Ochsner Health says an AI platform increased patient screening capacity by 41%, expanded follow-up of potential candidates and reduced manual screening work.
Ochsner Health is using an AI-enabled recruitment platform connected to electronic health records and clinical data to identify patients who may qualify for oncology trials. The Louisiana-based health system reported a 41% increase in the number of patients it could screen during the measured implementation period, alongside a 3.6-fold increase in the number followed for possible future eligibility and an approximately 75% reduction in manual screening workload.
The system addresses a logistical challenge: potential participants may receive care far from a trial team, and eligibility can change after a treatment, test result or shift in a patient’s condition. The platform screens a broader patient population against criteria for open studies, prioritizes records for review and retains visibility into people who might qualify later. Ochsner worked with Paradigm Health on the EHR-integrated platform.
The reported figures describe screening and workflow, not trial enrollment or improved clinical outcomes. Ochsner’s vice chief scientific officer emphasized that research coordinators review the candidates, verify criteria that require professional judgment and remain responsible for eligibility confirmation, patient communication and enrollment decisions. Patients must still meet protocol requirements and choose whether to participate.
The deployment illustrates how AI can help research teams search across a large health system and revisit potential candidates as their clinical information changes. Ochsner operates 47 hospitals and more than 370 clinics, making consistent identification across sites a substantial operational task. The health system said the broader screening process can help surface research opportunities that might otherwise be missed because of a patient’s location or the limits of manual chart review.
Ochsner is exploring whether the partnership can also support other trial operations, including matching new studies to its patient population and automating data collection. The reported results are an early operational case study; the article does not provide a controlled comparison or establish that the approach increased enrollment or changed patient outcomes.