As a human factors specialist frequently involved with the design of medical equipment, I am very familiar with the concept of alarm fatigue. For those not familiar, it refers to the cacophony of noise coming from much of the medical equipment that surrounds healthcare workers every day. Each piece of medical equipment has its own alerts and alarms to monitor, and while any one piece of equipment may not alarm all that frequently, when combined with those coming from the dozens or even hundreds of other devices in a hospital unit, the result can be a near constant din. In an ICU, it is not uncommon for nursing staff to be exposed to 150-400 alarms per patient in the course of a single shift.
Patient Facing Alarms
I’m not a clinician, but am a patient—and I have been experiencing my own personal version of alarm fatigue. Though perhaps alarm annoyance or maybe even the occasional alarm rage might be a better way to describe it. My life has become ruled by devices that yell at me. (Well, not literally, but that is often what it feels like.)
As an insulin-dependent diabetic, I wear both an insulin pump and a continuous glucometer (CGM) to help manage my blood sugar levels. Each of these devices issues audible alerts and alarms for various things. My CGM is set up to alarm whenever my blood sugar is too high or too low, or if it’s falling so quickly that I’m in imminent danger of hypoglycemic symptoms. These alerts are extremely useful and I am happy to have them.
The alarms from my insulin pump, however, drive me batty. I’ve turned off every alert and alarm that I have the option of turning off, but it still yells at me far too often for my liking.
For example, my insulin pump controller beeps 90 minutes after I put on a new wearable pump, which I have to do every 3 days. It is designed to remind me to test my blood, so that if the new wearable isn’t working well, I can catch it before my blood sugar fluctuates too violently. But because I also wear a CGM that alarms for those events, that 90-minute pump alarm is completely redundant and utterly unnecessary. And yet the device does not permit me to turn it off and it repeats every few minutes until I acknowledge it.
Last night, my pump failed around midnight and I awoke to an alarm, which meant I had to put on a new wearable pump. But that also meant that “helpful” pump alarm was going to sound in 90 minutes and wake me up again. If I’d instead opted to just not put on a new pump, in addition to a very high blood sugar in the morning, my controller would have beeped every 15 minutes: another “helpful” reminder that can’t be turned off. In the past when this has happened, I’ve moved the controller to a different room so I could sleep in peace—but because it will continue to alarm every 15 minutes for the rest of the night, I’ve learned it drains the controller’s batteries (which are not rechargeable). Neither option was ideal, or even good. So around 1:30 a.m., I was woken up yet again by my pump alarm. I immediately shut it off and then struggled to get back to sleep, cursing my pump while I lay there.
While that’s easily the most aggravating alarm from my pump controller, there are many others that I cannot turn off, even though I would dearly love to, and these alarms are often the bane of my diabetic existence. First, if I am within a certain number of hours of the wearable pump expiring, there’s a 3-beep alarm that sounds multiple times over the course of several minutes and does so repeatedly until I acknowledge it. So if I am driving, with my controller in my bag in the back seat and it starts beeping, I basically have to listen to it until I get wherever I’m going.
If my pump has “expired”, the same alarm sounds every hour, even though the pump continues to work perfectly fine for another eight hours (and the alarm sounds with increasing frequency during that last hour). When the insulin reservoir gets too low, there’s another 3-beep alarm that repeatedly sounds until I acknowledge it. And of course, if I don’t put on a new wearable pump right away, as previously noted, an alarm will sound every 15 minutes, with no way to snooze or otherwise indicate that there may be valid reasons why I don’t want to put a new one on right away.*
The ironic thing about all of this is that my CGM genuinely does have many useful alarms that I have the option to enable. But I’ve chosen not to because I hear far too many alarms as it is and any more might just send me over the edge. If I had the option to reduce the number of alarms I hear from my pump, I’d probably enable more of the CGM alarms, which could actually result in better control of my blood sugars. But I’ve reached my alarm threshold—I just can’t bring myself to do it.
Addressing Personal Alarm Fatigue
While the Joint Commission has urged hospitals and medical device manufacturers to address alarm fatigue for clinical personnel, we as designers of in-home medical devices and even consumer devices, sometimes neglect to do the same for our users. But as more and more devices become “smart” or “connected”, the proliferation of alerts, alarms, and even push notifications is likely to continue. Without offering users greater control over the ability to customize those alerts and alarms for their very individualized needs, we run the risk of our users choosing to not use our devices to their full potential—or worse, we risk our users choosing to not use our device at all and instead going to our competition.
When designing medical devices for in-home use, we need to be mindful of what alarms sound and when and from where else—otherwise what could seem like a helpful nudge from a designer’s perspective may very well end up being just another noise that drives the user nuts.
*I understand that the manufacturer of my insulin pump created all of these alarms out of concern about potential lawsuits should a patient’s health suffer while using the device. It’s not just a failing of the manufacturer, but a failing of our legal system that absolves people of personal responsibility. As a designer, I’d default these alarms to be enabled and require the user to acknowledge the risks prior to disabling them, but I firmly believe the user should have the option to disable them, especially when using devices with overlapping functionality, as I do.
[…] This article was originally published on the Daedalus Blog. […]