Risks of trying to be safe dept.

“Alarm fatigue” is said to be a growing peril in hospitals as “nurses become so desensitized to the constant beeping that they don’t hear or ignore important warnings that a patient’s condition might be worsening.” [Liz Kowalczyk, Boston Globe]


  • I’ve actually published a paper on dealing with this topic. Part of the problem is that much of the time the alarms are warning when there is really nothing wrong. When something bad happens the “experts” go back and ask, “why did you cancel the alarm?” and they answer is “that things always going off for no reason”.

    The industry puts an alarm on something to mitigate a hazard like “What if someone kicks the plug out” or “What if the O2 pressure goes to zero”. Who can blame them, they need to pass FDA.

    The worst thing you can do is blame the worker and not rework the system.

  • Us old geezers learned this lesson as children from the story “The Boy Who Cried Wolf” – if you raise too many false alarms, no one will listen when you raise a true one.

    I suppose the multitude of alarms are another response to the modern lawsuit culture – if a manufacturer raises an alarm then they’ve protected themselves from liability if something bad happens to a patient.

  • This is a variant of the warning sticker plague that I have been bitching about for decades, now. Everything is so plastered with yellow stickers that you don’t read the one that says, “This thing will blow up and kill you.”

  • Mannie raises a good point. But selling something that warns “This thing will blow up and kill you” is a poor business plan.

  • I work at a factory that has heavy forklift traffic. They set a beeper to go off when the forklift is in motion. The problem is, you hear a constant beeping, so you desensitize to it and the beeping loses its effect as a warning device.

  • Another good parallel is the California Prop 65 warning signs that are on every business that’s open to the public, from Starbucks to chemical refineries.

  • Same effect for the car alarms of Brooklyn circa the late 1990’s. The sound never stopped. Nobody paid the slightest attention.

  • This is a common problem in industry. Statistical process control has multiple methods of detecting a potential failure, each of which has false alarms as well as real alarms. The cost of responding to false alarms is a necessary calculation to make when setting your control limits.

    And when someone will die if the alarm is not answered, shock collars might be useful.

  • Another problem with all these alarms is the stress they put on the patient. When my dad was in the hospital after complications following heart surgery, he had alarms on his ventilator, the warmer for the vent oxygen, the pulse oximeter, both IV pumps, the feeding tube pump, the heart monitor, and the bed alarm. His roommate had a similar number of alarms. I spent many hours with my dad, and there was not a single hour without one or more alarms going off–if he coughed, changed position in bed, took a deeper-than-usual breath, or moved the hand with the pulse-ox, or if his roomie did similar things. Both men were seriously ill, both too weak to use their call-buttons, and both had tracheostomies, so couldn’t have told staff about the alarms even if they could use the call-buttons, so they had to lie there listening to the alarm(s) and wonder if they were about to dieThe one item that could have benefitted from some sort of alarm (and doesn’t have one) was his catheter, which was pulled out by accident when he was being moved. Ouch! (Google “Foley catheter” and note the balloon to get an idea how this might feel).

    I suspect that the alarms are on lots of these devices because they seem like a good idea to the purchasing agents, who probably have no idea what a seriously ill patient is hooked up to.

  • Hospitals could be empowered and encouraged to turn off alarms for non-critical systems, but tie them to a flashing yellow light outside the patient’s door and also next to his room number at a nursing station switchboard. Critical systems could remain wired to alarms and flashing red lights. Protocols could be set up for doctors and senior nurses to downgrade red to yellow (malfunctioning false alarm) or upgrade yellow to red on a case-by-case basis.