When you or a loved one\u00a0is\u00a0admitted to the hospital\u2019s medical-surgery\u00a0unit, St. Luke\u2019s University Health Network has an\u00a0award-winning extra\u00a0layer of care.<\/p>\n
All med-surg units at St. Luke\u2019s use\u00a0a proprietary artificial intelligence\u00a0(AI)\u00a0tool\u00a0to\u00a0predict,\u00a0identify\u00a0and\u00a0rapidly\u00a0address\u00a0the needs of their\u00a0admitted or under-observation patients to\u00a0reduce\u00a0unanticipated transfers to the ICU\u2013and save lives.\u00a0<\/p>\n
This tool is\u00a0like a superhero sidekick:\u00a0AI\u00a0that watches over people in the hospital every second of every day, awake or asleep.\u00a0<\/p>\n
The AI tool looks at metrics such as heart rate, blood pressure, breathing, lab work, patient\u00a0age\u00a0and nursing documentation in real time at all 16 St. Luke\u2019s campuses.\u00a0\u00a0<\/p>\n
Information about these metrics is transmitted to the St. Luke\u2019s Virtual Response Center, which is staffed\u00a0by a team of registered nurses who, depending upon a patient\u2019s deterioration score, can alert the patient\u2019s bedside nurse, a nursing supervisor, or in more rapidly deteriorating patients, a St. Luke\u2019s rapid response team (RRT) to identify and treat patients early, before they require transfer to the ICU or have a sudden cardiac arrest.<\/p>\n
Earlier this year, St.\u00a0Luke\u2019s\u00a0utilization of\u00a0the AI tool called\u00a0Epic\u2019s\u00a0Deterioration Index\u00a0was recognized by The Hospital and Healthsystem Association of Pennsylvania (HAP)\u00a0with a 2025\u00a0\u201cIn Safe Hands\u201d Achievement\u00a0Award for the initiative that provided\u00a0significant measurable gains in operational\u00a0quality\u00a0and financial metrics,\u00a0while reducing cardiac arrest\u00a0events, overall morbidity and mortality.<\/p>\n
This proactive approach led to a 34 percent reduction in cardiac arrests and a 12 percent drop in rapid response transfers\u00a0to the ICU, with survival rates climbing from 24 to 36 percent.<\/p>\n
Beyond clinical outcomes, the Deterioration Index\u00a0predictive model\u00a0has\u00a0helped to\u00a0deliver measurable\u00a0budgetary impact, saving $1.75 million in 2024 by avoiding unnecessary ICU transfers.<\/p>\n
\u201cHumans could essentially calculate\u00a0these\u00a0risk\u00a0scores,\u201d said Julie\u00a0Tanhauser, Information Technology Strategic Planner at St. Luke\u2019s, \u201cbut we\u00a0can\u2019t\u00a0continuously do it every second for every patient among\u00a0hundreds of\u00a0patients on\u00a0our\u00a016 campuses.\u00a0 That\u2019s\u00a0your AI\u00a0piece.\u201d<\/p>\n
The Deterioration Index transforms patient safety through predictive analytics.<\/p>\n
Today, the\u00a0initiative is deeply embedded in\u00a0St. Luke\u2019s workflow for\u00a0physicians,\u00a0nurses\u00a0and advanced practitioners, so they are alerted\u00a0immediately\u00a0when a\u00a0patient\u2019s\u00a0Deterioration Index\u00a0score crosses critical\u00a0thresholds before\u00a0something goes critically wrong, allowing providers and nurses a chance to intervene faster with the proper treatment.<\/p>\n
As part of the Network\u2019s broader AI strategy, the\u00a0Deterioration Index\u00a0exemplifies how machine learning can be responsibly integrated into care delivery, improving outcomes while\u00a0still\u00a0preserving\u00a0and enhancing\u00a0clinician judgment.<\/p>\n
This initiative has\u00a0enhanced\u00a0patient care outcomes by improving survival rates,\u00a0diminishing\u00a0unexpected ICU\u00a0transfers\u00a0and lowering\u00a0code blue\u00a0(cardiac arrest)\u00a0incidents over a period of three fiscal years.<\/p>\n