Participant Recruitment and Retention

Overview

Recruitment is the process by which potentially eligible people move from identification to informed enrolment. Retention concerns continued participation and outcome ascertainment after enrolment. These are related to sampling but are not the same.

A sound sampling design can fail if selected people are never reached, approached inconsistently or burdened by an impractical follow-up schedule. Conversely, an energetic recruitment campaign can rapidly enrol participants whose selection mechanism does not match the intended population.

This lecture presents the recruitment funnel, participant-centred recruitment and retention, operational monitoring, ethical safeguards and protocol adaptation.

Why this matters

Recruitment and retention affect:

• whether the planned sample size is achieved;
• how long the study takes;
• whether important subgroups are included;
• participant burden and trust;
• selection bias;
• missing outcome data;
• study cost; and
• the credibility of implementation claims.

The relevant question is not only “Did the study recruit enough participants?” It is also “Who was never identified, never approached, declined, withdrew or became unreachable, and why?”

Key concepts

The recruitment funnel

The recruitment funnel separates operational stages:

1. potentially eligible;
2. screened;
3. eligible;
4. approached;
5. provided information;
6. consented;
7. enrolled;
8. initiated study procedures; and
9. contributed primary-outcome data.

Each transition has a denominator and a reason for non-progression. Combining stages conceals problems. For example, a low enrolment count can result from poor frame coverage, low eligibility, insufficient staff approach, refusal or delayed consent.

Recruitment versus consent

Recruitment includes identification, approach and enrolment. Informed consent is the ethical process by which a person voluntarily authorises participation after receiving and understanding relevant information. Consent should not be treated as a signature obtained to complete recruitment.

Retention versus adherence

Retention means maintaining appropriate study contact and outcome ascertainment. Adherence refers to following an intervention or study procedure. A participant can discontinue an intervention but remain in follow-up. Preserving follow-up after discontinuation is often crucial for unbiased effect estimation.

Planning recruitment

Identify recruitment pathways

Potential pathways include:

• direct identification from a frame;
• clinical-team referral;
• research registry invitation;
• community outreach;
• public or targeted advertisement;
• mail, telephone or electronic invitation; and
• participant referral.

Each pathway should be mapped from exposure to invitation through consent. Gatekeepers and algorithms can shape who is seen.

Estimate rates at each stage

Suppose the target is 360 participants, 70% of eligible women are expected to consent and 85% of enrolled women are expected to provide the primary outcome. The team must not assume that screening 360 women is sufficient.

Planning should estimate:

• number in the source population;
• frame coverage;
• expected eligibility;
• approach rate;
• consent rate;
• enrolment completion; and
• primary-outcome retention.

These assumptions should come from relevant local data or pilots where possible.

Reduce avoidable barriers

Recruitment may be strengthened by:

• flexible approach times;
• concise, understandable materials;
• language support;
• accessible formats;
• private discussion;
• staff training;
• separation of clinical care from pressure to participate;
• remote or delayed consent where approved;
• reimbursement of reasonable expenses; and
• multiple contact modes.

Strategies should be evaluated for equity. A digital-only process may recruit quickly while excluding people with limited connectivity.

DECISION GUIDE

Diagnose a recruitment shortfall

• If few people enter the frame, examine coverage and referral.
• If many are ineligible, revisit assumptions and criteria.
• If eligible people are not approached, examine workflow and staffing.
• If refusal is high, review burden, trust, timing and information.
• If consented people do not start procedures, examine scheduling and handover.
• If follow-up is poor, review contact quality, visit burden, flexibility and participant experience.
• Do not respond to every shortfall by simply extending recruitment.

Illustrative recruitment dashboard

illustrative

Table 1. Monthly recruitment funnel

Stage	Number	Percentage of previous stage	Operational interpretation
Potentially eligible	160	—	Frame yield
Screened	150	93.8%	Screening coverage
Eligible	120	80.0%	Eligibility yield
Approached	108	90.0%	Approach performance
Consented	82	75.9%	Consent conversion
Enrolled	80	97.6%	Start-up completion
Primary outcome obtained	68	85.0%	Retention

Note: Values are illustrative. Rates should be interpreted with reasons and subgroup patterns.

