Should Eligibility Criteria Be Changed if Recruitment Is Slower Than Expected?

Should You Relax Your Eligibility Criteria When You Cannot Recruit Enough Participants?

Suppose you are conducting a randomized study evaluating an AI-supported academic writing intervention. Your protocol requires participants to be second-year undergraduate students enrolled in a particular writing course, with no previous experience using generative AI.

You planned to recruit 200 students within six weeks. After four weeks, only 35 participants have enrolled.

You consider expanding eligibility to first-year and third-year students, removing the restriction on previous AI experience, or allowing students from other academic programs to participate.

Would these changes solve the recruitment problem, or would they fundamentally alter the research question?

Similar difficulties arise in clinical trials, longitudinal studies, and qualitative research. Recruitment may be slower because eligibility criteria are unnecessarily restrictive, but it may also reflect limited access, inconvenient procedures, low participation willingness, or unrealistic feasibility assumptions.

The methodological challenge is determining whether eligibility criteria genuinely need revision or whether the recruitment strategy should be improved without changing the population the study was designed to investigate.

Sometimes, but Slow Recruitment Alone Is Not Sufficient Justification

How to Respond to Recruitment Difficulties Without Compromising the Study

Why Is Slow Recruitment Not Automatically an Eligibility Problem?

Recruitment depends on more than the proportion of people who satisfy inclusion and exclusion criteria.

A study may have a large eligible population but recruit slowly because potential participants never receive invitations, institutional permissions are delayed, or participation requires excessive time.

For example, students may qualify for an educational intervention but decline because sessions conflict with examinations or other academic responsibilities.

Changing eligibility would not necessarily address those barriers.

Researchers should distinguish the availability of eligible participants from their accessibility, willingness to participate, and ability to complete recruitment procedures.

The appropriate remedy depends on where participants are being lost.

How Can You Identify the Actual Recruitment Problem?

A recruitment funnel can help researchers distinguish inadequate population access from restrictive eligibility or low participation.

At minimum, the team should examine how many potential participants were identified, contacted, screened, found eligible, invited to enroll, and successfully recruited.

These quantities answer different questions.

A low screening eligibility rate may suggest restrictive criteria, but it may also indicate that recruitment is reaching the wrong population. The distinction should be investigated before modifying the protocol.

What Recruitment Metrics Should You Monitor?

Recruitment monitoring should distinguish eligibility from participation.

Useful descriptive quantities include the screening eligibility proportion, enrollment proportion among eligible candidates, and recruitment rate over time.

These measures require clearly defined denominators. For example, the enrollment proportion should specify whether all eligible candidates were actually invited and whether their enrollment decisions were known.

Recruitment rates may vary across academic calendars, clinical service schedules, and institutional approval periods.

Researchers should compare observed recruitment with realistic projections rather than treating the original schedule as proof that eligibility is defective.

When Might Eligibility Criteria Be Unnecessarily Restrictive?

Criteria may be unnecessarily restrictive when they exclude people who belong to the intended population without providing a sufficient scientific, methodological, or ethical benefit.

For example, requiring previous generative AI experience may be inappropriate for an intervention intended for both novice and experienced users.

Similarly, requiring personal laptop ownership may be unnecessary if institutional devices can support the intervention.

In clinical research, some exclusions involving stable comorbidities or concurrent treatments may warrant review when they do not create meaningful safety or interpretive concerns.

However, the existence of a restrictive criterion does not establish that removing it is appropriate.

Researchers should evaluate whether the combined eligibility restrictions unnecessarily narrow the study population.

Which Eligibility Criteria Should Not Be Relaxed Merely to Improve Recruitment?

Criteria necessary for participant safety, valid intervention delivery, or the scientific definition of the population should not be relaxed solely to meet enrollment targets.

For example, a clinical contraindication should not be disregarded because the study is behind schedule.

Similarly, including participants for whom an educational intervention or outcome has no meaningful interpretation may undermine the research question.

Some criteria are essential to the intended estimand, while others may be operational restrictions that can be modified.

The research team should distinguish these categories before considering changes.

Any safety-related revision requires appropriate evidence and oversight rather than a simple recruitment decision.

What Alternatives Should Be Considered Before Changing Eligibility?

Researchers should first consider whether recruitment can be improved while preserving the original population definition.

Possible approaches include additional approved study sites, better communication with eligible participants, more accessible recruitment materials, revised scheduling, and reduced participation burden.

For example, an educational study may recruit more effectively by avoiding examination periods or offering several approved assessment schedules.

Clinical studies may benefit from improving screening procedures or coordinating recruitment across suitable care settings.

These changes may require protocol or ethics review depending on their nature and the applicable requirements.

Recruitment improvements should not introduce coercion, inappropriate inducements, or unequal access to participation.

Can Additional Study Sites Solve Slow Recruitment?

