Clinical trials are the cornerstone of evidence-based medicine, yet a remarkable proportion of them fail not because the science is wrong, but because participants do not stay.
Retention failure, the attrition of enrolled subjects before a trial concludes, is one of the most persistent and costly problems in clinical research. Analyses of large trial registries consistently show that poor patient accrual is the leading cause of premature trial termination, accounting for roughly one-quarter to more than half of discontinued trials depending on therapeutic area [1] [2] [3].
Even trials that successfully reach their enrollment targets face a separate and underappreciated challenge of substantial medication discontinuation among enrolled participants. This problem is not confined to any single disease area: discontinuation rates among enrolled patients range from approximately 23–29% in completed cardiovascular outcome trials of shorter duration [4] [5] to 47–51% in longer cardiovascular trials [6] [7] to 74% in major psychiatric trials [8]. These rates persist regardless of whether the trial met its enrollment goals.
Despite the scale of this problem, retention has historically been addressed as an operational challenge such as a scheduling issue, a reminder call, or a small stipend adjustment. The emerging research tells a different story. Retention is fundamentally an experience problem. Participants who feel respected, informed, and genuinely cared for by trial staff stay. Those who feel like data points do not.
This white paper synthesizes current evidence on why subjects leave clinical trials and proposes actionable, evidence-informed strategies to improve retention by treating the subject experience as a core design priority rather than an afterthought.
1. The Scale of the Problem
The operational and scientific costs of retention failure are substantial. In a systematic analysis of over 88,000 clinical trials registered on ClinicalTrials.gov, more than 11% of nonsurgical trials and nearly 16% of surgical trials were prematurely discontinued with poor recruitment and dropout as the leading reasons cited [2].
The pattern is not confined to US trials. UK investigators synthesized evidence on retention strategies and documented attrition as a routine feature of trial conduct, with dropout rates varying widely by disease area, trial duration, and follow-up demands [9]. The central finding is striking: despite decades of effort, the evidence base for interventions that reliably reduce participant dropout remains thin, and the problem is structurally similar across health systems.
In long-duration trials, medication discontinuation compounds the problem even when subjects nominally remain enrolled. A 2019 analysis of IMPROVE-IT cardiovascular outcomes trial found that 46.7% of participants had discontinued study medication before the trial ended with the highest rates observed in the United States, and among women, non-white participants, and smokers.
This was substantiated in an analysis of the TIMI Study Group trial database of 11 Phase 3 & 4 cardiovascular outcome trials published after 2000 with a pooled population of >187,000 patients. These patterns are not random; they reflect structural inequities in who bears the burden of trial participation and who receives the least support during it.
Beyond the lost science, premature discontinuation carries ethical weight. When participants who have already accepted personal risk and inconvenience leave before a trial concludes, their contribution cannot be fully utilized, and the research question often remains unanswered [3]. This represents harm to individual participants and a waste of public and private investment that the research community has been slow to take seriously.
Retention is fundamentally an experience problem. Participants who feel respected, informed, and genuinely cared for by trial staff stay. Those who feel like data points do not.
2. Why Participants Leave
Qualitative and quantitative research has now produced a reasonably clear picture of the factors that drive participants to withdraw from clinical trials. These factors operate across three domains: design, logistics, and relationships.
2.1 Design-Level Factors
Many retention problems are built into trials before a single participant is enrolled. Eligibility criteria that are too narrow reduce the pool to a highly selected population that may not be representative of real-world patients with the disease.
Overoptimistic recruitment estimates, a frequently cited finding in qualitative studies of discontinued trials, lead to trials that are understaffed, underfunded, and ill-prepared to support participants through the full duration. Complex protocols with excessive visit requirements, invasive procedures, or demanding time commitments are consistently identified as deterrents to both initial enrollment and continued participation [3].
An underappreciated design failure is the absence of patient input in protocol development. Stakeholder interview studies have identified lack of patient engagement in trial design as a distinct, named cause of recruitment and retention failure that does not appear in published trial reports [3]. Participants who are enrolled into trials they find incomprehensible, inconvenient, or disconnected from their actual disease experience are more likely to disengage.
2.2 Logistical and Burden-Related Factors
The practical demands of participation including travel to study sites, parking costs, time away from work or family, and the need to plan months in advance around inflexible appointment schedules are among the most frequently cited withdrawal reasons across disease areas [3] [9].
