Randomised controlled trials : the basics

Randomized controlled trials
The Basics

Definition
 RCT is a study in which a group of
investigators study an intervention in a series
of individuals who receive the intervention in
a random order.
 Intervention to be tested is called the
experimental group

 The other group of participants is called the
control group.
 The control can be conventional practice, a
placebo, or no intervention at all

Schema of a simple trial
Eligible patients
Rx group 1
Rx group 2
Randomize

Why Randomize?
 We need to analyse groups at the end of the
trial
 To ensure that difference in groups is
because of the Rx
 For this you need comparable groups at the
start of trial
 Purpose of randomization is to make the
treatment groups comparable

Value of randomization
 it reduces the risk of serious imbalance in
unknown but important factors that could
influence the clinical course of the
participants.

RCT
 ‘the most powerful tool in modern clinical
research “
– Prospective
– Controlled
– unbiased

What is wrong with non-
randomized studies?
 Two main types of studies, those with and
those without concurrent control groups

Non-randomized studies II
 Without concurrent controls
 Case series studies
 Historical controls
 type of patient may change, due to eligibility
criteria
 environment changes
 data quality often quite different between
groups

Non-randomized studies III
 Controlled non-randomized studies
 Difficult to argue that one group is different from
another but allocation is predictable, so bias can
arise from selection of patients
 Randomization must be unpredictable

Random allocation
 all participants have the same chance of
being assigned to each of the study groups
 the purpose is to keep both groups as similar
to each as possible at the start of the trial.

Is coin tossing OK?
 OK for big trials
 For small trials, such ‘simple randomization’
can lead to imbalance in group sizes

Example: trial with 30 patients
 If 30 patients are in a trial randomized using
coin tossing there is a 14% chance of 15:15
split
 For 16:14 chance is 27%
 ‘Worse’ than 20:10 is 10%
 Why ‘worse’?
 Because imbalance leads to loss of power

We need randomization
 to be done properly
 to ensure similar numbers in groups
 To combine with stratification -in large trials-
to ensure comparability

Pseudo-randomisation
 Alternating record number
 Date of birth
 Geographical distribution

True randomization
 Need to separate the person who generates
allocation from those who assess eligibility
 Third party schemes
 Telephone randomization service
 Pharmacy randomization
 Web-based service?
 Envelopes
 Sealed envelopes (preferably opaque)

Blinding
 The best way to protect a trial against bias is
by keeping the people involved in the trial
unaware of the identity of the interventions
for as long as possible

Types of RCTs
 RCTs according to whether the investigators
and participants know which intervention is
being assessed
– Open trials
– Single blind trials
– Double blind trials
– Triple blind trials

Blinding is difficult
 Having placebo in the same shape , formula
and taste is very costly, and time consuming.
 The drug side effects e.g. local reaction at
the site of injection would partially unblind .
 Impossible if surgical and medical treatments
are compared.
 The need for urgent unblinding code in case
of serious side effects

Other types of RCTs
 RCTs according to how the participants are
exposed to the interventions
– Parallel trials
– Crossover trials
 Trials testing one variable or factorial design
e.g (2 X2 X 2)

Follow up
 During the trial
– Adherence to the study protocol
– Patients compliance with treatment
 After finishing the intervention, follow up of
participants should be sufficiently long and
complete

Analysis of clinical trials
Analysis of clinical trials
Intension to treat analysis Per treatment analysis
Sub group analysis

Disadvantages of RCTs
 expensive: time and money;
 volunteer bias;
 ethically problematic at times.

Interim Analysis
 Done in large RCTs
 To explore the results after recruiting of half
of the participants
 If marked difference is recognized , then trial
should be stopped
 Examples: WHI trial
Breech Trial

So, how to do RCT
 Set up a protocol
 Recruit your patients
 Randomize (try to be blind)
 Follow up
 Analyze your data
 Publish

RCTs
The gold standard for therapeutic research
Basis for Meta-analysis
Search for it first

Randomised controlled trials : the basics

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Randomised controlled trials : the basics (20)

More from Hesham Al-Inany (20)

Recently uploaded (20)

Randomised controlled trials : the basics