxie@hsph.harvard.edu
DepartmentofEpidemiology,
SchoolofPublicHealth,
HarvardUniversity
May29AMCourseOverview&EpidemiologyConcept
•CourseOverview
•CourseObjectives
•Toprovideintroductoryknowledgeinepidemiologyanditsapplicationinetiologyresearch.
•Toprovideanorientationtoepidemiologyasabasicscienceforpublichealthandclinicalmedicine.
•Toaddresstheprinciplesofthequantitativeapproachtoclinicalandpublichealthproblems.
•Toprovideparticipantsbetterunderstandingofepidemiologyresearchandmedicalresearchpapers.
•Togiveyouasetoftoolsandawayofthinkingasanepidemiologist
•Bottomline:
Ifyouneverconsidertheresultsofastudythesameagain,thiscoursewillhavebeenasuccess
•Readingsandtextbooks
•Seesyllabus
•Epidemiology≠流行病学
•EpidemiologyinMedicineisjustlikeEconomicsinbusiness,whichisatoolcanbeappliedtosolvemedicalquestions.
•Widelyusedinmedicalresearch:
Nutritional、Genetics、MolecularBiology、Healthpolicyandmanagement、Pharmaceutical
•Epidemiology
•Epidemiologyisthestudyofthedistribution,determinants,andfrequencyofdiseasesinhumanpopulation.
Rootword:
Epidemic(infectiousdisease)
•Epidemiology,"thestudyofwhatisuponthepeople",isderivedfromtheGreekterms:
epi=upon,among;
demos=people,district;
logos=study,word,discourse;
suggestingthatitappliesonlytohumanpopulations.
•流行病学(epidemiology)是研究特定人群中疾病、健康状况的分布及其决定因素,并研究防治疾病及促进健康的策略和措施的科学。
•Questionscanbesolvedbyepidemiology
•Canbraincancerbecausedbyuseofcellphones?
•Shouldwomenofchildbearingagebeconcernedaboutmercuryinfish?
•WilltakingvitaminDsupplementsreduceriskofchronicdiseases?
•Whatistheminimumamountofexercisewehavetodotolowerourriskofheartdisease?
diabetes?
cancer?
•Whatisthelongtermimpactofpesticidesinfood?
•因此,流行病学又称方法学,是广大基层卫生人员和乡村医生从事防治工作时必备的预防医学知识。
•同时在临床工作和药效评价方面也常采用流行病学的分析方法,探讨和解决存在的问题。
•Hypotheticalscenarios
•Accordingtoareviewof4,000patientswho
hadcancer,2/3ofthesepatientshad
atsometimeweighed15%ormoreabovethenormfortheirsexandage.
•Conclusion:
Obesityisassociatedwithcancer.
•Whatdoyouthink?
...
•Themortalityrate(死亡率)per1,000populationinGuangdongis4.8comparedto14.5inSichuan.
Conclusion:
Guangdongisamorehealthyenvironment.
•Whatdoyouthink?
...
•HandednessandLifeSpan
NEnglJMed,April4,1991,p998
•Observationthatproportionofleft-handersinthepopulationdecreasessubstantiallywithageledtosuggestionthatsinistralitymaybeassociatedwithdecreasedlifespan.
•Sentquestionnairestonextofkinof2000individualswhodiedinsouthernCalifornia.967responded,withinformationaboutwhetherpersonwasright-handedornon-right-handed(leftormixed).
•Meanageofdeath:
•right-handed:
75years
•left-handed:
66years
p<0.0001
•significantlymoreleft-handersdiedinaccidents,orinaccidentswhiledrivingavehicle
•Conclusion:
"Theseresultsareconsistentwithpredictionsbasedonimpliedpathologicalfactorsandenvironmentalinteractions,suggestingthatleft-handersareatgreaterriskofdeathatanygivenage."
•Whatdoyouthink?
...
