Oct 03, 2014
Terminology 101: Randomized controlled trial study design
Randomized controlled trial study design: An experimental research design in which the effectiveness of an intervention is evaluated to infer causality
Source: Gordis, L. (2009). Epidemiology. (4th ed.). Philadelphia: Saunders Elsevier
Randomized controlled trials (RCTs) are interventional (or experimental) studies that constitute the gold standard of clinical research design. Researchers can employ techniques in RCTs to eliminate selection, confounding and measurement biases, all of which are inherent limitations of observational studies. Elimination of such biases in RCTs allows one to infer causality and conclude that the outcome is the result of the intervention. In contrast, the various types of observational studies discussed in earlier columns can only determine whether there is an association between an exposure and an outcome; they cannot establish causality.
In their most basic form, RCTs compare two or more groups on a specific psychosocial, educational, behavioural or physiological outcome. Participants assigned to the control group usually receive either a sham intervention (placebo control) or the treatment that they would have received if they had not participated in the study (active control). Participants assigned to the experimental group receive the new treatment that is under study. RCTs can be conducted to establish the superiority, equivalence or non-inferiority of treatment options. Superiority RCTs, the most common type, usually hypothesize that the experimental group will experience the positive outcome at a significantly higher rate than the control group. For example, consider a nursing study designed to examine the effectiveness of chlorhexidine gluconate (CHG) oral rinse in preventing ventilator-associated pneumonia (VAP) in the intensive care unit. If those who receive CHG experience a lower rate of VAP than those who receive a placebo or standard of care oral rinse, one will conclude that CHG reduces the risk of VAP.
The RCT design maximizes the sound (true effect) and minimizes the noise (bias) through allocation concealment, randomization and blinding. Allocation concealment is the practice of keeping clinicians and participants unaware of which group assignment the next recruited patient will receive, so that selection bias can be eliminated. Randomization, or random assignment of participants, is the practice of giving each study participant an equal chance to be assigned to the control or experimental group. The idea is to produce study groups with similar demographic and prognostic characteristics and thus eliminate confounding bias. Blinding is the practice of hiding the nature of the intervention (e.g., control versus experimental treatment) from study participants and researchers alike, thus eliminating measurement bias. While blinding may not always be possible in RCTs, allocation concealment and randomization should never be ignored. If researchers do not adequately follow proper protocols for the latter two design components, the validity of the findings of their RCT will be thrown into serious question. All three of these design components will be discussed further in future columns.
Although RCTs are the most rigorous type of study, it is important to note that results from a single RCT are never sufficient to change or inform practice. Rather, cumulative evidence from repeated RCTs is needed.
NurseONE.ca resources on this topic
- Fitzpatrick, J. J., & Kazer, M. W. (Eds.). (2011). Encyclopedia of nursing research (3rd ed.).
- Gallin, J. I., & Ognibene, F. P. (Eds.). (2012). Principles and practice of clinical research (3rd ed.).
- Supino, P. G., & Borer, J. S. (Eds.). (2012). Principles of research methodology: A guide for clinical investigators.