Cohort Study

RCT

REF: Park 20″ ed p. 59

• Observational studies

           a  Descriptive studies

           b  Analytical studies

I. Ecological or correlational

II. Cross sectional or prevalence

III. Case control or case reference

IV. Cohort or follow up

• Experimental studies

1. RCT or clinical trials
2. Field trials or community intervention
3. Community trials

Case control study REF: Park 20th edition page 67

Repeat in December 2011

Cause to effect (disease) = prospective study (cohort study)

Effect (disease) to cause = retrospective study (case control study) RCT and ecological studies are also prospective studies

Case control study [Ref: Park 20/e p69]

• Recall bias is associated with Case-control studies.
• Park writes- “When cases and controls are asked questions about their past history, it may be more likely for the cases to recall the existence of certain events or factors than the controls who are healthy persons. For example, those who have had a myocardial infarction might be more likely to remember and recall certain habits or events than those who have not. Thus cases may have a different recall of past events than controls”
• Recall bias is an error due to the differences in accuracy or completeness of recall to memory of past events or experiences. As a hypothetical example, suppose one is studying the possible relationship of congenital malformations to prenatal infections. A case-control study is conducted and interview is done of mothers of children with congenital malformations (cases) and mothers of children without malformations (controls). Each mother is questioned about infections she may have had during the pregnancy.
• A mother who has had a child with a birth defect tries to identify some unusual event that occurred during her pregnancy with that child. She wants to know whether the abnormality was caused by something she did. Why did it happen? Such a mother may even recall an event, such as a mild respiratory infection, that a mother of a child without a birth defect may not even notice or may have forgotten completely. This type of bias is known as recall bias.
• Bias is said to have occurred when there is a systematic difference between the results from a study and the true state of affairs. Bias may be introduced at all stages of the research process, from study design, through to analysis and publication. Bias can create a spurious association (i.e. overestimation of an effect) or mask a real one (underestimation of an effect). While appropriate statistical methods can reduce the effect of bias, they may not be able to eliminate it entirely. Innumerable number of bias have been listed.
• Bias associated with cohort study:

1. Selection
2. Confounding
3. Interviewer’s bias

• Bias associated with Case-control study:

1. Selection
2. Confounding
3. Recall bias
4. Berkesonian bias
5. Interviewer’s bias

• Interviewer bias is a type of information bias that arises when an interviewer consciously or unconsciously elicits inaccurate information from study subjects. Interviewer bias can result in differential error, which can seriously distort disease-exposure associations, if the interviewer is aware of the disease status and exposure hypothesis in a case-control study, or if the interviewer is aware of the exposure status and outcome hypothesis in a cohort study. In the former case, the interviewer may probe more deeply for evidence of exposure among cases than among controls. In the latter case, the interviewer may try to elicit evidence of health effects more assiduously in exposed than in unexposed cohort members.
• Selection bias occurs when patients included in the study are not representative of the population to which the results will be applied.
• Confounding bias occurs when we find a spurious association between a potential risk factor and a disease outcome or miss a real association between them because we have failed to adjust for any confounding variables. A confounding variable or confounder is an exposure variable that is related to both the outcome variable (e.g. disease) and to one or more of the other exposure variables. For example in the study of the role of alcohol in the etiology of esophageal cancer, smoking is the confounding factor because it is associated with the consumption of alcohol and it is also an independent risk factor for esophageal cancer. Thus the effects of alcohol consumption on esophageal cancer can only be determined when the influence of smoking is neutralized by matching.
• Berkesonian bias: It occurs when patient admission rates vary between the experimental and control groups, especially in studies that use hospitalized patients for both groups. For example, this type of bias is present when hospitalized patients with the risk factor (experimental group) are admitted at a higher rate than the control group.

Odds ratio

Case control study

Less time consuming

Cause and effect cannot be determined by a case-control study. That is one of it’s drawbacks. 

Randomized control trials produce best evidence for evidence of causation.

But in reality, experimental data may not be available in human beings.

Cohort studies also provide strong evidence for cause to effect relationship.

Crossectional studies provide weaker evidence and ecological studies provide the least satisfactory evidence for cause effect relationship.

Disadvantages of cohort studies:

1. Have to follow large numbers of subjects for a long time.
2. They can be very expensive and time consuming.
3. They are not good for rare diseases.
4. They are not good for diseases with a long latency.
5. Differential loss to follow up can introduce bias.

