What Decentralization Actually Changed — and What It Didn't
The expansion of decentralized clinical trial (DCT) infrastructure from 2020 onward has had real and measurable effects on access and retention in certain therapeutic areas. Remote patient monitoring, telehealth-enabled consent and follow-up, direct-to-patient drug delivery, and home nursing visits have collectively expanded the geographic catchment for trial participation beyond the 50-mile radius that historically constrained it. For trials in chronic disease management, cardiovascular outcomes, and patient-reported outcome-heavy protocols, these capabilities represent genuine improvements in both enrollment reach and patient burden reduction.
What decentralization did not solve — and what is now becoming apparent as DCT infrastructure matures — is the patient identification problem. The DCT toolkit assumes that you have already found the patient. The question of how to surface the right patient, from the right population, with enough lead time to convert an interested candidate into a consented participant, is upstream of the telemedicine visit and the home kit shipment. And it's a question that pure DCT infrastructure addresses poorly.
This is the gap that hybrid recruitment models are now being designed to fill. The combination of decentralized trial infrastructure for patient access and participation, with EHR-based site matching for patient identification, is producing a different enrollment dynamic than either approach achieves alone.
Why Pure DCT Recruitment Struggles With Identification
Decentralized trial recruitment, in practice, has relied heavily on direct-to-patient advertising: digital advertising, patient advocacy network outreach, disease-specific social media communities, and patient registries. These channels have real value and represent a genuine broadening of recruitment beyond the traditional site-referral model.
The limitations are structural. Direct-to-patient advertising reaches patients who are actively engaged with their condition, online, and receptive to trial participation — a self-selected subset of the eligible population that tends to skew toward specific demographics. The Tufts Center for the Study of Drug Development and other academic research groups have observed that patient diversity in trials that rely predominantly on digital self-referral channels is often lower than in trials using site-referred populations, partly because digital health engagement itself is not demographically uniform.
More fundamentally, direct-to-patient advertising for complex protocols with stringent biomarker-based criteria is expensive and inefficient. A patient who self-identifies via a digital ad and contacts the trial team still requires eligibility screening — and in biomarker-selected trials, that screening often requires laboratory testing that must happen in a clinical setting. The "decentralized" patient has not actually been pre-screened; they've been pre-interested, which is a different and operationally weaker signal.
The Hybrid Architecture: EHR Matching at the Site, DCT for Participation
The hybrid recruitment model separates the identification and access functions and optimizes each independently. Identification uses EHR phenotyping at clinical sites — finding patients in the healthcare system who match the protocol's eligibility criteria before they are asked to self-refer. Access and participation uses DCT infrastructure — telehealth visits, remote monitoring, direct shipment — to reduce the geographic and logistical barrier to participation once the patient has been identified and consented.
In this architecture, the site network is still essential, but its role changes. Rather than serving as the sole point of patient contact across enrollment, consent, treatment visits, and monitoring, the site primarily functions as the patient identification engine and the anchor for procedures that require in-person clinical assessment. DCT infrastructure handles the longitudinal participation burden — which is often where patients disengage in traditional site-based trials.
Consider a Phase II/III trial in a chronic pulmonary condition where the key eligibility criteria include spirometry findings, specific maintenance medication history, and absence of acute exacerbations in the prior 60 days. Spirometry requires in-person equipment — a purely remote enrollment pathway is not viable. But follow-up visits for tolerability assessment and patient-reported outcomes do not require in-person attendance. The hybrid model finds the patient via EHR phenotype matching at respiratory medicine practices in the site network, conducts initial screening and baseline spirometry in person, then shifts to remote monitoring and telehealth follow-up for the remainder of the study period. The site does not need to handle 12 in-person visits per patient — it needs to handle 2-3.
