The Patient Access Problem in Clinical Trials

The enrollment gap nobody has solved

Every clinical trial is a promise: that a new treatment will be tested rigorously, safely, and in the people who will eventually use it. Yet the single most common reason that promise breaks is not science, safety, or regulation. It is enrollment.

Trials miss their enrollment timelines with a regularity that would be unacceptable in almost any other industry. Studies get extended, rescue sites get added, countries get opened late, and budgets climb. Industry research, including long-running work from the Tufts Center for the Study of Drug Development, has repeatedly found that a meaningful share of activated sites enroll few or no patients at all. Each of those sites represents months of start-up work, monitoring visits, and investigator time that produced nothing.

We spent four days at the SCRS Global Site Solutions Summit in Orlando this October talking with the people who live with this problem every day: site engagement leads at large sponsors, feasibility teams at global CROs, site network operators, independent investigators, and the coordinators who actually make the phone calls. The question we kept asking was simple. If trials are struggling to find patients, where are the patients?

The answers we heard were candid, sometimes frustrated, and in places genuinely hopeful. The short version: the patients exist. The industry is often looking for them in the wrong places, with the wrong tools, under protocols that make them hard to enroll even when they are found.

This article lays out what we learned. It is written for sponsors and CROs planning 2027 portfolios, for site leaders trying to grow, and for anyone who cares about getting new medicines to patients faster and more fairly. It is a companion to our earlier piece on clinical trial site sustainability after SCRS 2026, and the two are best read together: sustainable sites and accessible patients are two halves of the same problem.

Why trials still miss enrollment: four root causes

1. Feasibility predicts poorly because it asks the wrong questions

Most enrollment failures are baked in at site selection. A sponsor or CRO sends a feasibility questionnaire, the site estimates how many patients it can enroll, the estimate is scored, and the site is selected or not. The trouble is that almost every step of this process rewards optimism over accuracy.

Sites want studies, so they lean toward generous estimates. Questionnaires often ask about historical trial experience and infrastructure rather than about the specific patients the protocol needs. Scoring models favor sites with long track records, which tend to be the same dedicated research centers competing for the same patients in the same metro areas. Meanwhile, sites with large patient populations but less sponsor history can be scored out before anyone reads their response.

At SCRS, both sides of the table acknowledged the problem. Sponsors told us they distrust self-reported enrollment projections. Sites told us they distrust scoring that seems disconnected from reality, and several described feasibility scores that later turned out to be miscalculated. The industry is spending enormous effort to produce data that neither party believes.

2. Protocols are designed for the ideal patient, not the real one

Protocol complexity has grown steadily for two decades. More endpoints, more procedures, more visits, and tighter eligibility criteria all make sense individually, and each is usually defended by a reasonable scientific or regulatory argument. Together, they create a study that few real patients qualify for and fewer can realistically complete.

Site leaders described the pattern clearly. A protocol excludes patients on common concomitant medications, or with common comorbidities, or with lab values slightly outside a narrow range. Screen failure rates climb. The patients who do qualify face visit schedules that require repeated half days away from work or family. Retention suffers, and the patients who stay are rarely representative of the population that will eventually be prescribed the drug.

The frustrating part is that much of this is avoidable. Eligibility criteria are often copied forward from previous studies without fresh justification. Procedures are included for exploratory purposes that add burden without changing decisions. Very few protocols are tested against real patient data before they are finalized.

3. Geography: trials go where trials have always gone

Clinical trial activity remains concentrated in a relatively small number of metropolitan areas and academic medical centers. That concentration has a logic, since those are the places with experienced investigators and research infrastructure, but it also means that large populations live far from any trial site.

Distance is a powerful barrier. A patient who would happily participate in a study will often decline if it requires a ninety-minute drive for each of twelve visits. Rural patients, working parents, older adults without reliable transportation, and people who cannot take unpaid time off are systematically underrepresented as a result.

Decentralized and hybrid trial approaches were supposed to solve this, and in some cases they have helped. But the conversations in Orlando reflected a more mature, less utopian view. Fully remote trials work for some designs and fail for others. What patients often need is not a trial in their living room but a trial site in their own community, staffed by people they already know.

4. Trust is earned locally, and the industry has not earned enough of it

The final root cause is the one most often discussed in careful language: trust. Many patients, and particularly patients from communities that have historically been underserved or mistreated by medical research, are understandably wary of being approached about a trial by someone they have never met.