Participant-centred retention

Retention strategies should preserve autonomy and minimise burden. They may include:

• confirming preferred name, language, contact mode and time;
• obtaining permission for alternative contacts;
• verifying details at each interaction;
• reminders before visits;
• flexible scheduling;
• remote questionnaires or local testing;
• transport or parking support;
• culturally appropriate communication;
• brief newsletters or study updates;
• rapid response to participant questions; and
• tracing procedures proportionate to the study risk and importance.

The protocol should distinguish routine reminders from tracing after missed contact. Excessive messages can become intrusive.

Retention also depends on staff behaviour. Participants are more likely to remain when communication is respectful, appointments start on time and the study keeps its promises.

Monitoring recruitment and retention

PROTOCOL CHECKLIST

Recruitment and retention plan

• recruitment sources and responsible staff;
• approach location, timing and method;
• consent process and accommodations;
• expected numbers at each funnel stage;
• recruitment and retention targets by time period;
• screening and refusal categories;
• contact verification and preference recording;
• reminder and missed-contact procedures;
• reimbursement and participant support;
• withdrawal and intervention-discontinuation pathways;
• subgroup monitoring for equitable inclusion;
• criteria for corrective action; and
• governance process for protocol changes.

A dashboard should show trends, not merely cumulative totals. Overall rates can hide poor recruitment of rural participants, older adults or people using another language. Sites should compare actual with expected rates and investigate bottlenecks.

Adapting the recruitment plan

Recruitment problems are common, but adaptations must preserve scientific and ethical integrity. Before changing the protocol:

1. identify the stage causing the shortfall;
2. review subgroup and site patterns;
3. consult participants or public contributors where appropriate;
4. evaluate effects on the population and estimand;
5. determine whether ethics or governance approval is required;
6. update materials, training and registries; and
7. document the date and rationale.

Adding a new recruitment channel can change the source population. Extending the period can introduce seasonal differences. Widening criteria can change event rates and sample-size assumptions. These are methodological changes, not merely administrative fixes.

Ethical and governance considerations

Interface with statistical analysis

Recruitment determines the baseline sample; retention determines observed outcomes. Analysis planning should include:

• the primary analysis population;
• treatment of intervention discontinuation;
• outcome-specific denominators;
• reasons and timing of missingness;
• repeated attempts and alternative ascertainment sources;
• predictors of non-response;
• weighting or imputation where appropriate; and
• sensitivity analyses for missing-not-at-random mechanisms.

A simple attrition inflation during sample-size planning does not remove bias caused by differential loss. Prevention and outcome ascertainment remain primary.

Embedded recruitment and retention studiesMMed extension

A study within a trial can randomise eligible participants or sites to different recruitment or retention strategies, such as invitation formats or reminder modes. This provides stronger evidence than informal before–after changes. The embedded study requires its own objective, consent and governance considerations, outcomes and analysis plan.

Worked example

WORKED EXAMPLE

Redesigning postpartum follow-up

Problem: Only 69% of enrolled women have a known 12-week testing outcome.

Funnel review: Enrolment is adequate. Missing outcomes are concentrated among rural residents, women returning to work early and those who planned testing outside the hospital.

Adaptations:

• verify two contact modes before discharge;
• record preferred contact time and language;
• accept linked results from district and primary-care laboratories;
• offer a brief remote outcome contact;
• send one pre-due reminder and a prespecified sequence after the due date;
• train staff to continue follow-up after missed visits; and
• monitor retention by hospital, residence and employment return.

Governance: The team updates the protocol and participant materials before implementation.

Interpretation: The strategy addresses the actual retention mechanism rather than simply enrolling more women.

Suggested protocol or reporting wording

Evidence basis and critical appraisal

Systematic evidence reviews identify multiple strategies that may improve recruitment, although effects vary by setting and study design [1]. Retention reviews similarly show that no universal reminder solves all loss to follow-up; strategies must be proportionate and context-sensitive [2]. Earlier evidence highlights the value of contact information, participant relationships, flexibility and tracking systems [3].

Practical trial guidance encourages teams to diagnose recruitment and retention as study processes rather than blaming participants [4]. Current good clinical practice guidance places participant rights, safety, reliable results and proportionate quality management at the centre of conduct [5].

