CASE-CONTROL STUDY
It is a type of analytical study also called as Retrospective study.
It is commonly used as first approach to test hypothesis i.e to see whether
an association exists between the suspected causative factor and the
disease and what is the strength of the association.
Three distinct features of case-control study
1. Both exposure and outcome have occurred before the start of the disease.
2. The study proceeds backward from the effect to cause.
3. It uses a control or comparison group to support or refute an interference.
Basic steps
1. Selection of case and control
a. Selection of cases:-
Definition of cases -what exactly constitute a case should be specified
with the help of the diagnostic criteria and eligibility criteria before
start of the study.
Its should not be altered during the course of study.
Sources of case- hospital or general population.
b. Selection of control :-
Control should be free from disease, under study, and should be similar
to the case as far as possible except for the absence of disease under
study.
Sources of control – hospital, relatives, neighbourhood population and
general population.
2. Matching:
It is defined as the process by which we select a control in such a
way that they are similar to case with regard to certain pertinent
selected variables (eg- age), which are known to influence the
outcome of the disease and which if not adequately matched could
distort or confound the result.
A confounding factor is defined as factors that are known to be
associated with exposure and are independent risk factor for the
disease. E.g- while studying the role of alcohol in oseophageal
cancer, smoking which is the confounding factor should be
adequately matched in both the groups so as to elicit the
association between alcohol and oseophageal cancer. Smoking is a
confounding factor because the smokers are generally known to
consume alcohol and smoking itself is an independent risk factor for
oseophageal cancer.
Matching can be done either as group matching or pair matching.
3. Measurement of exposure :-
Definition and criteria about exposure are just important as those used to
define cases and controls.
What constitute an exposure should be well defined with the help of
specific definitions and criteria so as to avoid ambiguity in the collection of
data.
Information about exposure should be obtained in same manner for both
cases and control.
It may be obtained by interviews, questionnaire, studying past records like
hospital and employment records.
4. Analysis:-
Analysis of the data is done to –
Find whether association exist between the exposure and disease
by estimating the exposure rate (frequency of exposure) among
cases and controls.
Estimate the strength of association between disease and
exposure (suspected factor) by odds ratio.
Frame work of case-control study
Suspected Disease Total
cause Yes No
Present a b a+b
absent c d c+d
Total a+c b+d a+b+c+d
a
Exposure rate among cases = x 100
a+c
b
Exposure rate among control = x 100
b+d
If exposure rate among cases is greater that control that means there is
existence of association between disease and suspected factor.
Example -
Cases (with lung Control (without total
cancer) lung cancer)
Smoker 33(a) 55(b) 88(a+b)
Non-smoker 2(c) 27(d) 29(c+d)
total 35(a+c) 82(b+d) a+b+c+d
Exposure rate :
a 33
a) Cases = x 100 = x 100= 94.2 percent
a+c 35
b 55
b) Control = x 100 = x 100= 67.0 percent
b+d 82
P value less than or equal to 0.05 is regarded as “statistically significant”.
Smaller is the P value, greater is the statistical significance.
Estimation of risk
The estimation of disease risk associated with exposure is obtained by
an Index known as “Relative risk”/”risk ratio”- defined as the ratio
between the incidence of disease among exposed persons and
incidence among non-exposed.
Formula,
incidence among exposed a c
Relative risk = = /
incidence among non−exposed a+b c+ d
Odds ratio (cross-product ratio) = ad/bc
Odds ratio tells the strength of association between the causative
factor and the occurrence of the disease.
Advantages of case-control study
1. East to carry out
2. Inexpensive
3. Results are obtained rapidly
4. No risk to subjects
5. Requires less study subjects
6. Suitable for rare/chronic disease
7. Many factor can be studied together
8. Risk factor can be identified
9. No attrition problems.
10.Ethical problems are minimal.
Dis-advantages of case-control study
1. Problem of bias
2. Selection of control is difficult
3. Incidence cannot be measured
4. Does not differentiate between associated and causative factors
5. Not suited for evaluation of therapy or prophylaxis.
6. Another major concern is the representatives of cases and controls.
ODDS RATIO
Odds ratio measures the strength of association between risk factors and
outcome in a case-control study. It tells how much is the risk of developing
the disease amongst those exposed as compared to the non-exposed.
The derivation of odds ratio is based on three assumption:-
1. Disease being investigated must be relatively rare.
2. Cases must be representative of those with disease.
3. The controls must be representatives of those without the disease.
Calculation of odds ratio –
Step 1 :- construct 2x2 table
Exposure Cases Controls total
Present a b a+b
Absent c d c+d
Total a+c b+d a+b+c+d
Step 2 :- calculate odds ratio
Odds ratio = ad/bc
Step 3 :- Interpretation
Eg. - An odd ratio of 8 in a study of association of smoking and lung cancer
means that lung cancer cases are 8 time higher in smoker than in non-
smokers.
OR = 1 -- exposure does not affect odds of outcome
OR > 1 -- exposure associated with higher odds of outcome
OR < 1 -- exposure associated with lower odds of outcome
Importance of odds ratio :-
1. Used to compare the relative odds of the occurrence of the disease, with
different levels/magnitude of exposure.
2. It can be used to determine whether a particular exposure is a risk factor
for a particular outcome.
3. It can compare the magnitude of various risk factors for a given outcome.
BIAS
Bias is any systematic error in the determination of the association
between the exposure and disease.
The relative risk estimate may increase or decrease as a result of the bias.
It reflects some type of non-comparability between the study and control
groups.
Some types of bias are –
1. Bias due to confounding – this bias can be removed by matching in
case of case-control study.
2. Memory or recall bias –
3. Selection bias- it can be controlled by its prevention.
4. Berkesonian Bias – this bias arise due to different rates of admission
to hospital for people with different diseases.
5. Interviewer Bias – arises when the interviewer knows the hypothesis
and also knows the cases. It can be reduced by the -double binding.
Example of case- control study
a. Adenocarcinoma of vagina in USA
b. Oral contraceptive and thromboembolic disease by Doll and Vassey.
c. Thalidomide tragedy in Britian
