Too many medical device alarms are akin to crying wolf; too few could neglect to alert clinicians of an emergency.
When designing a system intended to monitor a patient’s wellbeing, deliver treatment, or both, knowing when and how to draw clinicians’ attention is one of the most complex challenges. If your device calls the clinician’s attention too loudly and too frequently, they will learn to ignore it. If it neglects to draw attention when urgently needed, the patient’s safety is at risk. A balanced approach can help to ensure that your medical device alarms are as effective as possible.
I recently participated in a panel at a recent FDA public workshop, “Physiological Closed-Loop Controlled Devices,” and shared the following insights with members of the medical device design community.
Where to Start
Begin by identifying tasks the clinicians need to perform in collaboration with your device. Look for instances when the clinicians need to take action to complete a successful treatment and/or ensure patient safety.
Imagine where the clinicians may be at these times (e.g., within a few feet of the device, up to 12 feet away, or possibly out of the room). Consider how many clinicians might be present at the time and the role each might play when intervention is required.
Refer to your use error analysis to identify instances when you’ll need to draw the clinicians’ attention in order to reduce the likelihood of an error that could lead to significant harm.
The Fewer, the Better
Sort through your potential alerts and alarms. Eliminate all but the few you believe to be crucial. When we design our devices, we work hard to provide useful information throughout the therapy. Sometimes we forget the complexity of a critical care environment and think it would be good for the device to ping the clinician periodically to provide “all is well” notices. Even if the device provides very different visual and auditory cues for “all is well” updates as compared to “take action now” alarms, when we draw the clinician’s attention regularly without the need for action, we dilute the effectiveness of all the device’s alerts and alarms.
Anticipate what the clinicians will need to know when the device requires their intervention. Imagine one clinician briefing another, and provide the information required to act with confidence, such as the following:
- Concisely state the action required.
- Provide instructions on how to complete the task.
- Display information about the recent therapy delivery and patient vitals.
- Use a progressive disclosure approach so that the clinician can drill down to the level of detail desired.
Test, Test, Test
Remember that your device is part of a larger system composed of many devices and clinicians. Take early prototypes of your device, with approximations of your intended alerts and alarms, to clinical training facilities. Ask clinicians to interact with your device as they would in an actual critical care environment. Just seeing your device perform in the context of this environment will provide you with a wealth of information you can use to create the delicate balance needed for effective medical device alarms.
|Don't miss Battelle director of electrical, software, and systems engineering Melissa Masters speak about the future of hospital and medical device interoperability at BIOMEDevice San Jose on December 3, 2015.|
Debbie McConnell is human factors lead at Battelle.
[image courtesy of STUART MILES/FREEDIGITALPHOTOS.NET]