How to Find Eligible Patients for Your Clinical Trial

Most research sites do not have a patient volume problem. They have an eligibility problem. A site with three hundred patients carrying a target diagnosis in its EHR might have twelve who actually meet a Phase 3 protocol's full inclusion and exclusion list.

Finding eligible patients is a different skill than finding patients. This guide covers where eligible patients actually come from (internal databases, referral networks, pre-screening design, registries, and digital channels) in the order most sites should build them.

Why “More Leads” Is the Wrong Target

Eligibility criteria filter hard. A Phase 3 protocol with a narrow biomarker requirement, a washout period, or a recent-event window can disqualify a majority of patients who otherwise carry the target diagnosis. Chasing lead volume against a filter that steep produces a queue of unqualified contacts and a coordinator spending hours on calls that end in a screen fail.

Screen fail rate is the number that exposes this. A site converting one in three pre-screened leads into a randomized patient is running a tight process. A site converting one in twenty is not short on patients. It is short on a filter that finds the right ones before the phone call happens.

The fix starts before outreach, not during it. Every channel below (internal database, referral network, digital) only performs as well as the eligibility filter applied to it.

Your Existing Patient Base Is the First Place to Look

Before building anything external, query what the site already has. Most EHR systems support search by diagnosis code, medication history, lab values, and visit frequency, which covers most inclusion criteria on the first pass.

A basic query (diagnosis code, plus an active prescription, plus a recent visit within the last 12 months) produces a starting list. Layer in lab value ranges or exclusion diagnoses to narrow further before a coordinator opens a single chart.

Chart review comes next, and it should follow the query, not replace it. Reviewing every chart in a large patient panel manually to find a dozen eligible patients wastes days. Reviewing the charts a query already narrowed to takes an afternoon.

The physician of record should sign off before outreach begins. A referral framed as “your doctor's office noted you might be a fit for a study” converts at a meaningfully higher rate than a cold call from an unfamiliar research coordinator, because the physician relationship is already established.

Physician and Specialist Referral Networks

Internal database mining runs out fast for narrow-eligibility protocols. A rare disease trial or a Phase 3 study with a tight biomarker window needs patients beyond the site's own panel, and that means specialist referral relationships.

Sites that build durable referral networks do three things: identify specialists in the relevant subspecialty within reasonable driving distance, hand those specialists a one-page eligibility summary instead of the full protocol, and close the loop after every referral. Was the patient eligible? Did they enroll? Specialists who never hear back stop referring.

This is slow to build and durable once built. A referral network that took months to establish keeps producing eligible patients for every study the site runs in that indication going forward, not just the current protocol.

Pre-Screening Is Where Eligible Patients Get Found, Not Where They Get Lost

A pre-screen built around the criteria that actually disqualify patients does the finding work, not just the filtering work. A pre-screen that only asks about diagnosis and age is a lead form, not a screening tool.

The strongest pre-screens open with the two or three criteria that eliminate the most patients for that specific protocol (a washout period, a prior treatment requirement, a lab value threshold) before asking anything else. A patient who fails on question two never reaches a coordinator's call queue, which protects the coordinator's time for patients who can actually enroll.

Response speed compounds this. A referral contacted within 10 minutes of applying is 4 times more likely to answer the screening call than one contacted a day later, based on Clinical Enroll client portal data across 30+ indications. The best pre-screen in the world does not help if the patient has moved on before anyone calls them back.

The guide on clinical trial enrollment challenges covers the structural reasons sites miss enrollment targets, including where pre-screening design breaks down most often.

The channel mix depends on the eligibility profile:

A high-prevalence, low-complexity indication like acne finds eligible patients primarily through volume and digital reach. A narrow-eligibility Phase 3 protocol like Type 1 Diabetes finds them primarily through referral networks and internal database precision. The channel mix should match the eligibility profile, not the other way around.

Registries, Advocacy Groups, and Community Partnerships

For indications with active patient communities, registries and advocacy organizations are a channel most sites underuse. Condition-specific nonprofits, disease registries, and community health centers maintain contact with patients who have already self-identified as living with the target condition.

These relationships take a different form than a referral network. A registry partnership usually means the organization distributes information about the study to its members directly, rather than the site contacting individual patients. This works best for indications with organized patient communities: autoimmune conditions, rare diseases, and chronic conditions with active advocacy groups.

Community health centers serve a different function: reaching patient populations that a single-specialty research site does not otherwise see, which matters directly for studies with diversity enrollment requirements.

Digital Channels Multiply Reach. They Do Not Fix a Small Population.

Digital recruitment (search, social, display) extends reach beyond the site's existing patient base and referral network. It is the right tool once the internal and referral channels above are in place, not a replacement for them.

Digital works best against high-prevalence conditions with a large addressable population, where volume compensates for a lower per-lead qualification rate. It works poorly against narrow-eligibility protocols, where the eligible pool is small enough that even a well-targeted campaign produces mostly unqualified leads.

The guide on digital recruitment for clinical trials covers channel selection, funnel benchmarks, and cost-per-patient math in depth. The short version here: digital finds patients faster once a site already knows what an eligible patient for its protocol looks like. It does not define eligibility.

Track Eligible Rate, Not Lead Count

Lead count measures activity. Eligible rate (the share of pre-screened contacts who pass the eligibility filter) measures whether the finding process is working.

A site tracking eligible rate by source (internal database, referral, registry, digital) can see which channel is actually producing patients who enroll, and which is producing volume that looks productive on a dashboard but converts into nothing.

This number belongs at the top of any recruitment report a site reviews, not buried under impressions and click-through rate. Impressions tell a vendor's story. Eligible rate tells the site's story. A deeper look at what to track is in the guide on feasibility assessment, which covers how to size the eligible population before a study even starts.