Skip to content

Alarm fatigue


It is easy to be bamboozled by the amazing technology we have at our fingertips – portable monitors, ultrasound machines, tricorders – but you only really need to know one button. It’s that one we all instinctively reach for before we have even assessed the patient. It’s the big button marked “Silence“.

Emergency departments, operating theatres and intensive care units are noisy places. Our ears are assaulted by the auditory artillery so often it is a wonder that we don’t all get alarm fatigue.

What is alarm fatigue?

With most physiological alarms completely irrelevant to the patient, it is no wonder that we quickly learn to ignore them. The high number of false alarms leads to a degree of alarm fatigue. Some of the alarms you might hear are relatively unimportant and may be viewed as a nuisance – the ECG lead has fallen off – and some – the patient has gone into VT – are critical.  And unfortunately, most of them sound the same.

A systematic review by Paine et al. found that whilst up to 26% of alarms in an adult ICU might be considered actionable, only around 1% are actionable in a paediatric ward setting. It is no wonder that alarm fatigue has been linked with a number of sentinel events if 99% of them require no action. Such sentinel events have led to ‘alarm hazards’ being ranked in the top three causes of technology-related death and have rightfully become a target of The Joint Commission’s National Patient Safety Goal.

Let’s take a look at this paper from JAMA.

Bonafide CP, Localio AR, Holmes JH, Nadkarni VM, Stemler S, MacMurchy M, Zander M, Roberts KE, Lin R, Keren R. Video Analysis of Factors Associated With Response Time to Physiologic Monitor Alarms in a Children’s Hospital. JAMA pediatrics. 2017 Apr 10.

This single-centre prospective cohort study examined factors associated with alarm response time. The authors analysed 551 hours of video related to the care of 100 children and involved 38 nurses. Whilst the study tries to extrapolate which factors might be related to delays in response time, I think it is worth starting with some of the raw data.

In the 551 recorded hours, the nursing staff were subjected to 11,745 alarms – or one every 168 seconds!! Only 50 of the 11,745 actually required any action. Actionable alerts or alarms were defined as those which identified physiological conditions that required either an intervention or consultation with a health care provider.

What factors were associated with a faster response time?

  • A potentially lethal arrhythmia
  • A one to one nurse to patient ratio
  • Previous interventions required
  • Less experienced nursing staff
  • No family members present

The reason for the faster response time seems intuitive in the first three cases. Perhaps less experienced nursing staff respond faster because they do not know which alarms to ignore.

What about a slower response time?

I think the most telling reason for a delayed response was how far into the shift the staff were. As each hour into the shift passed, a 15% delay in response time was noted. Perhaps the constant tintinnabulation of tachycardic tones leads to cognitive overload, or perhaps it leads to a degree of desensitisation. It is more likely that there is a complex interplay of the two.

What were the limitations of this study?

This was a single-centre study involving just a small number of nursing staff. The nurses were aware that they were being videotaped, and this may have affected their response times. However, one would suppose it would globally reduce the time it took to hit the big silence button.

How can we reduce alarm fatigue in the workplace?

Dealing with the problem requires both clever engineering solutions as well as some human ingenuity. Both software and hardware engineers are hard at work behind the scenes, working out how to use multiple physiological parameters combined to come up with a more accurate alarm. I’ve often wondered where the standardised charts of physiological data for children come from. Most of the data is based on small studies that are 50 years old. Using real-time data from hospital patients, analysts are able to calculate the new normal. They can then use this data to alter alarm parameters.

Is there anything I can do on my next shift?

Karnik and Bonafide developed a great framework to reduce alarm fatigue that you can read here.

Set some limitsMost monitoring devices allow you to set alarm parameters. Use them. If your monitors do not recognise age-related limits, you’ll need to set them yourself. Whilst a pulse rate of 160 might be normal for a 6-month-old, it is not for a 6-year-old. If you are lucky, your monitoring device has a (small) built-in delay. If the sats dip below 90% for just a few seconds before rebounding back, do you need to hear an alarm?

Stick to itSome of the invalid non-actionable alarms are due to sticky electrodes coming off, so take the time to attach them securely.

Take it off. Does the patient actually need to be monitored? This is not so much a problem in the paediatric world, thank goodness, where careful observation often outweighs technology. Do you need to use continual sats monitoring in a child with normal oxygen saturations or not requiring oxygen therapy?

Switch off. If monitors, ventilators, syringe drivers and infusion pumps all bleeping and pinging may lead to cognitive overload and sensory desensitization, please switch off your phone. Unless you need it for work, how can you complain about alarms going off if you reach for your pocket every time you get a Facebook notification?

I think Edgar Allen Poe got it right…

Hear the loud alarum bells-
Brazen bells!
What a tale of terror, now, their turbulency tells!
In the startled ear of night
How they scream out their affright!
Too much horrified to speak,
They can only shriek, shriek,
Out of tune,
In a clamorous appealing to the mercy of the fire,
In a mad expostulation with the deaf and frantic fire,
Leaping higher, higher, higher,
With a desperate desire,
And a resolute endeavor,
Now- now to sit or never,
By the side of the pale-faced moon.
Oh, the bells, bells, bells!
What a tale their terror tells
Of Despair!

