Technology overload: An alarming danger
Preventing alarm fatigue to improve safety
Beep. Beep. Beep.
The telltale noise of a hospital is the bleeps, the sweeps, and the creeps of the many machines hospital patients are hooked up to during their stay. These machines have alarms—as many as 400 per patient—and many are important. They alert staff if a patient needs immediate care.
But those alarms can become such an ingrained part of the background noise that frontline staff may no longer recognize when an important alarm is sounding. Just as a mother of three small children may no longer pay attention to their screaming, with so much noise, nurses may not hear what they need to in order to provide safe care.
Take, for example, a well-publicized event at Massachusetts General Hospital in January 2010. Federal investigators concluded that alarm fatigue contributed to the death of a heart patient. A bedside alarm had been turned off, but another alarm sounded at the nurses’ station—an alarm that 10 nurses failed to hear.
The problem is serious, says Kathryn Pelczarski, director of the applied solutions group at the ECRI Institute, a nonprofit organization that researches best practices for medical procedures, devices, drugs, and processes. ECRI listed “alarm hazards” as the No. 2 technology hazard for 2011 in its November 2010 issue of Health Devices. The list reflects what ECRI believes are risks that should receive priority after analyzing how often they occur, the severity of the consequences, recent literature and recalls, and the institute’s own investigative and consulting experience.
The report concludes that often when staff become overloaded with alarms, they will improperly modify the alarm settings—something that should only be done in concert with and after careful consideration of the patient’s condition.
“I think most hospitals would agree that alarm fatigue is a pervasive problem,” says Pelczarski. “To give you an example, in some of the critical care units that I have been in, I have actually seen that their physiologic monitors, their bedside alarms, may be as many as 150–400 physiologic alarms per patient per day. And in these units, the nurse-to-patient ratio is typically one nurse for one or two patients. If you think of a nurse just dealing with two patients, that’s 300–800 alarms. But the problem gets worse when they hear the alarms for all the other patients in that unit—typically alarms for 10 or 12 patients.”
Pelczarski says staff become so desensitized to the noise that it gets to the point where they don’t actually hear the alarms. Adding to the confusion, many of the machines sounding off are close to each other and sound similar.
The Joint Commission retired a National Patient Safety Goal on improving the overall effectiveness of alarms in 2004. The trend at that time, says Pelczarski, was to simply focus on ensuring that alarms were audible and not turned off. The Joint Commission told The Boston Globe in February 2010 that it has seen a resurgence in alarm-related incidents.
Addressing the problem
Hospitals may want to start at the source: the purchase of alarm technology.
“It’s certainly good to look at the machines, see how the alarms are set, how easy it is to set the alarms within the physiologic monitoring system that you’re purchasing, and to question whether you have the ability to change the alarm level,” says Pelczarski.
For example, if a patient’s blood pressure is out of the normal range, an alarm will sound. However, if it is determined that the patient is healthy outside the regular normal rate, it is wise to change the settings to avoid the alarm constantly sounding, which can promote alarm fatigue. Pelczarski gives the example of a marathon runner whose heart rate is naturally lower, often under 60 beats per minute. This would cause an alarm set for the average patient to go off even though the runner’s situation is noncritical.
Another important factor to consider is how clearly the alarm is displayed and whether a critical alarm’s sound is distinguishable over the noise of the other alarms in the facility.
Nuisance alarms (e.g., false alarms caused by a loose electrode) add to the problem of alarm fatigue. Tailoring alarms to the patient as much as possible is critical, as is ensuring proper skin preparation technique before placing electrodes on the patient. Frontline staff members should also strive to troubleshoot false alarms when they occur, not ignore them or find a way to work around them.
Still, these simple methods need to be held together by a core alarm management strategy.
“The only way to prevent alarm fatigue is through better alarm management,” says Pelczarski. “This is really a very complex issue. There are core strategies that can be effective, but they alone are not going to solve the problem. Each hospital and each care area within the hospital really has a unique set of circumstances with unique processes in place, unique vulnerabilities, and variations of many common problems.”
Elements such as patient population, clinical needs, staffing patterns and care models, the architectural layout of the care area, and the alarm coverage model will vary from place to place, says Pelczarski. Because all these elements vary, so do the specific underlying causes of alarm fatigue, she notes. A strategy needs to be developed, and just as an emergency management strategy needs to be facility specific, so does alarm management.
With a plan in place, Pelczarski recommends a few core strategies:
- Make all alarms actionable so that the nurses are only alerted to clinically significant alarms and system alarms that need a response. This can be addressed by analyzing whether the default alarm settings are appropriate for the patient population in question.
- Consider incorporating a brief delay in alarm notification (e.g., five to 10 seconds) so an alarm does not go off for a problem that quickly resolves. Pelczarski urges caution when implementing this strategy because hospitals need to think critically about what fits—it’s important to ensure that the delay does not jeopardize quick access to necessary critical care. She gives the example of one hospital that set a nine-second delay for ventilator alarms because alarms would sound every time a patient had a brief coughing spell. The delay ensured that alarms wouldn’t sound for brief episodes of coughing.
- Implement appropriate preventive maintenance. Routinely replace electrodes before they dry out. It may be advisable to replace them every 24 hours, depending on what types of electrodes are used.
- Address alarms as soon as they occur. Delayed alarm response contributes to alarm fatigue because the alarm perpetuates. Pelczarski notes this is especially important with “leads off” alarms, or alarms that sound when the patient is no longer being monitored because a lead fell off. The problem is twofold: the unaddressed constant alarm contributes to alarm fatigue, and the patient is not monitored until the alarm is addressed.
Where to start
A multidisciplinary team composed of a chief nursing officer or director of quality, key medical staff, clinical engineering, nurse managers, frontline nurses, patient safety officers, and IT staff should be gathered to begin the effort. The first step might be to analyze adverse events and near misses and whether they might be attributed to alarms. These data will allow the team to determine a starting point. It might also be a good idea to observe alarm coverage and survey staff members about their concerns regarding alarms, says Pelczarski.
Before implementing a new alarm notification system (e.g., the use of a mobile device for alarms), it’s important to think of all the potential failures of the new system, says Pelczarski. Think of what needs to be alarmed, what patients’ specific conditions are, and what your hospital’s current alarm culture is. The principles must remain the same, no matter what technology exists.