Patient recruitment remains one of the most persistent challenges in running a clinical trial. The Association of Clinical Research Professionals (ACRP) reports that 85% of clinical trials fail to recruit enough patients, while 80% experience recruitment-related delays. The common assumption is that eligible patients are scarce. In many health systems, that explanation does not hold up. The patients are often already documented in the health system with their qualifying details scattered across labs, notes, and codes in ways no one has had reason to connect. The detail that makes someone eligible might sit in a specialist’s note from months ago that no one has revisited, invisible to a standard query.
This problem is compounded as clinical research staffing has declined industry-wide even as the number of active studies has grown. Fewer people are covering more protocols, and the default response remains manual chart review, conducted one record at a time.
Recent federal funding pressure has sharpened the squeeze. As grants tighten and research budgets contract, hospital and health-system research teams are being asked to run more studies with fewer people, which makes the hours lost to manual chart review even harder to justify.
How Eligible Patients End Up Hidden in Plain Sight
Trial eligibility criteria rarely stop at a diagnosis code. A protocol may hinge on a lab value, an imaging measurement, or a clinical concept that only appears in the narrative text of a note rather than a discrete field—an ejection fraction in a cardiology note or a lesion size in a radiology report, for example.
This complexity is what makes a standard EHR query fall short. Eligibility is often dependent on layered criteria, complex logic, and timing requirements that a simple diagnosis or lab lookup cannot express. And much of the deciding detail sits in unstructured notes rather than discrete fields where a standard data query cannot ingest it. That gap is where eligible patients go unidentified even though the documentation already exists, and it’s why manual chart review remains the fallback: reading notes at scale doesn’t scale.
Why Patient Recruitment Breaks Down, Even When Eligible Patients Exist
With enrollment challenges so widespread across clinical research, the root cause is rarely a single issue. The gap between eligible patients and enrolled patients is often driven by a combination of recurring factors:
- Eligibility criteria increasingly depend on complex logic and unstructured notes, not codes that can be queried directly.
- Staffing hasn’t kept pace with trial volume, leaving little time for proactive screening.
- Clinical research staff turnover is high and costly, with manual, repetitive work like chart review and hand-entering data driving much of the burnout.
- Feasibility estimates often rely on assumption rather than data, so sites commit to targets they can’t hit.
- Funnel visibility is limited, so a stalling study may go unnoticed until it’s already behind.
These issues carry a real cost. Every day a trial sits stalled adds to its operating expense and delays patients’ access to treatment. The consequences extend beyond one study, too: sites that repeatedly under-enroll risk their standing with sponsors and future funded work; coordinators spend more time on chart review than on the patient-facing work that drew them to research; and patients who are never identified never get the chance to consider a relevant trial.
In my experience working with research-active health systems, the pattern is remarkably consistent: the issue is not a lack of effort, but a lack of resources and time in the day. Clinical trial staff are constantly forced to choose between engaging directly with patients and conducting more prescreening. Principal Investigators are torn between the never-ending demands of patient care and the need to refer patients to trials. Other providers are often unaware of relevant trials, even if an eligible patient is sitting in their clinic. This leads to my conclusion: eligible patients are often already being seen in the health system, but research teams do not have the right tools to help them find the proverbial needle in the haystack.
What Changes When Patient Identification Is Supported by Advanced Data Technology
Health systems making progress here start with the protocol itself: dense inclusion and exclusion criteria get translated up front into structured logic that checks automatically against a patient’s record, including lab values, imaging parameters, and clinical concepts buried in notes. Coordinators then work from a ranked list of matches with evidence attached, and leaders see where the funnel breaks down rather than just a headcount. Clinical judgment still matters; what changes is how much searching happens first.
Cohort Discovery Tools vs. Continuous Identification
Most EHR platforms include native cohort discovery, useful for a quick count but generally limited to a snapshot at the moment the query ran, and not purpose-built to handle the unstructured detail many protocols depend on. Continuous identification checks eligibility against the EHR on a recurring basis instead, so a patient who qualifies after their next visit or lab result is surfaced automatically, which matters most for studies with narrow, complex criteria.
What to Look for in Patient Recruitment Technology
Health systems evaluating this technology weigh a similar set of factors. Research teams prioritize how well a platform handles complex logic and unstructured notes, whether it simplifies their workflow with integrations and automated reporting, and how well they are supported in developing effective cohorts. IT and compliance teams focus on governance: audit logging, certifications like SOC 2 or HITRUST, and hosting optionality to meet stringent security requirements.
What This Means for Research Teams
None of this requires restructuring how a research team operates. It mostly changes where coordinators spend their time: less searching, more screening and enrolling the patients who were already there. That shift matters for retention as well as recruitment: much of the manual burden behind coordinator burnout and turnover is exactly the kind of repetitive searching this technology removes. Health systems that make this shift tend to see faster study starts, higher enrollment velocity, and more predictable delivery across their research portfolio, the kind of consistency that keeps sponsors coming back.
MRO built Clinetic to help health systems make exactly that shift, connecting to structured and unstructured EHR data so research teams begin each day with a ranked queue of eligible patients rather than a manual review process. Adoption isn’t something health systems have to figure out alone: MRO backs Clinetic with a team of coding and clinical experts who help build and refine study-specific cohorts as new studies open and criteria evolve.
Learn more about Clinetic, or read our practical guide to patient recruitment workflows for further detail on how research-active health systems are approaching this shift.
Frequently Asked Questions
What does it mean that a clinical trial patient is “already in the EHR”?
It means the patient’s diagnosis, lab values, or clinical history already satisfy a trial’s eligibility criteria, even though no one has yet identified them as a candidate. This information is often distributed across structured fields and unstructured notes that have not been systematically reviewed.
Is the recruitment problem primarily a lack of eligible patients?
Not typically. Most health systems already have patients who qualify for open studies. The more common challenge is identifying them systematically before a study’s enrollment window closes.
How does EHR-driven patient identification differ from a standard cohort discovery query?
A cohort discovery query provides a snapshot at a single point in time. EHR-driven identification runs continuously, acting as a co-pilot for clinical trial staff: it surfaces newly eligible patients automatically as patients’ records update and supports their workflow.
Why do clinical trials struggle with recruitment even when eligible patients clearly exist?
Staffing and visibility are the two most common factors. Coordinators are managing more protocols with less support and eligibility criteria increasingly depend on complex logic and unstructured notes that manual review cannot scale to address.
What should a health system look for in patient recruitment technology?
Key considerations include how effectively a platform processes complex logic and unstructured clinical notes, whether it simplifies the recruitment workflow from identification through enrollment, how it manages data security and PHI, and the support provided to configure useful cohorts.