How they clang, and clash, and roar!
What a horror they outpour
On the bosom of the palpitating air!
Yet the ear it fully knows,
By the twanging,
And the clanging,
How the danger ebbs and flows:
Yet the ear distinctly tells,
In the jangling,
And the wrangling,
How the danger sinks and swells,
By the sinking or the swelling in the anger of the bells-
Of the bells-
Of the bells, bells, bells,bells,
Bells, bells, bells-
In the clamor and the clangor of the bells!

From The Bells by Edgar Allen Poe c. 1848


Bonafide CP, Localio AR, Holmes JH, Nadkarni VM, Stemler S, MacMurchy M, Zander M, Roberts KE, Lin R, Keren R. Video Analysis of Factors Associated With Response Time to Physiologic Monitor Alarms in a Children’s Hospital. JAMA pediatrics. 2017 Apr 10.

Bonafide CP, Lin R, Zander M, Graham CS, Paine CW, Rock W, Rich A, Roberts KE, Fortino M, Nadkarni VM, Localio AR. Association between exposure to nonactionable physiologic monitor alarms and response time in a children’s hospital. Journal of hospital medicine. 2015 Jun 1;10(6):345-51.

Korniewicz DM, Clark T, David Y. A national online survey on the effectiveness of clinical alarms. American Journal of Critical Care. 2008 Jan 1;17(1):36-41.

Goel VV, Poole SF, Longhurst CA, et al. Safety analysis of proposed data-driven physiologic alarm parameters for hospitalized children. J Hosp Med. 2016;11(12):817-823.

Karnik A, Bonafide CP. A framework for reducing alarm fatigue on pediatric inpatient units. Hospital pediatrics. 2015 Mar;5(3):160.

Lawless ST. Crying wolf: false alarms in a pediatric intensive care unit. Critical care medicine. 1994 Jun 1;22(6):981-5.

Sowan AK, Reed CC. A Complex Phenomenon in Complex Adaptive Health Care Systems—Alarm Fatigue. JAMA Pediatrics. 2017 Apr 10.

Paine CW, Goel VV, Ely E, Stave CD, Stemler S, Zander M, Bonafide CP. Systematic review of physiologic monitor alarm characteristics and pragmatic interventions to reduce alarm frequency. Journal of hospital medicine. 2016 Feb 1;11(2):136-44

Ancker JS, Edwards A, Nosal S, Hauser D, Mauer E, Kaushal R. Effects of workload, work complexity, and repeated alerts on alert fatigue in a clinical decision support system. BMC Medical Informatics and Decision Making. 2017 Apr 10;17(1):36.



No data was found

Leave a Reply

Your email address will not be published. Required fields are marked *

4 thoughts on “Alarm fatigue”

  1. Regarding using simulation to train for this- it’s tough to simulate an entire busy clinical environment- for example, a nurse taking care of 4 patients, 2 are on a monitor and 2 are not, they are getting paged to do this and that, alarms are going off, they need to give meds on time and attend to parents with questions, etc. Would also require a much longer sim session than we usually do since the effects we are seeing can take hours to accumulate (slowing of response time etc) but not impossible. Hawthorne effect also comes into play. The closest I’ve seen published is here:

    A very different type, complexity, and scale of simulation than the 15 minute resus scenario we’re used to doing. But not impossible.

  2. Great stuff and so true. Have a think about the parallel situation of hospital exec. Each department has their “crisis” shoots off alarms as multiple emails cc everyone and creates a huge amount of noise. Looking for the true signal “what I have to act on” becomes difficult, both to the clinician trying to respond to it all, but I imagine also to hospital exec trying to work out whether each departments crisis really is a crisis.

    Learnt this on a different job:

    The bottom line- cc only those that need to know
    use language appropriate to the actual situation- don’t overplay or underplay
    Beware the boy who cried wolf

    This applies to managing up or down.

    From the sim side:

    Yes we can train to understand the effect of distraction, fixation and alternatively to understand and appreciate focus and concentration. We can shape our environment whether by re-setting alarms or removing or limiting distractions and interruptions. Andy I keep on thinking we are going to see you at a course like ACME that creates, discusses and debriefs this type of scenario. Shall I hold a spot for you my friend?

  3. Great stuff Andy

    I have been reading a bit about the “wet” side of this equation- our brains and how they filter all this sensory input.
    We actually do it incredibly well…. with practice
    REcall your first day as a Med Student in a critical care ward, the sheer volume of data is overwhelming- but with time, training and experience our brains are able to sort out the signal from the noise.
    We probably evolved this capacity in early primate stages- it comes in handy now
    Maybe we should train explicitly for this.? Would make a great Sim experiment