3. BACKGROUND
Both in public health and in clinical practice, the
main objective is to modify the natural history of
disease so as to prevent or delay death or disability
and to improve the health of patient or the
population.
The challenge is to select the best available
preventive or therapeutic measure to achieve this
goal.
To do so, research is carried out that determine the
value of these measures.
4. BACKGROUND
The first instance of random allocation of patients to experimental
and control conditions is attributed to James Lind, a naval surgeon,
in 1747.
Lind randomly assigned 12 sailors to 6 different treatments for
scurvy. The two patients who were given lemons and oranges
recovered most quickly, suggesting a beneficial effect of citrus.
5. BACKGROUND
Randomization was Contributed by
statistician R.A. Fisher in agriculture in
1923.
Randomized plots of crops to receive
different treatments
6. Background
The first RCT in medicine is credited to Sir A. Bradford Hill, an
epidemiologist for England's Medical Research Council.
The trial, published in the British Medical Journal in 1948, tested
whether streptomycin is effective in treating tuberculosis.
Because the drug was in short supply, he simply alternated the
assignment of hospital admissions to drug versus control .Later
he recognized it led to selection bias because the sequence was
too easy to predict. That realization led to the use of a random
numbers table to generate the numeric series by which patients
would be assigned to conditions.
7. Experimental/ Intervention studies
Involve an active attempt
Types
With patients as unit of
study
CLINICAL TRIALS
With healthy people as unit
of study
FIELD TRIALS/
COMMUNITY
INTERVENTION
STUDIES
With communities as unit of
study
COMMUNITY
TRIALS
9. RANDOMIZED CONTROLLED TRIAL
In a randomized controlled trial,
Participants are assigned to treatment conditions at random
(i.e., they have an equal probability of being assigned to any
group).
Procedures are controlled to ensure that all participants in all
study groups are treated the same except for the factor that is
unique to their group. The unique factor is the type of
intervention they receive.
10. Goal of RCT
Primary goal
to test whether an
intervention works by
comparing it to a control
condition, usually either no
intervention or an
alternative intervention.
Secondary goals may include:
identify factors that influence
the effects of the intervention
(i.e., moderators)
understand the processes
through which an intervention
influences change (i.e.,
mediators or change
mechanisms that bring about
the intervention effect)
14. Cross-over type of study design
Will patient do better on drug A or
drug B?
Removes “patient effect” reducing
variability and increasing
precision of estimation
Assumption of ‘no carryover
effects’ is difficult to test
Needs to determine appropriate
length of washout period.
“Period” effects : Progression of
disease, Dropouts
15. The acceptability of the female and male condom: a
randomized crossover trial.
Kulczycki, A, Kim, D-jin, Duerr, A, Jamieson, DJ, Macaluso,M
Perspect Sex Reprod Health 2004 May-Jun, vol. 36(3)
A sample of 108 women in stable relationships recruited
from an urban, reproductive health clinic were randomly
assigned to use 10 male or female condoms, followed by
use of 10 of the other type to analyze measures of the
methods' relative acceptability.
16. A Walnut Diet Improves Endothelial Function in
Hypercholesterolemic Subjects : A Randomized Crossover
Trial
Emilio Ros, Isabel Núñez, Ana Pérez-Heras, Merce Serra, Rosa Gilabert,
Elena Casals and Ramón Deulofeu.
Circulation 2004, 109:1609-1614
Randomized in a crossover design 21 hypercholesterolemic men and
women to a cholesterol-lowering Mediterranean diet and a diet of
similar energy and fat content in which walnuts replaced 32% of the
energy from monounsaturated fat.
Twelve participants followed the control diet first for 4 weeks and
then switched to the walnut diet for 4 weeks; 9 subjects followed the
same diets in reverse order.
Because diet-induced lipoprotein changes stabilize in <4 weeks, they
did not incorporate a washout period between diets.
17. Factorial design
Evaluates multiple factors
simultaneously
Major concern: interaction of
interventions
Patients must be willing and
able to take any of the treatment
combinations
may be hard
to determine
18. A Randomized Longitudinal Factorial Design to Assess
Malaria Vector Control and Disease Management Interventions
in Rural Tanzania
Randall A. Kramer et al
Int J Environ Res Public Health. 2014; 11(5): 5317–5332.