These burdens fall disproportionately on participants who are already disadvantaged: those in lower income brackets, those without flexible employment, those with caregiving responsibilities, and those living in rural or underserved areas [10].
The composition of this burden differs across health systems. In jurisdictions with universal healthcare coverage (most European countries), participants are typically shielded from direct medical and procedural costs associated with trial visits because those costs are absorbed by the health system.
In the United States, trial-specific procedures not covered by insurance or government programmes are typically covered by the sponsor, which similarly shields participants from direct out-of-pocket clinical expenditure. The meaningful difference between these two contexts is therefore not in direct medical cost but in indirect cost exposure (lost wages, childcare, eldercare, and practical consequences of attending visits).
In the US, this burden falls disproportionately on non-exempt workers, who do not receive paid time off for medical appointments and for whom each trial visit represents a measurable income loss [11].
In universal-coverage European systems the same indirect costs exist but are partially mitigated by stronger statutory sick-leave entitlements and, in some countries, a cultural and legal expectation that medical appointments are accommodated by employers, so the burden profile is dominated by time, travel, and indirect costs rather than out-of-pocket clinical expenditure [12].
This distinction matters when designing burden assessments and reimbursement strategies for multi-regional trials: a single global compensation model risks over-compensating in some regions and under-compensating in others and is rarely the right answer.
Participants consistently report that they experienced more burden than was communicated to them at the time of informed consent [4]. This expectation gap between what consent documents describe and what participation actually requires erodes trust and creates conditions for withdrawal, particularly in the early months of a trial. While a consent form may describe the visits and procedures to be performed, it rarely details the estimated time for completion of each visit [13] [14].
2.3 Relationship and Communication Factors
The most consistent predictor of retention across disease areas is the quality of the relationship between participants and study staff. Participants who received proactive check-in calls, had direct contact numbers for study personnel, and felt genuinely believed when they reported side effects or concerns were far more likely to continue [3]. Conversely, participants who felt dismissed, rushed, or treated as sources of data rather than people showed higher rates of disengagement.
Post-trial communication is also a significant factor. Participants who were not informed of trial results after completion reported feeling devalued and were less willing to consider future trial participation [3]. The absence of feedback about their own data, the trial’s outcomes, and what their contribution produced is experienced as a form of abandonment by many participants.
3. A Critical Gap: No One Is Measuring the Experience
One of the most striking findings from participant focus groups is that the vast majority of trial participants have never been asked about their experience of being in a trial. Not before, not during, and often not after [3].
The research community has invested heavily in measuring clinical endpoints on whether the drug works but has built almost no infrastructure for measuring whether the trial process itself is working for the people inside it.
This matters because patient-reported outcome (PRO) measures, commonly cited as evidence of patient-centeredness in trial design, measure something fundamentally different from participation experience. PROs capture how a patient’s disease or treatment affects their health and functioning. They do not capture how the patient feels about the consent process, whether they feel respected by staff, whether their out-of-pocket costs were covered, or whether they would enroll again. These are distinct constructs that require distinct measurement tools that are rarely deployed and have not been standardized [3].
Ethics oversight bodies, Institutional Review Boards (IRBs) in the US and Research Ethics Committees (RECs) in Europe and the UK, are charged with protecting participant welfare but are structurally limited in their ability to close this gap. Their oversight has historically been front-loaded at the protocol approval stage, document-based, and reliant on investigator self-reporting [15].
Regulation (EU) No 536/2014 and the associated Clinical Trials Information System, which became fully mandatory in 2025, have introduced more coordinated lifecycle oversight in Europe, including harmonized safety reporting and structured transparency obligations [16].
Even within this strengthened framework, however, no jurisdiction systematically requires ECs to interview participants, audit whether compensation adequately covers actual costs [17] or assess how the experience of participation is evolving during trial conduct. The measurement gap is therefore a global one, even where the regulatory architecture around it has improved.
4. Evidence-Informed Solutions
The evidence base for effective retention interventions is growing, though it remains less robust than the evidence base describing the problem. The following strategies reflect the strongest available signals.
4.1 Design Trials with Participants, Not Just for Them
Formal patient advisory panels, convened during protocol development and empowered to influence visit schedules, endpoint selection, and communication strategies, have been shown to reduce design-level burden and improve participant satisfaction. This is not tokenism. It requires genuine input at a stage when changes can still be made [3]. Sponsors and investigators should be prepared to modify protocol elements in response to patient input and should document how that input was incorporated.
