ED Accreditation Update: Citing reports of alarm-related deaths, The Joint Commission issues a sentinel event alert for hospitals to improve medical device alarm safety

Experts urge outside expertise, data-driven approach to make progress

With the proliferation of medical devices in recent years, hospital providers are now bombarded with a cacophony of sounds, signals, and other information emanating from these ubiquitous machines. While this type of messaging is well-intentioned, The Joint Commission (TJC) stresses that the sheer number of devices now in use is creating “alarm fatigue” among providers, which in turn is putting patients at serious risk. To draw attention to the issue, the accrediting agency has issued a Sentinel Event Alert, prompting hospital administrators and providers to thoroughly examine their practices related to the alarms on these devices, and make changes that will improve safety. (Also, see “Leadership, outside expertise needed to drive improvements in alarm safety,” below.)

This is not a new issue to many health care providers, but the scope of the problem is an increasing concern. TJC points out that between January 2005 and June 2010, 566 alarm-related deaths were reported in the U.S. Food and Drug Administration’s Manufacturer and User Facility Device Experience database. Further, TJC’s own sentinel event database includes reports of 80 alarm-related deaths and 13 serious alarm-related issues that occurred between January 2009 and June 2012.

Providers in some hospital units have to deal with thousands of alarm signals every day, and an estimated 85% to 95% of these alerts don’t require any intervention, according to TJC. This could be because the alarm conditions are set too tightly or the default settings are not suited to the patient or the patient population that is using the device. Providers may also be accustomed to alarms going off because the sensors have been positioned incorrectly or the electrodes are dried out. And with too many instances of “false” alarms, hospital staff can become desensitized to them. The danger is that care providers will fail to respond when a patient really does require immediate action.

Review alarms used in the ED

While alarm fatigue can put patients at risk in many hospital settings, the issue can pose particular problems in an environment as chaotic and noisy as a busy ED. “False alarms are frequent in the ED,” explains Sandra Schneider, MD, FACEP, professor and chair emeritus of the Department of Emergency Medicine at the University of Rochester, and an attending physician at Strong Memorial Hospital in Rochester, NY. Indeed, TJC cites the ED as one of the hospital sites where the majority of alarm-related events reported to the agency most often occurred, and it notes that the major contributing factors for these events were: absent or inadequate alarm systems, improper alarm settings, alarm signals that were not audible in all areas, and alarm settings that were inappropriately turned off.

Another problem that Schneider has observed is that many medical alarms that are used in the emergency setting are actually designed for intensive care unit (ICU) patients who typically stay in the unit for long periods of time. For example, she recalls one monitor system that had to “reboot” for 15 minutes between patients or it would automatically revert to the previous patient’s settings instead of the default setting. “This makes sense if you are working in an ICU, but in the ED there may only be a few minutes between patients,” she says.

Schneider advises ED managers to thoroughly familiarize themselves with the monitors and alarms that the department is using. “If possible, create default settings that work in the ED, and contact the manufacturers, if necessary, to customize alarms so that they work for your setting,” she says.

One issue that makes EDs particularly vulnerable to alarm fatigue is the number of patients that nurses often have to monitor. “Unlike an ICU where the nurses have two patients at the most, and they can be stationery, there is less often dedicated personnel that are in any one location who can watch the monitors [in an ED],” explains Terry Fairbanks, MD, MS, FACEP, who is the director of the National Center for Human Factors Engineering in Health Care at the MedStar Institute for Innovation, an attending emergency physician at MedStar Washington Hospital Center, and an associate professor of emergency medicine at Georgetown University in Washington, DC. “The nurse-to-patient ratio in most states is completely unregulated in the ED, although it is regulated on the inpatient units, so if we assign nurses the responsibility for watching the alarms, then they may have a high task load that makes this very difficult to do.”

Fairbanks notes that the types of alarms that administrators should be most concerned about in the ED are the dysrhythmia alarms on heart monitors, oxygen saturation alarms, and signals that a patient has a low respiratory rate. Ventilator alarms, which signal that a ventilator is not working or someone is not breathing, are also important, but Fairbanks explains that the alarms on ventilators tend to be very shrill and noticeable. “There is not really an alarm fatigue problem with ventilator alarms,” he says. “People tend to hear them and go to them because they don’t issue false alarms very often.”

Consider guidelines, training, and inspections

When issuing the Sentinel Event Alert on alarm safety in April, TJC made several recommendations for hospitals to follow, including:

• Ensure there is a process for safe alarm management and response in areas identified by the hospital as high risk.

• Inventory all medical devices with alarms that are used in high-risk areas or for high-risk conditions, and identify the default alarm settings and the limits that are appropriate for each care site.

• Create guidelines for alarm settings on medical devices that are used in high-risk areas and for high-risk conditions, and identify situations in which alarm signals are not necessary.

• Establish guidelines for customizing alarm settings and limits for individual patients. These should address circumstances in which limits can be modified to minimize alarm signals.

• Establish regular inspections and maintenance of alarm-equipped medical devices to ensure that they are operating as intended, and that the alarm settings are appropriate. Inspections and maintenance activities should be informed by manufacturer recommendations, risk levels, and experience.

