Analytical

Epidemiology
Subheading

Dr Deepthi Athuluru
 CLASSIFICATION OF EPIDEMIOLOGICAL STUDIES

Epidemiological
Studies

Observational Experimental

Community
Descriptive Analytical RCT Field Trials
Trials

Cross
sectional Ecological Case-control Cohort
 Analytical Studies
•Second major type of epidemiological studies
•Subject of interest is the individual or small groups
of individuals in contrast to descriptive studies where
researcher deals with large populations
•Tests the hypothesis
•Inference is to population
 Case- Control Studies
• A common first approach to test causal
hypothesis
• Increasingly used to know the causes of
diseases, especially rare diseases
• Other names:
case referent,
retrospective or
trohoc studies
 Features of …

– Both exposure and outcome have occurred

before the onset of the study
– Study proceeds backwards, from effect to cause
– Use a control or comparison group to support or
refute an inference
… Case- Control Studies

Design of a case-control study

Time

Direction of enquiry

Exposed

Cases

Not exposed

Population

Exposed

Controls

Not exposed
Retrospective study
… Case- Control Studies

Steps in …
1. Selection of cases and controls

2. Matching

3. Measurement of exposure

4. Analysis and interpretation

… Case- Control Studies

1. Selection of cases & controls

• Selection of cases

– Define the cases

• Diagnostic criteria

• Eligibility criteria

– Sources of cases

• Hospitals

• General population
… Case- Control Studies

• Selection of controls
– Crucial step in case-control studies
– Controls must be
• Free from the disease under study
• Be similar to the cases except for the
absence of the disease under study
• Sources of controls
– Hospitals
– Relatives
– Neighborhood controls
– General population
… Case- Control Studies

2. Matching

• Definition:
“The process by which we select controls in such a
way that they are similar to cases with regard to
certain pertinent selected variables, which are
known to influence the outcome of disease and
which, if not adequately matched for comparability,
could distort or confound the results.”
• Confounding factor
– One which is associated both with exposure and
disease; and is distributed unequally in study and
control groups

– Although associated with ‘exposure’ under

investigation, it itself is a risk factor for the disease
Eg1. Alcohol – oesophageal cancer : confounding
factor – smoking
 … Case- Control Studies - Matching

• Do not match the suspected etiological factor

• Methods of matching
– Group matching
– Matching by pairs
 … Case- Control Studies

3. Measurement of exposure
• Exposure can be measured by
– Interviews
– Questionnaires
– By studying past records
– Examinations

• Bias should be avoided while measuring the exposure

by blinding the investigator
… Case- Control Studies

4. Analysis

•Involves two steps

1.Exposure rates among cases and controls to
suspected factor
2.Estimation of disease risk associated with
exposure (odds ratio)
• 1. Exposure rates

Cases Controls

Tobacco chewing 33 55
(a) (b)
Non chewing 2 27
(c) (d)
Total 35 82
(a+c) (b+d)

Exposure rate among cases= (a/a+c)100 = (33/35) 100 = 94.2 %

Exposure rate among controls= (b/b+d)100 = (55/82) 100 = 67 %
… Case- Control Studies

Odds ratio (cross product ratio)

• It is a key parameter in the analysis of case
control studies

• A measure of the strength of the

association between risk factor and
outcome
• Derivation of odds ratio is based on 3
assumptions
– Disease under investigation is a rare one
– Cases are representative of those with
disease
– Controls are representative of those
without disease
 Cases Controls
Tobaccco 33 55
chewing (a) (b)
Non chewing 2 27
(c) (d)
Total 35 82
(a+c) (b+d)

Odds ratio = ad/bc = 33 X27/ 55X2 = 8.1

tobacco chewing have a risk of having oral cancer

8.1 times that of non smokers
… Case- Control Studies

ADVANTAGES
• Relatively easy to carry out
• Rapid and inexpensive (compared with cohort
studies)
• Require comparatively few subjects
• suitable to investigate rare diseases or about which
little is known.
• No risk to subjects
• Allows the study of several different aetiological
factors (e.g., smoking, physical activity and
personality characteristics in myocardial infarction)
• Risk factors can be identified. Rational
prevention and control
programmes can be established
• No attrition problems, because case control
studies do not require follow-up of
individuals into the future
• Ethical problems minimal
… Case- Control Studies

Disadvantages
• High chances for bias
• Validation of information obtained is
difficult or sometimes impossible
• Selection of an appropriate control group
may be difficult
• We cannot measure incidence, and can
only estimate the odds ratio
• Not suited to the evaluation of therapy or
prophylaxis of a disease
 Cohort studies
• Usually undertaken to obtain additional evidence
to refute or support the existence of an
association between suspected cause and disease
• Other names
– Incidence study
– Forward looking study
– Longitudinal study
– Prospective study
… Cohort studies

Features
1. Cohorts are identified prior to the appearance of
the disease under investigation
2. Study groups are observed over a period of time
to determine the incidence of disease
3. The study proceeds from cause to effect
• Cohort is defined as a group of people who share
a common characteristic or experience within a
defined time period
• Eg, age cohorts, occupational cohorts, exposure
to a drug cohorts, marriage cohort etc.
… Cohort studies

Indications
1. When there is a good evidence of an association
between exposure and disease
2. When the exposure is rare but the incidence of
disease is high among exposed
3. When the attrition can be minimized
4. When ample funds are available
… Cohort studies
Design of a cohort study
Time