Europe offers an unusually mature public infrastructure for this work. The European Patients’ Academy on Therapeutic Innovation (EUPATI) trains patient experts to contribute to medicines development; the IMI-funded PARADIGM project produced operational tools and recommendations for patient engagement at defined decision points in the lifecycle, including protocol design [18] and the UK National Institute for Health and Care Research (NIHR) provides national standards and infrastructure for public involvement in research [19].
Sponsors planning trials with European sites should integrate these networks, which materially shortens the time from intent to operational engagement and increases the credibility of the consultation in the eyes of regulators and ECs.
4.2 Conduct Honest, Plain-Language Burden Assessment
Before finalizing a protocol, sponsors should conduct a structured assessment of the time, financial, and procedural burden the trial will impose on participants, and test this assessment with actual patients in the target disease population [3]. Informed consent documents should reflect realistic participation demands including travel time, visit duration, and potential out-of-pocket costs. Expectation alignment at enrollment is one of the most actionable levers available for reducing early dropout [4].
4.3 Invest in Decentralized and Flexible Participation Models
Home health visits, remote monitoring, telemedicine check-ins, and patient-friendly scheduling where appointments can be booked further in advance and adjusted when life circumstances change have all shown positive effects on retention in trials that have implemented them.
In the DIAN-TU Alzheimer’s trial, 94.7% of participants who used home nurse visit services rated the experience positively, and the option was strongly associated with continued engagement among participants who lived far from study sites [20].
The European regulatory environment now actively supports these approaches. The 2022 joint Recommendation paper on decentralized elements in clinical trials, issued by the EMA together with the Heads of Medicines Agencies and the Clinical Trials Coordination Group, provides a harmonized framework for home health visits, electronic informed consent, direct-to-participant investigational medicinal product shipment, and remote monitoring across Member States [21]. The UK MHRA has issued complementary guidance, and several Nordic agencies have run successful national pilots [22].
For sponsors planning European trials, the question is no longer whether decentralized elements are permissible, but which combination best matches the protocol, the disease area, and the participating sites. These approaches are no longer experimental; they should be part of standard retention planning.
4.4 Measure Participation Experience Systematically
Sponsors and sites should deploy brief, validated experience surveys at multiple timepoints during trial conduct not only at the end, when it is too late to act on the findings [3]. Questions should address staff communication quality, clarity of trial information, adequacy of compensation, procedural burden relative to expectations, and overall satisfaction.
Critically, this data should be reviewed by study management and used to adjust in real time. Exit interviews with participants who withdraw should be standard practice, not an occasional research exercise.
A practical operationalization of this principle is to administer a short, structured questionnaire to the first two to three participants enrolled at each site at approximately a quarter of the subject’s study timeline, with a follow-up later in their participation.
The early timepoint matters: most withdrawal-relevant burden (visit demands, transport friction, perceived staff responsiveness) surfaces within the first weeks but typically goes unmeasured until the participant has either adapted or withdrawn. Querying the first few participants per site also generates site-level signal, which is what distinguishes protocol-design issues (consistent across sites) from site-execution issues (one or two sites, locally remediable) and allows the study team to act before patterns translate into withdrawals.
4.5 Address Compensation Adequacy Empirically
Current practice leaves ethics committees to evaluate compensation proposals without binding standards, validated benchmarks, or empirical data on what participation actually costs participants. Sponsors should conduct pre-trial cost analyses covering transportation, childcare, lost wages, and meal costs for a realistic participant profile, and structure reimbursement to cover these costs.
US IRBs have substantial discretion over subject compensation. US regulations do not specify what constitutes adequate payment, however US IRBs typically focus on restriction of payment rather than its adequacy of reimbursement of time and expenses [11]. A planning consideration that is frequently underestimated by sponsors is that European ECs apply a meaningfully different ethical baseline.
Most European countries permit reimbursement of demonstrable expenses (travel, accommodation, meals, and documented lost wages) but restrict or prohibit additional payment as an incentive in patient trials. The UK HRA guidance on payments and incentives in research is the clearest articulation of this position [23] and similar restrictions apply across most European jurisdictions.
Sponsors of multi-regional trials should therefore design two related but distinct strategies: in the United States, separation of expense reimbursement from compensation for time and inconvenience [24] remains appropriate within the latitude IRBs allow; in Europe, expense reimbursement should be designed generously and transparently rather than supplemented with incentive payments that ethics committees are likely to reject. Post-trial audits of whether compensation matched actual participant costs should inform future protocols [25].
