Biases-in-Studies-of-Screening-Programs在筛选程序研究的偏见-36页PPT资料.ppt

Overview nIntroductionTN BiasesDefintionsnProblems with observational studiesVolunteer biasLead time biasLength biasStage migration biasPseudodiseaseScreening tests:TN Biasesn“When your only tool is a hammer,you tend to see every problem as a nail.”nClinical care accounts for 95%of spending but only 20%of determinants of health*nBiggest threats are public health threatsnInterventions aimed at individuals are overemphasized because they are more profitable and we know how to do/sell them*Teutsch SM,Fielding JE.Comparative effectiveness:looking under the lamppost.JAMA 2019;305:2225-6Cultural characteristicsWe live in a wasteful,technology driven,individualistic and death-denying culture.-George Annas,New Engl J Med,2019What is screening?nCommon definition:testing to detect asymptomatic diseasenBetter definition*:application of a test to detect a potential disease or condition in people with no known signs or symptoms of that disease or condition.Disease vs.conditionAsymptomatic vs.no known signs or symptoms*Common screening tests.David M.Eddy,editor.Philadelphia,PA:American College of Physicians,1991 Screening tests may be history questionsScreening SpectrumRisk factorRecognized symptomatic diseasePresymp-tomatic diseaseUnrecognized symptomatic disease Decreasing numbers labeled and treated Decreasing difficulty demonstrating benefitExamples and overlapnUnrecognized symptomatic disease:vision and hearing problems in young children;iron deficiency anemia,depression nPresymptomatic disease:neonatal hypothyroidism,syphilis,HIVnRisk factor:hypercholesterolemia,hypertension nSomewhere between:prostate cancer,ductal carcinoma in situ of the breast,more severe hypertensionScreenedNot screenedMortality after RandomizationRD+D-D-D+Mortaltiy after RandomizationEvaluating Studies of ScreeningnIdeal Study:Randomize patients to be screened or notCompare outcomes in ENTIRE screened group to ENTIRE unscreened groupObservational studies:Patients are not randomizednCompare outcomes in screened vs.unscreened patients nOr among patients with disease:Compare outcomes in those diagnosed by screening vs.those diagnosed by symptomsCompare stage-specific survival with and without screeningKEY DIFFERENCE:Mortality vs.SurvivalnMortality:denominator is a population,most of whom never get the diseasenSurvival:denominator is patients with the diseasenBeware of any studies evaluating screening tests using survivalPossible Biases in Observational Studies of Screening TestsnVolunteer biasnLead time biasnLength time biasnStage migration biasnPseudodiseaseVolunteer BiasnPeople who volunteer for screening differ from those who do notnExamplesHIP Mammography study:Women who volunteered for mammography had lower heart disease death ratesMulticenter Aneurysm Screening Study(MASS;Problem 6.3)Men aged 65-74 were randomized to either receive an invitation for an abdominal ultrasound scan or not.MASS Within Groups Result in Invited GroupAvoiding Volunteer BiasnRandomize patients to screened and unscreenednOtherwise,try to control for factors(confounders)associated with both screening and outcome Examples:family history,level of health concern,other health behaviors,baseline health/illnessesLead Time Bias(zero-time bias)nScreening identifies disease during a latent period before it becomes symptomaticnIf survival is measured from time of diagnosis,screening will always improve survival even if treatment is ineffectiveLead time biasSource:EDITORIAL:Finding and Redefining Disease.Effective Clinical Practice,March/April 2019.Available at:ACP-Online acponline.org/journals/ecp/marapr99/primer.htm accessed 8/30/02Avoiding Lead Time BiasnOnly occurs when survival from diagnosis is compared between diseased personsScreened vs.not screened Diagnosed by screening vs.by symptomsnAvoiding lead time biasMeasure mortality,not survivalCount from date of randomizationFollow patients for a long time(20 years?)and use total,not e.g.5-year survivalLength Bias(Different natural history bias)nScreening picks up prevalent diseasenPrevalence=incidence x durationnSlowly growing tumors have greater duration in presymptomatic phase,therefore greater prevalencenTherefore,cases picked up by screening will be disproportionately those that are slow growingLength biasSource:EDITORIAL:Finding and Redefining