Metlit-10 (Cohort Study)

Cohort studiesFollowing groups of subject over timeSudigdo Sastroasmoro

Cohort = ancient Roman term = a group of soldiers that marched together into battleIn clinical study = a group of person followed up together over time

Cohort studiesDescriptive: To describe incidence of certain outcome over timeAnalytic: To analyze association between risk factors and outcomesThe only design to determine incidence rate

Prospective cohortSelect a sample from the populationMeasure predictor variable (present or absent)Follow-up the cohortMeasure outcome (present or absent)

Advantages of cohort studies Can assess several outcomesTime-order generally clearProspective control over exposure and outcome measurement possible (in prospective studies)Somewhat less potential for bias than case-control studies, but equal potential for confounding

Disadvantages of cohort studies Generally require large samplesNot useful for rare outcomesAs an observational study, can never be assumed to be free of confounding and biasMust usually control for potential confounding in the analysis, though can control in the design

Retrospective cohortBasically the same with prospectiveBasic measures, follow up, outcome all in the pastOnly possible if all data are complete and prepared for other purpose

Retrospective cohortAssemble cohort in the pastMeasure risk factorsFollow-upMeasure outcomesAnalyze

Retrospective cohort: strengthMuch less costly and efficientLess time consumingAll subjects (assumed) from same population

Retrospective cohort: weaknessSecondary dataMay not include necessary data

Steps in planning cohort studies:things to considerWhen to use cohort designChoosing among cohort designsSelecting subjectsMeasuring predictor and confounding variablesFollowing subjects and measuring outcomesAnalyzing cohort: incidence and relative risk

When to use cohort designAccurately describing incidenceMay be the only way to establish temporal sequence of risk and outcomesMalnutrition in chronic diarrhea may be the result rather than a causeOnly way to study certain rapidly fatal diseasesAvoiding survivor biasAllow investigator to study multiple outcomes / ever increasing outcomes

Choosing among cohort designsRetrospectiveQuickEconomicalProspectiveRapidly occurring outcome, e.g. discharge from nursery after bone factureWhen less expensive design fail to answer research question properlyWhen case control studies give conflicting resultsWhen key measurements must be performed before outcomes occur

Selecting subjectsDefine group of subjects at the beginning of the study (inception cohort), e.g., cervical cancer, stage 1-2Select samples with rapidly occurring outcomes, e.g., hip fracture elderly womanAdequate sample sizeControl in double cohort should be selected by random sampling (whenever possible)

Measuring predictor and confounding variablesThe quality of the result depend largely on the accuracy of measuring predictor and outcome variablesThe validity of the result (cause-effect relationship) also depends on the measurement and control of confounding variables

Following subjectsImportant: minimize loss to follow up!!!Strategies:During design:Restriction: exclude those likely to lossMovingUnwilling to returnPlanning for future trackingAddress, telephone, mobile, fax etcDuring follow-upPeriodic contactPhone, visits, etcOther relevant measures

Analyzing resultsDescriptive: Incidence rate, mean, proportions, SD, etcAnalytic: Relative risk (Risk ratio, Incidence ratio)Other analysis: Survival analysisMultivariate analysis as appropriate

Cohort study: exampleDisease -Exp. +Disease +50504552030Exp. -Disease -Disease +

Cohort study: analysisExp + 203050Exp - 54550Disease +Disease -Relative risk = incidence in expose/incidence in non-exposedRR = 20/50 : 5/50 = 4

RR = 4The probability of developing the disease in exposed group is 4 x the probability of developing the disease in non-exposed groupExposed individuals are 4 x more likely to develop the disease compared with non-exposedCI is recommended; if CI include 1, then statistically not significant (the probability that the result is caused by chance is high)

RR = 4 (95% CI 1.8 to 8.3)P value?

RR = 4 (95% CI 0.9 to 11.4)P value?

Famous cohort studiesPopulation-basedCardiovascularChild healthSpecial exposures

1. Cardiovascular diseaseFramingham, MaTecumseh, MiEvans county, Ga (biracial)Muscatine, IABogalusa, LA (children)

2. Child healthNational birthday trust studiesOne week of births in England and Wales in 1946, 1958 and 1970Project on premature infantsAll births < 1,500 g or < 32 weeks in Holland in 1983

The national childrens study http://www.Nichd.Nih.Gov/about/despr/despr.Htm

A study in Jakarta of 100,000 pregnancies with offspring followed to age 21?

3. Special exposuresAtomic Bomb Casualty Commission (ABCC):Hiroshima and Nagasaki survivors (effects of radiation)Dutch famine survivors (effects of starvation)

Seveso (effects of dioxin exposure)

Modification of cohort studiesDouble cohort study: exposed and not exposed subjects are constructed separately but with the same eligibility criteria except for variable being studiedSynonym: cohort with external controlsMore frequently usedCohort with ordinal or numerical outcomeHybrid cohort & case-control studies (with some further modifications):Case cohort studyNested case control study

Case-cohort design: purpose

The case-cohort design is used to reduce the costs of exposure assessment

Case-cohort design: approachA population at risk is identified and screened for disease, and prevalent cases are omitted.A case-identification procedure is developed to detect new cases of disease in the cohort. (So far all is the same as any cohort study)

Case-cohort design: approachThe whole cohort is subject to case-identification, but only a random sample (called the sub-cohort) receives detailed exposure assessment.The cases are those emerging in the population (both in and out of the sub-cohort); the controls are subjects in the sub-cohort who are not cases.Analysis is like a cohort study. Since the sampling fraction is known, and the entire population is sampled for caseness, true incidences and relative risks can be calculated.

Nested case-control studyA case-control study nested in the cohort studyPurpose: to reduce cost of exposure assessmentIn a cohort study (for other exposure), specimen is kept until the cohort study finishesSubjects who developed outcome are chosen as CASE; the CONTROLS are selected randomly from the subjects who did not develop outcomeAssess risk factors for case and controlsAnalysis is similar with case-control study