•Astudyof250patientswhohadanginapectoriswasinitiated.Fiveyearslater,200ofthesepatientswerecontactedtoobtainfollow-updata.Ofthese200patients,26%haddied,in20%symptomshadworsened,andin54%symptomsremainedthesame.
Conclusion:
Thefive-yearsurvivalrateforanginapectoriswas74%.
•Whatdoyouthink?
...
•流行病学研究方法类型
•Descriptivestudies(描述性研究)
•Analyticalstudies(分析性研究)
“流行病学基本上是一门归纳性的科学”(Frost,1941)。
流行病学从“描述”与“分析”两方面来体现它的归纳性。
在描述中注重分析,在分析中贯穿描述。
•Descriptivestudies(描述性研究)
•Descriptivestudiesareconcernedwithdescribingthegeneraldistributionofdiseaseinspaceandtime(cross-sectionalstudies).
•所谓描述性研究即人们常说的描述流行病学,是将所得资料按不同地区、不同时间以及不同人群特征分组,将疾病、健康或卫生事件的分布情况真实地展示出来。
人们往往会从其结论中获得启迪,引导人们进一步的探索与研究。
•Distribution
•Howcommon?
•Whoisaffected?
•Whendoesitoccur?
•Wheredoesitoccur?
•Howcommonorunusualisdiseaseoccurrence
•Rare(sporadic)disease
•Endemicdisease
•Epidemicdisease
•Pandemicdisease
•Rare(sporadic)diseasehasnoaccepteddefinition
•TheRareDiseaseActof2002definesararediseaseasonethataffectslessthan200,000personsintheUS
•Wilkipediadefinesararediseaseasonewhichoccurssoinfrequentlythatadoctorinabusygeneralpracticewouldnotexpecttoseemorethanonecaseayear
•NOTE:
diseasecanberareinsomeareasandcommonelsewhere.Itcanberarenowbutcommoninthepast(orfuture).
•Endemic:
usualpresenceofadiseasewithinagivenpopulation
•Epidemic:
occurrenceofadiseaseclearlyinexcessofnormalexpectancyinadefinedcommunityorregion
•Pandemic:
worldwideepidemic
•描述分布的常用测量指标(Measurementofdiseasefrequency):
发病率(Incidence)、现患率(Prevalence)、死亡率(Mortality)、病死率(morbidity)的意义、用途与计算方法。
描述疾病流行强度的常用语:
爆发、散发、流行和大流行;疾病的三间分布(时间、空间、人群)的概念及主要内容。
病因的定义及病因推断的原则。
•Analyticalstudies(分析性研究)
•Analyticstudiesareconcernedwithcausesandpreventionofdisease(case-controlstudies,cohortstudies).
•所谓分析性研究,从方法学上来看是指那些已发展成熟的方法如观察性研究(observationalstudies)、实验性研究(experimentalstudies)以及数学模型研究(statisticalmodeling)。
•Themostcommonhealth-relatedstatesandeventsinepidemiologicresearcharedeathanddisease
•Howdoyouknowsomeonehasadisease?
•Inindividuals
–Asktheperson
–Askthedoctor
–Reviewmedicalrecords
•Inpopulations
–Useactivesurveillance
–Relyonreporting
•Howdoyouknowsomeoneisdeadandwhatwasthecauseofdeath?
•Inindividuals
–Callandask
–Asknext-of-kin
–Reviewmedicalrecords
•Inpopulations
–Examplesincludeunderstandingofthemeaningofanoddsratio(OR),aconfidenceinterval(CI),anddifferentepidemiologicstudydesigns:
randomizedclinicaltrial,case-controlstudy,andcohortstudy.
•人们对某一医疗、卫生事件深入的认识都需要反复经历描述性研究与分析性研究的过程。
在描述性研究的基础上可以比较容易地开展分析性研究,而当一项或多项分析性研究结束后,往往需要对此在更高的层次上做描述性研究,如此循环不已,直至事物的本质得到充分的揭示,完成从具体描述到宏观描述,从初步分析到深入分析的飞跃。
•Timing
•Prospective
•Retrospective
MeasurementofDiseaseFrequency
Foursimplemathematicalparameters:
Count
Simplemeasureofquantity
Example:
NumberofcancerpatientsatSunYat-senUniversityCancerCenterin2009
Ratio
Obtainedbydividingonequantitybyanother.Thesequantitiesmayberelatedormaybeindependent.
