Comparing Continuous with Periodic Vital Sign Scoring for Clinical Deterioration Using a Patient Data Model

Area of Science:

• Medical Technology
• Clinical Monitoring
• Health Informatics

Background:

• Periodic early warning scores (EWS) have limitations due to large measurement intervals, potentially delaying the detection of clinical deterioration.
• Continuous vital sign monitoring with real-time algorithms, such as the Visensia Safety Index (VSI), may overcome these limitations.

Purpose of the Study:

• To compare the effectiveness of a minute-by-minute monitoring algorithm (VSI) against periodic EWS in detecting clinical deterioration and patient workload.
• To evaluate sensitivity, alert frequency, number of warnings needed to evaluate (NNE), and time from initial alert to escalation of care (EOC).

Main Methods:

• A prospective comparative data modeling cohort study (NCT04189653) involving 1529 admissions of continuously monitored medical and surgical inpatients.
• Comparison of continuous VSI alerts versus periodic EWS for detecting EOC events (Rapid Response Team activation, unplanned ICU admission, emergency surgery, or death).
• Analysis of VSI alerting minutes in patients with and without EOC.

Main Results:

• Continuous VSI alerted for 55% of EOC events, compared to 51% for periodic EWS.
• VSI had a higher NNE (152 alerts/EOC) than EWS (21 alerts/EOC), generating more frequent alerts (0.99/day/patient vs 0.13/day/patient).
• Time from detection to escalation was similar (8.3 hours for VSI vs 5.2 hours for EWS, P=0.074); however, VSI showed a higher percentage of alerting minutes in patients with EOC (2.36% vs 0.81%, P<0.001).

Conclusions:

• Continuous vital sign monitoring with VSI demonstrates potential for earlier alerts of clinical deterioration compared to periodic EWS, despite similar sensitivity.
• The increased percentage of VSI alerting minutes in patients with EOC suggests its utility in risk identification.
• Further research may refine algorithms to balance sensitivity and alert frequency for optimal clinical implementation.