This document outlines a case-control study focusing on odds ratio calculation within epidemiological research. It details the methodology for selecting cases and controls, the matching process, and how to measure exposure, alongside examples to calculate and infer risk factors related to various diseases. The document also discusses the advantages and disadvantages of case-control studies.

1. Class: III MBBS
Date: 22-04-2024
Case control study
Exercises on Odds ratio calculation

2. Epidemiological Studies
Observational Studies Experimental / Interventional Studies
Descriptive Studies Analytical Studies
1. Ecological studies.
2. Cross sectional studies.
3. Case Control Studies.
4. Cohort studies.
1. Randomized Controlled trials.
2. Field trials.
3. Community trials.
Epidemiological studies

3. Scenarios in the past
In early 1940s
Dr Alton Ochsner (A Surgeon in New Orleans)
observed that all the patients whom he
operated for Lung cancer had a history of
Cigarette smoking
He hypothesized that Cigarette smoking is
linked to lung cancer.
In 1940s
Sir Norman Gregg (An Ophthalmologist in Austria)
observed that infants and young children
developed unusual Cataract. These children were in
utero during a Rubella outbreak.
He hypothesized that Prenatal Rubella
exposure is linked to Cataract.
Can we conclude that the hypothesis is true?
What if:
• Most of the population smoked cigarettes among those who have not developed cancer?
• Most of the pregnant women during that time had Rubella infection but only some of their children
developed Cataract?

4. TIME
Direction of Enquiry
Population
CASES
People with the disease
CONTROLS
People without the disease
Exposed
Not Exposed
Exposed
Not Exposed
Basic premise of a Case control study

5. Case control study
3 distinct features
• Both exposure and outcome (disease) have occurred before the start of the study.
• Study proceeds from effect to cause.
• It uses a control or comparison group to support or refute an inference.

6. Basic steps in a Case control study
1 Selection of cases and controls.
2 Matching.
3 Measurement of exposure.
4 Analysis and interpretation.

7. Selection of cases and controls
Step 1
Selection of
Cases
Definition of
the CASE
• Diagnostic criteria
• Eligibility criteria: New vs
Old cases.
Sources of
CASES
• Hospitals: convenient but
source of bias.
• General population: ideal
in most situations. Done
using surveys or hospital
registries.

8. Selection of cases and controls
Step 1
Selection of
Controls
• Controls are similar to cases in all aspects other than having the disease under study.
Source of
controls
• Hospitals: cases with different disease.
• Relatives: spouse or sibling controls.
• Neighborhood controls: same locality / factory / school.
• General population: by survey or hospital registry usage.
Number of
controls per case
• Minimum of one control per case.
• Maximum of 4 controls per case.
Type of controls
• Same type:
Example: Cases are Lung cancer patients and Controls are case’s best friends
without lung cancer.
• Different type:
Case Controls
• Children with
Brain tumor
• Children with tumors other than Brain tumor.
• Normal children with no tumor.

9. Matching
Step 2
MATCHING
Process by which we select the controls
In such a way that they are similar to cases with regard to some pertinent
variables which are known to influence the outcome of the disease
AND
If not adequately matched for comparability could distort or confound the
results.

10. Matching
Step 2
Confounding factor
factor associated with both exposure and disease and is distributed
unequally in study and control groups.
• Example: Role of alcohol in esophageal cancer.
Smoking is confounding factor.
(associated with consumption of alcohol and is also an independent risk
factor for esophageal cancer)
Effect of alcohol consumption on esophageal cancer can be determined
only if the effect of smoking is neutralized.
Done by matching: Same number of smokers in cases and controls.

11. Matching
Step 2
Group matching Individual matching
Select all the cases
In cases
- 30% are aged more than 50 years.
- 45% are males.
- 25% are married.
Controls are selected so as to end up on
similar proportion like cases.
• Also called pair matching.
Case 1: 45 year old married Asian woman
with the disease under study.
Control 1: 45 year old married Asian woman
without the disease under study.
Case 2: 23 year old unmarried African male
with the disease.
Control 2: 23 year old unmarried African
male without the disease under study.