Several speakers and many of the site leaders we spoke with made the same point. Patients are far more likely to consider a trial when it is raised by a clinician or care team they already have a relationship with, at a location they already visit. A recruitment ad or a cold outreach call from an unfamiliar research center starts from a deficit. A conversation with a familiar provider starts from a position of trust.

This has profound implications for diversity and representation. Regulators in the US have pushed sponsors to plan for enrollment that reflects the populations affected by a disease, and the policy details have continued to evolve. Regardless of where the rules land, the operational reality is the same: you cannot enroll a representative population from sites that do not see a representative population.

Where the patients actually are

If the root causes are poor feasibility, burdensome protocols, geographic concentration, and a trust deficit, the solutions point in a consistent direction: bring research closer to where people already receive care. That idea ran through many of the most energizing conversations at SCRS 2026.

Community health systems and primary care

Most people do not live near an academic medical center, but most people do have some relationship with a community clinic, a primary care practice, or a regional health system. These settings manage the chronic conditions that dominate today's development pipelines, including type 2 diabetes, obesity, cardiovascular disease, chronic kidney disease, COPD and asthma, and women's health conditions.

When research is embedded inside these systems, the site does not need to find patients from scratch. It needs to identify the right patients within a population it already serves, and invite them through a clinician they already trust. Several sponsors told us they are actively expanding their networks to include health-system-embedded research programs for exactly this reason.

Urgent care and walk-in care networks

One of the less obvious themes we heard was growing interest in urgent care and walk-in care networks as research partners. At first glance, urgent care seems like an odd fit for clinical research. Patients come in for acute problems, they are seen quickly, and many do not return.

But look closer and the advantages are striking. Urgent care networks see very large volumes of patients, often including people who do not have a regular primary care physician and who would never be reached by traditional recruitment. They are a natural front door for infectious disease studies, where a trial needs to enroll patients within days of symptom onset for conditions such as influenza, COVID-19, RSV, or urinary tract infections. And many urgent care organizations have expanded into primary care, weight management, and occupational health, giving them longitudinal relationships with patients who have chronic conditions too.

Diagnostics developers were especially interested. Clinical performance studies for rapid tests need large numbers of samples from symptomatic patients, collected at the point of care, and an urgent care network can supply exactly that.

EHR-driven pre-screening

The single most practical tool for improving feasibility and enrollment is something many sites already have: their electronic health record. A de-identified query of the EHR against a protocol's key inclusion and exclusion criteria produces a far better estimate of eligible patients than any guess on a questionnaire.

The more advanced sites are going further. They use structured queries and, increasingly, AI-assisted chart review to flag potentially eligible patients ahead of scheduled visits, so the care team can raise the trial at the right moment. Done well, with appropriate privacy safeguards and IRB-approved processes, this turns recruitment from a broadcast exercise into a targeted, clinician-led conversation.

The caveat we heard repeatedly: pre-screening only helps if there are coordinators available to follow up. Identifying 400 potentially eligible patients is worthless if no one has time to call them.

Diversity as a design outcome, not a recruitment campaign

The most thoughtful sponsors we spoke with no longer treat diverse enrollment as a late-stage recruitment problem to be solved with targeted advertising. They treat it as a site selection and protocol design problem.

That means choosing sites based on the demographics of the populations they actually serve, not just their historical enrollment. It means reviewing eligibility criteria for exclusions that disproportionately remove certain groups without strong scientific justification. And it means reducing visit burden so that participation is realistic for people with jobs, caregiving responsibilities, and limited transportation.

Community-embedded sites are central to that strategy. They see the patients the rest of the industry struggles to reach, because those patients are already walking through their doors.

A practical playbook for 2027

Insight is only useful if it changes what people do on Monday morning. Here is how we would translate the lessons from SCRS 2026 into action.

For sponsors

1. Test the protocol against real data before you finalize it. Run draft eligibility criteria against de-identified EHR or claims data in your target regions. If the eligible population is tiny, change the criteria before a single site is activated, not after the first amendment.
2. Ask sites to review the schedule of assessments. Two or three experienced coordinators will tell you within an hour which visits patients will skip and which procedures will drive dropouts.
3. Diversify your site mix deliberately. Set a target for the share of sites that are community-based or health-system-embedded, and hold feasibility teams to it.
4. Fund recruitment at the site level. Sites that know their patients are better at reaching them than centralized campaigns. Budget for pre-screening, outreach, and coordinator time accordingly.
5. Reduce participant burden. Combine visits where possible, reimburse travel and time promptly, and allow local labs or home visits for procedures that do not require the site.

