International Journal of Research

Regional Anaesthesia

Randomized Controlled Trials

Nidhi Bhatia1 Anju Grewal2
1
Department of Anaesthesiology & Intensive Care, PGIMER, Chandigarh, India.
2
Department of Anaesthesia, Dayanand Medical College and Hospital, Ludhiana, Punjab, India.

Abstract
A randomized controlled trial (RCT) is a prospective, comparative, quantitative experiment / study that is
performed under controlled conditions with random allocation of interventions to comparison groups. Among all
the clinical study designs, evidence generated from RCTs is considered to be at top of the evidence pyramid. There are
many different RCT designs and they can be classified on the basis of interventions evaluated, participants’ exposure
and level of blinding. All RCTs should be planned prospectively, a research question should be formulated, sample
population approached and informed consent obtained from participants of the trial. These consented subjects are
randomly assigned to any of the study arms and the changes are then measured over time. The basic principles to
designing an RCT include formulating a research question, developing a protocol, randomization, allocation
concealment, blinding, sample size calculation and registering of RCTs. Appropriate guidelines for reporting RCTs
should be followed and RCTs should only be conducted if they are ethically viable, economical and clinically
worthwhile.
Keywords: Randomised Control trial (RCT)

Introduction
Clinical studies are conducted among human participants to generate new knowledge by evaluating the impact of
interventions. The main aim of all clinical studies is to evaluate interventions with respect to an associated outcome
[1]. There are many different clinical study designs and the quality of evidence generated by any study is determined
by its experimental design [1, 2]. Of all the clinical study designs, evidence generated from randomized controlled
trials (RCTs) is considered to be at top of the evidence pyramid.

What is Randomized Controlled Trial?

A randomized controlled trial is a prospective, comparative, quantitative experiment/study that is performed under
controlled conditions with random allocation of interventions to comparison groups [2]. While performing an RCT,
robust, high quality evidence can be generated that greatly helps in evaluating the safety and effectiveness of an
intervention as well as determining the cause-effect relation between an intervention and an outcome. Furthermore,
RCTs yield themselves well to systemic reviews and meta-analysis, thus providing strong base for synthesizing
evidence generated by such studies [2].

Why is Evidence Based on RCT Considered of Highest Quality?

RCTs are considered to be at the top of the evidence pyramid because:
- Concealed random assignment eliminates/greatly reduces confounding from known and unknown factors
- Blinding eliminates bias
In a prospective study, there are often systematic differences between the groups. As a result of these differences, the
outcome of the groups may be different. This is known as confounding. Bias, on the other hand, is defined as the
systematic tendency of any factors associated with the study to make the estimate of the effect of intervention deviate
Address of Correspondence: Dr. Anju Grewal, Professor & Head, Department of Anaesthesia, Dayanand Medical College and Hospital,
Ludhiana, Punjab, India.
E-mail: dranjugrewal@gmail.com

DOI: 10.13107/ijra.2021.v02i02.037 | www.ijrajournal.com |

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial-Share Alike 4.0 License
(http://creativecommons.org/licenses/by-nc-sa/4.0) which allows others to remix, tweak, and build upon the work non-commercially as long as
appropriate credit is given and the new creation are licensed under the identical terms.

107 © International Journal of Regional Anaesthesia | Published by Academy of Regional Anaesthesia of India
Bhatia N & Grewal A www.ijrajournal.com