Disease.Effective Clinical Practice,March/April 2019.Available at:ACP-Online acponline.org/journals/ecp/marapr99/primer.htmLength BiasEarly detectionHigher cure rateSlower growing tumor with better prognosis?Avoiding Length Bias nOnly present when survival from diagnosis is comparedAND disease is heterogeneousnLead time bias usually present as wellnAvoiding length bias:Compare mortality in the ENTIRE screened group to the ENTIRE unscreened groupStudy disease subgroups with a uniform natural historyStage migration biasOld testsNew testsStage migration biasnAlso called the Will Rogers PhenomenonWhen the Okies left Oklahoma and moved to California,they raised the average intelligence level in both states.-Will RogersnDocumented with colon cancer at YalenOther examples abound the more you look for disease,the higher the prevalence and the better the prognosisBest reference on this topic:Black WC and Welch HG.Advances in diagnostic imaging and overestimation of disease prevalence and the benefits of therapy.NEJM 1993;328:1237-43.A more general example of Stage Migration BiasnVLBW(2500 g)newborns exposed to Factor X in utero have decreased mortality compared with those not exposednIs factor X good?nMaybe not!Factor X could be cigarette smoking!Smoking moves babies to lower birthweight strataCompared with other causes of LBW(i.e.,prematurity)it is not as badStage Migration BiasLBWVLBWNBWNBWLBWVLBWUnexposed to smokeExposed to smokeAvoiding Stage Migration BiasnThe harder you look for disease,and the more advanced the technologythe higher the prevalence,the higher the stage,and the better the(apparent)outcome for the stagenBeware of stage migration in any stratified analysisCheck OVERALL survival in screened vs.unscreened groupnMore generally,do not stratify on factors distal in a causal pathway to the factor you wish to evaluate!PseudodiseasenA condition that looks just like the disease,but never would have bothered the patientType I:Disease which would never cause symptomsType II:Preclinical disease in people who will die from another cause before disease presentsnIn an individual treated patient it is impossible to distinguish pseudodisease from successfully treated asymptomatic disease nThe Problem:Treating pseudodisease will always look successfulTreating pseudodisease will always be harmfulExample:Mayo Lung ProjectnRCT of lung cancer screeningnEnrollment 1971-76n9,211 male smokers randomized to two study armsIntervention:chest x-ray and sputum cytology every 4 months for 6 years(75%compliance)Control:Tests at trial entry,then a recommendation to receive the same tests annually*Marcus et al.,JNCI 2000;92:1308-16Mayo Lung Project Extended Follow-up Results*nAmong those with lung cancer,intervention group had more cancers diagnosed at early stage and better survival*Marcus et al.,JNCI 2000;92:1308-16MLP Extended Follow-up Results*nIntervention group:slight increase in lung-cancer mortality(P=0.09 by 2019)*Marcus et al.,JNCI 2000;92:1308-16What happened?nAfter 20 years of follow up,there was a significant increase(29%)in the total number of lung cancers in the screened groupExcess of tumors in early stageNo decrease in late stage tumorsnOverdiagnosis(pseudodisease)Black W.Overdiagnosis:an underrecognized cause of confusion and harm in cancer screening.JNCI 2000;92:1308-16Looking for PseudodiseasenAppreciate the varying natural history of disease,and limits of diagnosisnImpossible to distinguish from successful cure of(asymptomatic)disease in individual patientnFew compelling stories of pseudodiseasenClues to pseudodisease:Higher cumulative incidence of disease in screened groupNo difference in overall mortality between screened and unscreened groupsEach year,182,000 women are diagnosed with breast cancer and 43,300 die.One woman in eight either has or will develop breast cancer in her lifetime.If detected early,the five-year survival rate exceeds 95%.Mammograms are among the best early detection methods,yet 13 million women in the U.S.are 40 years old or older and have never had a mammogram.39,800 Clicks per mammogram(Sept,04)Why is this misleadingnEach year 43,000 die,182,000 new cases suggests mortality is 24%n5-year survival 95%with early detection suggests 5%mortality,suggesting about 80%of these deaths preventablenActual efficacy is closer 20%for breast cancer mortality(lower for total mortality)Questions?