This study conducted in Mumbai is an example of a placebo-controlled, double-blind randomized clinical trial.

In this type of study, subjects with a particular disorder are randomized (the lottery) to receive either the treatment in question or a placebo. The information about whether the patient receives the treatment or the placebo is not known to either the subject or the investigator (double-blind).

This type of study may be used to infer causality, i.e., if the patients taking the medication have more CD4+ T cells than the other group, it is due to the drug in question.

A case-control study (Choice : Case-control study) is a retrospective study that pairs known cases of a disease with matched controls. This type of study gives information about the importance of risk factors. The lack of a prospective design makes inferences about causality difficult.

A case report (Choice : Case report) is a published report of a single incidence of a disease in a subject. This is usually done for rare diseases, or for rare associations of a disease with a particular risk factor.

A cohort study (Choice : Cohort study) is an observational (non-interventional) study that tracks subjects prospectively. This type of study gives a measure of incidence of a disease.

This is an example of a current cohort study or a prospective cohort study as it proceeds from cause to effect.

In this type of study incidence can be calculated.

A prospective cohort study is one in which the outcome (e.g. disease) has not yet occurred at the time the investigation begins.

It is also called longitudinal study, incidence study or forward-looking study.

Ref: Park’s Textbook Of Preventive And Social Medicine, By K. Park, 19th Edition, Pages 65, 66, 70, 71.

In a cohort study we can determine incidence rates directly in those exposed and those not exposed.

Incidence Rate is defined as the number of new cases occurring in a defined population during a specified period of time.

Only cohort study moves forward with time.

Here we study groups, who do not have the disease of interest initially but develops with time or not.

Ref: Park’s textbook of Preventive and Social Medicine, 21st edition, page 67-76

Only Cohort study follows patient for a definite time period.

It proceeds from cause to effect. So incidence can be calculated.

Besides relative risk as well as attributable risk are also derived from cohort study.

While case control study study yields only estimate of relative risk.

Cross sectional study yields prevalence rate.

Ans. is `d’ i.e., Field trial

Analytical epidemiology (Analytical study)

o Analytical studies are the second major type of epidemiological studies (after descriptive studies).

o It contrast to descriptive studies that look at entire population, in analytic studies, the subject of interest is the individual within the population (except in ecological study).

o The object is not to formulate, but to test hypothesis.

o There are four major types of analytical observational studies –

i)      Case control study Longitudinal studies

ii)     Cohort study

iii)    Cross sectional study

iv)    Ecological study

o Simplest form of observational Study —> Cross-sectional Study

o Most preferable observational/analytic Study design —> Cohort Study

o Least preferable observational/analytic study design —-> Ecological Study.

Ans. is ‘b’ i.e., Analytic study 

Cohort study

o Cohort study is a type of analytic observational study (like case control study).

o Cohort study proceeds forward from cause to effect, i.e., disease has not occured in subjects (In contrast to case-control study which proceeds backward from effect to cause).

Ans. is ‘b’ i.e., Cohort study 

Cohort study analyses the following data :‑

1)     Incidence rate among exposed and nonexposed

2)     Estimation of risk

a)     Relative risk (risk ratio)

b)     Attributable risk (risk difference)

c)     Population attributable risk

Incidence rate

o In cohort study, incidence rate can be determined directly in those exposed and nonexposed.

Ans. is ‘c’ i.e., Concurrent cohort study 

Concurrent cohort study or prospective cohort study is one in which the outcome (disease) has not occurrec at the time the investigation begins. In the given question the study began in 1970 and the outcome (occurrence of cancer) was studies between 1990 and 1995.  Thus the outcome had not occurred at the time the investigation begins. The study progressed from “cause to effect” is from ‘alcohol consumption’ to the occurrence of ‘cancer’ and hence is a classical prospective cohort study.

