Pilot Randomized Controlled Trial — Feasibility-First Experimental Design
Pilot Randomized Controlled Trial · Also known as: pilot RCT, feasibility RCT, pilot trial, preliminary RCT
A pilot randomized controlled trial (pilot RCT) is a small-scale, fully randomized experiment conducted before a definitive RCT to test the feasibility of study procedures, estimate key parameters such as recruitment rates and effect-size variability, and identify practical barriers. It uses the same randomization, intervention, and measurement protocol as the planned full trial but on a fraction of the target sample. The goal is not to confirm efficacy but to refine and justify the main trial design.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
+4 more
When to use it
Use a pilot RCT when the definitive trial would be large or costly and important logistical or methodological uncertainties remain — unknown recruitment rates, untested outcome measures, unverified intervention fidelity, or no reliable variance estimate for sample-size calculations. It is appropriate in early-phase clinical, educational, or behavioral intervention research where the evidence base is thin. Do not use a pilot RCT as a substitute for a definitive trial: pilot data alone cannot establish efficacy because the study is not powered for that purpose. Avoid when a full systematic review already provides reliable parameter estimates, or when stakeholders will misinterpret a non-significant pilot result as evidence that the intervention does not work.
Strengths & limitations
- Identifies fatal flaws in recruitment, retention, or protocol fidelity before they compromise an expensive definitive trial.
- Provides empirical variance estimates that make sample-size calculations for the main trial more accurate than assumptions alone.
- Tests the randomization procedure and blinding mechanisms under operational conditions.
- Allows refinement of the intervention, outcome measures, and data-collection procedures based on real participant feedback.
- Ethical: exposes fewer participants to an unproven intervention before efficacy is established.
- Generates preliminary data that can support grant applications for the definitive trial.
- Not powered to detect treatment effects — a non-significant result does not mean the intervention is ineffective.
- Small sample sizes produce imprecise and often inflated effect-size estimates; using the pilot effect size to power the main trial is a known methodological error.
- Publication bias: negative or inconclusive pilots are frequently unpublished, distorting the evidence base.
- Adds time and cost to the overall research programme before the definitive trial begins.
- Volunteer and Hawthorne effects may be more pronounced in small pilot samples, making feasibility estimates overly optimistic.
Frequently asked
What is the difference between a pilot RCT and a feasibility study?
The terms are often used interchangeably but have a technical distinction. A pilot RCT is a miniaturized version of the planned definitive trial, run under the exact same protocol. A feasibility study is broader — it may not include randomization and can address questions such as whether the intervention can be delivered at all, whether the target population exists, or whether stakeholders will accept the study. All pilot RCTs address feasibility, but not all feasibility studies are pilot RCTs.
Can I report efficacy results from a pilot RCT?
You can report preliminary estimates, but they must be clearly labeled as exploratory and interpreted with extreme caution. The study is not powered to confirm or refute efficacy, so p-values for the primary outcome are uninformative. The CONSORT extension for pilot trials recommends reporting confidence intervals rather than p-values, and explicitly stating that efficacy conclusions cannot be drawn.
How large should a pilot RCT be?
There is no universal rule, but common pragmatic recommendations range from 30 to 70 total participants (15–35 per arm). The sample should be large enough to estimate the key feasibility metrics with acceptable precision — typically meaning that 95% confidence intervals around recruitment and retention rates are narrow enough to make a credible go/modify/stop decision. Formal sample-size calculations for pilots based on desired precision of variance estimates are possible and recommended.
Should pilot RCT participants be excluded from the main trial?
Excluding pilot participants from the main trial (an external pilot) avoids contamination of the definitive sample but increases total recruitment burden. An internal pilot design includes pilot participants in the main analysis, which is more efficient but requires that the protocol not change after the pilot phase. The choice should be pre-specified in the protocol and reported in the pilot publication.
What reporting standard applies to pilot RCTs?
The CONSORT 2010 extension for randomised pilot and feasibility trials (Eldridge et al., 2016, Pilot and Feasibility Studies journal) provides specific reporting items for pilot studies. Key additions include explicit reporting of pre-specified feasibility criteria and the decision made on the basis of the pilot results.
Sources
- Thabane, L., Ma, J., Chu, R., Cheng, J., Ismaila, A., Rios, L. P., ... & Goldsmith, C. H. (2010). A tutorial on pilot studies: the what, why and how. BMC Medical Research Methodology, 10(1), 1. DOI: 10.1186/1471-2288-10-1 ↗
- Lancaster, G. A., Dodd, S., & Williamson, P. R. (2004). Design and analysis of pilot studies: recommendations for good practice. Journal of Evaluation in Clinical Practice, 10(2), 307-312. link ↗
How to cite this page
ScholarGate. (2026, June 3). Pilot Randomized Controlled Trial. ScholarGate. https://scholargate.app/en/experimental-design/pilot-randomized-controlled-trial
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Adaptive Randomized Controlled TrialExperimental design↔ compare
- Cluster Randomized Controlled TrialExperimental design↔ compare
- Crossover Randomized Controlled TrialExperimental design↔ compare
- Factorial Randomized Controlled TrialExperimental design↔ compare
- Randomized Controlled TrialExperimental design↔ compare