This document discusses case-control studies, a type of observational research focused on identifying associations between exposures and diseases through retrospective comparisons of cases and controls. It outlines potential biases in the selection of cases and controls, highlights the importance of precise definitions to avoid misclassification, and describes the interpretation of odds ratios as a measure of association. The document also emphasizes the strengths and weaknesses of case-control studies and includes examples of how different studies are designed and evaluated.

1. CASE CONTROL STUDIES
Drug information resources
Prepared by : Dr.Nada Essam
Under the supervision of : Dr/ Hesham Metwally

2. Clinical research design

3. Retrospective cohort study

4. [A] Case-control studies
*A type of observational study commonly used to look
at factors associated with diseases or outcomes.
*Compare groups retrospectively.
*They are often used to generate hypotheses that can
then be studied via prospective cohort or other studies.

5. Case-control study
The researcher compares the exposure history of cases
(individuals with the disease) and controls (individuals
without the disease) to identify factors that may be
associated with the disease.

6. Case-control study
The researcher then looks at historical factors (retrospectively) to
identify if some exposure(s) is/are found more commonly in the
cases than the controls.
If the exposure is found more commonly in the cases than in the
controls, the researcher can hypothesize that the exposure may
be linked to the outcome of interest.

7. Synonyms:
Exposure
Risk factor
determinant
= Independent variable
Outcome
Disease
Condition
Drug related event
= Dependent variable

8. Study begins here
Controls
Cases
Time
Exposed
Exposed
Not exposed
Not exposed

9. Example:
Study begins here
Controls [No peptic ulcer]
Cases [Peptic ulcer]
Time
NSAIDs use
NSAIDs use
Not NSAIDs use
No NSAIDs use

10. [B] Bias & validity of case control study

11. Sources of bias in case control study
1. Selection of cases:
Misclassification bias
2. Selection of controls:
Selection bias
3. Exposure data:
a. Information bias
b. Recall bias
c. Interviewer bias

12. 1. Selection of cases:
Misclassification bias
When individuals are assigned to a different category
than the one they should be in.
Avoid by: Precise outcome definition
e.g., Obesity (measured BMI vs self reported)

13. 2. Selection of controls:
Selection bias
A distortion in a measure of association due to a control
selection that does not accurately reflect the target
population (cases).
Avoid by: Controls should be selected in:
1. Population-based (e.g. same geographical area of
cases)
2. Hospital or clinic (e.g. from same institution of cases
except not having the disease of interest)

14. 3. Exposure data:
a. Information bias
Due to errors in data collection.
Avoid by:
1. Accurate measurement/recording of data
2. Identical way of collecting data from cases & control
Note: Objective data obtained from records preferred
than subjective data obtained by asking the patient

15. 3. Exposure data:
b. Recall bias
Patient may have different recalls of past events
than controls (memory).

16. 3. Exposure data:
c. Interviewer bias
The expectations or
opinions of the
interviewer interfere
with his objectivity,
either negatively or
positively, clouding their
judgment of the person
being interviewed.

17. Case-control study
Strengths
• Fast/Easy/Cheap
• Rare/ Long Latency
Diseases
• Multiple Risk Factors
Weaknesses
• Misclassification bias
• Selection bias
• Recall bias
• Information bias
• Interviewer bias
• Confounding?

18. Confounding
Confounding is a special type of bias.
It is defined as a distortion of the association between an
exposure and an outcome.
It occurs when the study groups differ with respect to a
third factors that influence both the outcome & exposure.
Age & gender are common confounder.
Controlling can be by matching or restriction.

19. Confounding

20. [C] Odds Ratio (OR)
*The odds ratio is commonly used in case-control studies
*Where the odds of exposure are compared between
cases (individuals with the outcome) and controls
(individuals without the outcome)
*To estimate the strength of association between the
exposure and the outcome.

21. Odds Ratio (OR)
OR = (a/b)/(c/d)= ad/bc
a = number of exposed individuals with the outcome
b = number of exposed individuals without the outcome
c = number of unexposed individuals with the outcome
d = number of unexposed individuals without the
outcome
Outcome

22. Odds Ratio (OR)
OR = (a/b)/(c/d)= ad/bc
OR = (80/30)/(20/70)
OR = 9.33
Outcome
Exposure
Control
Case

23. Interpretation of OR
No association
Odds of exposure is the same
1
Positive
association
Odds of exposure is greater in
diseased group
>
1
Negative
association
Odds of exposure is lower in
diseased group
<
1
Measure association between an exposure and an outcome:

24. [D] Cases

25. Case 1: OR ratio
Investigating the association between caffeine
consumption and the risk of developing heart
palpitations. a study of 500 people, 250 of whom
regularly consume caffeine and 250 of whom do not.
Among the group that consumes caffeine, 50 people
develop heart palpitations, while 200 do not. Among
the group that does not consume caffeine, 25 people
develop heart palpitations, while 225 do not. What is
the odds ratio?

26. Case 1: OR ratio
Heart palpitation No heart palpitation
Caffeine 50 200
No caffeine 25 225
OR = (a/b)/(c/d)= ad/bc
OR = (50/200)/(25/225)
OR = 2.25

27. Case 2

28. 1. Is there an association?

29. 2. Identify “exposure” and “outcome”

30. 3. Any efforts to address potential
sources of bias?
1
2

31. 3. Any efforts to address potential
sources of bias?
1. Using death registry for information about both
cases and controls (identical)>> avoiding information
bias
2. Precise definition of both cases & controls >>
avoiding misclassification bias

32. 4. Is there a positive association
between an exposure and an outcome?

33. 4. Is there a positive association
between an exposure and an outcome?
Odds of exposure (non-use of bicycle helmets) is
greater in those who died from fatal head
injury.
Positive association!

34. Case 3

35. Multiple risk factors can be studied

36. 1. Is there an association?

37. 2. Identify “exposure” and “outcome”

38. 3. Any efforts to address potential
sources of bias?
1
3
2

39. 3. Any efforts to address potential
sources of bias?
1. Using medical records to obtain data about both
cases and controls >> avoiding information bias
2. Cases and controls from same institution except for
the exposure variable >> avoiding selection bias
3. Matching >> avoiding confounding

40. 4. Is there a positive association between a
Tracheobronchial stenosis and an outcome?
Odds of exposure (Tracheobronchial stenosis ) is
greater than 1.
Positive association!

41. THANK YOU!