Overview of tool
TriageGO is a clinical support tool that utilizes AI to help recommend inpatient acuity levels. This tool is embedded into existing EMR’s without disrupting workflows and helps with front-end prioritization with limited patient information. It ultimately replaces limited ESI criteria to support objective and risk-based triage-decision-making
Background/Method
The AI model is trained on 100,000+ patient records at each site and is adapted to specific location demographics, resources, and triage objectives. It utilizes inputs such as demographics, arrival mode, chief complaints, vital signs, and EMR medical history.
What makes this solution stand out
This addresses a two-fold problem present in modern-day triage methods: 1) high variability and lack of patient differentiation, 2) limited ability to detect high-risk patients.
Key Outcomes
TriageGO has seen significant success in the time saved per patient and the detection of critical care outcomes.
While triage can be a highly subjective process, this has helped to identify nuances in patient conditions and minimize ambiguous ESI Level 3 designations that lead to sub-optimal resource use and unnecessary waiting.
Growing pains and remaining concerns
This tool indirectly impacts time saving and resource utilization. Those that do not operate split-flow models may not see as significant operational impact when implementing this tool.