How UMass Memorial Health Improved Patient Safety by 44% with KATE AI

Hospital decreased safety intelligence (SI) reports filed from 50 to 28 annually, by reducing clinical risk at triage.

Download Case Study
UMass Memorial Health Case Study
Ken Shanahan

It's like having a second set of eyes. And in emergency care, that can be life-saving.

Ken Shanahan
Senior Director of Emergency Medicine, UMass Memorial Health

UMass Memorial Health At A Glance

138,200
Annual ED Visits
781
Licensed Beds
Epic
EHR System
Level 1
Academic Medical Center & Trauma Center
The Challenge

Reducing Risk Events and Administrative Burden at Triage

UMass Memorial Health wanted to improve patient safety at triage by reducing risk events. In 2021–2022, staff filed 50 Safety Intelligence (SI) reports, identifying a clear opportunity to improve high-risk identification at the front door. This volume of reporting was exacerbated by subjective disagreements over triage accuracy, even when assessments were correct. Consequently, the manual reporting process became a significant administrative burden, consuming 10-15 minutes per incident and diverting valuable nursing time away from direct patient care.

50
Safety Intelligence Reports
Filed (2021–2022)
10–15 min
Consumed Per
Incident Report
Manual
Reporting Process Diverting
Nurses From Patient Care
The Solution

KATE AI: A Real-Time Safety Net for High-Risk Patients

UMass Memorial Health deployed KATE AI to objectify triage accuracy and provide a real-time safety net for high-risk patients. Integrated directly into the Epic EHR, the system applies clinical risk intelligence to identify anomalies in care and escalate cases that merit a second look. By providing an unbiased, data-driven validation of nurse acuity assignments, the hospital increased clinical confidence and preemptively aligned the care team, significantly reducing the errors and safety events.

  • Real-time clinical risk intelligence at triage
  • Direct Epic EHR integration
  • Unbiased validation of nurse acuity assignments
  • Automated anomaly detection & escalation
  • Objective clinical documentation trail

Measurable Results

Significant improvements across safety, efficiency, and risk mitigation.

Reduction in Safety Intelligence Reports
44%
SI reports dropped from 50 to 28 annually, signaling higher confidence in initial patient assessments
Reduction in Reporting Time
67%
Time per report reduced from 10–15 minutes to 2–5 minutes, freeing nurses for patient care
Medical Malpractice Claim Avoided
Validated
Objective clinical record leveraged to avoid a potentially costly malpractice claim

A Safer, More Efficient Emergency Department

The initiative achieved a 44% reduction in Safety Intelligence reports, dropping volume from 50 to 28 annually. This shift signals a fundamental improvement in safety culture and triage precision, as the reduction in reports correlates directly with higher confidence in initial patient assessments.

This operational improvement also delivered significant efficiency gains, reducing the time spent on reporting by 67% (reduced from 10-15 minutes to 2-5 minutes). Furthermore, the medical center successfully leveraged this objective clinical record to avoid a potentially costly medical malpractice claim, validating the system's ability to protect both patients and the institution.

KATE AI gives our triage nurses an unbiased, data-driven second opinion in real time. It's not replacing clinical judgment—it's reinforcing it, and the results speak for themselves.

Ken Shanahan
Senior Director of Emergency Medicine, UMass Memorial Health

More Customer Stories

UMass Memorial Health

AI Assistant Online

Select your role (optional)

This helps us provide more relevant information

Ask anything about Mednition's KATE platform, clinical validation, safety controls, or EMR integration. Questions about KATE? Get product, clinical, and integration answers, or request a demo.

Want to speak with a Mednition expert? Just say the word.

This assistant provides product and clinical information. It does not provide medical advice.

AI Assistant Online

How can I help you today?

AI can make mistakes. Check important info.

l,..