from its true value [2]. The only way to best eliminate these given to each participant.
differences is by randomization and blinding. Blinding refers 3. Double-blinded RCT: Two group of individuals involved
to concealment of group allocation from one or more in the trial (usually patients and treating physician) do not
ind iv idual s involved in cl inical research study. know which intervention is given to each participant.
Randomization is allocating each individual to one or the
other intervention at random. Thus, the probability of any When and How Should RCTs be Planned?
individual receiving one or the other intervention is decided All RCTs need to be planned prospectively and are suitable
solely by chance. As a result of random allocation, if the for pre-clinical and clinical research. The participation in the
sample size is adequate, all factors influencing outcome are trial needs to be carefully controlled with the experiment in
likely to be distributed equally between groups [1, 2]. This mind. RCTs cannot be conducted retrospectively [3]. Its
can minimize the differences in characteristics of the groups other salient features include: [2]
that may influence the outcome. - An RCT tests the effectiveness of one or more
interventions.
RCT Designs - The intervention being tested is allocated to two or more
The simplest RCT design has one treatment group (or ‘arm’) study groups.
and another control group. There can be variations in design - The control group may receive no intervention, a standard
by having multiple treatment arms or a factorial design [3]. treatment or a placebo.
The two key features of an RCT is ‘random sampling’ and - There should be sufficient uncertainty about the utility of
‘random assignment’. Random sampling refers to how the intervention, known as equipoise.
samples are drawn from one or more populations, whereas, - The question to be answered by RCT design should be safe
random assignment refers to how groups or individuals are for the participants.
assigned to either a treatment group or a control group [3]. - The research question should be ethically appropriate.
- In an RCT it is impractical to study outcomes that are
RCT Classification on The Basis of Interventions extremely rare or may take a very long time to develop.
Evaluated: [4]
1. Explanatory Trials: These trials are designed to evaluate Designing an RCT [1, 2, 3, 4]
whether an intervention works. The basic framework to conducting an RCT is fairly simple.
2. Pragmatic Trials: These trials not only determine if a trial The research question is formulated, the sample population
works, but also describe all consequences of the intervention. is approached, and patients consented for participating in the
3. Efficacy Trials: Evaluate effectiveness or efficacy of an trial. The consented subjects are randomly assigned to any of
intervention. the study arms and the changes are then measured over time.
4. Phase 1, 2, 3, 4 Trials: These are the phases of a trial The basic principles to designing an RCT include the
conducted for introduction of a new intervention, following:
traditionally a new drug.
Formulating Research Question: A research question
RCT Classification on The Basis Of Participants’ guides the research. A good study should have one research
Exposure: question and one hypothesis, which should be precise. The
1. Parallel Design: In this each group of participants is research question should be in the ‘PICOT’ format: P-
ex posed to only one of the study inter ventions. Population of Interest; I-Intervention; C-Comparator
2. Crossover Design: Each of the participants are subjected to Inter vention; O - Outcome; T-Time duration for
all the study interventions over successive periods. intervention/outcome. In addition, ‘FINER’ criteria are used
3. Factorial Design: Two or more experimental interventions in the development of a good research question, according to
are not only evaluated separately but also in combination and which the research question should be Feasible, Interesting,
against a control. Novel, Ethical, Relevant. These criteria highlight relevant
and useful points that increase the chances of developing a
RCT Classification on The Basis of Level of Blinding: successful research project.
1. Open RCT: Everyone involved in the trial knows which
intervention is given to each participant. Developing a Protocol: The protocol should be developed
2. Single-blinded RCT: A group of individuals involved in the in collaboration with other researchers in the team,
trial (usually patients) do not know which intervention is statisticians, investigators, necessary regulatory authorities

108 | International Journal of Regional Anaesthesia | Volume 2 | Issue 2 | July-December 2021 | Page 107-110
 Bhatia N & Grewal A www.ijrajournal.com