Example:
Numberofstillbirthsperthousandlivebirths.
Proportion
Aratioinwhichthenumeratorisincludedinthedenominator
Example:
Thenumberoffetaldeathsoutofthetotalnumberofbirths(expressedas%)
Rate
Ameasureofhowquicklysomethingofinteresthappens
Example:
ThenumberofbirthsinayearinGuangzhou
MeasuresofDiseaseFrequency
Incidence(I):
Measuresnewcasesofadiseasethatdevelopoveraperiodoftime.
Prevalence(P):
Measuresexistingcasesofadiseaseataparticularpointintimeoroveraperiodoftime.
•Incidence
Reflectstransitionfromdisease-freetoaffectedstate
•Mortality
Reflectstransitionfromalivetodead
•Incidence
Reflectstransitionfromdisease-freetoaffectedstate
•CumulativeIncidence(CI)
•ProblemswithusingCI
•ToaccuratelycalculateCIweneedtofollowtheentirepopulationforthespecifiedtimeinterval.
•Thisisrarelypossiblefortwomainreasons
–Peoplemoveinandout(dynamiccohort)
–Peoplemaydiefromdiseasesotherthandiseaseofinterest(competingrisks)
•Incidencerate(IR)
•Rates
•WhatistheRiskofDiseaseX?
•WhatistheRiskofDiseaseX?
•ConfusingRiskwithRate
Whichonesofthese“rates”aretruerates?
•Attackrate
•Incidencerate
•Five-yearsurvivalrate
•Infantmortalityrate
•Mortalityrate
•Prevalencerate
•Age-specificincidencerate
•Case-fatalityrate
•Cause-specificmortalityrate
•ConfusingRiskwithRate
•Prevalencevs.Incidence
•Prevalencecanbeviewedasdescribingapoolofdiseaseinapopulation.
•Incidencedescribestheinputflowofnewcasesintothepool.
•Fatalityandrecoveryreflectstheoutputflowfromthepool.
•Prevalence
•Measuresexistingcasesofahealthcondition
•Primaryfeatureofacross-sectionalstudy
•TwotypesofPrevalence
•Pointprevalence
•Periodprevalence
•PointPrevalence
C=#ofobservedcasesattimet
N=Populationsizeattimet
Measuresthefrequencyofdiseaseatagivenpointintime
•PointPrevalence
Example
Supposethereare150individualsinapopulationand,onacertainday,15areillwiththeflu.Whatistheestimatedprevalenceforthispopulation?
•PeriodPrevalence
•C=the#ofprevalentcasesatthebeginningofthetimeperiod.
•I=the#ofincidentcasesthatdevelopduringtheperiod.
•N=sizeofthepopulationforthissametimeperiod.
•Prevalence
Usefulfor:
•Assessingthehealthstatusofapopulation.
•Planninghealthservices.
NotUsefulfor:
•Identifyingriskfactors
•Example
Supposewefollowedapopulationof150personsforoneyear,and25hadadiseaseofinterestatthestartoffollow-upandanother15newcasesdevelopedduringtheyear.
Whatisthepointprevalenceatthestartoftheperiod?
25/150=0.17=17%
Whatistheperiodprevalencefortheyear?
(25+15)/150=0.27or27%
Whatisthepointprevalenceattheendofperiod?
Notknown
Whatisthecumulativeincidencefortheoneyearperiod?
15/125=0.12=12%
May29PMEpidemiologicStudyTypesandDesigns
Exposure
Alcoholconsumption
Rawhamburger
Smoking
HealthOutcome
BreastCancer
E.coliinfection
LungCancer
•TypesofEpidemiologicStudies
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 哈佛 博士 流行病学 培训 讲义