By Emma Yasinski
BU News Service

Walking past patient’s rooms on the cardiac floor of Boston Medical Center, I could hear nurses ruffling papers at the central station. The wheels of a monitor scratched against the hard, white floors. Nurses and doctors spoke at relaxed volumes. A nurse walked by and I heard a quiet beep-beep sound. “Was that an alarm?” I asked Deborah Whalen, a clinical nurse manager at Boston Medical Center. “No that’s just a beeper.” We’d been standing on the cardiac floor for 15 minutes and had yet to hear an alarm.

This hall sounded very different 15 months ago.

Nurses here used to be overwhelmed by alarms. In most other hospitals, they still are. Barbara Drew, RN, PhD, FAAN at the University of California San Francisco recently counted 2.5 million alarms during a month-long period in their Intensive Care Unit. Each patient triggered an average of 45 alarms per hour. The study also found that 88% of these alarms were false. Over time, the alarms became background noise for caretakers. “When I worked on my undergrad…Georgetown was right near a national airport. [At first] I would hear every plane landing and taking off,” said Whalen, describing her first year in a hospital setting. “Then, by the end of my freshman year, I didn’t even hear a plane. You get accustomed to the sound and therefore you block it out.” The decibel level of the background alarms was so high that not even a plane taking off would get her attention anymore.

This is the type of environment that investigators say led to the death of a patient in the cardiac unit at Massachusetts General Hospital in 2010. When there are so many alarms blaring, nurses sometimes lower their volumes, or even turn them off in an effort to reduce noise. The patient was recovering from surgery and had been hooked up to a variety of monitors. When his heart rate dropped, not one of the 10 nurses on the floor heard the alarm or came to his aid. The investigation showed that the alarm didn’t sound at all. A nurse had purposely turned the alarm off.

The Joint Commission, a group responsible for accrediting hospitals, listed alarm fatigue as a crucial problem in patient safety in 2014. There have been instances, such as that in 2010, where desensitization to alarms led to disastrous consequences.The Joint Commission reported 80 deaths related to alarm issues between 2009 and 2012. Reporting these events is optional for hospitals, so it is likely that the actual number of deaths related to alarms is much higher. To be accredited in 2016, hospitals will need to improve their alarm management.

Boston Medical Center had two main types of audible alarms: crisis alarms and warning alarms. The difference between the two alarms is a single beep. A warning alarm is two high-pitched beeps, while a crisis alarm is three high-pitched beeps. Each beep sounds the same. The warnings don’t usually require immediate intervention. With such a high volume of alarms, it can be difficult for nurses to distinguish between a crisis and a warning. After observing the pilot floor, Whalen and her team discovered that most of the noise was coming from warning alarms, but that nurses rarely answered them. They were much more likely to respond to the higher level crisis alarms, but the distraction and background noise from the warning alarms still distracted them and made it difficult for them to identify a true crisis.

Boston Medical Center had just begun studying the effects of alarm fatigue when the patient at Massachusetts General Hospital passed away. After determining that they too might be vulnerable to the effects of alarm fatigue, Whalen and her colleagues began developing a plan to safely reduce the noise on the hospital floors. The first change they made was removing superfluous warning alarms. The few warning alarms that were deemed urgent were upgraded to crisis alarms. Now the condition that would have triggered a two-beep warning before still showed up in the patient’s history, but do not make distracting noise. The audible alarms are reserved for true emergencies.

Next, the hospital had to decide who had the power to change the alarm settings on a specific patient and when. Every individual patient is different. A heart rate that may signal an emergency in one patient may not signal an emergency in another. For example, the hospital’s default settings will sound the alarm sound if a patient’s heart rate falls below 45 beats per minute. However, many athletes have lower-than-average heart rates when they are resting or sleeping, meaning their alarms would sound frequently, but not signify a problem. In this case, two nurses at Boston Medical Center can come together and choose to set new parameters on this individual’s monitor. When the two nurses agree on the appropriate setting, they change it and then report it to a doctor on staff. Putting this power in the hands of the nurses on staff allows for an immediate reduction in noise volume. Whalen’s reasoning for requiring two nurses to make the change is that “Two nurses together making a bad decision (knock on wood) has never happened.” These changes, along with increased training, helped reduce alarms from approximately 90,000 per week to 10,000 per week. At the end of the six week pilot, the nurses begged to keep the changes in effect.

Drew, whose study on the prevalence of alarms was funded by General Electric, is working with manufacturers to identify places where the technology can improve. “There’s a lot of things hospitals can do,” she said “but manufacturers also need to step up.” Drew argues that hospitals need smarter alarms, able to take data from several sources on a single patient, such as heart rate and oxygen levels. The alarm would be able to identify patterns that often occur together before and during emergencies. Then, only a single alarm would sound in the event of a specific emergency. However, these smarter alarms could present legal issues. Rather than reporting on specific parameters, a smart alarm is responsible for diagnosing certain conditions. This has the potential to make the alarm manufacturer liable in the event of a misdiagnosis.

Whalen, Drew, and a few others have become leaders in the field. Hospitals around the country have been requesting their help in understanding and implementing these strategies. While the Joint Commission is not requiring that alarm fatigue be reduced until 2016, it is requiring that hospitals show that they have identified it as a problem and are starting to work on it this year. “We are trying to take an incremental approach,” said Gerald Castro, MD, PhD, Project Director of Safety Initiatives at the Joint Commission. “What we want them to do is to first address the problem within their organization.” One of the greatest difficulties with addressing the problem, is that the data regarding it is incomplete. Researchers aren’t certain how often alarm fatigue leads to problems for patients, because hospitals are not required to report it. Plus, if they do choose to report it, they can send the report to any of several organizations, including the Joint Commission, the FDA, or another Patient Safety Organization. These organizations haven’t been able to combine their data yet. Castro describes this as the next big step for the Joint Commission.

As we waited on the floor, still no true alarms sounded. Whalen offered to set one off for me just so I could hear the three shrill beeps and understand what it sounded like. She did, and while several nurses looked her way, they saw her turn it off and continued about their business. We watched another nurse enter a new patient’s information into the monitor and set his parameters. She typed his last name and first initial into the small blank spaces, and clicked on options from some drop down menus. After she walked away, I heard “Beep! Beep! Beep!” Whalen turned around, scoured the monitor. She clicked a button and the alarm stopped. “That one was junk,” she said, before quietly walking me back to the elevator.