19. Cluster design
Groups or clusters randomly
assigned, not individuals. E.g.
: villages, classrooms ,
platoons
20. Home-based versus Mobile clinic HIV testing & counseling in rural
Lesotho: A Cluster-Randomized Trial
Labhardt ND et al
PLoS Med. 2014 Dec 16;11(12):e1001768.
21. Types of RCT
Clinical trials
Preventive trials
Risk factor trials
Cessation experiments
Trial of etiological agents
Evaluation of health services
22. Basic steps in conducting RCT
1. Drawing up a protocol
2. Selecting reference and experimental populations
3. Allocation of study subjects: Randomization
4. Intervention / manipulation
5. Follow –Up
6. Assessment of Outcome
23. Drawing up protocol
One of the essential feature of RCT is that it is conducted under
a strict Protocol.
Once protocol is evolved, it should be strictly adhered to
throughout the study.
Prevents bias and reduces source of errors in the study.
Preliminary or pilot test runs of protocol can be held so to see
whether it contains any flaw.
Final version of protocol should be agreed upon by all concerned
before the trial begins.
24. Selecting reference and study population
I. Reference /Target
Population:
It is the population to which
findings of the trial ,if found
successful , are applicable.
e.g. whole population,
population of school children,
population of a city, industrial
workers or social groups
II. Experimental /Study
Population:
Derived from reference
population.
Ideally, it should be chosen
randomly from reference
population so that it is
representation of reference
population.
Otherwise it may not be
possible to generalize the
findings of the study to
reference population.
25. Selection of study subjects
Inclusion & exclusion criteria : for determining who will or will
not be included in the study must be spelled out with great
precision, and in writing.
To ensure the replicability by others, just as is the case with
laboratory experiments
26. Allocation of study subjects to groups
To derive a causal inference regarding relationship of
intervention and outcome, comparison is important.
Types of controls :
Historical
Simultaneous non-randomized controls
Randomized controls
27. Historical controls
Comparison group from past.
We go back to records of patients who were treated before new
treatment became available.
Simple
Demerit:
Comparability can not be assured
Recall bias
Quality : Data for medical purpose not for research purpose
Changes in many factors over calendar time
28. Simultaneous non-random controls
Day of month of admission-odd/even days
Alternate assignment into study & control group
Predictable by investigator- Selection bias
No. TB deaths
No. %
Vaccinated 445 3 0.67
controls 545 18 3.30
No. TB deaths
No. %
Vaccinated 556 8 1.44
controls 528 8 1.52
Results of a trial of BCG vaccination :
Am Rev Tuberculosis 53:517-532,1946
29. Randomized controls
Randomization - a statistical procedure
by which the participants are allocated to
“Study” and “Control” groups.
The critical element of randomization is
the unpredictability of the next
assignment.
It makes RCT the GOLD STANDARD
design for performing clinical trials.
30. “RANDOMIZED, DOUBLE-BLIND,
CONTROLLED TRIAL” is considered as
research design par excellence and
“GOLD STANDARD” amongst research
designs with which results of other
studies are often compared. Deviation
from this standard has potential
drawbacks
31. Sequence Generation
Flipping a coin?
Rolling dice?
Shuffling cards?
Table of random numbers
Computer random number
generators
• Random
• Reproducible
Preferable & recommended
• Random but tempt
investigators toward non-
randomness
• Adequate methods but not
optimal
• Cannot be checked – no audit
trail
Not recommended
32. Allocation procedures
the probability of being
assigned to any
intervention stays
constant over the course
of the trial
the allocation probability
changes in response to
the balance, composition,
or outcomes of the groups
controversial because they
allocate patients not purely
at random
Aim: increase the sample's
probability of being
assigned to the best
treatment
FIXED ALLOCATION
PROCEDURES
ADAPTIVE PROCEDURES
34. Simple (Complete) Randomization
Elementary form of randomization, in which, every
time when there is an eligible participant, the
investigator flips a coin to determine whether the
participant goes into the intervention or control group.