4.6 Build and Sustain Staff Relationships
Recruitment and retention training for study coordinators and investigators should explicitly address the inter-relational dimensions of participant support not just protocol compliance and data collection. Participants should have named, accessible contacts on study teams, not generic email addresses. Sites should monitor coordinator turnover and have transition protocols in place to maintain continuity of the participant relationship when staff changes occur [3] [9].
The retention benefit of staff continuity has been demonstrated empirically. In a randomized analysis of physician continuity within a clinical trial, participants who saw the same physician across all study visits had a completion rate of 50% versus 29% for those who did not [26].
The mechanism is consistent with the broader relationship-and-communication evidence: participants who experience their trial care as continuous with their routine care relationship are materially more likely to stay. The operational implication is that “continuity” should be designed into site-level staffing models from activation onward, not treated as a happy accident when staff happen not to leave.
5. Strategic and Planning Considerations
The most consequential retention decisions are made before the protocol is finalized and before the first site is activated. Country selection and site selection are the first preventive levers available to a sponsor: a trial run in the right countries, with the right sites, in the right healthcare context will retain participants better than a trial run anywhere else, regardless of how well the downstream retention tactics are executed. Treating geographic strategy as a retention question (rather than purely as a recruitment speed or cost question) is the single most underused preventive intervention available to sponsors today.
5.1 Country selection as the first retention decision
Countries differ systematically in the structural drivers of participant retention: the proportion of trial-related cost absorbed by the health system, the maturity of patient organizations, the cultural baseline of trust in clinical research, the geographic dispersion of the eligible population, and the ease of accessing care between protocol-mandated visits. Sponsors planning multi-regional trials should therefore assess candidate countries against retention-relevant criteria explicitly, not only against recruitment-rate forecasts.
Countries with universal or near-universal healthcare coverage (EU member states, the UK, Canada, Australia, Japan, and several Latin American countries) tend to perform well on the cost-burden dimension because participants are typically not exposed to direct medical or procedural costs associated with trial visits [7].
In the United States, sponsor coverage of trial-specific procedures plays an analogous role for direct medical costs, although the indirect-cost burden on non-exempt workers remains a distinct US-specific challenge [11].
The underlying structural lever is therefore not unique to any single region: it is the degree to which the system, rather than the participant, absorbs the financial cost of participation. The maturity of patient organizations and the cultural baseline of trust are similarly distributed across any single region. The UK, the Netherlands, the Nordic states, and Spain have particularly well-established patient-involvement traditions [12]; the US has a strong patient-advocacy ecosystem in oncology and rare disease; Canada, Australia, and Japan have mature public-research-participation cultures.
These advantages are not uniformly distributed within any region, and they are not adequately captured by the recruitment-rate metrics that typically dominate country selection. A retention-aware country assessment should therefore complement the conventional feasibility process.
An additional structural factor worth weighing in country selection is that the degree of integration between trial care and routine care varies meaningfully between countries. In healthcare systems where patients are routinely referred to trials by the specialist who continues to manage their care at the same institution (common in many European public systems and in integrated US networks such as Kaiser Permanente or Veterans Affairs), staff continuity becomes the default rather than the exception.
The retention benefit of physician continuity has been demonstrated empirically [19]. The question for sponsors is not which region delivers this by default, but which candidate sites in any region operate within care models that support it.
5.2 Site selection: choosing for retention, not only for recruitment speed
At the site level, the predictors of strong retention are partly distinct from the predictors of strong recruitment. Sites that recruit quickly may do so by enrolling participants who are subsequently more likely to withdraw (particularly when recruitment is achieved through pressure-driven processes, or where the site’s catchment population faces structural barriers to sustained participation).
Sponsors should therefore evaluate prospective sites against retention-specific indicators alongside the usual recruitment metrics: historical screen-failure-to-enrolment ratios, withdrawal rates in prior trials of comparable duration and burden, coordinator continuity, the site’s track record of conducting exit interviews, and the geographic accessibility of the site to the eligible population.
Where this information is available from prior trials in the same region, it should be weighed at least equally with recruitment-rate projections. Where it is not available, sponsors should consider it a question worth asking during qualification, rather than an unknown to be tolerated.