Types of cohort studies:

o Prospective cohort study:

        Known as ‘Current cohort study’ or ‘Concurrent cohort study’

       Outcome has not yet occurred when the study has begun: Only exposure has occurred; we look for development of same disease in both exposed and non-exposed groups

     Examples :- Framingham heart study, Doll & Hills prospective study on smoking & lung Cancer.

o Retrospective cohort study :

     Known as ‘Historical cohort study’ or Won-concurrent cohort study’

       Combines advantages of both Cohort study and Case control study

       Both exposure as well as outcome have occurred when the study has begun: First we go back in time and take only exposure into consideration (cohorts identified from past hospital/ college records), then look for development of same disease in both exposed and non-exposed groups

       Sample size required is same as that of prospective cohort study

        Usually used in occupational cancer studies

        Examples : – Effect of fetal monitoring on neonatal death; PVC exposure and angiosarcoma of liver.

o Combined prospective-retrospective cohort study

        Known as Mixed cohort study’

        Combines designs of both prospective cohort study and retrospective cohort study

       Both exposure as well as outcome have occurred when the study has begun: First we go back in time and take only exposure into consideration (cohorts identified from past hospital/ college records), then look for development of same disease in both exposed and non-exposed groups; later cohort is followed prospectively into future for outcome

        Examples :– Court-Brown & Doll Study on effects of radiation therapy.

Ans. is b  i.e., Cohort study

Example of cohort sutdies

1)         Framingham heart study is a classical example of cohort sutdy.

2)         Smoking and lung cancer study.

3)        Oral contraceptives and health study.

Ans. is ‘b’ i.e., Odd’s ratio can be calculated & c i.e., For rare disease

o Cohort study is more reliable than case-control study for association between a suspected risk factor and subsequent disease because relative risk can be estimated by cohort study, while case – control study measures only an estimate of relative risk (odds ratio).

Why Case Control Study is suitable for rare disease but not Cohort Study?

o In cohort study we proceed from effect to cause and if the study is for rare disease we may get very few cases or no case at the end of study. For example, if a rare disease has incidence rate. 01 per 1000 (1 per 100000) population and we take the sample of 100 people to expose the risk factor, there will be very few cases at the end of the study as the disease is very rare and has low incidence of 1 per 100000 population. (you can expect, How low will be the incidence in sample of 100 people).

o On the other hand, in case-control study we can choose controls for the few available cases and the history of possible/suspected exposure (s) can be explored.

Ans. is ‘b’ i.e., Lesser time spent 

o Cohort study is more time consuming.

o Cohort study is prospective study.

o Chances of bias are high in case-control study.

Ans. is ‘b’ i.e., Cohort study 

Ans. is ‘a’ i.e., Case series report 

Case series study: It is one of the research study used for non experimental epidemiology studies. Case series study simply describes presentation of a disease in a no. of patients. It does not follow the patients for a period & it uses no control or comparison group. Therefore it cannot establish a cause effect relationship & it’s validity is entirely a matter for reader to decide, whether it is useful or worthless.

Case report study :a special term form case series in which only one patient is described – this too may be very valuable or virtually worthless.

Case control study: It is a analytical study in which comparison of people who do have the disease (the cases) with otherwise similar people who don’t have the disease (controls) is made. These are retrospective studies. Case control studies start with outcomes or dependant variables. They then look back into past for possible independent variable (that may have caused the disease) to see if a possible risk factor was present more frequently in cases thean controls.

Cohort Studies: This focusses on factors related to development of a disease. A cohort (group of people with common characteristics) that does not have the disease of interest is selected & then observed for an extended period. Some of the members will already have been exposed to a suspected risk factor for the disease & others will eventually become exposed. Incidence of disease can be studied. It is also known as followup, or longitudinal incidence or prospective study.

Clinical trial: It is an experiemtnal method of epidemiological study. Clinical trials aim to isolate one factor (such as use of a drug) & examine its contributions to patients health by holding all other factors as constant as possible. They also utilise control groups & involve randomisation so also known as randomised controlled clinical trials.

Ans. is `b i.e., Cohort study 

o Incidence rate is estimated by cohort study.

Ans. is ‘b’ i.e., Cohort study 

Miscellaneous type of biases

i) Attention bias (hawthorne effect) —-> Study subject systemically alter their behavior when they are being observed as in follows up of Cohort study.

ii)Apprehension bias —> Certain parameters (e.g., pulse, BP, RR) may alter from their usual levels when the subject is apprehensive.

iii) Lead time bias (Zero time shift bias) —> Bias due to over-estimation of survival time, due to backward shift in starting point, as by screening procedures.