A limitation is that random assignment is truly
random.
A random process can result in the study winding
up with different numbers of subjects in each group.
This is more likely to happen if sample size is small
36. Suppose we want to compare 2 treatments(A & B)
Random Number Table can be used in a no. of ways:
a) we will consider every odd number an assignment to
A and every even number an assignment to B
b) we could say that digits 0 to 4 would be treatment A,
and digits 5 to 9 treatment B.
c) If we are studying three groups, we could say that
digits 1 to 3 are treatment A, digits 4 to 6 treatment
B, digits 7 to 9 treatment C, and digit 0 would be
ignored.
d) Prepare a series of opaque envelopes that are
numbered sequentially on the outside: 1, 2, 3, 4, 5,
and so on.
37. Restricted Randomization
Random assignment to achieve balance between
study groups in size or baseline characteristics.
Restricted randomizations guarantee balance
1. Permuted-block
2. Biased coin (Efron)
3. Urn design (LJ Wei)
542-04-#37
38. Blocked Randomization
Blocked randomization reduces the risk that different
numbers of people will be assigned to the treatment
(T) and control (C) groups.
Patients are randomized by blocks.
The order is chosen randomly at the beginning of the
block.
In randomly permuted blocks, there are several
block sizes (e.g., 4, 6, and 8), and the block size and
specific order are chosen randomly at the beginning
of each block.
39. Permuted-Block Randomization
Example
0 Block size 2m = 4
2 Trts A,B } 4C2 = 6 possible
0 Write down all possible assignments
0 For each block, randomly choose one of the six
possible arrangements
0 {AABB, ABAB, BAAB, BABA, BBAA, ABBA}
ABAB BABA ......
Pts 1 2 3 4 5 6 7 8 9 10 11 12
40. Blocked randomization
Advantage
A balance in the number
of cases assigned to T
versus C at any point in
the trial (which could be
valuable if the trial needs
to be stopped early).
Disadvantage
With fixed blocks,
predictability of the group
assignment of patients being
randomized late in the block
by research staff .
Reduced by using method of
randomly permuted blocks
and blinding of research
staff to the randomization
process
41. Stratified randomization
To ensure that the treatment and control groups are
balanced on important prognostic factors that can
influence the study outcome (e.g., gender, ethnicity,
age, socioeconomic status).
Before doing the trial, the investigator decides which
strata are important and how many stratification
variables can be considered given the proposed
sample size.
A separate simple or blocked randomization schedule
is developed for each stratum.
Large trials often use randomly permuted blocks
within stratification groups.
43. Minimization Method
o Minimization corrects (minimizes) imbalances that
arise over the course of the study in the numbers of
people allocated to the treatment and control.
o An attempt to resolve the problem of empty strata
when trying to balance on many factors with a small
number of subjects
o Balances Trt assignment simultaneously over many
strata.
o Used when the number of strata is large relative to
sample size as stratified randomization would yield
sparse strata.
o Logistically more complicated
44. Biased Coin Randomization
o In this procedure, if the imbalance in treatment assignments
passes some threshold, the allocation is changed from
chance to a bias in favor of the under-represented group.
o For example... If after 10 randomizations, there are 7 patients
assigned to intervention and 3 assigned to control, the coin
toss will become biased.
o Then, rather than having 50/50 chance of being assigned to
either condition, the next patient will be given a 2/3 chance of
being assigned to the under-represented condition and a 1/3
chance of being assigned to the overrepresented one.
o This procedure requires keeping track of imbalances
throughout the trial. In smaller trials, imbalances can still
result
45. Urn Randomization
This procedure tries to correct imbalances after each
allocation.
For example...
The investigator starts off with an urn containing a red
ball and a blue ball to represent each condition.
If the first draw pulls the red ball, then the red ball is
replaced together with a blue ball, increasing the odds
that blue will be chosen on the next draw. This
continues, replacing the chosen ball and one of the
opposite color on each draw.
The procedure works best at preventing imbalance
when final sample size will be small.
46. Timing of randomization
Actual randomization should be delayed until just
prior to initiation of therapy after consenting.