12. Measurement of exposure
Step 3
• Definition and criteria about exposure should be determined.
• Data is collected using:
- Interviews.
- Questionnaires.
- Past records: Employment / hospital records.

13. Analysis and interpretation
Step 4
Suspected or Risk
factor
Cases Controls
Present a b
Absent c d
a+c b+d
A) Exposure rates among cases and controls to suspected factor.
B) Estimation of disease risk associated with exposure (Odds ratio).

14. • Cases: a / (a+c) = 33/35 = 94.2%.
• Controls: b / (b+d) = 55/82 = 67%.
Inference: Frequency of lung cancer is higher among smokers than non smokers.
Suspected or Risk factor Cases
(with lung cancer)
Controls
(without lung cancer)
Total
Smokers
(Less than 5 cigars / day)
33
a
55
b
88
(a+b)
Non smokers 2
c
27
d
29
(c+d)
35
a+c
82
c+d
Step 4 A) Exposure rates

15. • Odds ratio: measures strength of association between risk factor and
outcome.
• Odds ratio is cross product ratio.
Suspected or Risk factor Cases
(with lung cancer)
Controls
(without lung cancer)
Smokers
(Less than 5 cigars / day)
33
a
55
b
Non smokers 2
c
27
d
35
a+c
82
c+d
Odds ratio = (a/b) / (c/d)
= ad / bc.
= 33x27 / 55x2
= 8.1.
• Inference: Odds of smoking (less than 5 cigars/ day) was 8.1 times higher among lung
cancer patients than among those with no lung cancer.
Step 4 B) Estimation of risk

16. Odds ratio interpretation
• 1.0 (or close to 1.0) indicates that the odds of exposure among case-
patients are the same as, or similar to, the odds of exposure among controls.
The exposure is not associated with the disease.
• Greater than 1.0 indicates that the odds of exposure among case-patients
are greater than the odds of exposure among controls. The exposure might
be a risk factor for the disease.
• Less than 1.0 indicates that the odds of exposure among case-patients are
lower than the odds of exposure among controls. The exposure might be a
protective factor against the disease.

17. Scenario 1
There were about 200 cases of acute myocardial infarction, of which
145 were smokers. Among 200 individuals recruited during the same
study period, 100 were smokers.
• Construct a 2×2 table and calculate and interpret the odds ratio.

18. Suspected or Risk
factor
Cases
(Acute MI)
Controls
(No Acute MI)
Smokers 145
a
100
b
Non smokers 55
c
100
d
200
a+c
200
c+d
Odds ratio = (a/b) / (c/d)
= ad / bc.
= 145x100 / 55x100
= 2.6
• Inference: Odds of smoking was 2.6 times higher among Acute MI patients than among
those without it.
• Smoking might be a risk factor for Acute Myocardial infarction.
Scenario 1 answer

19. Scenarios
Scenario 2 Scenario 3
A study was conducted to measure the
protective effect of BCG among children under
age 5 against tuberculous meningitis. Eighty
cases of tuberculous meningitis were
diagnosed in 10 hospitals in Mangalore. An
equal number of controls of similar ages were
selected. Trained workers assessed the BCG
vaccination status of the cases and controls by
looking for a typical scar over the left deltoid
region. One-fourth of the cases and half of the
controls had BCG scar. Calculate the odds
ratio. Calculate and interpret the protective
effect of BCG against tuberculous meningitis
A case-control study was carried
out to determine the risk of
cervical cancer among oral
contraceptive users. Among 100
women with cervical cancer, 17
used oral contraceptive pills.
Among 400 women without
cervical cancer, about 39
reported the use of oral
contraceptive pills. Construct an
appropriate 2×2 table and
calculate and interpret the odds
ratio.