and funding bodies [2, 5]. The protocol should have clearly ‘significant’ if the probability of this difference arising solely
defined aims and objectives, inclusion and exclusion criteria, by chance is < 0.05 [2]. This is the probability or p-value. The
consent process, interventions and outcome definitions. It chance that a difference will be found even if there is no real
should be reviewed and approved by an independent ethics difference is known as Type-I error. This is usually fixed at
committee and prospectively registered with a publicly 0.05, though other levels of significance (0.01 or 0.005) may
available trial registry prior to recruiting any participants. also be chosen. Type-II error is the inability to demonstrate a
significant difference even when one does exist. This is set at
Randomization: Randomization ensures that each patient 0.2 to 0.05. Power of the study is the chance that the study will
has an equal chance of receiving the intervention. The main be able to demonstrate a significant difference if it is present.
purpose of randomization is to eliminate selection bias and Conventionally, power of the study is fixed at 0.8-0.95 (80-
balance known and unknown confounding factors. Methods 95%). In RCTs where effect size is measured in proportion,
of randomization include using a table of random numbers or sample size can be calculated using these values. However, if
a computer program that generates random number [6]. the effect size is measured as a continuous variable, then in
Generation of random sequence should be done by an addition to Type-I and Type-II values, one would need mean
independent person, usually a statistician. Block and standard deviation of the variable in each group to
randomization is designed to randomize participants into calculate the sample size [2].
groups that result in equal sample sizes. Cluster
randomization can be used when randomization of Ethical Considerations: While conducting RCTs strict
individual participants is not feasible, in which case hospitals, ethical principles should be followed. Evaluation of risks and
clinics, geographical areas, etc can be used as units for benefits to the society and the participants, obtaining ethical
intervention allocation [2]. approval and taking written, informed consent is important.
Further, before an RCT is conducted, there must be
Allocation Concealment: This is an essential component of equipoise (genuine doubt whether one course of action is
RCT and is useful in situations where blinding is not better than the other) [2, 4]. While conducting medical
possible. It means that neither the investigators nor the research, it is the duty of the physician, conducting the
participants are aware of whether the next eligible participant research, to protect the health, life, privacy as well as the
will be receiving control or treatment intervention. It is dignity of human subjects. This is guided by a statement of
important that the person who generates allocation sequence ethical principles in the ‘Declaration of Helsinki’, which was
is not the person who determines eligibility and entry of developed by the World Medical Association for medical
participants [2, 5]. research involving human subjects, including research on
identifiable human material and data [6].
Blinding: Blinding helps in eliminating bias. It refers to
keeping the trial participants, assessors and/or investigators Registering RCTs: RCTs are considered the premium
unaware of the assigned intervention. Blinding maybe single- methodological design in science. However, there is
blinded, double-blinded, or triple-blinded (double-blind considerable variance in the ways RCTs are reported in the
trial that also maintains a blind data analysis). Although peer-review literature. In order to ensure systematic
blinding is important, however, it may be difficult or standards in the reporting of evidence as well as to have
impossible in certain trials. greater public transparency in the way RCTs are conducted
and reported, it is now mandatory that all RCTs are
Sample Size: The study or the trial should have an adequate prospectively registered, prior to starting the trial/enrolling
sample size and power. The conclusions generated from the first participant, in a database open to public access [7].
studies with adequate sample size can be applied to larger India launched its own free online public registry, the Clinical
populations. The sample size required to test a hypothesis Trial Registry of India (CTRI) on July.20, 2007, for
should be governed by effect size and derived from previous registration of clinical trials conducted in India.
observations and trials. For calculating sample size, it is
important to know baseline estimate of outcome rate in Reporting of RCT
control group. Further, it is important to know what It is important to be aware of the quality of reporting RCTs
percentage of patients are expected to benefit from the and the limitations of the research methods. Appropriate
intervention. The difference in primary outcome between guidelines for reporting RCTs should be followed. The
the ‘control’ and ‘intervention’ groups is said to be CONSORT (Consolidated Standards of Reporting Trials)

109 | International Journal of Regional Anaesthesia | Volume 2 | Issue 2 | July-December 2021 | Page 107-110
Bhatia N & Grewal A www.ijrajournal.com

Statement, first published in 1996 was designed to assist research. Although formulating, designing, and conducting
reporting of RCTs and is still followed [4, 5, 8]. them may require more effort, however they are still
considered a ‘gold standard’. Nonetheless, RCTs should only
Conclusion be conducted if they are ethically viable, economical, and
RCT is one of the most robust research methods used to clinically worthwhile.
generate evidence in basic, translational, and clinical

Conflict of interest: Nil Source of support: None

References
1. Zabor EC, Kaizer AM, Hobbs BP. Randomized Controlled Trials. Chest 6. World Medical Association. World Medical Association Declaration of
2020; 158: S79-S87. Helsinki: Ethical Principles for Medical Research Involving Human
Subjects. JAMA. 2013;31:2191–2194. .
2. Bhide A, Shah PS, Acharya G. A simplified guide to randomized
controlled trials. Acta Obstet Gynecol Scand 2018;97:380-387. 7. Elliott TR. Registering randomized clinical trials and the case for
CONSORT. Exp Clin Psychopharmacol. 2007 Dec;15:511-8.
3. White H, Sabarwal S, De Hoop T. Randomized Controlled Trials
(RCTs): Methodological Briefs 2014; Impact Evaluation No. 7, 8. Akobeng AK. Understanding randomised controlled trials. Arch Dis
Methodological Briefs no. 7. Child 2005 ;90:840-4.
4. Stolberg HO, Norman G, Trop I. Randomized controlled trials. AJR Am
J Roentgenol 2004;183:1539-44.
5. Thiruvenkatachari B. Randomized controlled trials: The technique and
challenges. J Indian Orthod Soc 2015;49:42-7.

How to cite this article: Bhatia N, Grewal A | Randomized Controlled Trials | International Journal of Regional
Anaesthesia | July-December 2021; 2(2): 107-110.

110 | International Journal of Regional Anaesthesia | Volume 2 | Issue 2 | July-December 2021 | Page 107-110