Feasibility Study

A feasibility study is a preliminary investigation conducted to evaluate whether a proposed research project, clinical trial, or intervention can be carried out as planned. It focuses on practical aspects such as participant recruitment, data collection procedures, resource requirements, and the acceptability of the intervention to participants and implementers, rather than on estimating treatment effects.

Many research projects fail not because the hypothesis is wrong but because the study cannot be executed as designed. Recruitment targets may be unrealistic, procedures may be too burdensome for participants, instruments may not work in the target population, or the intervention may not be deliverable as planned. Feasibility studies identify these problems early, before significant resources are invested in a full scale trial. They save time, money, and effort by revealing practical barriers that can be addressed in the planning stage. Funding agencies (such as the NIHR in the UK and the NIH in the US) often look for evidence of feasibility before funding large scale trials, and some have dedicated funding streams specifically for feasibility and pilot studies.

A feasibility study addresses specific practical questions rather than testing a hypothesis about effectiveness. The questions typically include: Can enough participants be recruited within the planned timeframe? Will participants consent and complete the study procedures? Is the intervention deliverable as designed? Are the outcome measures acceptable and practical to administer? Are the data collection systems functional? What are the resource requirements in terms of staff time, equipment, and costs?

Feasibility studies are often conducted without a control group, since the purpose is not to estimate a treatment effect but to assess whether the trial machinery works. When the uncertainty includes randomization itself, such as whether people will accept being randomized or whether the control condition is acceptable, a randomized pilot trial with a control group is used instead; it is still not designed to estimate the treatment effect. Sample sizes are typically small (often 20 to 50 participants) and are justified based on the precision needed for feasibility estimates rather than on power calculations for hypothesis testing.

The findings are reported as rates and proportions (such as recruitment rate per month, retention rate, completion rate for outcome measures) rather than as effect sizes. Qualitative data from interviews with participants and staff can supplement quantitative feasibility data by revealing the reasons behind any problems. The results inform the design of the definitive trial, including modifications to the protocol, recruitment strategy, outcome measures, or intervention delivery.

The terms feasibility study and pilot study have been used in different and overlapping ways. One older usage treats a pilot study as a smaller version of the planned main study, using similar methods and procedures, and a feasibility study as work on specific aspects of feasibility without running a miniature version of the full trial. The conceptual framework of Eldridge and colleagues (2016) instead uses feasibility study as the overarching term: a feasibility study asks whether something can be done, whether to proceed with it, and if so how, and pilot studies are a subset of feasibility studies in which a future study, or part of it, is conducted on a smaller scale. Pilot studies may be randomized or non randomized, and some feasibility studies are not pilot studies at all, such as a study that asks clinicians whether they would be willing to recruit patients.

Randomized pilot and feasibility trials are reported using the CONSORT extension for randomised pilot and feasibility trials (2016). It asks authors to report any prespecified criteria used to judge whether, or how, to proceed to the definitive trial, and states that formal hypothesis testing for effectiveness is not recommended, because a pilot trial does not aim to assess effectiveness and is usually underpowered to do so.

A feasibility study is appropriate before any large scale trial or research project that involves novel interventions, unfamiliar populations, complex procedures, or untested measurement instruments. It is particularly valuable when the research team is uncertain about recruitment rates, participant willingness to comply with the protocol, or the practicality of the intervention in real world settings. It should be conducted early enough that its findings can inform the design of the full study.

A common mistake is treating a feasibility study as a small efficacy trial, reporting effect sizes and drawing conclusions about whether the intervention "works." Feasibility studies are not designed or powered to test effectiveness, and preliminary effect sizes from small samples are unreliable estimates of the true effect, so they should not be used to set the main trial's sample size; that calculation should rest on the smallest effect that would be clinically important. Another error is failing to specify feasibility criteria in advance (such as "recruitment is feasible if at least 70% of eligible patients consent"). These are often called progression criteria, and they can set separate thresholds for proceeding unchanged, proceeding with changes, and stopping. Without predefined criteria, the interpretation of feasibility data becomes subjective. Some researchers also skip feasibility studies to save time, only to encounter insurmountable problems in the main trial that a feasibility study would have identified.

Before launching a 500 participant trial of a mindfulness app for chronic pain, a research team conducts a feasibility study with 30 patients from two pain clinics. Recruitment takes 3.5 months, and each patient uses the app for 8 weeks. Against criteria set in advance, they assess recruitment (4.3 patients per clinic per month, against a target of 5), retention (87% completed the follow up, against a target of 80%), app usage (participants used the app a median of 4.1 days per week, against a target of 3), and outcome measure completion (93% completed all questionnaires). The feasibility study reveals that one outcome measure takes too long to administer and is replaced before the main trial, and because recruitment fell short of its target, the team plans to recruit from more clinics in the main trial.