Ayaz Bhatti 1

What is Hypothesis?

Hypotheses is a statement about the population characteristics (parameter),

formulated with prior knowledge or based on theoretical consideration.

2
Formulation of etiological hypothesis. From case study,
case series, cross sectional study and descriptive
study
A hypothesis is a supposition arrived from
observation or reflection.
HYPOTHESIS SHOULD SPECIFY

1. The population

2. Specific cause being considered

3. The expected out come (disease)

4. Dose response relation ship

5. Time response relationship

3
Direction of Inquiry in analytical
studies (individual based)

CROSS-SECTIONAL

Case-Control TODAY Cohort

Historical
Cohort
4
Case –Control studies
(Retrospective)

5
Case-Control Studies

• an “observational” design comparing exposures in disease cases vs. healthy

controls from same population
• exposure data collected retrospectively

• most feasible design where disease outcomes are rare

6
 Case-Control Studies

It investigates the cause of disease (rare condition). Two

groups of individuals are compared, one with the Disease

(cases) with those who are free from the disease (controls).

The occurrence of the possible cause is then compared

between cases & controls i.e.

“it measures the prevalence of exposure in cases & control”.

7
Case-Control Studies
Distinct features:
• Both exposure & outcome have occurred before the start of study
• The study proceeds backward from effect to cause
• It uses a control or comparison group to support or refute an inference

8
Case-Control Studies

Cases: Disease
Controls: No disease

9
 factor present
Cases
(disease)
factor absent
Study
population
Case-Control Design

factor present
Controls
(no disease)
factor absent
present
past

time

Study begins here 10

Design of a Case - Control Study
Time
Direction of inquiry

Exposed
Cases
People with disease
Not exposed
Population
Exposed Control
People without
Not exposed disease
11
Methodology
1. Stating the hypothesis
2. Define Cases
3. Selection of cases
4. Selection of controls
5. Matching of controls
6. Determination of the start & duration of exposure
7. Minimize Bias
8. Confounding factors
9. Analysis

12
Odds Ratio (Analysis)

Cases Control

Exposed A B

Not C D
Exposed

Total A+C B+D

AD
Odds Ratio =
BC 13
Exercise
Deep vein Thrombosis

Present Absent

+ A= 40 A+C
B= 20

OC

- C= 20 D= 30 B+D

A+C B+D

OC + History of oral contraceptive use

OC - No history of oral contraceptive use
Odds ratio = A x D
BxC 14
The odds ratio in this study is:

A x D = 40 x 30 = 1200 = 3
B x C 20 x 20 400

Therefore, the odds ratio states that, according to

this retrospective case control Study, women with
a history of oral contraceptive use were three times
as likely to develop deep vein thrombosis as women
without this history of birth control.

15
 Advantages of case-control Study
1. Well suited to the study of rare diseases
2. Relatively quick to conduct.
3. Relatively inexpensive.
4. Requires comparatively fewer subjects.
5. Existing records can occasionally be used.
6. No risk to subjects
7. It allows to study multiple potential causes of a
disease

16
Disadvantages of case-control study

1.Relies on recall or records for information

on past exposures.
2.Validation of information is difficult or
sometimes impossible.
3.Selection of an appropriate comparison
group may be difficult.
4.Rates of disease in exposed and unexposed
individuals cannot be determined
17
Matching
A process by which we select control in such a way that
they are similar to cases with regard to certain pertinent
selected variables e.g. age, sex.. Which are known to influence
the outcome of disease and which, if not adequately matched
for comparability, could distort or confound the result

Confounding Factor
A characteristic which is associated both with exposure
and disease and is distributed unequally in study and control
groups. Although it is associated with exposure under
investigation, is itself, independently of any such association,
a risk factor of the disease.

18
Effect of Confounding

Smoking Cancer of the Larynx

Alcohol Consumption

Confounding Variable

19
Potential errors in epidemiological
studies.
• Random error
Random error is the divergence, due to chance
alone, of an observation on a sample from the
true population value, leading to lack of precision
in the measurement of an association. This can
be because of:
1. Biological variation
2. Sampling error
3. Measurement error

20
 Systematic error or Bias
Is a systematic error in determination of the
association between the exposure and disease . It
occurs in epidemiology when there is a tendency to
produce result that differ in a systematic manner from
the values.

• Confounding Bias
• Memory Bias
• Selection Bias
• Measurement Bias

21