Transcript
Cohort Study
Dr Naveen PhuyalMBBS,MD
Synonyms
• Longitudinal study• Prospective study• Incidence study• Forward-looking study
What is a Cohort?
• Group of people who share a common characteristic or experience within a defined time period.
AgeOccupationExposure to drug or vaccinePregnancyInsured person
• An ancient Roman military unit, comprising six centuries, equal to one tenth of a legion.
• A group of people banded together or treated as a group.
What is a Cohort?
When is a cohort study warranted?
• Good evidence of association between exposure and outcome as shown by descriptive and case control studies
• When exposure is rare but incidence is high among exposed
• When attrition of people can be minimized
• When ample of funds are available
Research methods
Observational
Descriptive
Case series, case reports,
CS, cohort
Analytical
Ecological Cross-sectional
Case control Cohort
Experimental
ControlledUncontrolled,Non-random
7
Position in the evidence hierarchy
Design of a cohort study
Cohort study
Features of Cohort study
Features of Cohort study
• Cohorts are identified prior to appearance of disease under investigation
• Study groups are observed over a period of time to determine the frequency of disease among them
• Study proceeds forward from cause to effect
• Key element of Cohort study is time.
• Following the exposed and unexposed groups over time cohort studies are uniquely equipped to describe process and mechanisms by which exposures relate to the development of disease.
• Cohort study is the primary tool to study time and medicine ( Samet 2000)
Features of Cohort study
• They provide the data to describe when disease occur and track their consequences over time.
• Factors that cause disease or early signs of disease can be monitored over time.
• Diversity of individuals in cohort study: provides data to identify risk factors that make certain individual more susceptible.
Features of Cohort study
• Data collected in cohort studies are useful to describe the prognostic markers of exposure.
• Multisite cohort studies may serve additional role of characterizing where disease occur and to what extent the diseases are spread in different locations.
• A cohort study with adequate sample and follow up enables us to understand natural history of disease.
Features of Cohort study
• Once a measure of frequency of disease occurrence (incident cases in person-years) is adopted, cohort studies allow the direct comparison of the risk of becoming ill in several groups.
• This comparison can be relative or absolute– Relative: how many times higher or lower is the risk between
exposed and unexposed (relative risk)
– Absolute: how much difference in risk is there between exposed and unexposed ( attributable risk)
Features of Cohort study
• Relative risk will give you causal relationship between disease and exposure.
• Attributable risk measures the change of incidence due to exposure in question ( it quantifies the burden of disease that an exposure exerts in a population.)
• Identification of exposures and risk factors for a disease forms the basis for prevention.
Features of Cohort study
Cohort based associations that have resulted in prevention strategies
• Lung Ca and smoking: avoid cigarette• Unprotected sex and HIV: avoid• IV drug use and HIV: avoid• To prevent heart disease: avoid high LDL and
low HDL levels• To prevent Cervical cancer: avoid Papilloma
infection
General consideration
• Cohorts must be free from disease under study.
• Both the study and control groups should be equally susceptible to the disease under study
• Both groups should be comparable in respect to all possible variables
• Diagnostic and eligibility criteria of disease must be defined beforehand.
Types of cohort studies
Types of cohort studies
1. Prospective cohort studies
2. Retrospective cohort studies
3. Combination of retrospective and prospective cohort studies
Prospective cohort study
Retrospective cohort study
Retrospective plus prospective cohort study
Elements of cohort study
Elements of cohort study
• Selection of study subjects• Obtaining data on exposure• Selection of comparison groups• Follow up• Analysis
Elements of cohort study
• Selection of study subjects• Obtaining data on exposure• Selection of comparison groups• Follow up• Analysis
1. Selection of study subjects1. General population
when exposure (cause) of death is frequent in general population, residing in well defined geographical, political and administrative areas.Eg. Framingham heart study
2. Special groupsSelect groups: doctors, nurses, lawyers, teacherscollege alumni, employees, volunteers.
Exposure groups: special exposure to physical, chemical and other agents.
1. Selection of study subjects
Elements of cohort study
• Selection of study subjects• Obtaining data on exposure• Selection of comparison groups• Follow up• Analysis
2. Obtaining data in exposure
• Cohort members: interviews/ questionnaires
• Review of records: dose of radiation/ kinds of surgery/ details of medical treatment
• Medical examination or special tests: blood pressure, serum cholesterol, ECG
• Environmental surveys: where the cohort lived or worked
• Information about exposure will allow classification of cohort members:
1. According to whether or not they have exposed to suspected factor
2. According to degree of exposure at least in broad class in case of special exposure groups
2. Obtaining data in exposure
Elements of cohort study
• Selection of study subjects• Obtaining data on exposure• Selection of comparison groups• Follow up• Analysis
3. Selection of comparison groups
i. Internal comparisons
ii. External comparisons
iii. Comparison with general population rates
Internal comparison
• No outside comparison group required.
