Wednesday, July 26, 2017

Alarm Fatigue

I remember being a Pediatrics resident in a Neonatal Intensive Care Unit during my training and hearing many alarms going off simultaneously. Team members were frequently slow to respond to the alarms or they just didn’t respond at all. This was because they went off all the time and it was usually because the leads had become disconnected from the baby or the alarm was set to go off for something that we knew was not really an emergency…. like a momentary high heart rate because the baby was crying.  This concept, not reacting to an alarm because we have learned that they go off for reasons that do not require immediate action, has been termed alarm fatigue. Can you imagine being on a commercial jet where the pilot and the co-pilot were suffering from alarm fatigue?

“False” alarms breed contempt for the alarm system and teach people to ignore the alarm.  Do you always leave a building as soon as you hear a fire alarm? Probably not, because you have witnessed fire alarms going off when there was no fire. If this is the case, you have begun to assume that the alarm is false and the alarm is no longer serving its intended purpose. At GBMC, our standard work is to announce the testing of a fire alarm before we test so as to not create alarm fatigue and put our people at risk.

When I was a Pediatrics resident I received no training in complex systems, human factors or the Swiss Cheese Model of error. I did not know about the high-reliability concept of preoccupation with failure. No one taught me to be aware of the so-called “weak signals” that a catastrophe was brewing. No doctor or nurse told me that if I ignored cardiac monitors long enough eventually some baby might get hurt. Today, I know better but only because I have been involved in some cases and have read about many others where smart, well-trained and incredibly hard working clinicians got caught in the trap of alarm fatigue. Allowing devices on a hospital unit to alarm and ignoring the alarms is a hole in the Swiss cheese that is just waiting for a sick patient and hardworking and smart but human staff members to create a real problem.

So, I am asking all of my GBMC colleagues to not ignore alarms. If you hear an alarm respond to it. If the device is defective, then work with your manager to get it fixed or replaced. If devices are alarming because we don’t have good standard work (e.g. the leads are off the patient because they’ve gone for a test, but no one has shut off the device) then create the standard work. If devices are alarming because we have set the device to alarm at too low or too high a rate…reset the device.  If alarms are going off frequently for no good reason, we have to reduce the number of times this occurs.

I know how hard everyone works in our health care system. We should also want to reduce the number of alarms to reduce the stress on us and our patients as well.

Thanks for everyone’s help on this. Please tell me what you think.

New Executive Vice President for Medical Affairs and Chief Medical Officer
Congratulations to Harold Tucker, M.D. who was recently appointed as GBMC’s Executive Vice President for Medical Affairs and Chief Medical Officer. He is replacing the retired John Saunders, M.D. Dr. Tucker has been an active member of the GBMC's medical staff since 1984 and in that time has taken on several important roles. He was the Chief of the Medical Staff for six years and did an outstanding job of advocating for physicians and advanced practitioners and helping to improve patient care. Dr. Tucker will also continue to serve as the President of Greater Baltimore Medical Associates. Please join me in congratulating him as he takes on this important position.


  1. Howard J. Birenbaum, MDJuly 28, 2017 at 8:48 AM

    Hi Dr. Chessare,

    I wanted to share with you the work we did in NICU regarding alarm safety and nuisance alarms. This was begun a few years ago as a collaborative with VON's iNICQ. Pulse oximeters were identified as the most common false or nuisance alarm in NICU, resulting in increased noise levels, as well as a risk that staff would tune out and ignore true alarms.

    We identified those infants who did not require such monitoring upon admission to the NICU, and developed criteria for those who no longer required continuous pulse oximeter monitoring.

    We achieved about a 40% decrease, and submitted a poster to the PAS two years ago which was well received.

    There is much more that can be done with more up to date monitoring systems that could be tied in to nursing smart phones, as well as differentiating critical from non critical alarms. Allowing for differing alarm triggers by type, duration, and individual patient status would be helpful as well.
    Monitors that provided histograms of day, time of day, and type of alarm would be a useful quality improvement activity. Our current monitors do not have this capability.

    1. Thank you for your leadership in this area, Howard.


Thank you for taking time to read "A Healthy Dialogue" and for commenting on the blog. Comments are an important part of the public dialogue and help facilitate conversation. All comments are reviewed before posting to ensure posts are not off-topic, do not violate patient confidentiality, and are civil. Differing opinions are welcome as long as the tone is respectful.