Study Design
• The design of a study is the
investigator’s plan of action for
answering the research question(s).
• The objective in selecting a study
design is to minimize possible errors
by maximizing the reliability and
validity of the data.
•? Why
were
•? What could be
warning
Hurricane Katrina
the consequences
ignored
•Why this
devastation
•? What is happening
happened
•? How many victims
Classification of Study Design
•
It can be classified into 3
subheadings
Cross-sectional
1. Descriptive
Longitudinal
Case-control
1. Analytical
3. Interventional
Cohort
Experimental
Quasiexperimental
What type of study will we do?
• It depends on
1.the research question
2. study objective
3. feasibility
If the Research Question is
• who is affected/
• What do they know, believe,thinks about
the problem
And the research objective is
• To recognize the problem and to set up
a hypothesis
We will go for a Descriptive type of
study.
Cont.....
• R.Q:
1. Are certain factors indeed associated
with the problem?
2. What is the cause of the problem?
•
so we want to demonstrate the cause and
effects
We will go for an Analytical type of study.
Cont......
R.Q:
• 1. Will the removal of a particular factor
prevent or reduce problem?
• 2.What is the effect of a particular
intervention/ strategy?
To test the effectiveness of control
measures in small scale project drug
trial.
We will go for an Experimental type of
study.
• Descriptive study:
1. Merely describe a problem in terms
of time,place and person.
2. It doesn’t have a formal comparison
group.
3. As there is no comparison group
there is no scope for analysis.
4. It can only done for establishing
hypothesis.
cont......
• 5. No conclusion can be drawn about
the association between exposure and
outcome.
Descriptive study has many
name,
• Cross-sectional
• Survey
• Prevalence study
Advantages:
• 1. Generalizability.
• 2. It takes short time.
• 3. low cost.
Limitations:
• Cause and effect relationship cannot
be measured.
• Some of the diseases of long duration
shows a high prevalence rate.
• Chance of missing of acute diseases
like asthma.
• Patient who are under treatment can
be missed.
• Longitudinal Studies
When observations are repeated on
the same population over a period of
time.
Exp: serveillance.
• Case- Studies
• Case control study
Exposure to
risk factor
Disease
Yes
sample with
disease/cases
No
TIME
Population at risk
Research
Yes
No
sample without
disease/controls
selection of cases and controls
Inclusion criteria for the mother were as
follows:
1.Woman who gave birth to live- birth singleton
infants.
2. Woman having normal vaginal delivery.
Exclusion criteria for the mother were as
follows:
1. Non antenatal care card holders.
2.Those who gave birth to premature babies.
3.Subjects who gave birth to still born babies.
Cont....
4. Pregnant women who had any medical
complication (e.g Diabetes Mellitus, Heart
Disease , Chronic Lung Disease,
Jaundice etc).
5. Eclamptic and pre-eclamptic subjects.
6. Multiple pregnancies.
7. Caesarian section cases.
8. Congenital abnormal babies.
9. Post mature babies.
•
•
•
•
•
•
•
•
Advantages:
Short duration
easier
less expensive
suitable to investigate rare diseases about
which little is known/ also for common
disease.
no risk of subjects
require comparatively few subjects
risk factor can be identified.
Ethical problems are minimal.
Disadvantages:
• problem of bias
• Selection of control group
• we cannot measure incidence rate
only odds ratio
• representative ness of cases and
control
Odds Ratio
• The odds that a case is exposed
divided by the controls is exposed.
Exposed
Not
Exposed
Cases
A
C
A+C
Odds ratio=
[{A/(A+C)}/ {C/ (A+C)}]/
[{B/(B+D)}/ {D/ (B+D)}]
Noncases
B
D
B+D
A+B
C+D
Cohort study
Exposure to
NSAIDS
Renal failure
Yes
Yes
Population
TIME
sample
No
T I ME
sample
Population at risk
Yes
No
No
Advantages:
• estimate risk or rate directly because
of the availability of population at risk.
