cohort study 2

8/14/2019 Cohort Study 2

1/28

COHORT STUDY COHORT STUDY

DR. A.A.TRIVEDIDR. A.A.TRIVEDI(M.D., D.I.H.)(M.D., D.I.H.)

ASSISTANT PROFESSORASSISTANT PROFESSORemail :email : [email protected][email protected]
mailto:[email protected]:[email protected]:[email protected]


2/28

EpidemiologyEpidemiology

Defined by John M. Last in 1988Defined by John M. Last in 1988Study of Distribution andStudy of Distribution andDeterminants of health related stateDeterminants of health related stateor event in a specified populationor event in a specified populationand the application of this study toand the application of this study tothe control of health problem.the control of health problem.

We measure We measure Disease frequencyDisease frequencyDiseases distributionDiseases distributionDeterminants of disease.Determinants of disease.


3/28

TYPES OF EPIDEMIOLOGICAL TYPES OF EPIDEMIOLOGICALSTUDIESSTUDIES

1. OBSERVATIONAL STUDIES

A. DESCRIPTIVE STUDY

DESCRIBE DIESEASE BY

TIME

PLACE

PERSON

B. ANALYTICAL STUDIES

ECOLOGICAL STUDY

CROSS SECTIONAL STUDY

CASE-CONTROL STUDY

COHORT STUDY2. EXPEREMENTAL STUDIES

RANDOMIZED CONTROLLED TRIAL (RCT)

FIELD TRIAL

COMMUNITY TRIAL


4/28

Descriptive EpidemiologyDescriptive Epidemiology

Describe the disease byDescribe the disease by Time Time

PlacePlacePersonPerson


5/28


6/28

Cohort study is undertaken to support theCohort study is undertaken to support theexistence of association betweenexistence of association betweensuspected cause and diseasesuspected cause and disease

A major limitation of cross-sectionalA major limitation of cross-sectionalsurveys and case-control studies issurveys and case-control studies isdifficulty in determining if difficulty in determining if exposure

exposure oror riskrisk

factorfactor preceded thepreceded the diseasedisease oror outcomeoutcome ..Cohort Study:Cohort Stud

y:

Key Point:Key Point:

Presence or absence of risk factorPresence or absence of risk factoris determinedis determined beforebefore outcomeoutcomeoccurs.occurs.


7/28

WHAT IS COHORTWHAT IS COHORT

Ancient RomanAncient Romanmilitary unit, Amilitary unit, Aband of warriors.band of warriors.

Persons bandedPersons bandedtogether.together.Group of personsGroup of persons

with a commonwith a commonstatisticalstatisticalcharacteristic.characteristic.[Latin][Latin]

E.g. age, birthE.g. age, birth


8/28

Cohort studieslongitudinalProspective studiesForward looking study IIncidence study

starts with people free of disease

assesses exposure at baselineassesses disease status at follow-up


9/28

INDICATION OF A COHORTINDICATION OF A COHORTSTUDYSTUDY

When there is good evidence of When there is good evidence of exposure and disease.exposure and disease.

When exposure is rare but incidenceWhen exposure is rare but incidenceof disease is higher among exposedof disease is higher among exposedWhen follow-up is easy, cohort isWhen follow-up is easy, cohort isstablestableWhen ample funds are availableWhen ample funds are available


10/28


11/28


12/28

b+d

Frame work of Cohort studies

cc+d

aa+b

Total Yes

Disease Status

Yes

No

Exposure Status

b

d

a+c N

No

Studycohort

Comparisoncohort


13/28

General consideration whileselection of cohorts

Both the cohorts are free of thedisease.Both the groups should equallysusceptible to diseaseBoth the groups should becomparableDiagnostic and eligibility criteria forthe disease should be defined well inadvance.


14/28

Elements of cohort studyElements of cohort studySelection of study subjectsSelection of study subjectsObtaining data on exposureObtaining data on exposureSelection of comparison groupSelection of comparison group

Follow upFollow upAnalysisAnalysis


15/28

Selection of study subjectsSelection of study subjects

General populationGeneral populationWhole population in an areaWhole population in an areaA representative sampleA representative sample

Special group of populationSpecial group of populationSelect groupSelect group

occupation group / professional group (Dolls study )occupation group / professional group (Dolls study )

Exposure groupsExposure groupsPerson having exposure to some physical, chemicalPerson having exposure to some physical, chemicalor biological agentor biological agent

e.g. X-ray exposure to radiologistse.g. X-ray exposure to radiologists


16/28

Obtaining data on exposureObtaining data on exposure

Personal interviews / mailed questionnairePersonal interviews / mailed questionnaireReviews of recordsReviews of records

Dose of drug, radiation, type of surgery etcDose of drug, radiation, type of surgery etc

Medical examination or special testMedical examination or special test

Blood pressure, serum cholesterolBlood pressure, serum cholesterolEnvironmental surveyEnvironmental survey

