Experimental Study Design

1
 Experimental Studies
• Investigator can “control” the exposure
• akin to laboratory experiments except living populations are
the subjects
• Generally involves random assignment to groups
• Clinical trials are the most well known experimental design
• The ultimate step in testing causal hypotheses

2
 Experimental Studies
• In an experiment, we are interested in the
consequences of some treatment on some
outcome.
• The subjects in the study who actually receive
the treatment of interest are called the
treatment group.
• The subjects in the study who receive no
treatment or a different treatment are called
the comparison group.

3
 Experimental Studies

• Randomized Controlled Trials (RCTs)

– a design with subjects randomly assigned to
“treatment” and “comparison” groups

– provides most convincing evidence of

relationship between exposure and effect

– not possible to use RCTs to test effects of

exposures that are expected to be harmful, for
ethical reasons 4
 RANDOMIZATION outcome
Intervention
Experimental Design
no outcome

Study
population
outcome
Control

no outcome

baseline
future

time
Study begins here (baseline point) 5
 Experimental Studies

• Randomized Controlled Trials (RCTs)

– THE “GOLD STANDARD” OF RESEARCH DESIGNS

– PROVIDES MOST CONVINCING EVIDENCE OF

RELATIONSHIP BETWEEN EXPOSURE AND EFFECT

• trials of hormone replacement therapy in

menopausal women found no protection for heart
disease, contradicting findings of prior
observational studies
6
 Experimental Intervention Studies

THE RESEARCHER ATTEMPTS TO CHANGE

A DISEASE DETERMINANT, SUCH AS EXPOSURE,

BEHAVIOR OR THROUGH TREATMENT IN ONE

OR MORE GROUPS OF PEOPLE.

7
Aims
• It provide scientific evidence about the
etiological or risk factors
responsible for the condition.
• It also measure the effectiveness and efficiency of
the health services.

8
Types Experimental Study Design

1. Randomized control trials/Clinical trials

1. Parallel design
2. Cross over design
2. Field trials
3. Community trials or Quasi study design

9
 Design of a randomized controlled trial
Study
population

Selection by
defined criteria
Potential Non participants
participants (do not meet
selection criteria

Invitation to
participate

Non participants
participants

Randomization

CONTROL
Treatment 10
Randomized control trial

Methodology
1. Formulation of hypothesis & specification of
Objective
2. Selection of Study population from reference/target
population
3. Establish an Inclusion or Exclusion Criteria.
4. Informed Consent
5. Ethical Considerations
11
Randomized control trial (contd.)

6. Allocation to treatment regimen or Intervention

1. Randomization (the determination of assignment to treatment
group is based on probability alone and is not influenced by the
preference of researcher and patient)
2. Blinding (Blinding means the treatment assignment is not known
to certain person.
Table of single, double and triple blind may be put here:
Blinding Patient Researcher Data Likely Bias
Analyst
Single Blinded Not Blinded Not blinded Assessment bias
Double Blinded Blinded Not blinded Assessment bias
Triple Blinded Blinded Blinded No bias
The researcher ensures that the medications are disguised and labeled in such a
manner that in a triple blind study the patient, the researcher and
also the data analyst do not know, which drug is given to which group. 12
7. Experimental Component of the study
8. Measure the outcome
9. Analysis / Interpretation

13
Designs Used in Experimental Studies

• Parallel Design
• Cross Over Design

14
a
Exposed to
Random Specific R Observation
Assignment

Patients Compare
outcome
Unexposed to
Specific R

Time

Observation
Exposed to
Random Compare
Specific R
Assignment Outcome
Exposed and
Patients Unexposed to
R
Unexposed to
Specific R

Time 15
Advantages of Experimental Studies
• Exposure is under the control of investigator
• Randomization
• Blinding eliminates bias
• Control on time span
• Confounding factors can be controlled
• Best method to study causal relationship
• We can confirm or refute etiological hypothesis on evidence.
• Evaluate effectiveness and efficiency of Health services

16
Disadvantages of Experimental Studies

• Subject exclusion may limit ability to generalize findings to other patients.

