When a physician refers a patient to a clinical trial site, the coordinator currently handles the first qualification step: reviewing the patient's details against the inclusion and exclusion criteria to determine whether a full screening visit is warranted. For sites receiving 20 to 50 referral inquiries per month, this pre-screening review consumes significant coordinator time, particularly for referrals from physicians who are early in the referral relationship and have not yet developed an accurate sense of which patients fit the protocol.
AI-assisted pre-screening changes this step. Instead of the coordinator reviewing each referral individually, a structured AI workflow evaluates the patient's reported details against the protocol criteria and returns a routing recommendation — schedule, needs clarification, or does not qualify — before the coordinator sees the referral in their queue.
How the Pre-Screening Works in Practice
When a physician referral arrives — through an email reply, a referral form, or an SMS response — the AI model receives the patient's key details: age, diagnosis, current medications, relevant medical history indicators, and any other protocol-specific information the referral form captured. The model evaluates these details against the study's inclusion and exclusion criteria, which have been configured as evaluation rules at the start of the study.
For clearly eligible referrals — patient meets all primary criteria — the AI routes the referral to the scheduling queue with a pre-screening summary that the coordinator can review at a glance. For referrals with uncertain eligibility — patient meets some criteria but has a flag on a secondary criterion — the AI routes the referral to a coordinator review queue with the specific uncertain criterion highlighted. For clearly ineligible referrals — patient is outside the age range, has a disqualifying comorbidity, or is on an excluded medication — the AI generates a decline notification for coordinator review before any patient contact is made.
What This Changes for Coordinator Workload
A coordinator without AI pre-screening review every referral individually — spending 10 to 20 minutes per referral gathering additional patient information, looking up medication interactions against exclusion criteria, and writing a pre-screening assessment. With AI pre-screening, clearly eligible referrals arrive with a summary already prepared. Clearly ineligible referrals are filtered before consuming coordinator time. The coordinator's time is concentrated on the uncertain cases that require actual clinical judgment — the segment where human review adds the most value.
The Referral Feedback Loop
AI pre-screening also enables faster feedback to referring physicians. Because the eligibility assessment happens in near-real-time rather than waiting for coordinator availability, the physician can receive a referral acknowledgment within hours rather than days. Faster feedback produces better physician engagement and more consistent future referrals.