Abstract 19717: Patient Characteristics Associated with False Arrhythmia Alarms in Intensive Care

Abstract

Introduction: A high rate of false arrhythmia alarms leads to clinical alarm fatigue, i.e. desensitization and inappropriate silencing of alarms. Objective: To determine patient characteristics associated with frequent false arrhythmia alarms in the intensive care unit (ICU). Methods: This descriptive, observational study prospectively enrolled every consecutive patient admitted to 1 of 5 adult ICUs (77 beds) at an urban medical center during March 2013. All monitor alarms and continuous waveforms were automatically stored on a secure server. Expert nurse scientists used a standardized protocol to annotate 6 clinically important arrhythmia alarms (asystole, pause, ventricular fibrillation, ventricular tachycardia, accelerated ventricular rhythm, ventricular bradycardia) as true or false. Medical records were examined to acquire patient characteristic data. Results: A total of 461 unique patients (mean age=60) were enrolled, generating 2,507,822 alarms, including all levels of arrhythmia, parameter, and technical alarms. There were 48,404 hours of patient monitoring time, resulting in an overall alarm rate of 52 alarms/hour. Investigators annotated 12,674 arrhythmia alarms; 11,137 were determined to be false (88%); 250 patients generated at least 1 of the 6 annotated alarm types; 2 patients generated 6,930 alarms (55%). Monitoring time for each patient was measured and number of false alarms/hour calculated for the clinically important arrhythmia alarms (range 0 - 8 false alarms/hour). Patient characteristics were compared with false alarm rates using nonparametric statistics to minimize the influence of outliers. Significant associations included: older age, ≥ 60 years (p=0.005), confused mental status (p<0.001), breathing with mechanical ventilation (p<0.001), and wide QRS complex due to bundle branch block, BBB (p=0.006) or ventricular paced rhythm (p=0.001). Conclusion: Patients likely to trigger a higher number of false arrhythmia alarms are those with confused mental status, on mechanical ventilation, or who have a wide QRS complex due to BBB or pacing. Algorithm improvements need to focus on better noise reduction (motion artifact with confused state) and distinguishing BBB and paced rhythms from ventricular arrhythmias.