Randomized Controlled Trial

A randomized controlled trial, or RCT, is an experimental study in which participants, or sometimes groups of participants, are assigned by a random process to different study groups and their outcomes are compared. Typically, one group receives the intervention being tested while a control group receives another intervention, usual care, a placebo when appropriate, or another defined comparison .^1,2

Randomization means that assignment is determined by chance rather than by participant characteristics or an investigator's preference. When properly implemented, this tends to reduce systematic differences between study groups in factors that could influence the outcome, including factors that researchers may not have measured. Comparing outcomes between randomized groups therefore provides a strong basis for estimating the causal effect of an intervention .^1,2

The word “controlled” refers to the presence of a comparison group against which the intervention group is evaluated. RCTs may also use blinding, in which participants, researchers, or outcome assessors do not know the assigned intervention, to reduce certain forms of bias. Blinding is not what makes a study randomized, however, and some randomized trials cannot practically be blinded .^1,2

An RCT differs from an observational study because researchers actively assign the intervention in an RCT. In an observational study, investigators observe exposures or treatments that occur without random assignment. This distinction is important because nonrandomized comparisons are generally more vulnerable to confounding, in which differences between groups other than the intervention contribute to differences in outcomes .¹