Scribd
Upload
416 views

Cohort Study Design

This document describes cohort study design. It defines a cohort study as an analytic study that follows groups of individuals over time to examine exposure-outcome relationships. Cohorts can be prospective, following disease-free individuals, or retrospective, identifying exposure status in the past. Advantages are the ability to establish temporal sequence and directly calculate incidence rates. Disadvantages include cost and potential loss of participants over time. Selection of cohorts and comparison groups, data sources, analysis including risk ratios and confidence intervals, and issues like bias and confounding are discussed.

Uploaded by

Ameerah Mantawil
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
416 views

Cohort Study Design

This document describes cohort study design. It defines a cohort study as an analytic study that follows groups of individuals over time to examine exposure-outcome relationships. Cohorts can be prospective, following disease-free individuals, or retrospective, identifying exposure status in the past. Advantages are the ability to establish temporal sequence and directly calculate incidence rates. Disadvantages include cost and potential loss of participants over time. Selection of cohorts and comparison groups, data sources, analysis including risk ratios and confidence intervals, and issues like bias and confounding are discussed.

Uploaded by

Ameerah Mantawil
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 25

Cohort Study Design

BL
August 15, 2015

Learning objective
At the end of the session, the students will be
able:

To describe and draw diagrammatic representation of


a cohort study design.
To distinguish between retrospective and prospective
cohorts.
To identify factors that are considered in the selection
of study cohorts.
To of data for cohort study.
To perform simple analysis of data from cohort
studies.
To identify issues in the interpretation of results of
analysis of data from cohort studies.

Introduction
analytic study design that is a more
straightforward method for drawing causal
inferences
This involves a design in which
information about the study factor (E =
exposure and E= non exposure) is known
for all subjects at the beginning of the
follow-up period

Introduction
These subjects are free of the outcome of
interest (usually disease-free individuals if
the outcome is incidence of disease) are
followed for a given period through
examinations or population surveillance,
during or after which new cases or deaths
(D) are identified.

Description
The cohort study has been called by a
variety of names:
Follow-up study
Prospective study
Incidence or mortality study
Longitudinal study

Cohort schema
C

ED
E C

ED
C
ED
EC
ED

Advantages
Since one starts with individuals free of the
outcome (disease free), the temporal sequence
between the exposure and outcome is more
clearly established (one is sure that the exposure
occurred before the outcome, which is a
requirement for the exposure to be a causal
factor).
The cohort study design allows for the direct
calculation of exposure specific incidence rates.
With cohort, one can look into the multiple effects
of a single exposure.

Lung Cancer
Chronic bronchitis
Smoking

Asthma
Other diseases

Note: In Case-Control Studies


Menarche
History of BC
BRCA
Other
exposures

Breast
Cancer

Advantages
Suitable for studies of rare exposures (those
identified as belonging to these group of
individuals who are exposed to this rare
exposure are taken as a cohort group and
compared to the general population).
Certain forms of biases that occur in other
analytic designs can be minimized, especially
with regard to the ascertainment of exposure

Disadvantages
Generally more expensive - this is brought
about by the need to ascertain the
development of the outcome at different
time points during the follow-up period.
Cohort studies require big sample sizes in
studies of rare disease.
Members of the cohort may be lost during
the follow-up period without knowing their
actual disease status.

Types of Case-Control Studies


Prospective Studies
Retrospective Cohort (Historical
Cohort)

Retrospective Studies
In this study, a group or groups of
individuals who are free of the outcome of
interest at some point in the past are
identified in terms of their exposure level
The cohorts may also be fixed or dynamic
he outcome status of these individuals up
to a specified time in the past or up to the
present time is determined through
existing records, e.g. morbidity or mortality
records

Retrospective Studies
In some retrospective cohort studies, the
follow-up period may spill over to the
future. Some authors call this
ambispective cohort studies

Selection of the Study Cohorts


depends on scientific considerations, feasibility
considerations and frequency of exposure
Asbestos worker
Those who have undergone a particular medical
therapy such as x-ray treatment or repeated
fluoroscopic examination
Individuals living near a suspected environmental
hazard (i.e. nuclear dumpsite in Clark Air Base)
Groups with unusual dietary or lifestyle practices such
as the Seventh Day Adventist
Those who are exposed by being at a given event
such as the dropping of the atomic bomb at
Hiroshima.

Selection of Comparison Group


These comparison groups should be
similar in all respects to the exposed
population except for the exposure
Another approach is sampling from the
general population

Sources of Data
For exposure information, examples of
sources of data include:
interview of cohort members,
medical examination of cohort members,
existing medical records and measures of the
environment.

Analysis of Cohort Studies


Data Layout

Cumulative Incidence of the Exposed (CIe) =


a/n1
Cumulative Incidence of the Unexposed (CIu) = c/n2
Cumulative Incidence Ratio (CIR) or Risk Ratio (RR) is:
a/n1
CIR = RR = CIe / CIu =
c/n2

95% Confidence Interval for CIR or


RR

Example

CIe = 143 / 160 = 0.89375


CIu = 3 / 55 = 0.0545455
CIR = RR = 16.4
Those who are exposed to general anesthesia are 16.4 times
more likely to develop post-operative pulmonary infection than
those who are exposed to spinal anesthesia.

95% Confidence Interval

LL = 5.45
UL = 49.3

Reject Null Hypothesis !!!

Reject H0

5.45

Reject H0

RR = 16.4

Reject H0

49.3

Issues in Interpretation
Selection Bias
Follow-up Bias
Misclassification Bias
Confounding

You might also like