页数:65页文件大小:783.8 KB来源: 流行病学ppt课件
本课件讲解筛检的概念与实施条件,围绕筛检试验的评价指标展开,说明灵敏度、特异度、假阳性率与假阴性率的计算,并讲解预测值、ROC曲线与联合试验的评价方法,以及筛查项目中的偏倚。
筛检Screening
Stage ofsusceptibilityStage ofSubclinicalDiseaseStage ofClinical DiseaseStage of Recovery,Disability, or DeathExposurePathologicalChangesOnset ofSymptomsUsual Timeof Diagnosis疾病自然史(Natural History of Disease)
目录
筛检是运用快速、简便的试验、检查或其他方法,将健康人群中那些可能有病或缺陷,但表面健康的人、同那些可能无病者鉴别开来Screeningis defined as theuse of quick and simple testingprocedures to identify andseparate persons who areapparently well, but who maybe at risk of a disease, fromthose who probably do nothave the disease.第一节概述Introduction
Apparently Well Population to be TestedScreeningTestDiagnostic TestDisease or RiskFactor PresentDisease or RiskFactor AbsentNegatives(Normal) (Persons presumedto be free of disease)Negative on screeningPositive on screen, no diseasePositive on screen, disease presentPositives(abnormal)presumed tohave diseaseTreatmentIntervention筛检试验流程图Flow diagram of screening test
筛查方法的评价与选择程序Procedures of evaluation and selection for screening test金标准(gold standard)目标人群Targetpopulation病人Patient非病人Non-patient待评价筛查方法Screening to beevaluated评价指标Evaluation indicator+-+-
筛检试验Screening test识别健康人群中未被发现的某病患者或可疑患者,或高危个体Screening is the detection of patients,suspected patient and high risk of thedisease in healthy population
筛检试验Screening test问卷、常规体格检查、物理学检查、实验室检验、分子生物学技术Questionnaire, routine physicalexamination, physical examination,laboratory tests, molecular biologytechniques
根据所用的筛检方法的数量多少
根据筛检对象的范围不同
简单性廉价性快速性安全性可接受性筛检试验应具备五个特征Five characteristics that screening testshould haveSimpleCheapQuickSafeAcceptable
一级预防二级预防疾病的自然史流行病学检测筛检的目的Primary preventionSecondary preventionNatural history of diseaseEpidemiologicalsurveillancePurposes of screening
筛检的疾病是当地一个重大的公共卫生问题
有进一步确诊的方法与条件
该病具有较长的潜伏期或领先时间
伦理学问题Ethical Issues尊重个人意愿Respectforperson’swill有益无害Usefulandharmless公正Fairness
一、筛检试验的评价方法
金标准所谓“金标准”是指当前临床医学界公认的诊断疾病的最可靠方法GoldStandardThe best available test thatis universally recognized byclinical medical circles iscalled the GOLDSTANDARD.
病例组纳入各种类型的病人对照组最好选择需要与研究疾病鉴别的其他病人,即所选择的对照组与病例组具有许多相似的条件研究对象Study SubjectsCase groupInclude appropriatespectrum of patientsControl groupNeed to be distinguished tothe disease studied, that is thecontrol has co-morbidconditions with case
当待评价筛检试验的灵敏度和特异度均接近50%Both sensitivity and specificity of screeningto be evaluated approach 50%n: the number of abnormal (diseased) or normal (healthy) patientsin the study.δ: admissible error.P: the estimation of sensitivity and specificity for testing.ua: the u value for the cumulative probability equal to a/2.
当待评价筛检试验的灵敏度和特异度小于20%或大于80%Both sensitivity and specificity of screening to be evaluatedless than 20% or greater than 80%
本来有病,被筛检为阳性本来有病,被筛检为阴性本来无病,被筛检为阳性本来无病,被筛检为阴性评价结果Evaluation Results
评价结果Evaluation ResultsDisease Status(Gold Standard)-+TestPositiveFree of disease,test positiveb(FP)Have disease,test positivea(TP)+TestResultTestNegativeFree of disease,test negatived(TN)Have disease,test negativec(FN)-Freeof diseaseHavedisease
False Negative RateLikelihood of a negative result when patient actuallyhas diseaseFalseNegativesSensitivity
FalseNegativesSensitivityLikelihood of a negative result when patient actuallyhas diseaseFalse Negative Rate
False negative rate increases with decreased sensitivitySensitivityFalseNegativesFalse Negative RateLikelihood of a negative result when patient actuallyhas disease
False Positive RateLikelihood of a positive result when patient doesnot have the diseaseFalsePositivesSpecificity
False Positive RateLikelihood of a positive result when patient doesnot have the diseaseFalsePositivesSpecificity
False Positive RateLikelihood of a positive result when patient doesnot have the diseaseFalse positive rate increases with decreased specificitySpecificityFalsePositives