20. Factor Cases
(TB Meningitis)
Controls
(No TB Meningitis)
BCG vaccine
received
20
a
40
b
BCG vaccine
not received
60
c
40
d
80
a+c
80
c+d
Odds ratio = (a/b) / (c/d)
= ad / bc.
= 20x40 / 40x60
= 0.33.
• Inference: Odds of BCG vaccination is 0.33 times lesser among TB meningitis patients
than among those without TB Meningitis.
• BCG vaccine might be protective against TB meningitis.
Scenario 2 answer

21. Risk Factor Cases
(Cervical cancer)
Controls
(No Cervical cancer)
Oral contraceptive
pill use present
17
a
39
b
Oral contraceptive
pill use absent
83
c
361
d
100
a+c
400
c+d
Odds ratio = (a/b) / (c/d)
= ad / bc.
= 17x361 / 83x39
= 1.9.
• Inference: Odds of using OCP is 1.9 times higher among those with Cervical cancer than
those without Cervical cancer.
• OCP might be a risk factor for Cervical cancer.
Scenario 3 answer

22. Scenarios
Scenario 4 Scenario 5
A case-control study was conducted
to determine the association between
Tobacco chewing and oral cancer.
About 250 cases of oral cancer and
500 controls were recruited for the
study. Tobacco chewing was present
in 175 oral cancer patients and 150
individuals without oral cancer.
Calculate exposure rates among cases
and controls, construct an appropriate
2×2 table, and calculate and interpret
the odds ratio.
A researcher investigates the influence of
parental smoking on the development of
Otitis media among children under 5 years
of age. The researcher recorded that
among 100 children with Otitis media, 39
of the parents smoked, and among 200
children without Otitis media, 42 of the
parents smoked. What are the odds of the
presence of Otitis media among children
less than age 5 with parental smoking
compared to parents without smoking?
Construct an appropriate 2×2 table, and
calculate and interpret the odds ratio.

23. Risk Factor Cases
(Oral cancer)
Controls
(No Oral cancer)
Tobacco chewing
present
175
a
150
b
Tobacco chewing
absent
75
c
350
d
250
a+c
500
c+d
Exposure rates
• Cases: a / (a+c) = 175/250 =
70%.
• Controls: b / (b+d) = 150/500 =
30%.
Odds ratio = (a/b) / (c/d)
= ad / bc.
= 175x350 / 150x75
= 5.4
• Inference: Frequency of exposure to tobacco chewing is higher among Oral cancer cases
than among those without oral cancer.
• Odds of tobacco chewing is 5.4 times higher among those with oral cancer than those
without Oral cancer.
• Tobacco chewing might be a risk factor for Oral cancer.
Scenario 4 answer

24. Factor Cases
(Otitis media)
Controls
(No Otitis media)
Parental smoking
present
39
a
42
b
Parental smoking
absent
61
c
158
d
100
a+c
200
c+d
Odds ratio = (a/b) / (c/d)
= ad / bc.
= 39x158 / 42x61
= 2.4
• Inference: Odds of parental smoking is 2.4 times among children with Otitis media than
those without parental smoking.
• Parental smoking might be a risk factor for Otitis media among their children.
Scenario 5 answer

25. Advantages of Case control study
1. Relatively easy to carry out.
2. Rapid and inexpensive (compared to Cohort study)
3. Requires fewer subjects.
4. Suitable to investigate a rare disease or diseases about which little is known.
5. No risk to subjects.
6. Allow the study of different etiological factors.
7. Risk factors can be identified.
8. No attrition.
9. Minimal ethical problems

26. Disadvantages of Case control study
1. Problem of bias since dependent on recall and past records.
2. Selection of appropriate control group may be difficult.
3. Cannot measure incidence.
4. Does not differentiate between causes and associated factors.
5. Not suited for evaluation of therapy or prophylaxis of disease.
6. Representativeness of cases and controls may not be appropriate.

28. Thank You