(1236) Distribution of Continuous Vital Sign Monitoring Alerts at the Bedside

Introduction: Monitoring alerts are ubiquitous, and may lead to alert fatigue. We sought to describe the numbers and types of alerts experienced on patient care units in a large multi-year sample.

Methods: Continuous monitoring of heart rate (HR), respiratory rate (RR), pulse oximetry (SpO2) and arterial blood pressure (ABP) and intermittent noninvasive blood pressure (NIBP) were performed on patients in one stepdown unit and 3 intensive care units of an urban tertiary care center. We analyzed all alerts generated by the commercially available bedside monitoring system over 5 years (2018-2022 inclusive). We evaluated the alerts due to vital sign exceedances (high and low HR, RR, mean ABP, or mean NIBP, and low SpO2) according to numeric thresholds set by bedside caregivers.

Results: There were 8,282 patients in the sample, and a total of 40,047 monitoring hours. Overall, there were a total of 4,032,502 alerts due to vital sign exceedance. Of those, the frequency distribution from highest to lowest was low SpO2 (41.33%), high HR (22.25%), low RR (10.44%), low HR (10.14%), high RR (8.55%), low mean ABP (3.40%), high mean NIBP (1.99%), low mean NIBP (1.49%), and high mean ABP (1.41%). When viewed as categories, respiratory related alerts (all RR and SpO2 alerts) comprised ~60% of alerts, while cardiovascular-related alerts (all HR and blood pressure alerts) comprised ~40% of alerts.

Conclusions: This study evaluating over 4 million alerts across multiple years demonstrates that in this sample the majority of monitoring alerts were respiratory related in nature. Heightened awareness of clinicians towards anticipating respiratory decompensation in advance of threshold exceedance may be helpful.

Funding: R01NR013912