By obtaining the data of exposure we canBy obtaining the data of exposure we canclassify cohorts asclassify cohorts asExposed and non exposed andExposed and non exposed andBy degree exposure we can sub classifyBy degree exposure we can sub classifycohortscohorts


17/28

Selection of comparison groupSelection of comparison groupInternal comparisonInternal comparison

Only one cohort involved in studyOnly one cohort involved in studySub classified and internal comparison doneSub classified and internal comparison done

External comparisonExternal comparisonMore than one cohort in the study for theMore than one cohort in the study for the

purpose of comparisonpurpose of comparisone.g. Cohort of radiologist compared withe.g. Cohort of radiologist compared withophthalmologistsophthalmologists

Comparison with general population ratesComparison with general population rates

If no comparison group is available we canIf no comparison group is available we cancompare the rates of study cohort with generalcompare the rates of study cohort with generalpopulation.population.Cancer rate of uranium miners with cancer inCancer rate of uranium miners with cancer ingeneral populationgeneral population


18/28

Follow-upFollow-up To obtain data about outcome to be To obtain data about outcome to bedetermined (morbidity or death)determined (morbidity or death)

Mailed questionnaire, telephone calls, personalMailed questionnaire, telephone calls, personalinterviewsinterviewsPeriodic medical examinationPeriodic medical examinationReviewing recordsReviewing recordsSurveillance of death recordsSurveillance of death recordsFollow up is the most critical part of the studyFollow up is the most critical part of the study

Some loss to follow up is inevitable due toSome loss to follow up is inevitable due to

death change of address, migration,death change of address, migration,change of occupation.change of occupation.Loss to follow-up is one of the draw-backLoss to follow-up is one of the draw-backof the cohort study.of the cohort study.


19/28

ANALYSISANALYSIS

Calculation of incidence rates amongCalculation of incidence rates among

exposed and non exposed groupsexposed and non exposed groups

Estimation of riskEstimation of risk


20/28

Incidence rates of outcomeIncidence rates of outcome

N

dc

ba

Yes No

Disease Status

Yes

No

Exposure Status

a+b

c+d

b+da+c

Total

Studycohort

Comparisoncohort


21/28

Incidence rateIncidence rate

Incidence among exposed =Incidence among exposed =aa

a+ba+bIncidence among non-exposed =Incidence among non-exposed =

cc

c+dc+d


22/28

Estimation of riskEstimation of risk

Relative RiskRelative Riskincidence of disease amongincidence of disease among

exposedexposedRR = ______________________________ RR = ______________________________

Incidence of disease among non-Incidence of disease among non-exposedexposed

a/a+ba/a+b== _________ _________

c/c+dc/c+d


23/28

Estimation of RiskEstimation of Risk

Attributable RiskAttributable RiskIncidence of disease among exposedIncidence of disease among exposed incidence of disease among non incidence of disease among nonexposedexposed

AR = _______________________________ AR = _______________________________ Incidence of disease among exposedIncidence of disease among exposed

a/a+b c/c+da/a+b c/c+dAR = _______________ AR = _______________

a/a+ba/a+b


24/28

10000992773

300029973NO

7000693070YES

NOYES

TotalLung cancerSmoking

Find out RR and AR for above data


25/28

Incidence of lung cancer among smokersIncidence of lung cancer among smokers70/7000 = 10 per 100070/7000 = 10 per 1000

Incidence of lung cancer among non-Incidence of lung cancer among non-smokerssmokers

3/3000 = 1 per thousand3/3000 = 1 per thousand

RR = 10 / 1 = 10RR = 10 / 1 = 10(lung cancer is 10 times more common(lung cancer is 10 times more commonamong smokers than non smokers)among smokers than non smokers)

AR = 10 1 / 10 X 100AR = 10 1 / 10 X 100= 90 %= 90 %

(90% of the cases of lung cancer among(90% of the cases of lung cancer amongsmokers are attributed to their habit of smokers are attributed to their habit of smoking)smoking)


26/28

Types of Cohort Study Types of Cohort Study

Prospective cohort studyProspective cohort studyRetrospective (historical) cohortRetrospective (historical) cohortstudystudyCombination of Retrospective andCombination of Retrospective andProspective cohort study.Prospective cohort study.


27/28

Cohort studiesCohort studies StrengthsStrengths

We can find outWe can find outincidence rate andincidence rate andriskriskMore than oneMore than onedisease related todisease related tosingle exposuresingle exposurecan establish causecan establish cause- effect- effectgood whengood whenexposure is rareexposure is rareminimizes selectionminimizes selection

and informationand informationbiasbias

WeaknessesWeaknesseslosses to follow-uplosses to follow-up

often requiresoften requireslarge samplelarge sampleineffective for rareineffective for rarediseasesdiseaseslong time tolong time to

completecompleteexpensiveexpensiveEthical issuesEthical issues


28/28

THANK YOU

cohort study 2

Documents