Randomized Controlled Trial: a brief introduction


Randomized controlled clinical trials (RCTs) are the top of evidence-based medicine for individual studies, the leading and most authoritative source of information. The most famous RTCs application is related to evaluating the effectiveness, tolerability, and safety of a new treatment or a new drug compared to a gold-standard procedure or a well-known pharmacological medication. Rigorous development, strict methodology, and well-conducted data evaluation and interpretation allow RCTs to achieve optimal and well-supported evidence to support clinicians’ and policymakers’ decisions. This post aims to conduct you into the world of RCTs through a simple guide that allows you to familiarize yourself with some essential elements of randomized controlled clinical trials.

Step 1: Randomized Controlled Trial – Design a suitable protocol

The first step of each RCT is to predispose a suitable research protocol. The protocol is characterized by a proper structure (background, objectives, hypotheses, method, sample calculation, and statistical analysis) that allows researchers to focus on their goals and identify procedures or methodology, evaluating their clarity and appropriateness. In many cases, designing a research protocol coincides with presenting a request for the study’s approval to the Ethics Committee (EC) or Internal Review Board (IRB). Therefore, predisposing a research protocol is a mandatory phase for every RCT independently from the need of EC or IRB approval request. Through a research protocol, we may understand what is actually realizable and what could go wrong during the experimental stage.

Why is the EC or IRB approval necessary?

Unlike other studies, such as systematic review and meta-analysis, randomized controlled trials need, even before being implemented, to obtain approval from an Ethics Committee or an Internal Review Board. This preliminary and necessary check allows investigators to understand if and how well their study is focused on their goals and if the designed methodology is suitable and rigorous.

The approval of the study by a commission outside the research team is both a step to safeguard patients (in compliance with the supreme principle of “don’t harm” and the Principles of Good Clinical Practice and the Declaration of Helsinki) and an actual first test for the study quality. The protocol must be constructed under the most recent and relevant scientific literature and guidelines without neglecting negative aspects. Obtaining the approval of the Ethics Committee is the first official step to start the work. Very often, the road to getting such consent is not easy. Inexperience, lack of knowledge of statistical fields, lack of awareness of the study’s limitations, and implementation problems can discourage researchers.

Step 2: Conducting the study: what to do and what not to do

After passing the first step of approval, we can begin the experimental phase. Experimenting means implementing what has been declared in the protocol as rigorously as possible. It is fundamental to respect the “rules of engagement” fixed and approved in the research protocol. For example, suppose you have established to apply a specific kind of patients’ randomization and allocation. In that case, you cannot perform another type of randomization and participant allocation process or even eliminate these procedures. 

In some cases, after acceptance of the study, you can realize that the RCTs needed some changes. Making mistakes or changing the first set of an investigation is normal in each research process. The adaptability of the methodological approach when clinical contexts change is the rule of each research.  

Sample size

A typical case of error is related to the low sample size. The first question that is asked to a statistician about the creation of a new study is: “How large must my sample be? “, and, from my experience, this request is generally accompanied by a: “Please do not give me a high number, because we do not know if we can enroll a high number of patients.” If you have small samples, it is not a problem. The problem is how much the small sample size prevents the results from correctly describing the entire reference population. And if this is true for statistics in general, it becomes fundamental for RCTs. Incorrect sample size will turn the study execution phase (especially processing results and interpretation) into a huge problem. 

Downgrade the study

Whether during a randomized controlled clinical trial you understand that there is a mistake, the first instinct is to downgrade the study. The thought is, “Okay, if I can’t do an RCT, then I convert it to a prospective or cross-sectional study,” but this is not the correct approach. If there is an error, this must be corrected. 

To give you an immediate idea of what it means to change the study design after a mistake identification without first rectifying the error, it is as if, after starting to build a house and discovering that the foundations are not well placed, you convert the house into a warehouse. If the foundations fail, the building falls. Regardless of whether the study is an RCT or a prospective study, the study will not be equally reliable if there is an error.

If you notice an error, the only possible remedy is to correct it. But be careful; fixing does not mean doing it without rules, a sort of “let’s make do and see what comes out“; it is a matter of remedying it in a rigorous, serious, and scientific way. Moreover, it is helpful to remember that each structural change should be communicated to the Ethics Committee or IRB to obtain consent to that specific change.

Step 3: Combining the information for a scientific publication

After enrolling the participants, collecting the data, and processing the statistical analysis, it is time to present the results. Preparing the material to be presented in a scientific publication is complex. Reporting the randomized controlled trials results means strictly respecting the CONSORT guidelines and, in particular, those reported in the CONSORT Statement. The statement is composed of 25 points to be followed step by step, explaining what has been done in detail. The checklist is divided into six sections: Title/Abstract, Introduction, Methods, Results, Discussion, and Other Information. 

Last step: Write, write, write

Write, write well and in a scientifically correct way so that the scientific community can understand and, if want, can replicate the study is mandatory. A well-conducted and presented RCT is the key to the success of the entire scientific community. After the long road of planning, experimentation, and evaluation, carried out respecting the highest levels of methodology and rigor, the writing step is a complex task. Communicating results is as tricky as processing them. The last part of the publication process requires meticulous and rigorous work—a necessary coming and going of controls suitable to ensure the highest scientific quality

The work to conduct an RCT starts from an idea and is articulated in a series of increasingly detailed and rigorous activities in which scientific research, knowledge, and forecasting are intertwined in a single plot that, combined with the researchers’ passion, gives rise to a scientific work of art.