Critical appraisal should assess both quantity and pathway:

• Did investigators report numbers at every funnel stage?
• Were all selected or potentially eligible people screened consistently?
• Were refusal reasons collected without pressuring participants?
• Did recruitment strategies change during the study?
• Did the change alter the source population?
• Were retention procedures applied consistently across prognosis and treatment groups?
• Was outcome collection continued after treatment discontinuation where appropriate?
• Were site and subgroup differences investigated?

Recruitment forecasting

Operational forecasting can be expressed as a chain of proportions. If a site identifies 100 potentially eligible people per month, screens 90%, finds 70% eligible, approaches 90% and enrols 70% of those approached, expected monthly enrolment is:

$$100 \times 0.90 \times 0.70 \times 0.90 \times 0.70 = 39.7$$

The calculation is only as useful as its local assumptions. Forecasts should be updated with blinded operational data and displayed by site. A recruitment shortfall caused by low eligibility requires a different response from one caused by missed approaches.

Retention quality indicators

Retention monitoring should include more than the final percentage. Useful indicators include:

• proportion with verified contact details;
• reminder delivery success;
• completion within the intended window;
• number of attempts before outcome ascertainment;
• use of alternative outcome sources;
• withdrawal from intervention versus withdrawal from follow-up;
• participant complaints or burden signals; and
• retention by site and key population groups.

These indicators identify preventable process failure before primary outcomes become extensively missing.

Participant and public involvement

People with relevant lived experience can identify burdens and communication problems that the research team overlooks. Their contribution may improve invitation timing, consent materials, visit schedules and acceptable contact procedures. Involvement should occur early enough to change the design rather than only endorse a completed plan.

Recruitment and retention materials should be version-controlled so that dates, sites and participant groups exposed to each version can be reconstructed.

This documentation also supports later interpretation of changing recruitment patterns.

Common mistakes

Knowledge check

Why should approach rate be separated from consent rate?

Answer and reasoning

A low approach rate indicates a workflow or staffing problem, whereas a low consent rate may reflect burden, trust, timing or study acceptability. Combining them prevents targeted correction.

Does stopping an assigned intervention require stopping outcome follow-up?

Answer and reasoning

Not necessarily. If the participant agrees and the protocol permits, continued outcome ascertainment is valuable and may be required for the primary estimand.

Why is replacing lost participants insufficient?

Answer and reasoning

Replacement may restore the nominal sample size but does not remove selection bias if loss is related to outcomes. It also cannot recover longitudinal information for those already lost.

What is a participant-centred retention strategy?

Answer and reasoning

A strategy that reduces avoidable burden and respects preferences and autonomy, such as flexible scheduling, approved remote options and clear, limited contact procedures, while preserving valid outcome ascertainment.

Key take-home messages

SUMMARY

Key take-home messages

• Separate identification, screening, approach, consent, enrolment and outcome retention.
• Monitor conversion and reasons at every recruitment-funnel stage.
• Design retention before enrolment and make procedures participant-centred.
• Continue appropriate follow-up after intervention discontinuation where consent permits.
• Investigate subgroup patterns rather than relying on one overall rate.
• Treat recruitment adaptations as methodological changes requiring review and documentation.

References

1. Treweek S, Pitkethly M, Cook J, Fraser C, Mitchell E, Sullivan F, et al. Strategies to improve recruitment to randomised trials. Cochrane Database Syst Rev. 2018;2(2):MR000013. https://doi.org/10.1002/14651858.MR000013.pub6
2. Gillies K, Kearney A, Keenan C, Treweek S, Hudson J, Brueton VC, et al. Strategies to improve retention in randomised trials. Cochrane Database Syst Rev. 2021;3(3):MR000032. https://doi.org/10.1002/14651858.MR000032.pub3
3. Robinson KA, Dennison CR, Wayman DM, Pronovost PJ, Needham DM. Systematic review identifies number of strategies important for retaining study participants. J Clin Epidemiol. 2007;60(8):757–765. https://doi.org/10.1016/j.jclinepi.2006.11.023
4. Murphy E, Shiely F, Treweek S. Recruitment and retention in clinical trials: a practical guide. Trials. 2023;24:756. https://doi.org/10.1186/s13063-023-07773-4
5. International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use. ICH harmonised guideline E6(R3): guideline for good clinical practice. Geneva: ICH; 2025. https://www.ich.org/page/efficacy-guidelines