FRANKLIN, Tenn., April 01, 2026 (GLOBE NEWSWIRE) -- Xsolis, an AI-driven technology company that reduces administrative waste by enabling collaboration between healthcare providers and payers, today highlighted newly published peer-reviewed research evaluating its artificial intelligence platform across three major U.S. healthcare systems. The independent studies report measurable, real-world performance across level-of-care accuracy, observation discharge rates, and operational forecasting within hospital-based utilization management workflows.
As healthcare organizations face increasing scrutiny around artificial intelligence performance and return on investment, peer-reviewed research provides an objective benchmark for assessing real-world results. According to a 2025 Sage Growth Partners survey of 101 health care C-suite executives, 83% believe AI could improve clinical decision-making — yet only 12% consider today's AI algorithms robust enough to rely on. A systematic review published in JAMA helps explain that gap: of 519 health care AI studies examined, only 5% used real patient data, highlighting the significant disconnect between how AI is typically tested and how it performs in actual clinical environments.
Studies published in Baylor University Medical Center Proceedings, the Journal of Doctoral Nursing Practice, and the Journal of Clinical and Translational Science, examined the use of AI-driven decision support tools at Baylor Scott & White Health, Yale New Haven Health, and Mayo Clinic Health System. Each evaluation was conducted by clinical and research teams within those organizations. Across the studies, AI was used as a clinical decision support tool within existing utilization management workflows rather than deployed as a standalone system.
“The question health system leaders are asking isn’t whether AI is promising — it's whether it works in practice,” says Joan Butters, CEO and Co-Founder of Xsolis. “These studies answer that question with data, not projections.”
Key findings from the published research include:
- Baylor Scott & White Health: A real-time AI Care Level Score achieved 86% correct classification for inpatient designation at a defined threshold and demonstrated stronger predictive performance than traditional commercial screening tools.
- Yale New Haven Health: Observation discharge rates declined from 16.69% to 12.75% following AI integration within a nurse-led utilization management department, with improved identification of comorbidities and strengthened assessment of medical necessity.
- Mayo Clinic Health System: AI-driven diagnosis-related group prediction reached 81% accuracy and predicted patient length of stay within 0.14 days of actual outcomes.
Independent evaluations across multiple major health systems offer a data-driven foundation for assessing predictive analytics in hospital utilization management. The findings contribute to a growing body of research examining how artificial intelligence performs in live clinical settings.
For more information about the independent published studies, visit https://www.xsolis.com/peer-reviewed-studies/.
About Xsolis
Xsolis is an AI-driven technology company that reduces administrative waste by enabling collaboration between healthcare providers and payers. Dragonfly®, its AI-driven proprietary platform, is the first and only solution to use real-time predictive analytics to continuously assign an objective medical necessity score and assess the anticipated level of care for every patient, enabling more efficiency across the healthcare system. Xsolis is headquartered in Franklin, Tennessee. For more information, visit www.xsolis.com.
