By: Ismael.

A (PHD fellow in
Epidemiology)
Session Learning Outcomes

1. Define research designs

2. Describe types quantitative research design.
3. Identify advantages and limitation of each design
4. Discuss different types of experimental and nonexperimental
research designs.
5. Define best level of evidence in each design.
6. Describe the characteristics of quantitative research design.
 Study designs…
 A study design is the process that guides researchers on
how to;
o Collect
o analyze and
o interpret observations.
 It is a logical model that guides the investigator in the
various stages of the research
o Its blueprint for the collection, measurement, and analysis of data
5
6
 Observational

 Researcher just observes and analyses researchable objects

or situations but does not intervene.
 Quantitative study design…
Descriptive studies:
 Describe the patterns of disease occurrence and other
health-related conditions by person, place and time.

 Personal variables: age, sex marital status or occupation,

as well as the consumption of various types of food or
medication use.

 Place : geographic distribution of disease, variation

among countries or within countries(urban and rural
areas).

 Time: examine seasonal patterns in disease onset, etc.

Ecological design

 Also known as correlational study

 Population level measurement is made for

health outcomes in relation to some factors
of interest
– Exposure, and or health outcome data are
collected at a group level.
– Not an individual level

• Generally ecologic studies use a group level

of measurement
9
Example:
Study air pollution and its health outcome.

• Average air pollution level would be

measured at city level and death from
chronic lung disease in the same city

10
Correlational Study

11
Advantage
– Ecologic studies have lower cost and are convenient

• Data can be publicly available(registry, Census)

– Ecologic studies are useful for evaluating the impact of

community level interventions

• Fluoridation of water
– Ecologic study is that this study design can maximize

exposure differences
• where minimal within community differences render
individual risk studies impractical
– Generate hypothesis 13
 Case report and case series
 Case Report/case study

– Its a careful, detailed report of the profile of a single

patient

– Document unusual medical occurrence

– Can represent the first clues in the identification

of new diseases.

14
 Case report and case series…
Case series
 The individual case report can be expanded to a case
series, which describes characteristics of a number of
patients with a given disease.
– A series of cases - treated or untreated
– Describing signs, symptoms or Patho-physiological
parameters in the series of patients
– Can lead to generating hypothesis.
Case report and case series…
Cross-sectional study design
 In this study, a sample of persons from a

population is enrolled and their exposures and

health outcomes are measured

simultaneously.

17
Cross-sectional Study….

18
Longitudinal Study

19
20
Advantage
– Multiple outcomes and exposures can be
studied

– important in public health for assessing the

burden of disease in a specified population
and in planning and allocating health
resources

– Good for descriptive analyses and for

generating
hypotheses

– Inexpensive and less time consuming 21

Limitation
– “Chicken or egg dilemma” or temporal ambiguity or
antecedent-consequent bias effects
• Since exposure and outcome are measured
simultaneously it is not possible to determine causal
relationship
– Length biased sampling
• Disease with long duration are oversampled while disease
with short duration are less represented
• Show association of survival not risk of development
– Not suitable for rare disease or disease with short
duration
22
 Note

 For factors that remain unaltered over time,

such as sex, blood group, etc., the cross-
sectional survey can provide evidence of a
valid statistical association.
 In general, measure prevalence rather than
incidence
 Case-control study
Hierarchy of observational analytical designs
validity Highest

Cohort

Case-control

Cross-sectional

Lowest

25
 Case-control
Group of subjects with the disease- cases

group of subjects without the disease- controls.

Examine the association between disease and potential risk

factors
o Concerned with identifying potential risk factors of a disease

Information about previous exposures are obtained for cases

and controls, and frequency of exposure compared for the two
groups.

26
 Case-control study…
Epidemiological studies are carried to answer
different question

Cohort Study

Case-control
 Case-control study…

Past Present

28
 Case-control study
The first most important step is selection of cases and controls

Precise diagnostic criteria and definition of cases

 Clinical, laboratory, or other criteria

 It is important to produce as much homogeneous disease

entity as possible.

 Once clear diagnostic criteria and definition has been

established, cases will be selected from the sources.

30
 Case-control study…
Selection of controls
 Controls should be
 Population at risk and representatives of the source
population
 Comparable to the cases
 Sources of controls
 General populations
 Hospital
 Special (i.e. friends, neighbors, relatives)
 Proxy control 31
 Case-control study…

Ratio of control to cases

 When larger cases are available, the cost and difficulty of

finding the controls and cases are similar and controls
represent study base.

 Maximum statistical efficiency can be obtained with 1:1

ratio

 If the above conditions are not meet

o Up to a maximum of 4:1 ration can be considered
o Beyond 4:1 ratio results negligeable statistical power

32
 Case-control study…
Information on exposure status

 Exposure should be clearly defined -defined when to say

exposed.

 An appropriate time window should be considered.

 Consider history, duration and intensity of exposure

 Information about exposure
 Study subjects, surrogate or records

 The manner used to obtain information must be similar for

cases and controls for comparable accuracy.
33
 Case-control study…

Unmatched case-control vs. Matched case-

control

E.g. if the potential confounder is sex

Selecting male control of each male cases and

female control for each female cases

34
 Cohort study
 Is a follow up study
 In this study
– Groups of individuals are selected on the basis of
exposure to suspected risk factors.

– The groups must bee free of the disease.

– Researchers Followed POPN at-risk over a period

to assess the occurrence of the disease

– Determine the subsequent risk or rate of disease

or health outcome.
 Cohort study…

Study Study End

start

37
Closed populations
Cohort study…
• A closed population adds no new members over time
and loses members only to death/risk free

• They begin with a certain number of individuals and

shrink over time
Open population Cohort study …
– May gain members over time
• Through immigration or birth
– May lose members over time
• Through emigration or death
– An open population that is in a steady state is said to be
stationary
– Over time, open populations may grow, remain the
same size, or shrink.
Types of cohort study
 Cohort study…
• Classified as retrospective and prospective
– Temporal relationship between study start and
disease occurrence
• Both starts with exposed and non exposed
• The difference is
– In retrospective
• Both the outcome and the exposure has already
occurred when the study is initiated
– In prospective
• The exposure may or may not occurred but the
outcome is not certainly occurred

41
prospective Cohort Study

42
Retrospective Cohort Study

43
 Selection of
• controls
The major principle
– The non-exposed group should be as similar as
exposed group related to the disease except the
determinants.
Source of data
• Sources could be
– Study subjects
– Surrogates
– Environmental study
– Medical records etc.
 To obtain adequate information it is a wise decision to use
multiple sources 44
 Approaches to Follow-
• up
In cohort study we trace all participants to determine
weather they develop the outcome
– A failure to do this is a significant source
of bias
Three important concepts are
– Event
– Censored
– Lost to follow up
• Loss to follow up
– Affect the validity and limit the power of the study
– Serious if
• Related to exposure, outcome or both
• If the proportion is larger in one group 45
 Limitati
• Timeon
consuming

• Expensive

• Prone to bias associated with loss to follow

up and ascertainment of exposure

46
Summary
The End!!

48