• A long period of time is often required to reach a conclusion.
• A large number of participants may be required.
• Financial costs are typically high.
• Ethical concerns may arise.
• Subjects may not comply with treatment assignments.
• Exposure or treatment alternatives should be acceptable to both groups

17
 Example
Randomized controlled trial of early hospital discharge after myocardial infarction.

Myocardial patients
(507)

Uncomplicated Complicated, excluded

(179) (328)

Randomized Not included in study

(80) (99) Q#1 What is the null hypothesis?
Q#2 What is the alternate hypothesis?
Early discharge Late discharge Q#3 How many subjects entered the study?
(40) (40) Q#4 What were the results?
Q#5 Support or refute the null hypothesis?
Outcome:
0 0 Deaths
6 10 Hospital readmission
0 5 Re-infarctions
3 8 Patients with angina 18
Field trials and community trials

Field trials (Healthy individuals) and community

trials (community based not individuals) are

other experimental designs, in which the

participants are, respectively, healthy people

and communities.
19
 Randomized Controlled Trials

• Disadvantages
• Very expensive

• Not appropriate to answer certain types of questions

• it may be unethical, for example, to assign persons to

certain treatment or comparison groups

20
 NON-RANDOMIZED TRIALS: -
•When RCT is not possible on ethical,

administrative grounds.

·When preventive measures can be applied on

community basis.

·When disease frequency is low and natural

history is long.

·When cost and logistic is limited.

 ASSOCIATION AND CAUSATION

“Association may be defined as the concurrence of two variables more often than would be
expected by chance”

Association can be broadly grouped under three headings:

1. Spurious Association

2. Indirect Association

3. Direct (Causal) Association

a. One to one Causal Association

b.07/29/2024
Multifactorial Causation 22
 CAUSATION & ASSOCIATION
1. Spurious Association
Sometimes an observed association between a disease and
suspected factor may not be real.

2. Indirect Association

The indirect association is a statistical association between a

characteristic (or variable) of interest and a disease due to
the presence of another factor known or unknown that is
common to both the characteristic and the disease.

A
Altitude

C
Iodine Deficiency

B
07/29/2024 23
Endemic Goitre
 CAUSATION & ASSOCIATION

3. Direct (Causal) Association

a). One to one Causal relationship

Two variables are said to be causally related. If a change in A is followed by a change in
B. This model suggest that when the factor A is present the disease B must result.

A B

b). Multifactorial Causation

Factor 1

Factor 2 Reaction at
Cellular level Disease
07/29/2024
Factor 3 24
 CAUSATION

• Ideally, the proof of causation is through experiments

• Henle-Koch Postulates for Causation:The putative

etiologic agent can be:

– isolated from the affected animal in pure culture

– propagated in artificial culture medium
– produce the same disease in a second animal
– isolated from the second animal

07/29/2024 25
 AUSTIN BRADFORD HILL CRITERIA FOR CAUSATION
– Consistency
– Strength of association
– Specificity
– Dose response
– Temporal relationship
– Biological plausibility
– Coherence
– Experiment

07/29/2024 26
 CAUSATION & ASSOCIATION

CRITERA FOR JUDGING CAUSALITY

1. TEMPORAL ASSOCIATION
Does the suspected cause precede the observed effects (Drinking contaminated water and
diarrhea).

2. STRENGTH OF ASSOCIATION
The strength of association is based on answers to two questions
Relative risk -------------------- is it large?
Is there a dose response- duration response relationship?
The larger the relative risk the greater the likelihood of a causal association.

3. SPECIFICITY OF ASSOCIATION

The concept of specificity implies a one to one relationship between the cause and effect. The
reasons are
a. A single cause or factor can give rise to more than one disease.
b. Most diseases are due to multiple factors with no possibility of demonstrating one-to-
07/29/2024 one relationship. 27
 CAUSATION & ASSOCIATION

4.CONSISTENCY OF ASSOCIATION

The association is consistent if the results are replicated when studied in different settings and
by different methods.

5. BIOLOGICAL PLAUSIBILITY
The association agrees with current understanding of the response of cells, tissues, organs
and systems to stimuli.

6. COHERENCE OF THE ASSOCIATION

Coherence with the known facts that are thought to be relevant e.g. rising consumption of
tobacco and the rising incidence of lung cancer.

07/29/2024 28
 Review Questions

• Describe the link between exposure and disease

• Describe study design sequence
• Describe strengths and weaknesses of each design

29