Randomized Controlled Trials (RCTs)

1. Randomized Controlled Trials
(RCTs)
Dr Amit Agrawal
Professor of Neurosurgery
Narayana Medical College and Hospital
Nellore (AP)

2.  Objectives
 What are randomized controlled trials (RCTs)?
 Why we do randomized trials?
 How to randomize?
 Types of studies
 Evidence hierarchy
 RCT: Definition
 Steps in conducting RCT
 Trial monitoring
 Ethical considerations
 Conclusions
Outline

3. Types of studies
 1. Observational studies
 a. Descriptive studies
 b. Analytical studies
 (i) Ecological
 (ii) Cross-sectional
 (iii) Case-control
 (iv) Cohort
 2. Experimental/Intervention
studies
 a. Randomized controlled
trials*
 b. Field trials
 c. Community trials
*RCTs are a subset of possible experimental designs

5.  " An epidemiological experiment in which subjects in
a population are randomly allocated into groups,
usually called study and control groups to receive and
not receive an experimental preventive or therapeutic
procedure, maneuver or intervention"
-John M Last,2001
Definition

6.  Designing a protocol
 Selecting reference and experimental population
 Randomization
 Intervention
 Follow up
 Outcome analysis
 Publishing
Steps in RCT

7.  Purpose of randomization is to make the treatment groups
comparable
 Ensures that only difference in groups is due to trial
treatments
Why Randomization?

8.  The process by which each subject has the same chance of
being assigned to either intervention or control.
 Must be purely by chance
 Each group is a random sample of eligible patients, so both
are representative of that same population
What is random allocation?

9.  Inclusion and exclusion criteria are the conditions that must
be met in order to participate in a clinical trial
 Age
 Gender
 The type and stage of a disease
 Treatment history
 Any other medical conditions
Inclusion and exclusion criteria

10.  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
 Serially numbered, sealed and opaque
Who should randomize?

11. Types of randomization
 Simple Randomization
 Unequal Allocation Randomization
 Stratified Randomization
 Blocked Randomization

12. Simple Randomization
 Participants have an equal chance of being allocated to
different treatment groups
 This method increases statistical power and meets the
assumption of equipoise in the randomly allocated groups


13.  Unequal allocation randomization means randomly
assigning participants into disproportionately sized groups
 The acceptable ratios for unequal allocation are 2:1, 3:1, and
4:1 controls to cases
 This method is useful for studying rare outcomes, adverse
events, or emerging therapies
 However equipoise is not possible
Unequal Allocation
Randomization

14.  Randomly assigning participants to treatment groups based
on predefined and objective characteristics (strata e.g. age
gender)
 Each stratum is randomly allocated to treatment groups.
Stratified Randomization

15.  For smaller studies
 Participant are randomly assigned to treatment groups in
small "blocks" of four or six participants
 Ensures that there are equally sized groups
 Groups meets the assumption of equipoise
Blocked Randomization

16.  Essentially the right thing to do
 Try not to do it in front of the patient
 OK for big trials
 For small trials, such ‘simple randomization’ can lead to
imbalance in group sizes
Toss a coin

17.  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%
Example: trial with 30 patients

18. Table of Random numbers
 Computer generated list
 According to date of birth (odd or even years)
 The number of their hospital records
 The date at which they are invited to participate in the study
(odd or even days)
 Alternatively into the different study groups

19.  Blinding requires placebo or use of alternate treatment that
cannot be distinguished from treatment
 Blinding is done to minimize participant or observer bias
 Unblinded (cannot conceal randomization, e.g., surgical
interventions)
 Single blinding (observer but not participant knows the arm of
randomization, e.g., cluster-level trials)
 Double blinding (neither observer or participant know the arm of
randomization)
 Complete blinding: Everyone involved in the study blinded to
treatment
Blinding

20. Schema of a
simple
randomized
trial

21. Trial Monitoring
 Trials must be approved by and monitored by Institutional
Review Boards (IRBs) for ethics and safety
 Trials should have an independent monitoring system to
periodically review data and ensure participant safety
 Data monitoring committee (DMC) or Data Safety and
Monitoring Board (DSMB) should have the authority to
terminate or change the trial procedures
 Trials should report all adverse events, especially serious
adverse events, and unexpected events

22.  It can be unethical to deny a new treatment to the
placebo group
 “Equipoise” a genuine uncertainty or doubt about
whether one intervention or treatment is superior to
another
 Protection of the study participants (including physical,
mental and social well being) is the most important
responsibility of the researcher
 The protection of the vulnerable population (whose
decision–making capacity is impaired or diminished)
 Informed consent
Ethical issues in randomized
trials

23. RCTs
 Unbiased distribution of
confounders
 Blinding more likely
 Randomisation facilitates
statistical analysis
 Expensive: time and
money
 Volunteer bias
 Ethically problematic at
times
Advantages Disadvantages

24.  The trial should ideally reflect real-world conditions
 Two groups that are comparable and free from bias
 RCT is gold-standard in terms of identifying causality
 Need to weigh theoretical advantages against practicalities
 Randomization method
 Unpredictable
 Balanced
 Simple
Summary