This prevents randomizing participants who drop out
before participating in any of the study.
This is important because everyone who gets
randomized needs to be included in the study's
analysis.
47. Operational mechanics of randomization
1. Sequenced sealed envelopes (prone to tampering!)
2. Sequenced bottles/ packets
3. Phone call to central location
- Live response
- Voice Response System
4. One site PC system
5. Web based
Best plans can easily be messed up in the implementation
48. Multi-institutional Trials
Often in multi-institutional trials, there is a marked
institution effect on outcome measures.
In multi-site trials, randomization usually occurs at a
centralized location.
Using permuted blocks within strata, adding institution as
yet another stratification factor will probably lead to
sparse cells (and potentially more cells than patients!)
Use permuted block randomization balanced within
institutions
Or use the minimization method, using institution as a
stratification factor
49. Allocation Concealment
Allocation concealment means that the
person who generates the random
assignment remains blind to what condition
the person will enter.
Preferably, randomization should be
completed by someone who has no other
responsibilities in the study. Often, the
study statistician assumes responsibility for
performing the randomization.
If allocation is not concealed, research staff
is prone to assign "better" patients to
intervention rather than control, which can
bias the treatment effect upward by 20-
30%
50. Follow-up
An adverse event (AE) is an undesirable health
occurrence that occurs during the trial and that may or
may not have a causal relationship to the treatment.
A serious adverse event (SAE) is defined as something
life-threatening, requiring or prolonging hospitalization
and/or creating significant disability.
E.g. A suicide attempt -- SAE in a study of any treatment.
The SAE needs to be reported regardless of whether it
bears any relationship to the treatment or the problem
being studied
Depending on the severity and frequency of adverse
events, investigators and data safety monitors may have
to decide to terminate the trial prematurely.
51. Attrition
Rate of loss of participants from the study that differs
between the intervention and control groups.
Can compromise study findings by reducing the
power of a study
52.
53. Data Analysis and Results
Which participants will be analyzed?
Intention to treat
Per protocol analysis
Subgroup Analyses
Statistical Power of study
Data Analytic Techniques
Continuous Outcome Variables
Categorical Outcome Variables
54. Intention to treat (ITT) analysis
Basic Principle - “Analyze What is Randomized”
All participants who were randomized and entered the
trial need to be included in the analysis in the
condition to which they were assigned, regardless of
whether they completed the trial, or may even have
switched over to receive the incorrect treatment.
The ITT analysis addresses the question of whether
the study treatment, if made available to the
population, would be superior to an alternative
intervention. For that reason, the disposition of the
sample from the moment they learn their allocation is
relevant in evaluating the treatment.
55.
56. Per protocol analysis
Opposite end of the spectrum from ITT analysis
Includes in the analysis only those cases who
completed treatment.
Its results represent the best case treatment results
that could be achieved if the study sample were
retained and remained compliant with treatment.
Should not be used alone/main analysis
57. Subgroup Analyses
Planned subgroup analyses. In a few instances, a study
may have been designed and powered to test whether a
treatment works better for one demographic group (e.g.,
females) than another (e.g., males). In that case, testing a
hypothesized treatment by demographic group interaction
would be a primary aim that definitely needs be tested.
Exploratory subgroup analyses. More often, many
different treatment-by-subgroup interactions will be explored.
Those analyses can support hypothesis generation. They are
done in the context of discovery rather than confirmation. Any
findings require replication in another trial.
58. Statistical power
Power is ability to find a difference when a real
difference exists.
The power of a study is determined by three factors:
Alpha level
Sample size
Effect size:
Association between DV and IV
Separation of Means relative to error variance.
59. Power and effect size
As the separation of means
increases, the power of study
increases.
As the variability about a mean
decreases power also
increases
60. Effect size
Effect size refers to the magnitude (i.e., size) of a
difference when it is expressed on a standardized
scale.
An effect size is exactly equivalent to a 'Z-score' of a
standard Normal distribution. For example, an effect
size of 0.8 means that the score of the average
person in the experimental group is 0.8 standard
deviations above the average person in the control
group.
