Unit VII: Epidemiological Methods

Objectives
At the completion of this unit learners will be
able to:
• Define study design,
• Discuss classification of study design.
• Describe the Descriptive study designs.
• Discuss the Analytical study designs
 Study design
A study design is a specific plan or protocol for
conducting the study, which allows the
investigator to translate the conceptual
hypothesis into an operational one.

3
 Study design
A set of defined steps required to carry out research
on a problem under study. The design will define:
• How study subject are selected?
• Method of sample size estimation
• Procedure of data collection
• Procedure of data analysis
• Types of statistical test required

4
 Cont
….
• The proof for evidence-based medicine is all
collected via research, which uses a variety of
study designs.
• Different study designs provide information of
different quality.
• Therefore, you need to understand the
strengths and limitations of each type of
study design, as applied to a particular
research purpose.
Classification of Epidemiological Research/
Study Designs

Descriptive Analytical
Research
Research

6
 Descriptive Research

Individual
Population based
based
Case reporting

Case series
Ecological /
Correlational Cross-sectional surveys
7
 Analytical Research

Observational Experimental /

Interventional

Cohort study Randomized Control

Trials
Case–control study

Cross-sectional study Quassi 8

 Objectives at various levels

Study Types Objectives

1. Knowing the frequency of
DESCRIPTIVE disease
STUDIES 2. Knowing the distribution
3. Developing the hypothesis
OBSERVATION 1. Testing the hypothesis
AL ANALYTICAL 2. Establishing
association
EXPERIMENTAL
OR 1. Strength of
INTERVENTIONAL association
STUDIES 2. Establishing the 9
 The
The researcher
Study Types studies, but does
not alter, what
researcher
intervenes to
change reality,
then observe what
occurs
happens
 Descriptive Studies
• Describe only; do NOT examine associations
between Exposure (E) and health Outcome (O).

• Generally the purpose is to describe the variability in

a health outcome and/or formulate hypotheses.

• A descriptive study involves describing the

characteristics of a particular situation event or case.

• Descriptive studies can be carried out on a small or

larger scale.

11
 Types of Descriptive Studies
Individual Based
Case Study
A study of one diseased individual, providing a detailed
description of an uncommon disease; provides timely
or rare information.
OR
A single patient’s clinical history is described in detail,
and then discussed in relation to the literature.
Almost always a rare unusual, or atypical case.
 Types of Descriptive Studies
Individual based
Case Series :
A study of multiple occurrences of unusual cases that
have similar characteristics.
Investigators can calculate the frequency of symptoms
or characteristics of people with the disease.
Results may generate causal hypotheses. Neither a
case study nor a case series includes a comparison
group.
Descriptive Study
Designs
Case One case of
Repor t unusual finding

Case Multiple cases

Ser ies of finding
 Types of Descriptive Studies
Individuals Based
Cross sectional Surveys
– Subjects or institutions are surveyed in order to
describe the prevalence of health outcomes and /
or characteristics of a population

15
 Descriptive Studies
Population Based
• Ecological
– An ecological study focuses on groups of people
(rather than individuals) as the units of
analysis.
– The variables include measurements taken at the
group level e.g. infant mortality rates of different
countries.

16
Types of Observational Analytical
Studies

17
 Analytical
(Non-
Intervention)
Studies

Cross-
sectiona Cohort
l studies
studies
Case-
control
studies
 Cross-sectional study
⚫ A cross sectional study measures the
prevalence of health outcomes or
determinants of health, or both, in a
population at a point in time or over a short
period.
 CROSS-SECTIONL STUDY

Information is collected from each subject at one point of time

Used to provide a snapshot of a population at a point in time

The main out-come measure is prevalence

Limited to the measurement of risk factor and out-comes at

one simultaneous point in time

Examples: screening surveys

knowledge attitude and practice (K.A.P.) surveys
 Begin with: Target
Population I

Sampl Determine presence or

absence of exposure &
e presence or absence
of disease

Gather Data on Exposure and

Disease

Exposed; Exposed; Not Not Exposed;

Have Do not Exposed; Do not have
Disease have Have Disease
Disease Disease

4 groups are possible

 Disease No disease II
Exposed a b
Not
Exposed
c d

Disease No disease Disease No disease

Exposed a b Exposed a b
Not c d Not c d
Exposed Exposed
 III
Disease No disease Disease No disease

Exposed a b Exposed a b
Not c d Not c d
Exposed Exposed

Prevalence of exposure Prevalence of disease

compared in diseased and compared in exposed and
non diseased non exposed
a b a c
vs. OR vs.
a+b c+d
a+c b+d
 Advantages of cross-sectional
• Outcomes and exposures measured at the
same time
• Uncovers associations for further study
• Useful for hypothesis generation
• Quick & cheap (no follow up)
• Best way to determine prevalence
• Questionnaire/interview based
• Useful for assessing practice, attitudes, knowledge,
beliefs , utilisation of services etc
 Advantages of Cross-Sectional study

• Can be conducted to assess the health care needs of

the population
• Helpful in measuring access and utilization of health
services
• Provides information between disease and various
variables
• Provides information regarding distribution of a
disease
• Determines burden of the diseases in a population.
So helpful for planning purposes
 Limitations of Cross-Sectional study

• No temporal or time sequence

so gives no information whether which comes first e.i. Cause or Disease

• Gives no idea about natural history of the disease or etiology

• Gives no measure of new cases occurrence

• Not useful for rare exposures or rare outcomes
 COHORT
STUDY
• Cohort are also called “ Fo l l o w - u p o r
Istudies
n c i d e n c e St udies” .

