Barbara J. Drew, RN, PhD, FAAN, FAHA
David Mortara Distinguished Professor in Physiological Nursing Research, Clinical Professor of Medicine, Cardiology
University of California, San Francisco (UCSF)
Department of Physiological Nursing
San Francisco, CA 94143-0610
Medical Research: What is the background for this study?
Dr. Drew: Physiologic monitors used in hospital intensive care units (ICUs) are plagued with alarms that create a cacophony of sounds and visual alerts causing “alarm fatigue.” Alarm fatigue occurs when clinicians are desensitized by numerous alarms, many of which are false or clinically irrelevant. As a result, the cacophony of alarm sounds becomes “background noise” that is perceived as the normal working environment in the ICU. Importantly, alarms may be silenced at the central station without checking the patient or permanently disabled by clinicians who find the constant audible or textual messages bothersome. Disabling alarms creates an unsafe patient environment because a life-threatening event may be missed in this milieu of sensory overload.
To date, there has not been a comprehensive investigation of the frequency, types, and accuracy of physiologic monitor alarms collected in a “real-world” ICU setting. For this reason, nurse and engineer scientists in the ECG Monitoring Research Laboratory at the University of California, San Francisco (UCSF) designed a study to provide complete data on monitor alarms.
Medical Research: What are the main findings?
Dr. Drew: Our study is the largest prospective study to date on the alarm fatigue problem. We found a staggering total number of alarms in our 5 adult ICUs (>2,500,000 in one month) when counting all audible and text alarms. Although many of these alarms were configured to be inaudible text messages, we still found a high audible alarm burden of 187 audible alarms per bed per day. A noisy alarm environment interrupts patients’ sleep and invokes fear in patients and their families.
We found the excessive number of ICU monitor alarms is primarily due to 3 factors:
1) inappropriate alarm settings by the clinical staff,
2) certain patient conditions like pacemaker devices or other ECG abnormalities like atrial fibrillation or bundle branch block, and
3) algorithm deficiencies by the monitor manufacturers such as failing to analyze all the available ECG leads to find a lead without motion artifact during patient movements.
We analyzed 12,671 arrhythmia alarms and found that 88% of them were false alarms. This is like having a car alarm in your neighborhood that goes off incessantly because it is triggered too easily. Neighbors get irritated by the noise but they ignore it so if someone is really breaking into a car, no one would report it.
Medical Research: What should clinicians and patients take away from your report?
Dr. Drew: What hospitals can do right now is to review their alarm settings, both the hospital default settings and the settings that nurses set up at the bedside for an individual patient. Our article gives practical advice about how to reduce the excessive number of alarms. For example, we discuss what settings are safe to change from an audible alarm to a text message alarm.
Our job as clinical nurse scientists is to inform and work with engineer scientists at UCSF and industry to improve these algorithms. Because computer devices have the potential to be more reliable than humans, an opportunity exists to use these devices to reduce the alarm fatigue problem. For example, in our published article, we make recommendations about how monitors could be more interactive to help nurses tailor the alarms better for individual patients so we can avoid the “one size fits all” approach that results in too many alarms.
Medical Research: What recommendations do you have for future research as a result of this study?
Dr. Drew: There is inadequate data on commercially-available electrodes and how often they should be changed to prevent false alarms due to electrode failure. It is also unknown what impedance measurement would indicate electrode failure and how best to measure impedance continuously. There is also insufficient data on the best skin prep regimens that will decrease electrode impedance without causing skin breakdown.
Adding alarm delays before an alarm is triggered would be effective in reducing alarm fatigue for many vital sign measurements. However, future research is required to determine whether these alarm delays are safe as well as effective in reducing alarm burden. Any arrhythmia algorithm change should be tested to determine the effect on false alarm rates and identification of any unintended adverse consequences.
Insights into the Problem of Alarm Fatigue with Physiologic Monitor Devices: A Comprehensive Observational Study of Consecutive Intensive Care Unit Patients
Barbara J. Drew, Patricia Harris, Jessica K. Zègre-Hemsey, Tina Mammone, Daniel Schindler, Rebeca Salas-Boni, Yong Bai, Adelita Tinoco, Quan Ding, Xiao Hu