Ans. is ‘c’ i.e., Case control study 

Ans. is ‘c’ i.e., Cohort study 

o Amongst the given options Cohort study is best to test the association between risk factor and disease.

o Here are the different epidemiological studies with decreasing order of accuracy to test the association between risk factor and disease : ‑

1. Systematic review and meta-analysis —> Overall most reliable
2. Randomized controlled trials (controlled clinical trails)       Most reliable individual study.
3. Retrospective (Non-concurrent/historic) Cohort study.
4. Prospective (concurrent) Cohort study.
5. Case control study
6. Cross-sectional study
7. Ecological study

Ans. is ‘b’ i.e., Ecological study 

To test the association between risk factor and disease

o Most reliable, overall —> systematic review and meta-analysis of RCTs

o Most reliable individual study —> Randomized controlled trial.

o Most reliable observational analytic study         Cohort study

o Least reliable study and least reliable observational analytic study —> Ecological study.

Ans. is ‘a’ i.e., Case Control study 

o In a case control study we proceed from the effect to cause therefore we can study many causal associated factors at a time.

o Whereas in cohort study we proceed from the cause to effect, therefore we can get to several diseases associated with the cause but we can not study many causal factors at a time.

Ans. is ‘b’ i.e., Etiology for lung Ca is multifactorial

o Smoking is not specific for lung cancer because

i)  Not everyone who developed lung cancer has smoked –> Other etiological factors (other than smoking) are also involved in causation of lung cancer, i.e., etiology of lung Ca is multifactorial.

Not everyone who smoke develops lung cancer.

Ans. is ‘a’ i.e., Case control study 

o Case-control study proceeds backward from effect to cause, i.e. disease has already occured in subjects.

o All other studies proceeds forward from cause to effect, i.e. disease has not occured in subjects.

Ans. is ‘c’ i.e., RCT 

Ans. is ‘b’ i.e., Cohort study 

Ans. is ‘d’ i.e., Always prospective 

Ans. is ‘d’ i.e. Prospective cohort study 

Ans. is ‘c’ i.e., Randomized controlled trials 

Ans. C: Relative risk

Odd’s ratio can be derived from a case control study, which is a measure of the strength of the association between risk factor and outcome.

Ans. is ‘c’ i.e., Meta-analysis

Here are the different epidemiological studies with decreasing order of accuracy to test the association between risk factor and disease : ‑

1.  Systematic review and meta-analysis → Overall most reliable
2. Randomized controlled trials (controlled clinical trails) → Most reliable individual study.
3. Retrospective (Non-concurrent/historic) Cohort study.
4. Prospective (concurrent) Cohort study.
5. Case control study
6. Cross-sectional study
7. Ecological study

Ans. is ‘d’ i.e., More than one outcomes can be studied

Ans. is ‘a’ i.e., Cohort study

In case control study we proceed from effect to cause by taking history. It is possible to take history of exposure of multiple risk factors → multiple risk factors can be identified for a single disease.

In cohort study we start by exposing the study cohort to a particular risk factor and then follow up the cohort for the occurence of disease. So, several possible outcomes related to exposure can be studied simultaneously → we can study the association of the suspected factor with many other disease in addition to the one under study.

• Case control study → can identify multiple risk factors.
• Cohort study → can identify multiple outcome (disease).

Ans. is ‘b’ i.e., Cohort study

Ans. is ‘c’ i.e., Cohort study

Longitudinal studies are used to study natural history of disease. Among the given options, only cohort study is a longitudinal study.

Ans. is ‘b’ i.e., Meta-analysis

Here are the different epidemiological studies with decreasing order of accuracy to test the association between risk factor and disease :‑

1. Systematic review and meta-analysis → Overall most reliable
2. Randomized controlled trials (controlled clinical trails) → Most reliable individual study.
3. Retrospective (Non-concurrent/historic) Cohort study.
4. Prospective (concurrent) Cohort study.
5. Case control study
6. Cross-sectional study
7. Ecological study

Ans. is `c’ i.e., Case report

Case report:

In medicine, a case report is a detailed report of the symptoms, signs, diagnosis, treatment, and follow-up of an individual patient. Case reports may contain a demographic profile of the patient, but usually describe an unusual or novel occurrence. Since it involves information on single case it does not require matching.

Others i. e. case control study, cohort study and randomized control trials are based on the comparative study of two groups.

To decrease the bias in the observations in the two groups under consideration it is essential that the groups be matched for all the characteristics except for the one under study. Thus matching is essential in these epidemiological studies.

Ans. is ‘b’ i.e., Case-control study