Advancing Clinical Trials Worldwide

We are committed to strengthening Clinical Trials, as a United Nations Vendor aligned with the World Health Organization’s “Joint statement on strengthening clinical trials 2025.”

Please contact us if you are interested in joining or funding a trial.

About this diagram: Randomized Controlled Trial Design

The randomized controlled trial is the primary design for establishing causal efficacy of an intervention. Random allocation of participants to intervention and control groups is the critical mechanism: it distributes both known and unknown confounders equally between arms in expectation, so that differences in outcomes can be attributed to the intervention rather than pre-existing differences between groups. Block and stratified randomization ensure balance on key prognostic factors (age, disease severity) even in moderate sample sizes of 200 to 1,000 participants per arm. Allocation concealment (hiding the allocation sequence from those enrolling participants) and blinding (masking arm assignment from participants and outcome assessors) are additional safeguards against selection bias and ascertainment bias respectively. A double-blind trial, where neither the participant nor the clinician knows the allocation, minimises placebo effects and assessment bias simultaneously.

Intention-to-treat (ITT) analysis is the pre-specified primary analytic strategy in well-designed RCTs: all participants are analysed in the arm to which they were randomized, regardless of whether they actually received the treatment (crossovers, non-compliers, dropouts). ITT preserves the comparability guaranteed by randomization; a per-protocol analysis that excludes non-compliers can reintroduce confounding because compliance is itself associated with prognosis. The CONSORT reporting standard (Consolidated Standards of Reporting Trials, widely adopted since 1996) requires a participant flow diagram showing screening, randomization, and completion numbers, which is the direct visual origin of this diagram. Sample size calculations for RCTs are driven by the expected effect size, desired power (typically 80 to 90%), and significance threshold (alpha = 0.05); for a drug reducing a binary event rate from 10% to 7%, approximately 900 participants per arm are needed for 80% power, illustrating why phase III trials routinely cost tens of millions of dollars.

About this diagram: Crossover RCT Schematic

A crossover RCT randomises each participant to a treatment sequence rather than to a single arm. In the canonical 2×2 design, sequence 1 receives drug A for a defined period, then a washout, then drug B for an equal period; sequence 2 receives B then A. Because each subject experiences both treatments, the within-subject difference (response on A minus response on B) becomes the unit of analysis. Between-subject variance, the dominant source of noise in most clinical outcomes, is removed by this pairing, which typically allows the crossover to detect the same effect with roughly half the sample size of an equivalent parallel-arm trial.

The price for this efficiency is carryover risk. If drug A is still pharmacologically active or has produced lasting physiological change when period 2 begins, the period-2 measurement contaminates the comparison. Practical defenses are a washout of at least five half-lives of the longer-acting agent, balanced randomisation to both sequences so any residual carryover affects both arms symmetrically, and a formal sequence-by-period interaction test in the analysis. Crossover designs suit chronic stable conditions and short-acting reversible interventions (analgesics, antihypertensives, bronchodilators) and are inappropriate for cures, irreversible outcomes such as mortality or surgery, and disease states that drift rapidly over the trial window.

About this diagram: Parallel-Arm RCT Schematic

The parallel-arm RCT is the canonical experimental design in clinical research. An eligible population is enrolled, assessed for inclusion criteria, and then randomly allocated to one of two arms: intervention or control. Both arms are followed forward in time under identical conditions apart from the treatment itself, and the same outcome is measured in both groups at the same time points. The difference in outcomes is attributed to the intervention because random allocation distributes both measured and unmeasured confounders evenly across arms in expectation.

The strength of the design rests on three operational details. Allocation concealment prevents recruiters from steering particular participants into a preferred arm. Blinding of participants, clinicians, and outcome assessors prevents expectation effects from biasing the measured outcome. Intention-to-treat analysis, which counts every randomized participant in their assigned arm regardless of adherence or crossover, preserves the balance that randomization established. When any of these break down the trial can still be informative but the causal claim weakens. A well-conducted parallel-arm RCT with adequate power sits near the top of the evidence hierarchy because no other design rules out confounding so cleanly.