Developing a real-time detection tool and an early warning score using a continuous wearable multi-parameter monitor

Area of Science:

• Medical Technology
• Patient Monitoring
• Clinical Decision Support

Background:

• Current early warning tools for clinical deterioration suffer from low specificity and infrequent measurements.
• There is a need for pre-symptomatic and real-time detection of patient deterioration.

Purpose of the Study:

• To develop and evaluate a multi-parameter real-time warning score (MPRT-WS) for early detection of patient deterioration.
• To compare the performance of MPRT-WS against the National Early Warning Score (NEWS).

Main Methods:

• Collected over 2 million physiological measurements from 521 participants (361 high-risk patients, 160 controls).
• Utilized parameters including heart rate, SpO2, respiratory rate, blood pressure, body temperature, stroke volume, cardiac output, and systemic vascular resistance, recorded every 5 minutes.
• Defined risk levels based on specialist-defined cutoffs and compared with NEWS.

Main Results:

• MPRT-WS and NEWS showed similar overall risk level distribution (p=0.863).
• Among 39 deteriorating patients, MPRT-WS identified 30 with 'High' or 'Urgent' scores significantly earlier (42.7 ± 49.1h prior) than NEWS (only 6 patients).
• Key abnormal vitals for MPRT-WS alerts included SpO2, SBP, SV, DBP, and SVR.

Conclusions:

• MPRT-WS offers timely, pre-symptomatic alerts for patient deterioration.
• The tool leverages high-frequency monitoring to improve early detection capabilities.
• This can enhance medical teams' ability to intervene proactively.