• Because the data on exposure and disease refer to

different points in time, cohort studies are also
longitudinal.

• Cohort studies have also been called “ P r o s p e c t i v e

St u di e s” .
 Cohort studies

• The observation of a cohort over time to

measure outcome(s)
• Synonymous terms (Last’s)
◦ Follow-up
◦ Longitudinal
◦ Prospective
 Cohort studies
□ Birth cohort: all individuals in a defined
geographical area born in the same period (usually
a year)
□ Inception cohort: all individuals assembled at a
given point based on some factor e.g. Workplace
 Cohort studies
□ Exposure cohort: a group of individuals that
potentially share a common exposure e.g.
Radiation

□ Disease cohort: a group of individuals with a

specific disease.
 STEPS IN COHORT
STUDY

• Cohort studies are conducted in three fundamental

steps:

1. Identify cohorts of exposed and unexposed

individuals who are free of the disease/outcome of
interest at the beginning of the study.
2. Observe each cohort over time for the development
of the outcome(s) of interest.
3. Compare the risks of outcomes between the cohorts.
COHORT STUDY
DESIGN
 Cohort studies
Recruitment

Prospective
Exposure Outcome

Retrospective
Exposure Outcome

Exposure Outcome
Ambidirectional
Exposure Outcome

Time
 COHORT STUDY
DESIGN
• Cohort study measure:

i. Incidence rate
ii. Relative Risk
iii. Attributable Risk
 DESIGN OF A
COHORT STUDY

Then Follow to see

whether
Disease Disease Total Incidence
Develop Does Rate of
not Disease
Develo
p
First Expose a b a+b a/a+ b
Select d

Not c d c+d c/c + d

Expose
d
 INCIDENCE
RATE

• Incidence in exposed group = a/ a + b

• Incidence in unexposed group = c/ c

+d

• Incidence in total (exposed + unexposed)

• a+c
= a+b+c+d
 RELATIVE
RISK

• Cohort study determine whether there is an

association between exposure to a factor and
development of a disease.

• Relative Risk = Incidence in exposed

Incidence in unexposed

= a/ a + b
c/ c + d
 ATTRIBUTABLE
RISK
• This is determined by the “ A t t r i b u t a b l e Ri s k ”,
which is defined as “ t h e a m o u n t o r p r o p o r t i o n
of diseases incidence ( o r disease risk) t h a t c a n
b e a t t r i b u t e d t o a spec ific e x p o s u r e ” .

• Attributable Risk is calculated as follow:

• Risk Difference = (Incidence in exposed group ) –

(Incidence in non-exposed group [Background risk]
 Advantages of cohort studies

• Useful for rare exposures

• Useful for more than one outcome
• Incidence of the outcome (and incidence rates)
• Temporal relationship between exposure and
outcome is clear as exposure status defined at start of
study
• If prospective, minimises bias in
measurement of exposure
• Sometimes the only ethical or legal way to do
study
 Disadvantages of cohort studies

• Not good for study of rare outcomes

• If retrospective they rely on the adequacy of records
• Exposed may be followed more closely
than unexposed
• If prospective they can be very expensive
and slow
• As they are follow up studies, the validity of
results is highly sensitive to losses to follow up
(migration, withdrawal, lack of participation, death)
CASE-CONTROL STUDY
DESIGN
 Case-control studies
• An analytical epidemiologic study design in which
individuals who have the disease under study, also
called cases, are compared to individuals free of
disease (controls) regarding past exposures.
• Exposure differences between cases and controls
are helpful to find potential risk or protective
factors. The purpose is to determine if there are
one or more factors associated with the disease
under study.
 CASE-CONTROL
STUDY
• To examine the possible relation of an exposure
to a certain disease, we identify;

1. A group of individuals with the disease

( c a l l e d c a s e s ) and for purpose of
comparison,

2. A group of people without the disease or

outcome variable ( c a l l e d c o n t r o l s ).