61. Effect size
Cohen’s d : expressed on a standardized scale that
ranges from -3.00 to + 3.00
It is just the standardized mean difference between
the two groups. In other words:
62. Effect size
“Effect-size r,” which is simply the Pearson
Correlation Coefficient (r)
R is also expressed on a standardized scale, -1.00 to +1.00
R values can also be averaged while weighting the avg. to
take into account varying sample size
63. Measures of effect size for ANOVA
Measures of association
Eta-squared (2)- proportion of the total variance that is attributed
to an effect
R-squared (R2)- proportion of variance explained by the model
Omega-squared (2)- estimate of the dependent variable
population variability accounted for by the independent variable
Measures of difference
Cohen’s f - averaged standardised difference between the 3 or
more levels of the IV
Small effect - f=0.10; Medium effect - f=0.25; Large effect - f=0.40
64. Maximizing Validity and Minimizing Bias
History: external events that occur during the course of a study
that could explain why people changed. E.g. death in the family.
Negative life events, such as these, may offer an alternative
explanation for study outcomes. Consequently, it is very useful that
random assignment equalizes these occurrences across the
treatment and control conditions.
Maturation: processes that occur within individuals over the
course of study participation that provide an alternate explanation
of why they changed. E.g. depression increases after the onset of
puberty. Thus, if more children in the control than the treatment
group reached puberty during the course of the study, that might
explain why the control group finished the study with more
depression than the treated group.
Temporal Precedence: in order to establish a causal relation
between the intervention and outcome, the intervention must occur
before the outcome.
65. Blinding/ Masking
Blinding is an attempt to reduce bias
arising out of errors of assessment
a) Single blind trial: Participant not aware
b) Double blind trial: Neither doctor/
investigator nor the participant is aware
c) Triple blind trial: Doctor/investigator ,
participant and the person analyzing the
data are all not aware of the assigned t/t
66. Minimizing Threats to External Validity
To what extent can the results be extended to people,
settings, interventionists different than those used in this
particular study?
Sample Characteristics: External validity can be enhanced
by having a broadly representative sample. It enables the
findings to be generalized to a diverse population.
Setting Characteristics: (e.g., clinic, therapists, study
personnel).
Effects Due to Testing: refers to the potential for participants
to respond differently because they know they are being
assessed as part of research
67. Testing Efficacy vs. Effectiveness
An efficacy trial (also k/a explanatory trials)
answers the question: "Does this intervention work
under optimal conditions?”
An effectiveness trial(also k/a pragmatic trials)
answers the question: "Does this intervention work
under usual conditions?"
Very few trials actually fall wholly into one of these
categories, but rather fall along a continuum of
pragmatic-explanatory.
68. Significance Testing and Beyond
Estimation vs. Statistical Significance Testing
Assessing the Effect Size of an Intervention
Assessing Clinical Significance
69. Significance Testing and Beyond
Estimation vs. Statistical Significance Testing
Whether the effect of a treatment reaches a conventional
significance level (p < 0.05) depends heavily on factors
such as sample size.
Assessing the Effect Size of an Intervention
An effect size describes the magnitude of an intervention's
effect on the study outcome.
In the case of RCTs, the effect size represents the
magnitude of the difference between the control and
intervention conditions on a key outcome variable adjusted
for the standard deviation of either group.
70. Significance Testing and Beyond
Assessing Clinical Significance
When testing interventions that address health problems
The Number Needed to Treat (NNT) expresses the
number of patients who need to receive the intervention to
produce one good outcome compared to control. NNT is a
widely used index of clinical significance
NNT= 1 _
(Rate in untreated gp) - (Rate in treated gp)
71. Example
The number of deaths and patients treated were
3671/50496 (7.26%) in the test group
3903/50467(7.73%) in the control group.
Pt – Pc =0.47%= 0.0047
NNT = 1/0.0047 = 213
Hence, such treatment saves about 5 lives /1000
treated
72. Reporting Results
The Consolidated Standards of Reporting Trials
(CONSORT) has become the gold standard for
reporting the results of RCTs.
A checklist and flow diagram.
The most up-to-date revision of the CONSORT
Statement is CONSORT 2010.