• A single cohort enters the study and on basis of information obtained, classified into several comparison groups according to degree or levels of exposure.
External comparison
• When information on degree of exposure is not available
• Eg. Smokers and non-smokers• Cohort of radiologists with cohort of opthalmologists• The study and control cohorts should be similar in
demographic and possibly important variables other than those under study.
Comparison with general population rates
• Frequency of lung cancer among uranium mine workers vs frequency of lung cancer in general population
• Asbestos workers vs general population cancers
Elements of cohort study
• Selection of study subjects• Obtaining data on exposure• Selection of comparison groups• Follow up• Analysis
4. Follow up
• Periodic medical examinations • Review of records: physician/ hospital• Routine surveillance of death records• Mailed questionnaires, phone calls, periodic home
visits- preferably all three every year
• Loss to follow up occurs due to death, change of residence, migration, withdrawal of occupation
• Achieve 95% follow up as much as possible
Elements of cohort study
• Selection of study subjects• Obtaining data on exposure• Selection of comparison groups• Follow up• Analysis
5. Analysis1. Incidence rates
1. Among exposed2. Among not exposed
2. Estimation of risk
3. Relative risk
4. Attributable risk
5. Population attributable risk
Disease
Exposure Yes No Total
Yes a b a+b
No c d c+d
Total a+c b+d a+b+c+d
Incidence rates among exposed= (a/a+b)*1000
Incidence rates among non-exposed= (c/c+d)*1000
Relative riskDisease
Exposure Yes No Total
Yes a b a+b
No c d c+d
Total a+c b+d a+b+c+d
Relative risk is the ratio of incidence among exposed to incidence among non-exposed
RR=( a/a+b)/( c/c+d)
Relative risk
• Direct measure of strength of association between suspected cause and effect
• RR=1 (no association)
• RR>1 (positive association)
• If RR=2 ( incidence rate of disease is 2 times higher among exposed than non exposed or 100% increase in risk)
Cigarette smoking
Lung Ca developed
No Lung Ca devloped
Total
Yes 70 6930 7000
No 3 2997 3000
RR= (70/7000)/ (3/3000)= 10Smokers are 10 times at greater risk Of devloping lung Ca than non smokers
Attributable risk
• Also called as ‘ risk difference’• AR is the difference in incidence rates of disease
(or death) between an exposed group and non-exposed group.
• AR= Incidence of disease rate among exposed – incidence of disease rate among non-exposed
Attributable risk percentageAR%= Incidence of disease rate among exposed
– incidence of disease rate among non-exposed
incidence of disease among exposed
In above example AR= (10-1) X 100 = 90%10
90 percent of lung cancer among smokers was due to smoking.This is the amount of disease that might be eliminated if the factor under study is controlled or eliminated.
X 100
Population attributable risk
PAR= It – INE
PAR %= It – INE
It
PAR % provides the estimate of the amount by which the disease could be reduced in that population if the suspected factor was eliminated or modified.
X 100
RR vs AR
• The size of RR is a better index than AR for assessing the etiological role of a factor in disease
• Larger the RR, stronger the association between cause and effect
• AR gives better idea on impact of successful preventive or public health programme might have in reducing the problem
Advantages of cohort studies
• Incidence can be calculated.• Several outcomes can be studied simultaneously.• Provide direct estimate of relative risk.• Dose response ratios can be calculated.• Certain bias can be minimised.
Disadvantages of cohort studies• Involves large no. of people and unsuitable for study of
diseases with low incidence.• Takes long time to conduct the study.• Loss of staff, loss of fund.• Requires extensive record keeping.• Loss to follow up.• Selection of comparison group is difficult.• Expensive• Study may alter the participants behavior.• Ethical problems.
Case control
• Effect to cause• Starts with disease• Tests weather suspected
cause is more frequent among diseased or undiseased.
• Fewer subjects• Quick results• Suitable for rare disease• OR• Relatively cheap
Cohort
• Cause to effect• Starts with exposure• Tests weather suspected
disease occurs more frequenty among exposed or not exposed
• Large no. of subjects• Long follow up• Suitable for rare exposures• RR/ AR• Expensive
References
• Epidemiology, Fifth edition. Leon Gordis• Park.s textbook of Prevntive and social
medicine, 23rd edition. K. Park• Maxcy-Roseneu-Last, Public Health and
Preventive medicine, 15th edition. Robert B Wallace.
• Oxford textbook of Public Health,Fifth edition
Thank You
top related