3. The study compares the occurrence of the

possible cause in cases and in controls.
DESIGN OF A CASE-
CONTROL STUDY
 Advantages of case-control Studies

• Can be carried out quickly and quite cheaply

• Useful for rare diseases and outcomes
• Can study multiple exposures for a
single outcome
• Case control studies can be ideal for the
study of rare diseases or those with a long
latency
• Compares odds of exposure between
cases and controls
 Disadvantages of case-control studies

• Selection of control population, overmatching

• Information bias as exposures – similar status
determined after outcome has occurred e.g. Recall
• Selection bias especially regarding controls
• Cannot establish sequence of events
(temporal relationship)
• Not good for rare exposures
• Cannot usually be used to estimate
incidence rates, relative risks or attributable
EXPERIMENTAL
STUDY
DESIGN
 Randomised controlled trial (RCT)

”An epidemiological experiment in which subjects

in a population are randomly allocated into groups,
usually called study and control groups to receive
and not receive an experimental preventive or
therapetuic procedure, maneuver, or
interventition”
Randomized Control Trials
(R. C.T)
Randomized:
Allocation of participants to various groups
in random fashion

Control:
A control groupis used to compare the
effects of a particular treatment

Trials:
An experiment conduction.

60
 RCT
Reference population

Taraget population

Sample

Random Allocation

Changed group
Intervention group during study Control group

Loss to
Loss to Outcome measure follow up
follow up
 RANDOMIZED CONTROLLED
TRIAL
• The true experimental study design (RCT) has three
characteristics:

1. R A N D O M I Z A T I O N - the researcher takes care to

randomly assign subjects to the control and
experimental groups.

• (Each subject is given an equal chance of being assigned

to either group.)
 RANDOMIZED CONTROLLED
TRIAL

2. C O N T R O L - the researcher introduces one or more

control group(s) to compare with the experimental
group.

3. M A N I P U L A T I O N - the researcher does something to

one group of subjects in the study.

• Note: The strength of experimental studies is that by

randomization of confounding variables.
 RANDOMIZED CONTROLLED
TRIAL

• In Randomized Controlled Trial (RCT), we begin

with a defined population.

• Subjects in the study population are randomly

allocated to intervention and control groups, and
the results are assessed by comparing outcomes.

• The basic design of RCT is given below;

 Allocation of study subjects -
randomization
• Random = governed by chance

• Randomization = allocation of individuals to

groups by chance

• Each sampling unit has the same chance of

selection
 BLINDING IN
RCT
• Blinding represents an important, distinct aspect of
randomized controlled trials.
• The term blinding (masking) refer to keeping the trial
participants ,healthcare provider
, participants healthcar
• piThe
nr voevterm
si dt iegrasbt)ol irns do(iru
n sgaus(asmlelasys ok ri n
s g ) ( refers
t h o s e to keeping e
, outcome data) unaware of an collectin
intervention, so that they are not influencedgbyassigned that
knowledge.

• Blinding prevents bias at several stages of a trial.

 TYPES OF
BLINDING
• Single Blind
– The subjects are not knowing the group to which
they are belonging .

• Double blind trials

– Neither the subject nor care giver is aware about
the groups

• Triple blind trials

– The subject, the care giver (nurse or doctor) and
the person doing the analysis are not aware about
the groups in. 75
75
 Advantages of RCT

• Exposure in under control.

• Due to randomization both intervention and
control groups have similar characteristics.
• By blinding the study, the observer and selection
bias can be eliminated.
• If properly designed & conducted, it can reduce
the confounding.
• Can confirm or refute etiological hypothesis.
• Can evaluate the efficacy / effectiveness /
efficiency of health services.
• Best method for studying causal
relationship.

76
 Disadvantages of RCT
• Ethical problems
Due to adverse effects
Due to benefits of intervention in the treated
group
Provision of Placebo

• Relatively expensive

77
 QUASI EXPERIMENTAL
STUDY

• In a Quasi Experimental Study, at least one

characteristic of a true experiment is missing, either
randomization or the use of a separate control group.

• A quasi experimental study, however, always

includes manipulation of an independent variable
that serves as the intervention.
 QUASI EXPERIMENTAL
STUDY
• One of the most common quasi experimental designs
uses two (or more) groups, one of which serves as a
control group in which no intervention takes place.

• Both groups are observed before as well as after the

intervention, to test if the intervention has made any
difference.

• The subjects in the two groups (study and control

groups) have not been randomly assigned.
 QUASI EXPERIMENTAL
STUDY
• Another type of design that is often chosen because it
is quite easy to set up uses only one group in which an
intervention is carried out.

• The situation is analyzed before and after the

intervention to test if there is any difference in the
observed problem. This is called a "Before- After"
study .
STUDY TYPES & STRENGTH OF
EVIDENCE
• Analytic Study involves the systematic of
evaluation suspected relationships, for
example, between an exposure and a
health outcome.

• Analytic studies typically provide

stronger evidence concerning particular
relationships.

• An experimental design is the only type of study

design that can actually prove causation.
 Reference
•
s
Principles of Epidemiology in Public Health Practice,
Third Edition An Introduction to Applied
Epidemiology and Biostatistics

• http://www.cdc.gov/

• Jhonhopkin university epidemiology lectures

 THANK YOU
for supporting me
Best of luck in your exam