Extensions of the CONSORT Statement have been
developed for other types of study designs,
interventions and data.
73.
74.
75.
76. Ethical Issues
Investigators are responsible to uphold ethical
standards and guidelines
Declaration of Helsinki- (developed by the World
Medical Association)- This set of ethical principles
guides medical researchers in conducting research on
human subjects.
78. Advantages of RCT
The use of randomization provides a basis for an
assumption-free statistical test of the equality of
treatments
Random assignment ensures that known and
unknown person and environment characteristics that
could affect the outcome of interest are evenly
distributed across conditions.
Random assignment equalizes the influence of
nonspecific processes not integral to the intervention
whose impact is being tested. Nonspecific processes
might include effects of participating in a study, being
assessed, receiving attention, self-monitoring, positive
expectations, etc.
79. Pros
Random assignment and the use of a control
condition ensure that any extraneous variation not
due to the intervention is either controlled
experimentally or randomized. That allows the study's
results to be causally attributed to differences
between the intervention and control conditions.
In sum, the use of an RCT design gives the
investigator confidence that differences in outcome
between treatment and control were actually caused
by the treatment, since random assignment
(theoretically) equalizes the groups on all other
variables.
80. Cons
Drawbacks of conducting an RCT are:
Time- and energy- intensive
Expensive
May not be feasible for all interventions or settings
(e.g., Some institutions have policies that prohibit
random assignment)
84. Four types of comparisons in RCT design
Parallel RCT design is most commonly used,
which means all participants are randomized
to two (the most common) or more arms of
different interventions treated concurrently
85. Superiority trials
A new treatment is more effective than a standard
treatment from a statistical point of view or from a
clinical point of view,
Its corresponding null hypothesis is that: The new
treatment is not more efficacious than the control
treatment by a statistically/clinically relevant amount.
86. Equivalence trials
The objective of this design is to ascertain that the
new treatment and standard treatment are equally
effective.
The null hypothesis of that is: Both two treatments
differ by a clinically relevant amount.
87. Non-inferiority trials
Non-inferiority trials are conducted to show that the
new treatment is as effective but need not superior
when compared to the standard treatment.
The corresponding null hypothesis is: The new
treatment is inferior to the control treatment by a
clinically relevant amount.
One-sided test is performed in both superiority and
non-inferiority trials, and two-sided test is used in
equivalence trials.
88.
89. Assuming RCT has two comparison groups
and both groups have the same size of
subjects
90. Parameter definitions
N=size per group;
p=the response rate of standard treatment group;
p0= the response rate of new drug treatment group;
zx= the standard normal deviate for a one or two sided x;
d= the real difference between two treatment effect;
δ0= a clinically acceptable margin;
S2= Polled standard deviation of both comparison groups
96. Problem: The research question is whether there is a
difference in the efficacy of mirtazapine (new drug)
and sertraline (standard drug) for the treatment of
resistant depression in 6-week treatment duration. A ll
parameters were assumed as follows: p =0.40;
p0=0.58;α=0.05;β=0.20; δ=0.18; δ0=0.10.
Parameter definitions N=size per group; p=the
response rate of standard treatment group; p0= the
response rate of new drug treatment group; zx= the
standard normal deviate for a one or two sided x; d=
the real difference between two treatment effect;δ0= a
clinically acceptable margin; S2= Polled standard
deviation of both comparison groups.
99. Problem: The research question is whether there is a
difference in the efficacy of A CE II antagonist (new
drug) and A CE inhibitor (standard drug) for the
treatment of primary hypertension. Change of sitting
diastolic blood pressure (SDBP, mmHg) is the primary
measurement, compared to baseline. All parameters
were assumed as follows: mean change of SDBP in
new drug treatment group=18 mm Hg; mean change
of SDBP in standard treatment group =14 mm
Hg;α=0.05;β=0.20; δ=4 mmHg; δ0=3 mm Hg; s=6mm
Hg.
100.
101. DISCUSSION
Firstly, the researcher should specify the null and
alternative hypotheses, along with the type I error rate
and the power (1- type II error rate).
Secondly, the researcher can gather the data of
relevant parameters of interest but sometimes a pilot
study may be required.