Additional sites may increase the accessible population without changing the substantive eligibility criteria.

For example, recruiting students from another university offering a comparable writing course may increase enrollment while preserving the intended educational population.

However, new sites can introduce differences in curriculum, institutional resources, intervention delivery, or outcome assessment.

Researchers should evaluate whether these differences affect the estimand or require design and analytical changes.

Site expansion is therefore not methodologically neutral, even when eligibility rules remain unchanged.

The selection of additional sites should be justified rather than based exclusively on convenience.

When Is Expanding Eligibility Scientifically Defensible?

Eligibility expansion may be defensible when the original restrictions were narrower than the research question required.

For example, an intervention designed for undergraduate academic writing may not require participants to be exclusively second-year students.

Including other undergraduate year levels could remain consistent with the intended population if the intervention and outcome are meaningful across those levels.

However, the expansion may introduce different baseline characteristics, writing tasks, or treatment responses.

Researchers should determine whether the original and newly eligible participants belong to a coherent population and whether the analysis can accommodate relevant variation.

Eligibility expansion should follow substantive reasoning rather than an assumption that broader criteria are always preferable.

Can Eligibility Be Narrowed During Slow Recruitment?

Narrowing eligibility would generally reduce the number of potential participants and is unlikely to solve a simple shortage of eligible candidates.

However, narrowing may occasionally improve recruitment efficiency when the original population was poorly defined and recruitment resources were being directed toward unsuitable candidates.

For example, a study requiring a specific clinical intervention might focus screening on patients for whom the treatment is actually indicated.

Such a change should be justified by the research question and feasibility evidence.

It should not be confused with improving the number of eligible participants by relaxing criteria.

How Can Eligibility Changes Alter the Target Population?

Eligibility criteria determine which individuals can enter the study.

Changing them may expand or narrow the population represented by the resulting evidence.

For example, changing an intervention study from first-year students to all undergraduate students may alter the educational population and potentially the intended treatment-effect estimand.

Removing a prior-experience restriction may similarly introduce participants with different baseline characteristics and intervention responses.

However, not every operational change necessarily alters the substantive target population.

Researchers should distinguish correcting an unnecessarily restrictive eligibility rule from deliberately redefining the population.

The consequences of eligibility changes for the target population require explicit evaluation.

What Happens When Participants Are Recruited Under Different Eligibility Rules?

When criteria change during recruitment, participants enrolled before and after the amendment may differ systematically.

For example, early participants may all be novice AI users, while later participants include experienced users.

Differences may also arise because recruitment periods correspond to different academic conditions, institutional settings, or clinical practices.

Researchers should document the eligibility version and recruitment period associated with each participant.

Where relevant, analyses may examine whether findings are sensitive to recruitment period or eligibility version.

However, post-amendment subgroup analyses should not automatically be interpreted causally, because eligibility version and calendar time may be confounded.

Should Previously Excluded Individuals Be Reconsidered?

Sometimes, when revised criteria make previously excluded individuals eligible and recontact is ethically and operationally permissible.

However, researchers should not assume that all previously screened individuals can be contacted again.

Consent, privacy permissions, record retention arrangements, and recruitment procedures may restrict recontact.

If recontact is allowed, a consistent procedure should be applied to relevant candidates rather than selectively approaching those expected to produce favorable outcomes.

Researchers should document whether earlier screening decisions were reconsidered and how this affected recruitment.

Do Eligibility Changes Require Research Ethics Approval?

Substantive changes to approved eligibility criteria generally require consideration under the applicable ethics review and protocol amendment procedures.

The exact requirements depend on the jurisdiction, institution, study type, and governing regulations.

Changes affecting participant safety, consent, intervention exposure, or vulnerable populations may require particularly careful review.

Researchers should not assume that an amendment is exempt from review because it is intended to improve recruitment.

Where prospective approval is required, revised criteria should not be implemented before authorization is obtained.

Urgent changes necessary to eliminate an immediate hazard may be subject to specific exceptions and reporting requirements under applicable rules.

Should Clinical Trial Registrations Be Updated?

When eligibility changes affect registered study information, the relevant registration record should be updated according to the registry's requirements.

For example, ClinicalTrials.gov maintains requirements and procedures for updating registered study information.

Researchers should distinguish the original registration, approved protocol amendments, and the current study design.

Registration updates do not substitute for required ethics approval.

Likewise, ethics approval does not automatically update a public trial registration.

Transparent documentation helps readers identify whether the published study followed its original or amended eligibility framework.

How Do Eligibility Changes Affect Sample Size and Statistical Power?

Eligibility changes can alter expected outcome variability, event rates, intervention effects, and participant availability.

For example, including students from additional year levels may increase recruitment but introduce greater variation in baseline writing proficiency.

The resulting effect on statistical power depends on the estimand and analysis.

