By

Dr: Ibrahim Ali kabbash

Prof., of Public Health & Community Medicine
Faculty of Medicine – Tanta University
The type of study chosen depends
on:
◦ the type of problem;
◦ the knowledge already available
about the problem; and
◦ the resources available for the
study.
I. Descriptive
◦ Case Report
◦ Case Series
◦ Ecologic Studies
◦ Cross sectional surveys
A) Observational
 Cross-sectional or prevalence Studies
 Retrospective (case control, Case
referent)
 Prospective
 Cohort or concurrent prospective.
 Historical or retrospective or non-
concurrent prospective.
B) Experimental
 Clinical Trials
◦ Therapeutic trials
◦ Intervention trials
◦ Prevention trials

 Community Trials
◦ (controlled interventions)
◦ (comparative experimental)
1. It gives snapshot of the problems among
studied population
2. It measures both exposure and outcome at the
same point of time
3. Needed usually to get better understanding of
the problem
4. Provide baseline data for policymakers.
 Can be converted to cohort study if the subjects
included in the study were followed up.
 Can be converted to case control study (nested
case control)
 Used to evaluate the presence of diseases
(prevalence) or other health related events.
Cross-Sectional Study
Ask: Do characteristics of the
exposure factor coexist with the
health problem?

Prevalence data; no risk statement

Advantages:
 Gives general description or scope of
problem
 Useful in health service evaluation
 Baseline for prospective study
 Identify cases and controls for
retrospective study
 Get data all at once, so inexpensive
Disadvantages:
 No calculation of risk
 Temporal sequence unclear
 Not good for rare disease
 Selective survival can bias
 Selective recall can bias
An observational study in which :
 diseased and un-diseased subjects
are identified
 then compared regarding specific
characteristics to determine possible
association or risk for the disease .
 The case-control design is uniquely well
suited to diseases with long induction
period;
 Uniquely suited to the study of rare
diseases.
 It shares the same logical framework of
inference as the prospective study.
Objectives:
 To provide valid, reasonably precise, estimate
of the strength of at least one hypothesised
cause-effect relationship.
 To evaluate several hypotheses - several
different etiologic factors both as
independent and interacting causes for a
given disease.
Exposed Non- Exposed Non-
Exposed Exposed
a c b d

No Disease
Disease
CONTROLS
CASES
Strengths:
 Can simultaneously evaluate several causal
hypotheses.

 Permits evaluation of interaction (extent to

which two or more factors modify the
strength of one another).

 Permits the evaluation and control of

confounding.

 If a population-based series of incident

cases have been assembled, it is possible to
estimate incidence rates.
Strengths (cont’d)
 Is efficient in time and cost.

 Can be used to study rare diseases.

 Can be used to evaluate the causal

significance of a rare exposure. (This is
especially true if a factor is rare, but
accounts for a high proportion of the
disease.)
 It is individual-based as is the prospective
study and allows estimates of relative risk.
 Under favourable circumstances, results
from case-control studies are consistent
with those from prospective studies.
Limitations of the Case-Control
Strategy:

 Is highly susceptible to selection bias which

creates non-comparability between cases
and controls.

 Can not estimate the true risk, relative risk.

 Recall bias.
Prospective or Cohort Study
In a cohort study, the investigator selects a
group of exposed individuals and a group of
non-exposed individuals, and follows up both
the groups to compare the incidence of
disease (or rate of death from the disease) in
the two groups.
 Cohort Studies

EXPOSED NON-
GROUP EXPOSED
GROUP

Develop Do Not Develop Do Not

Disease Develop Disease Develop
Disease Disease
a b c d
Strengths
 Provides a logical basis for inferring "cause”
because it proceeds from cause to effect.
 Allows direct assessment of risk (incidence
rates) for various groups and provides
information as to the level of the rate of
disease (or health problem) in the study
groups.
 Allows direct derivation of risk estimates,
relative risk and excess risk.
 Allows to obtain information on change in
exposure over time and its effect on
outcomes.
 Avoids certain biases such as selective survival.
Strengths (cont’d)
 Allows study of multiple potential effects
of a given exposure and can provide
information on benefits as well as risks.

 Allows framework for quality control in the

measurement of study variables.

 Allows measurement of physiological

status before the appearance of disease,
but care must be taken that pre-clinical
disease does not influence the measure.
Limitations:
 Prospective studies may suffer
from bias as do other designs,
but type, degree and means of
assessing bias varies among the
studies. Loss to follow-up.

 "Exposures" do change--difficult
to take this change into account.
Limitations:
 Prospectivestudies imply commitment
over many years for participants and for
researchers - Loss to follow-up

 It
may be difficult to obtain estimates of
population attributable risk if the study is
based purposely on groups with a much
higher prevalence of the relevant exposure
than the general population.
Definition:
An intervention study is a research design in
which the investigator manipulates a
factor(s) and measures the subsequent
outcome.
Elements of a “complete” experiment:
 Manipulation of independent variable:
 Use of a control group
 Ability to randomize subjects to treatment
groups.
Experimental Studies
 Prospective direction
 Ability to randomize subjects
 Appropriate temporal sequence of
cause and effect
 Ability to control extraneous variables
 “Best” evidence of causality
 Expensive in time, personnel, facilities
and cost.
 Ethical constraints
 Contrived situations
 Impossible to control human behavior
 External validity still uncertain
Possible experimental outcomes:
 Symptoms
 Laboratory test results
 Morbidity
 Mortality
Quasi experimental 
studies?!
THANK YOU