Randomized Controlled Trial

A randomized controlled trial (RCT) is an experimental study in which participants are allocated by chance to an intervention group or a comparison group, so that the groups differ on average only in the intervention received. Randomization is what allows an RCT to attribute differences in outcome to the intervention rather than to confounding, which is why well-conducted RCTs sit at the top of most evidence hierarchies for questions of effectiveness.

Definition

A randomized controlled trial is a prospective experimental design in which eligible participants are assigned to intervention or control conditions by a random process, and outcomes are compared between the resulting groups to estimate the causal effect of the intervention.

Scope

This topic covers the defining features of the RCT as a design: random allocation, allocation concealment, blinding, the comparison group, and intention-to-treat analysis, together with how trials are reported and graded. It is a methodological reference within evidence-based practice and does not provide clinical instructions.

Core questions

• Why does random allocation support causal inference?
• What roles do allocation concealment and blinding play in protecting a trial from bias?
• When is an RCT feasible and appropriate, and when is it not?

Key concepts

• Random allocation
• Allocation concealment
• Blinding (masking)
• Control or comparison group
• Intention-to-treat analysis
• Selection and performance bias
• Internal validity

Mechanisms

Random allocation distributes both known and unknown prognostic factors across groups by chance, so that with adequate sample size the groups are comparable at baseline and any outcome difference can be attributed to the intervention. Allocation concealment prevents foreknowledge of the upcoming assignment, protecting the randomization from manipulation, while blinding of participants, clinicians, and assessors limits performance and detection bias after assignment (Schulz & Grimes, 2002). Analysis by intention to treat preserves the benefit of randomization by keeping participants in their assigned groups regardless of adherence.

Clinical relevance

RCTs supply much of the effectiveness evidence underpinning health technology assessment and clinical recommendations, and appraising them is central to evidence-based practice. This entry explains how such evidence is generated and judged; it is not a basis for individual treatment decisions.

Evidence & guidelines

The CONSORT statement provides the consensus standard for transparent reporting of parallel-group RCTs (Schulz et al., 2010), and GRADE treats randomized trials as a high starting point for certainty of evidence that can be downgraded for risk of bias, imprecision, inconsistency, indirectness, or publication bias (Guyatt et al., 2008). Empirical comparisons have examined how RCT and observational estimates relate (Concato et al., 2000).

History

Controlled comparison with random allocation entered medicine in the mid-twentieth century, with the Medical Research Council streptomycin trial for pulmonary tuberculosis often cited as a landmark early randomized trial. Over subsequent decades the design became the reference standard for evaluating interventions, and reporting was standardized through the CONSORT statement (Schulz et al., 2010) and grading through frameworks such as GRADE (Guyatt et al., 2008).

Debates

Are RCTs always the most appropriate or most generalizable source of evidence?

RCTs maximize internal validity, but strict eligibility and controlled settings can limit external validity, and for some questions randomization is infeasible or unethical; empirical work has questioned whether observational designs systematically overstate effects, motivating a question-dependent rather than absolute view.

Key figures

• Kenneth Schulz
• Douglas Altman
• David Moher
• Gordon Guyatt

Seminal works

• schulz-2010-consort
• schulz-2002-allocation
• guyatt-2008

Frequently asked questions

What is the difference between allocation concealment and blinding?

Allocation concealment protects the randomization process by hiding the upcoming assignment until a participant is enrolled, whereas blinding hides which intervention a participant actually received after assignment; the first guards against selection bias and the second against performance and detection bias.

Why analyze trials by intention to treat?

Analyzing participants in the groups to which they were randomized, regardless of whether they completed the assigned intervention, preserves the baseline comparability created by randomization and avoids bias from non-random dropout or crossover.