正确指数(Youden’s index)=(特异度+灵敏度)-1=1-(假阳性率+假阴性率)真实性(validity)合计(total)金标准(gold standard)筛检试验(screening)非患者(non-patient)患者(patient)R1假阳性(FP)B真阳性(TP)A阳性(positive)R2真阴性(TN)D假阴性(FN)C阴性(negative)NC2C1合计(total)
真实性(validity)合计(total)金标准(gold standard)筛检试验(screening)非患者(non-patient)患者(patient)R1假阳性(FP)B真阳性(TP)A阳性(positive)R2真阴性(TN)D假阴性(FN)C阴性(negative)NC2C1合计(total)取值范围:-1和+1之间Range: between-1 and +1Kappa值判断标准Kappa值一致性强度<0弱0~0.2轻0.21~0.40尚好0.41~0.60中度0.61~0.80高度0.81~1最强
Kappa值
标准差与变异系数
影响可靠性的因素
阳性预测值
灵敏度、特异度、患病率与预测值的关系Relationship between sensitivity, specificity,prevalence and predict value
sensitivity and specificity in different blood sugar levelblood sugarSN SPblood sugarSN SP(mg/100ml)(﹪)(﹪)(mg/100ml)(﹪)(﹪)80 100.0 1.2 150 64.3 96.190 98.6 7.3 160 55.7 98.6100 97.1 25.3 170 52.9 99.6110 92.9 48.4 180 50.0 99.8120 88.6 68.2 190 44.3 99.8130 81.4 82.4 200 37.1 100.0140 74.3 91.2Relationship between sensitivity and specificity灵敏度与特异度的关系
非病人病人某项诊断标准的数值截断点Threshold三、筛检试验截断值的确定Determination of screening cut-off point理想情况Ideal situation人数NumberNormalDisease
非病人Normal病人Abnormal某项诊断标准的数值Cut-off三、筛检试验截断值的确定Determination of screening cut-off point实际情况Actual situation人数Population
Number of personsMeasurementScreening tests use an arbitrary cut-off toseparate two populations(normal vs. abnormal)NormalAbnormalacbdNormalacdb
NumberMeasurementWhere cut-off isplaceddeterminessensitivity andspecificityScreening tests use an arbitrary cut-off toseparate two populations(normal vs. abnormal)
NumberMeasurementWhere cut-off isplaced determinessensitivity andspecificityNormalAbnormalFew False NegativesHigh Sensitivity-result+ resultScreening tests use an arbitrary cut-off toseparate two populations(normal vs. abnormal)
NumberMeasurementWhere cut-off isplaced determinessensitivity andspecificityNormalAbnormalFew False PositivesHigh Specificity-result+ resultScreening tests use an arbitrary cut-off toseparate two populations(normal vs. abnormal)
NumberMeasurementWhere cut-off isplaced determinessensitivity andspecificityNormalAbnormalTrade off betweenFalse–and False +-result+ resultScreening tests use an arbitrary cut-off toseparate two populations(normal vs. abnormal)
NumberMeasurementTest with HighSensitivityNormalAbnormalA negative resultallows you to rule-outwhatever you aretesting-result+ resultSnNoutScreening tests use an arbitrary cut-off toseparate two populations(normal vs. abnormal)
NumberMeasurementTest with HighSpecificityNormalAbnormalA positive result allowsyou to rule-in whateveryou are testing-result+ resultSpPinScreening tests use an arbitrary cut-off toseparate two populations(normal vs. abnormal)
用真阳性率和假阳性率作图得出的曲线,可用于确定筛检试验的截断值。
A receiver operating characteristic (ROC),or simply ROCcurve, is a graphical is a graphical plot of the sensitivity,or true positives, vs. (1 − specificity), or false positives, fora binary classifier system as its discrimination threshold isvaried.受试者工作特性曲线Receiver operator characteristic curve, ROC
特异度血糖测定诊断糖尿病的ROC曲线截断点Cutoff pointROC curve for diagnosis of diabetes using blood glucose
第三节筛检效果评价
二、筛检的生物学效果评价Biological effectiveness evaluation病死率、死亡率、生存率筛检组Screened group未筛检组Unscreened group比较Fatality rate、mortality rate、survival rate
三、筛检的卫生经济学效果评价
领先时间偏倚
领先时间偏倚(lead time bias)领先时间:筛检诊断时间和临床诊断时间之差Leadtime:Intervalbetweenthediagnosisofadiseaseatscreeningandtheusualtimeofdiagnosis(bysymptoms)Diagnosisby screeningDiagnosisvia symptomsLead Time
Assumes survival
Diagnosis
病程长短偏倚(length bias)疾病开始筛检查出症状开始死亡ODPYOBiologicalonset ofdiseaseYSymptomsBeginDDeathPDiseasedetectablevia screening
志愿者(volunteer bias)Type of bias where those who choose to participate arelikely to be different from those who don’t.筛检参加者与不参加者之间,某些特征可能存在不同而导致的偏倚。
谢谢!
下载地址:{{downurl}}
{{{a}}}
{{#b}}{{{b}}}
{{/b}} {{/title}} {{#nolog}}登陆注册
{{/nolog}} {{#djsx}}点我刷新
{{/djsx}} {{#gocz}} {{/gocz}} {{#ffqr}} {{/ffqr}}