• Temporal relationship
• Reduction of Bias
• rare exposure
• Multiple outcome
Disadvantages:
•
•
•
•
•
•
large sample size
losses to follow up
multiple exposure
ethical problems
cost
time
type of cohort study
• Prospective
• Retrospective
Exp:Records of exposure of members
of the armed services to radio-active
fallout at nuclear bomb testing sites
are now being used to examine the
possible causal role of fall-out in the
development of cancer over the past
30 years.
• Retro-prospective
Relative Risk
• RR= (Incidence rate among the exposed)/
(Incidence rate among the exposed)
· If RR = 2, There is twice chance of
occurrence of the disease among the exposed
then the unexposed.
· If RR = 1, There is equal chance of
occurrence of the disease among the exposed
then the unexposed.
· If RR = 1<, The exposure is rather
protective.
Intervention studies
• When researcher manipulates a
situation and measures the effect of
manipulations.
• types:
1. Experimental
2. Quasi-Experimental
Experimental
• only type of study design that can
actually prove causation.
• individuals are randomly allocated to
at least two groups.
• Manipulation is there.
• Randomization.
2. Quasi-Experimental
• all the three main characteristics
are not followed.
• Manipulation must be present.
quasi-experimental control group
Study group
before
Intervention
Study group
after
Compare
control group
before
Control group
before
quasi-experimental no control group
Study group
before
Intervention
Compare
Study group
after
• In drug test,
we measure
sensitivity and specificity of the test.
• sensitivity : The proportion of people with
the disease who have a positive test for
the disease.
Rarely miss people with the disease.
• specificity:The proportion of people
without the disease who have a negative
test for the disease.
Rarely misclassify people without the
disease as diseased.
Potential Error in Epidemiological
Studies
• Errors can be
1. Random Error
2. Systematic 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.
• There are 3 major sources of
Random Error
1.Individual Biological variation.
2. Sampling Error.
3. Measurement Error.
• Sampling error occurs during the
process of selecting study
participants who are always a
sample of a larger population and
best way to reduce it by increasing the
sample size.
• Matching
is 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
aetiological factor and which if not
matched act as a confounder.
• Systematic error(or bias) occurs in
Epidemiology when there is a tendency to
produce results that differ in a systematic
manner from the true value.
• Systematic error is a particular hazard
because epidemiologists usually have no
control over participants in studies unlike
the situation in laboratory experiments.
• It is difficult to have representative
samples of source population.
• Some variables of interest are difficult to
measure, like personality type, alcohol
consumption habits and past exposures
to rapidly changing environmental
conditions .
• The principal Systematic Error
• 1. selection Bias:occurs when there is a
systematic difference between the
characteristics of the people selected for a
study and the characteristics of those who
are not.
• 2. Measurement bias: occurs when the
individual measurements or classifications
of disease or exposure are inaccurate.
( they do not measure correctly what they
are supposed to measure).
Sources of Measurement bias
• Biochemical or physiological
measurements are never accurate and
different laboratories often produce
different results on the specimen.
• Recall Bias in case-control study.
• Validity
is an expression of the degree to
which a test is capable of
measuring what it is intended to
measure.
•a study is valid if its results
correspond to the truth,
• there should be no systematic
error and random error should be
as small as possible.
• Internal validity- is the degree to which
the results of an observation are correct
for the particular group of people being
studied.
• Internal validity can be threatened by all
sources of systematic error but can be
improved by good design and attention to
detail.
• External validity or generalizability is the
extent to which the results of a study apply
to people not in it.
• External validity is assisted by study
designs that examine clearly stated
hypothesis in well defined people.
Ehical Issue
Descriptive
c/s
Information taken
single time
Descriptive
Longitudinal
Information taken
several time
Descriptive
Longitudinal
cohort
Not cohort
Cohort
• 1. persons who are having high fat diet
are 2.5 times more prone to develope bowl
cancer than those of not having high fat
diet.
• 2. High fat diet in takers are 2.5 times
more at risk of developing bowl cancer
than the non high fat diet in takers.
Diet habit
Contains of
the food
+
_
cases
bowl
ca
cases
pop
+
_
control
_
High fat diet
+
_
Cohort
pop
Not high fat
diet
+
_
Bowl ca