Researchers should reconsider planning assumptions when the population changes materially.

However, sample size revisions should not be driven by unblinded outcome comparisons without an appropriately justified design.

Where relevant, an independent statistical review or a prespecified blinded sample size reassessment may be appropriate.

Can Recruitment-Driven Eligibility Changes Introduce Bias?

Yes, depending on the timing, decision process, and relationship between eligibility and outcomes.

For example, selectively relaxing criteria for participants expected to respond well to an intervention may distort the sample.

Changing criteria after examining treatment outcomes creates additional concerns about outcome-driven design decisions.

Even when changes are made without outcome knowledge, combining participants recruited under different rules may complicate interpretation.

Researchers should establish a transparent rationale, document the timing, and avoid using observed effects to determine eligibility.

The consequences should be evaluated according to the intended estimand rather than assuming every amendment is biased.

What If the Recruitment Target Was Unrealistic?

Slow recruitment may reflect an overly optimistic feasibility estimate rather than inappropriate eligibility criteria.

For example, the original plan may have assumed that most eligible students would consent despite substantial time commitments.

Researchers should revisit the assumptions used to estimate the eligible population, participation rate, and recruitment duration.

Possible responses include extending recruitment, adding appropriate sites, reducing unnecessary participation burden, or reconsidering the study's feasibility.

However, reducing the sample size without evaluating statistical precision or power may leave the study unable to answer its principal question adequately.

Recruitment feasibility and scientific adequacy should be assessed together.

When Should Recruitment Be Extended Instead of Changing Eligibility?

Extending recruitment may be preferable when eligibility criteria remain scientifically appropriate and the recruitment rate suggests the target can be reached with additional time.

For example, a study recruiting students during an examination period may improve enrollment after academic schedules become less demanding.

However, extending recruitment can introduce temporal differences in participants or study conditions.

In rapidly changing educational technology settings, intervention tools or institutional policies may evolve during a prolonged recruitment period.

Researchers should evaluate whether extending the study affects the relevance or comparability of observations.

Required administrative, funding, ethics, and registration extensions should also be considered.

What If Eligibility Criteria Are Scientifically Necessary but Recruitment Remains Inadequate?

Necessary eligibility criteria should not be abandoned simply because recruitment is difficult.

If an intervention is appropriate only for a specific population, broadening eligibility beyond that population may make the study scientifically incoherent or unsafe.

Researchers may need to expand suitable sites, extend recruitment, revise the study design, or reconsider whether the project is feasible.

In some circumstances, stopping an inadequately recruitable study may be more defensible than producing an underpowered or poorly aligned investigation.

Such decisions should consider the scientific value of the available data, participant commitments, ethical obligations, and applicable oversight requirements.

How Should Eligibility Amendments Be Documented?

Documentation should identify the original criterion, revised wording, reason for the change, date of implementation, and applicable approvals.

Researchers should explain whether the amendment changes the target population, intervention suitability, measurement, or estimand.

Screening and enrollment records should identify which eligibility rules applied to each participant.

Where appropriate, statistical analysis plans should be updated before relevant outcome information is examined.

Changes should be reported transparently in the final publication rather than presented as though the revised criteria had always been in place.

How Should Recruitment Difficulties Be Reported?

Methods sections should describe the recruitment period, study sites, eligibility criteria, and substantive amendments.

Participant flow should distinguish individuals screened, excluded, eligible, enrolled, and analyzed, as appropriate to the design.

Explain important changes to recruitment channels or eligibility and their implications for population coverage.

For randomized trials, applicable CONSORT guidance supports transparent reporting of eligibility, recruitment, and participant flow.

For observational research, STROBE provides relevant reporting guidance.

Readers should be able to understand how the final study population was assembled and whether recruitment difficulties influenced its composition.

Deciding Whether to Broaden Eligibility in an AI Writing Intervention Trial

Suppose the team concludes that previous AI experience was unnecessarily restrictive because the intervention is intended for all eligible undergraduate writing students.

Removing that restriction may be scientifically defensible, provided the intervention and outcomes remain appropriate and the necessary amendment procedures are followed.

However, extending eligibility to postgraduate students might require a more substantial reconsideration because their writing tasks and educational context may differ.

The appropriate change is therefore not necessarily the one that produces the greatest number of additional participants.

It is the change that improves recruitment while preserving a coherent and ethically defensible research question.

Common Misconceptions About Changing Eligibility During Recruitment

How to Decide Whether to Change Eligibility Criteria

Begin by diagnosing the recruitment bottleneck and determining whether the original criteria remain scientifically justified.

The objective is to address the reason recruitment is failing, not simply to increase the number of people classified as eligible.

What to Verify Before Changing Eligibility Because of Slow Recruitment

Frequently Asked Questions About Changing Eligibility During Recruitment

Change Eligibility Only When the Research Justifies It

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.