In addition, TJC stresses that hospitals should equip providers and staff with training on the organization’s approach to safe alarm management as well as ongoing training regarding new medical devices or updates on existing devices. Further, the agency urges health care organizations to share information about alarm-related events as well as prevention strategies that have proven effective at minimizing or eliminating these events. (Also see “Embed specific goals in an action plan to drive improvement,” below.)

“One very important, impactful way to reduce false alarms is just by reducing the number of people who are on monitors in the first place,” observes Fairbanks. “We have this feeling that by putting someone on a heart monitor it reduces our risk because then in the small chance that their heart stops, we would find out. But what we forget is that in the bigger picture, it increases our overall risk if everybody and their brother is on a heart monitor because instead of saving [these devices] for the people who are truly at risk for sudden death, we cause this problem of constant alarms.”

While TJC already has several accreditation standards in place that are related to alarm safety, the organization says it is considering the implementation of a National Patient Safety Goal (NPSG) to further encourage health care organizations to address the issue. The agency conducted a field review of a proposed NPSG in February, and is now reviewing public comments on the measure.


Terry Fairbanks, MD, MS, FACEP, Director, National Center for Human Factors Engineering in Health Care at the MedStar Institute for Innovation, Attending Emergency Physician, MedStar Washington Hospital Center, and Associate Professor, Emergency Medicine, Georgetown University, Washington, DC. E-mail: terry.fairbanks@medicalhfe.org.

• Leah Lough, Executive Vice President, Association for the Advancement of Medical Instrumentation (AAMI), and Executive Director, AAMI Foundation, Arlington, VA. E-mail: llough@aami.org.

• Sandra Schneider, MD, FACEP, Professor and Chair Emeritus, Department of Emergency Medicine, University of Rochester, and Emergency Physician, Strong Memorial Hospital in Rochester, NY. E-mail: sandra_schneider@urmc.rochester.edu.

ED Accreditation Update: Leadership, outside expertise needed to drive improvements in alarm safety

When issuing its Sentinel Event Alert on medical alarm safety in hospitals, The Joint Commission (TJC) noted that addressing this problem will require leadership. It recommends that facilities establish a cross-disciplinary team that includes clinicians as well as representatives from clinical engineering, information technology, and risk management to address alarm safety in all patient care areas.

However, Terry Fairbanks, MD, MS, FACEP, the director of the National Center for Human Factors Engineering in Health Care at the MedStar Institute for Innovation, an attending emergency physician at MedStar Washington Hospital Center, and an associate professor of emergency medicine at Georgetown University in Washington, DC, believes TJC missed the mark with this recommendation in one respect.

“There is no one on that cross-disciplinary team that they are recommending that has any experience or training in alarm management, which is very complex, because finding the right threshold between having too many alarms and not enough alarms is quite involved,” explains Fairbanks.

People who work in safety science have PhDs in alarm management, observes Fairbanks. Further, he notes that the signal-to-noise ratio is a huge problem in safety throughout many complex, high-risk industries, so the issue of alarm safety is not unique to hospitals. “I think any ED trying to do this right should engage a community member who does alarm fatigue and alarm management in other industries as a part of what they do for living,” he says.

Academic medical centers might find such a person through the industrial/systems engineering or the cognitive psychology departments of the university with which they are affiliated, recommends Fairbanks. Other facilities might need to network with outside organizations.

The task of improving alarm safety could be made easier with more leadership on this issue among national groups, stresses Fairbanks. “We keep trying to do things at a local level,” he says. “Each ED forms a committee, and they are all spinning their wheels in the same way rather than trying to do an intelligent solution that everyone can do nationally.”

ED Accreditation Update: Embed specific goals in an action plan to drive improvement

One industry group that has focused considerable attention on the issue of alarm safety is the Association for the Advancement of Medical Instrumentation (AAMI), headquartered in Arlington, VA. The group partnered with The Joint Commission (TJC), the Food and Drug Administration, the American College of Engineering, and the ECRI Institute in 2011 to sponsor a summit on alarm safety, and it continues to work on the issue with these groups in making recommendations.

Leah Lough, the executive VP of AAMI and the executive director of the AAMI foundation, says that it is clear from the white papers the organization has published on the subject that for hospitals to improve their performance with respect to alarm safety, they must first understand the problem from a quantitative standpoint. “For example, understand how many alarms you have per bed, per unit, per day,” says Lough. “Pick a baseline measure so that you have a quantitative baseline to determine what it is you are trying to address.”

For instance, a hospital or ED that is interested in reducing the number of nuisance alarms — alarms that are not critical to the patient — may set a goal to reduce these types of alarms by three-quarters or to eliminate them, says Lough. With a concrete end-point in mind, administrators can then identify specific steps that will get them there.

Lough emphasizes that it is important for administrators to share their goals with staff, higher-ups, and everyone who will be impacted. “Make sure they are all on board,” she says.

Also, administrators need to consider the impact such goals will have on clinical staff and the tasks they are expected to carry out, advises Lough. “What is going on in the unit with the nurses, and how will your goal impact their workflow?” she says.

All of these issues can be identified and dealt with in an action plan that outlines specific steps a hospital or unit will take to improve alarm safety performance. “State the goals, get your baseline data, and that way you can benchmark your improvements,” she says.