This an overview of epidemiological study designs with strengths and weaknesses. It also underscores the fact that one does not necessarily need an experimental study to prove causality, as described by the Bradford-Hill criteria for causality. Further, it highlights the hierarchy of study designs
1. STUDY DESIGNS
Ephron Ephraim Yerembede, B.MLS, M.Sc.
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
1
2. • Describe the differences between observational and
experimental study designs
• List and describe the different sub-types of study
designs with strengths and weaknesses of each type of
study design
• Identify ideal study designs for particular situations
• List and explain the Bradford Hill Criteria for
causality
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
2
Learning Objectives
3. • A road map or blueprint and strategy of investigation
so conceived to find answers to research questions or
problems as objectively, validly, accurately, and
economically as possible
• Study designs direct how the investigation/research
is conducted
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
3
Definition
4. • Research questions
• Research goals
• Research philosophy and paradigms
• Ontology – what is reality?
• Epistemology – How can I know reality?
• Methodology – how do you go about finding out?
• Methods – what techniques do you use to find out?
• Researcher skills
• Time and funds
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
4
Factors to consider
5. Classification of Study Designs
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
Study Design
Observational
Descriptive
Case
report/Case
series
Cross-sectional
Surveys
Analytical
Case control Cohort
Experimental
RCCT RCFT
Ecological
5
6. • Observational: Investigator does not assign
exposure; they merely observe phenomena in their
natural settings
• Descriptive
• Analytical
• Experimental: This involves manipulation of the
exposure and randomization of study subjects to two
or more groups, usually treatment and control
• The main aim of various study designs:
- Descriptive studies: to generate/formulate
hypothesis.
- Analytical studies: to test hypothesis.
- Experimental studies: to prove hypothesis.
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
6
Principle
7. Observational Study
Investigator is passive; has no control over exposure
• Descriptive: used to describe the occurrence of disease or
health outcomes in a specific population or study group;
basically, 3 questions are asked when, where, and
who i.e., time, place, and person distribution.
- Case reports/case series
- Cross-sectional
- Ecological
• Analytical: describe the association between
exposure and outcome
- Case-control
- Cohort
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
7
8. • Describe in-depth the presenataion of a single case
• Generally, report a new or unique finding, e.g.,
- previously undescribed disease
- unexpected new therapeutic effect
- adverse events (drug reaction)
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
8
Case Reports
9. • Furnishes information about exposures, outcomes,
and other variables of interest among multiple
similar cases
• Experience of a group of patients with a similar
diagnosis
• Cases may be identified from a single or multiple
sources
• Generally, report on new/unique condition
• May be the only realistic design for bizarre
disorders/diseases
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
9
Case Series
10. Advantages
• Good for hypothesis generation
• Provides information for very rare diseases with few
established risk factors
• Characterizes averages for disorders
Disadvantages
• Cannot study cause-and-effect relationships
• Cannot assess disease frequency
Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
10
12. Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
• An observational study and the investigator has no
control over the exposure of interest (e.g., diet).
• It includes identifying a defined population at a
particular point in time, measuring a range of
variables on an individual basis, and at the same
time measuring the outcome of interest.
• A design that looks at exposure and disease status at
a single point in time (a cross-section of the
population)
“A Snapshot”
12
Cross-sectional Study
16. Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
• Conducted when individual-level data would be
difficult or impossible to collect
- e.g., effect of air pollution
- Legislation
• Useful for:
- Correlating population disease rates with factors of interest, such as
healthcare use
- Demonstrating changes in mortality over time (time series)
- Comparing the prevalence of a disease between different regions at a single
point in time (geographical studies)
• The unit of analysis is the population
16
Ecological study
17. Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
• Case-control studies are relatively simple and cheap
• They include cases (people with disease of interest)
and a suitable control group (people without that
disease or outcome variable).
• The occurrence of the possible exposure is
compared between cases and controls.
• Case-control studies are also called retrospective
studies since the investigator is looking backward
from the disease to a possible exposure
• Matching of cases with control is critical
• Effect measure: odds ratio
17
Case-control Study
23. Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712 Cohort Study
23
• A group of people with similar characteristics
proceeding together in time (birth, age, occupation,
etc)
• Attempts to study the relationship between a
purported exposure or risk factors and the
subsequent risk of developing health outcomes
• Conceptually longitudinal; the study groups so
defined are observed over a period to determine the
frequency of disease among them
• Begin with disease-free individuals
• Classify patients as exposed/unexposed
• Record outcomes in both groups Compare
outcomes using risk ratio or relative risk (RR)
29. Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712
Experimental Study
29
• Similar in approach to cohort studies, except that
the study conditions are under the direct control of
the investigator
• There is some deliberate action, intervention or
manipulation of the exposure
• Aim
- to provide for the scientific proof of causation
- To evaluate new forms of therapy or prevention methods
• Can be conducted on animal or humans
• Human experiment:
- Randomised controlled studies- clinical trials
- Field trial
- Community trial
- Non-randomised
30. Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712 Randomised Controlled Clinical Trials
30
• A planned experiment designed to test the efficacy
and/or effectiveness of new prophylactic,
diagnostic, and therapeutic agents, devices,
regimen, services, etc., applied on human subjects
• Basically, it involves comparing the outcomes in
two groups of individuals
1. Group 1: new treatment or intervention
2. Group 2: standard treatment or control
• Types of design include parallel design and cross-
over design
32. Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712 Randomised Controlled Clinical Trials
32
Randomization
• The heart and soul of the RCT
• Ensures that participants have an equal chance of
being assigned to one of two or more groups
• Participants are allocated into study (new
treatment or intervention) and control groups
(standard treatment or placebo) on random
basis.
• It eliminates bias and allows comparability.
• Randomization is best done by using table of
random numbers.
33. Global
Health
and
Infectious
Diseases
Control
Institute
–
NSUK
Public
Health
Research
Methods
–
GHI
712 Randomised Controlled Clinical Trials
33
Blinding
• Another quality improvement process in RCT
• Can be at the level of:
- Participants only (single blinding)
- Participants and investigator (double blinding)
- Partcipants, investigator, and analyst (triple blinding)
• Helps eliminate
- Co-intervention
- Biased outcome ascertainment
“Randomized, double-blind, controlled
trial” is considered as research design par
excellence and “Gold standard” amongst
research designs.