5.3 Using regulatory and infrastructural assets to support retention
The revised ICH E6(R3) GCP guideline, which reached Step 4 in January 2025, places explicit emphasis on participant-centered trial design, proportionate risk-based quality management, and the meaningful involvement of patients in protocol development [27].
In parallel, both major regulatory regions now mandate post-trial results disclosure within 12 months of trial completion: the FDA Amendments Act of 2007 and its 2017 Final Rule require summary results to be posted on ClinicalTrials.gov no later than 12 months after the primary completion date for applicable trials [28] and the EU Clinical Trials Regulation imposes a parallel obligation through the Clinical Trials Information System [16].
The two regimes differ in one respect that is directly relevant to retention: the EU CTR additionally requires a lay summary written in plain language in the languages of participating Member States, whereas the US obligation is satisfied by a structured technical summary. Read together, these regulatory regimes structurally support the closure of the post-trial communication gap that participants consistently identify as a retention failure [3].
The European Clinical Research Infrastructure Network (ECRIN) provides coordinated operational support across more than ten countries and is particularly valuable for rare-disease and academic-sponsored trials where the burden of multi-country coordination is otherwise prohibitive [29]. Comparable national infrastructures exist in other regions (the US NIH-supported Clinical and Translational Science Awards (CTSA) network, Canada’s Clinical Trials Ontario and the broader CIHR-supported infrastructure, and Australia’s Clinical Trials Activity initiative) though their scope and operating model differ from ECRIN.
Sponsors who plan their trials with these assets in mind materially reduce the operational friction that otherwise erodes the retention practices the evidence supports.
5.4 Designing the operational model around the chosen geography
Country and site selection should also drive the operational model rather than the other way around. A geographic footprint that includes countries with strong decentralized-trial frameworks supports a hybrid operational model that is well-matched to the participant burden profile in those countries.
Major regulators have converged on broadly compatible guidance over the last three years: the EMA/HMA/CTCG recommendation paper of December 2022 [21] the UK MHRA position [22] the FDA’s September 2024 guidance on the conduct of clinical trials with decentralized elements [30] and several Nordic national pilots collectively establish that decentralized participation is now an accepted regulatory option across most major trial markets, not a regional specialty.
Conversely, a protocol that mandates frequent on-site visits without a flexibility plan will retain participants poorly even in otherwise well-selected countries.
Diversity strategy needs to be locally calibrated, but in 2026 it is the US framework that is the most prescriptive of the major regimes. The Food and Drug Omnibus Reform Act of 2022 (FDORA) requires sponsors of applicable clinical studies to submit Diversity Action Plans covering age group, sex, and racial and ethnic representation, and the FDA’s June 2024 draft guidance operationalizes this statutory requirement [31].
The EMA’s reflection papers and ICH guidance on inclusion of women, older people, and under-served populations focus on demographic and clinical characteristics most relevant to the European population [32] and adopt a less prescriptive approach.
Sponsors of multi-regional trials should therefore design a diversity strategy that satisfies the most demanding applicable regime (typically FDORA) while remaining locally calibrated to the demographic and clinical realities of each participating country.
6. Conclusion
Subject retention in clinical trials is not a mystery. The research community has accumulated substantial evidence about why participants leave, what makes them stay, and what the experience of trial participation actually feels like from the inside [3] [6]. What has been slow is the translation of that evidence into practice.
The path forward requires a fundamental shift in how retention is conceptualized: not as a logistics problem to be managed after enrollment, but as an experience problem that must be addressed at the beginning of trial design. Trials that treat participants as partners who deserve honest communication, genuine flexibility, adequate compensation, and meaningful feedback on what their contribution produced will retain them [3] [7] [9].
Those that treat participation as a contractual transaction will continue to lose them at rates that compromise both the science and the participants who trusted the research enterprise enough to enroll [2] [6].
The tools to do this already exist or are within reach. What is needed is the commitment to use them.
Rob King, Executive President, US Affiliate – Pivotal
David Fuentes, Director Strategy – Pivotal
Lourdes Huarte, SVP Regulatory and Clinical Operations – Pivotal
References
La mejor actitud que podemos adoptar es la de trat...
El equipo de investigadores observó cambios en el...
El gen AtCDF3 promueve una mayor producción de az...
En nuestro post hablamos sobre este interesante tipo de célula del si...
GENESIS Biomed ha lanzado el programa GENESIS CEO Academy, una iniciat...
