OhioHealth

OhioHealth, an eight-plus hospital system with 35,000 associates, struggled with persistent patient throughput and discharge coordination challenges too complex to address internally. Traditional coordination methods like standup meetings broke down during COVID-era labor shortages, leaving care teams unable to efficiently manage the many moving parts required to prepare patients for discharge — skilled nursing facility placement, transportation, pending diagnostics, and precertification.

OhioHealth deployed Qventus' Q inpatient AI platform, which integrates directly with the EHR and uses generative AI, machine learning, and behavioral science to predict discharge bottlenecks and automate coordination tasks. The system prompts care teams on actions such as initiating SNF precertification, arranging patient transportation, and prioritizing pending MRI orders. Clinicians retain full override authority, and the model continuously learns from community resource availability and team feedback.

Within the first month of phase-one deployment in late March 2024, OhioHealth saved nearly $500,000. By mid-2024, cumulative savings reached nearly $1.7 million. The third-generation solution can automate nearly twice as many discharge decisions as prior versions, compressing the timeline to meaningful operational improvement.

• Early, accurate discharge planning — and actively managing barriers to that plan — is the single highest-leverage intervention for reducing excess patient days.
• AI that learns from clinician overrides creates a self-improving feedback loop, making recommendations more reliable over time without removing human judgment.
• Large health systems with many hospitals benefit disproportionately from AI-driven coordination because complexity and fragmentation multiply the inefficiencies the platform can eliminate.