For CROs

1. Replace questionnaires with conversations. Use short, protocol-specific feasibility forms that ask for evidence, such as EHR query counts, rather than estimates. Follow up by phone with promising sites before scoring them out.
2. Audit your scoring models. If your model consistently favors the same sites, it may be optimizing for familiarity rather than enrollment. Check scoring outputs against actual performance.
3. Close the loop. Tell every site that completes a feasibility survey whether it was selected and why. The goodwill pays back in faster responses next time.
4. Build an on-ramp for new sites. Capable sites with strong patient access but limited sponsor history are a resource, not a risk, especially when paired with extra start-up support.

For sites

1. Quantify your patient access. Know your active patient counts by condition, age, and demographics. Be ready to run a de-identified query for any protocol within 48 hours.
2. Lead with patients, not infrastructure. Every site has exam rooms and a centrifuge. Not every site can show 5,000 patients with the target condition within a 20-minute drive. If you can, say so first.
3. Engage your clinicians. The trusted provider is your best recruiter. Make it easy for clinicians to identify and refer patients, and close the loop with them when patients enroll.
4. Invest in coordinator capacity. Pre-screening generates leads; coordinators turn leads into randomized patients. Grow the two together.
5. Track and share your metrics. Screen-fail rates, randomization rates, retention, and time to first patient are the numbers sponsors care about. Report them proactively.

The bottom line

The clinical trial enrollment problem is not a mystery. Patients are not missing; they are in community clinics, primary care practices, urgent care centers, and health systems across the country, mostly unaware that a trial might be right for them. The industry's challenge is to design protocols they can realistically join, to select sites that actually serve them, and to let trusted local clinicians make the invitation.

The conversations at SCRS 2026 made us optimistic that this shift is underway. The sponsors, CROs, and sites that lean into it will enroll faster, generate data that better reflects real-world patients, and build the kind of partnerships that make the next study easier than the last.

Frequently asked questions

Why do so many clinical trials fail to meet enrollment targets? The main causes are inaccurate feasibility estimates, overly restrictive eligibility criteria, high participant burden from complex visit schedules, trial sites concentrated far from many patients, and limited trust between patients and research organizations they do not know.

What is clinical trial feasibility? Feasibility is the process sponsors and CROs use to assess whether a site, region, or country can recruit enough eligible patients and run a study successfully. It typically involves questionnaires, historical performance data, and increasingly, de-identified queries of patient records.

How can community-based sites improve clinical trial enrollment? Community-based sites, including research programs embedded in primary care, health systems, and urgent care networks, already serve large patient populations. Patients are more willing to participate when a trusted local clinician raises the trial, and participation is easier when the site is close to home.

Can urgent care centers run clinical trials? Yes. Urgent care networks with dedicated research staff and investigators can run trials, and they are particularly well suited to acute infectious disease studies and diagnostic device studies that need patients soon after symptom onset. Many also support chronic disease studies through affiliated primary care and weight management programs.

What is EHR pre-screening in clinical trials? EHR pre-screening uses a site's electronic health records to identify patients who may meet a study's eligibility criteria, under IRB-approved processes and privacy safeguards. It improves feasibility estimates and lets care teams raise a trial with the right patients at the right time.

How can sponsors improve diversity in clinical trials? The most effective levers are selecting sites that serve diverse populations, removing eligibility exclusions that lack strong scientific justification, reducing visit burden, and funding community-level outreach through trusted local clinicians.

Work with Seven Health

Seven Health helps sponsors, CROs, and research sites close the gap between protocol and patient. We advise on evidence-based feasibility, site network strategy, community and health-system site development, recruitment planning, and business development for sites seeking sponsor and CRO partnerships.

If you are planning 2027 studies and want to find patients faster, or you run a site with strong patient access and want sponsors to notice, get in touch with Seven Health. And if you missed it, read our companion article: Clinical Trial Site Sustainability in 2026: 9 Lessons from the SCRS Global Site Solutions Summit.