Site Selection in a Hybrid Context
Hybrid trial design changes the site selection calculus in a specific way. Sites selected for a pure DCT trial need primarily strong operational infrastructure and protocol compliance history — patient volume matters less because patient identification is handled externally. Sites selected for a pure site-based trial need strong patient volume and coordinator bandwidth to handle full in-person visit schedules.
Hybrid trials need both: sites with strong patient population fit for the identification function, and sites with infrastructure capable of supporting remote monitoring coordination — patient communication platforms, home kit dispatch logistics, data systems compatible with remote patient monitoring device integration. This is a narrower operational profile than either pure model, which means site selection for hybrid trials benefits disproportionately from upfront patient population analysis.
A mid-size oncology sponsor we're aware of encountered this problem in a hybrid Phase II program for a maintenance indication: they selected sites based primarily on historical enrollment performance, without assessing remote monitoring infrastructure readiness. Three of eight sites had no experience managing patients in a remote monitoring protocol. The onboarding burden fell on coordinators who were accustomed to in-person visit management, and two sites showed extended lag between consent and first remote data submission — a problem that propagated into data quality issues for the primary endpoint. Site selection for hybrid trials should include an explicit infrastructure readiness assessment, not just a patient volume assessment.
The Diversity Dimension of Hybrid Recruitment
One area where hybrid recruitment shows genuine promise for improving trial diversity is geographic and mobility access. Traditional site-based enrollment creates real participation barriers for patients who live more than 45-60 minutes from a research site, who have mobility limitations, who lack reliable transportation, or who cannot take frequent time off from work for in-person visits. These barriers are not uniformly distributed across demographic groups — they fall disproportionately on populations that are already underrepresented in clinical trials.
Hybrid models reduce these barriers for patients who are identified through EHR matching at a site they already attend for care. The patient does not need to travel to a distant research center for every visit; they consent and do baseline procedures at a site they're already connected to, then participate remotely thereafter. This structural access improvement is more targeted and reliable than trying to reach underrepresented populations through advertising.
We're not saying hybrid trials are a complete solution to diversity gaps in clinical research — protocol eligibility criteria, the burden of research participation itself, and the trust dynamics between research institutions and historically underserved communities are all independent factors. We're saying that removing the geographic access barrier through hybrid participation design is an operationally meaningful step, particularly when paired with site network composition that deliberately includes sites serving diverse patient populations.
FDA's Current Stance on Decentralized and Hybrid Trials
FDA's 2023 draft guidance on decentralized clinical trials (later finalized as guidance for industry in 2023) established a framework for DCT conduct that is relevant to hybrid trial planning. The guidance addresses electronic informed consent, remote assessment methods, local healthcare provider involvement, and data integrity requirements for remotely collected data. Sponsors designing hybrid trials need to ensure that their protocol and informed consent documentation explicitly address which procedures are conducted remotely and under what oversight structure.
The FDA guidance does not endorse any specific technology platform or remote monitoring approach, but it does establish that the sponsor is responsible for ensuring data integrity across all collection modalities — which in a hybrid trial means having clear audit trail documentation for both in-person and remotely collected data points. This has implications for the site monitoring plan and for the data management infrastructure supporting remote data collection.
Where Hybrid Works Best
Not every Phase II/III indication benefits equally from hybrid recruitment design. The strongest use cases share several characteristics: procedures-heavy baseline assessments that require in-person attendance, combined with longer-duration follow-up periods where visit frequency can be reduced without compromising endpoint data quality; patient populations that are geographically dispersed beyond the catchment of major research centers; and therapeutic areas where patient mobility or employment constraints create documented enrollment barriers in traditional site-based trials.
Indications where hybrid design adds less value: short-duration trials with few visits (where the participation burden reduction is small); protocols requiring frequent in-person procedures throughout the study period; and early-phase oncology trials where safety monitoring intensity requires high-frequency in-person contact regardless of patient preference. The model is a tool, not a default — and the sponsor's job is to apply it where the access improvement justifies the additional operational complexity of managing both site-based and remote participation modalities.