Direction of enquiry

Disease

Exposed

No disease
Healthy
Population

Disease

Not exposed

No disease
… Cohort studies

Types of …

• Prospective cohort study

• Retrospective cohort study

• Combination of retrospective and

prospective cohort studies

… Cohort studies

Steps in …
• Selection of study cohorts

• Obtaining data on exposure

• Selection of comparison groups

• Follow up

• Analysis
… Cohort studies

Selection of study subjects

• Cohorts can be selected from
• General population
– Special groups
• Select groups (eg. Doctors, lawyers,
teachers, etc.)
• Exposure groups
… Cohort studies

Obtaining data on exposure

• Information can be obtained from
– Cohort members
– Review of records
– Medical examination or special tests
– Environmental surveys
• Information about exposure should facilitate
classification of cohort members
– According to whether or not they were exposed
– According to the degree of exposure
… Cohort studies

Selection of comparison groups

•Internal comparisons
•External comparisons
•Comparison with general population
… Cohort studies

Follow up
•Periodic medical examination of each member
•Reviewing physician and hospital records
•Routine surveillence of morbidity and
mortality records
•Mailed questionnaires, telephone interviews,
periodic home visits
… Cohort studies

Analysis
• Data is analysed interms of
i. Incidence rates of outcome among
exposed and non-exposed
ii. Estimation of risk
• Relative risk
• Attributable risk
… Cohort studies

Incidence rates
• Incidence can be Cigare Lung No Total
measured directly tte cancer lung
smoki cancer
• Incidence rate among ng
smokers = 70/7000 = 10 Yes 70 6930 7000
a b a+b
per thousand
• Incidence rate among non- No 3 2997 3000
smokers= 3/3000 = c d c+d

1 per 1000
 Relative risk
• The ratio of incidence of Cigarett Lung No lung Total
e cancer cancer
disease among exposed smokin
g
and incidence among
Yes 70 6930 7000
non-exposed
a b a+b
• Also called ‘risk ratio’
No 3 2997 3000
c d c+d

Incidence among exposed

• RR= Incidence 10/1 = 10
among non-exposed =

• RR is the direct measure of strength of association

between suspected cause and effect
Attributable risk:

• The difference in incidence rates between exposed and

non-exposed groups

AR = Incident rate among exposed – incidence rate among non-exposed

Incident rate among exposed

(10-1/10) X 100 = 90%

=

•It indicates to what extent disease can be attributed to

the exposure
•Suggests the amount of disease that might be
eliminated if the factor could be controlled
Advantages

Allow the possibility of measuring directly

the relative risk of developing the condition for those
who have the characteristic, compared to those who do
not.
Allows for a conclusion of cause-effect relationship.
Because the presence or absence of the risk factor is
recorded before the disease occurs, there is no chance
of bias being introduced due to awareness of being sick
as encountered in case-control studies.
•Dose- response ratios can be calculated.

•Cohort studies are capable of identifying other

diseases that may be related to the same risk factor.

•Bias can be avoided.

•Unlike case-control studies, cohort studies provide

the possibility of estimating attributable risks, thus
indicating the absolute magnitude of disease
attributable to the risk factor.
Disadvantages

- Not always feasible.

- Relatively inefficient for studying rare
conditions.
- Sample sizes required for cohort studies are
extremely large, especially for infrequent
conditions; it is usually difficult to find and
manage samples of this size.
-The most serious problem is that of attrition
- There may also be attrition among investigators who
may lose interest, leave for another job, or become
involved in another project.
- Over a long period, many changes may occur in the
environment, among individuals or in the type of
intervention, and these may confuse the issue of
association and attributable risk.
- They are very costly in time, personnel, space and
patient follow-up.
Case control study Cohort study
Proceeds from effect to cause Proceeds from "cause to effect"

Starts with the disease Starts with people exposed to risk factor or
suspected cause.

Tests whether the suspected cause Tests whether disease occurs more
occurs more frequently in those with frequently in those exposed, than in
the disease than among those without those not similarly exposed.
the disease.

Usually the first approach to the Reserved for testing of precisely

testing of a hypothesis, but also useful formulated hypothesis
for exploratory studies

Involves fewer number of subjects Involves larger number of subjects

Yields relatively quick results Long follow-up period often needed,
involving delayed results.

Suitable for the study of rare diseases Inappropriate when the disease or
exposure under investigation is rare.

Generally yields only estimate of RR Yields incidence rates, RR as well as AR.

(odds ratio)

Cannot yield information about Can yield information about more than
diseases other than that one disease outcome.
selected for study

Relatively inexpensive Expensive.

 BIAS
Systemic error in determination of the association
between the exposure and disease.
 Types of Bias:
• Bias due to confounding
• Selection bias:
Prevalence and incidence bias (selective survival)
Admission rate (Berksons/Berkesonian bias)
• Information bias: memory or recall bias
telescopic bias
interviewers bias

Confounding
Confounding factor:
factor: One
One which
which isis associated
associated both
both withwith
exposure
exposure and
and disease;
disease; and
and isis distributed
distributed unequally
unequally in
in study
study
and
andcontrol
controlgroups
groups
Blinding:
Blinding is the concealment of group
assignment to either the treatment or
control
 To recapitulate……..
• Case control studies
• Cohort studies
• Bias
The method of knowledge is experiment
- William Blake

Thank you