Alarm Fatigue: MedSun Small Sample Survey Summary
MedSun: Newsletter #65, October 2011
Survey Topic: Alarm Fatigue - Summary of Responses
Year Conducted: 2011
This survey is an effort to learn about clinical experiences with medical device alarms, particularly the issue of alarm fatigue. In addition, FDA wants to understand other related issues that may occur with medical device alarms such as alarm discriminability, alarm volume and alarm activation thresholds.
Alarm fatigue is a clinical scenario that may occur when alarms sound so often that responders become desensitized to them and may not respond quickly enough or not at all. Alarm discriminability refers to the clinician’s ability to distinguish one medical device alarm from another to respond correctly to the actual alarm. Problems with alarm volume may occur when the alarm sound is not loud enough, can’t be adjusted to be loud enough for the responder to hear, or when settings have been changed and not reassessed frequently enough. Issues with alarm activation thresholds occur when the sensitivity levels for a given medical device alarm is adjusted based on a clinical situation or environment and is left at that setting and not readjusted for a new patient or new clinical situation.
The survey involved nine healthcare professionals from nine MedSun facilities. Two of the facilities are pediatric hospitals. Survey respondents include risk managers, staff nurses, nurse managers and biomedical engineers. The information that follows is a summary of responses.
During their shifts hospital staff encounter numerous medical device alarms and alerts which may lead to alarm fatigue. Six of nine respondents are familiar with the term alarm fatigue and describe it as becoming desensitized or immune to medical device alarms because there are so many. Those who are not familiar with the term say they are aware of the problem.
There are numerous medical devices with alarms. Respondents report the most commonly used devices with alarms are IV pumps, feeding pumps, ventilators, cardiac monitors, sequential compression devices, dialysis machines, nurse call systems, hospital beds, and chairs with exit alarms. Some additional devices in use with alarms include medication dispensing systems, extracorporeal membrane oxygenation (ECMO), and pocket phones, pagers, and telephones as secondary alarms. Seven respondents report that because of the barrage of sounds, instances of alarm fatigue occur on their nursing units every day.
Alarm fatigue is so common that it makes the ability to distinguish one alarm from another more challenging. When hospital staff hear a single alarm, or multiple device alarms they have several methods for identifying one from another. These methods include checking central monitors in the nursing station, searching room to room, using split screens in patient rooms, and recognizing sounds. The majority of respondents say their device alarms are loud enough to be heard from the central location on the unit. However, they report there are several factors that make it more difficult to hear alarms. These factors include the size and configuration of units, closed doors to patient rooms, and high noise levels on units.
Nearly all say they can distinguish urgent or critical alarms from all others most of the time because of a distinct difference in tone, sound, or rhythm. Many memorize the sounds to know the differences and say that over time staff learn the different tones. However, some respondents say that after staff are listening and responding to alarms for the duration of their shifts, the tones begin to blend together. In addition, one respondent reports that some device alarms sound identical, for example, patient beds and infusion pumps, making it even more difficult to distinguish which device is alarming.
In addition to auditory alarms, all respondents report that most of their devices have visual alarms. Visual alarms include blinking and/or colored lights, and monitors with pictures or text. Nearly all respondents find visual alarms helpful. Some respondents describe a visual “red” (critical) alarm that flashes along with the room call light at the central monitor that helps to quickly locate the alarm and the patient room. However, none thought that voice alarms on devices would be helpful. They believe that voice alarms would add to the noise and confusion on the nursing unit and think it would be even easier to tune out voice alarms than standard alarms. Also, they believe it may upset patients and their families.
Alarm activation thresholds are another critical concern involving medical device alarms. Half of the respondents say volume and parameter threshold settings are adjusted based on a patient’s clinical status and clinical judgment. Others say that default parameter settings are adjusted depending on the type of nursing unit such as the cardiac intensive care unit. Six of nine respondents say parameters are adjusted by unit nurses or other clinical staff. The remaining respondents have their biomedical engineering department make the parameter adjustments. When discussing silencing alarms, seven respondents say they are able to temporarily silence alarms for purposes of troubleshooting for a period of thirty seconds to three minutes. After this time frame, the alarm usually resets and turns back on. The majority of respondents say that serious alarms, such as those for lethal heart rhythms cannot be permanently disabled. And, according to nearly all respondents, once a patient is discharged from the device the alarm settings revert back to their default settings.
Respondents have several ideas about alarm modifications that may improve the safety and effectiveness of medical device alarms. One suggestion is increasing the use of “smart alarms.” An example of this is an IV pump occlusion alarm that self-corrects when the occlusion is the result of a patient who temporarily bends an arm with the IV or rolls on the tubing. Another suggestion is linking the apnea and oxygen saturation monitor alarms because apnea monitors are very sensitive and alarm even when oxygen saturations values are within set parameters. Also, many oxygen saturation alarms do not distinguish between values that are high or low. For example, the alarms for values of 89% and 30% are the same sound so different sounds would be more helpful.
Another suggestion is providing an escalating alarm or two different tones on cardiac monitors specifically for detecting bradycardia in infants. Having two distinct tones for bradycardia and tachycardia particularly for infants in neonatal intensive care units will help clinicians identify problems faster.
Also, several respondents say that when multiple alarms are sounding at once, it would be beneficial if the device could indicate the nature of the problem so the clinician can determine the type of response required. Examples of this include having different tones for a “leads-off patient” alarm versus an alarm for a critical patient issue, or an alarm that specifically indicates a high heart rate versus a low heart rate for an individual patient.
Respondents’ suggestions also include designing better technology to improve the safety and effectiveness of device alarms. These include providing clinicians with the ability to receive a text message about a device that is alarming on a smart phone, designing noise-cancelling technology for unit hallways, providing portable monitoring through use of a pad or tablet, and improving algorithms in monitor software for individual patients that are more accurate and can help eliminate false alarms. In the event of an alarm that is disabled, some suggest having a question appear on a monitor screen asking, “Do you want this alarm to remain off?” Also, many respondents believe that in addition to improving technology, working with physicians to establish criteria for patients that need monitoring and those that do not will also help reduce alarm fatigue and keep patients from being monitored unnecessarily.
Special Studies and Surveys are two of many tools the Agency is using to evaluate the public health impact of the potential problems associated with the use of medical devices. Additionally, FDA continues to receive adverse event reports from its Medical Device Reporting program. FDA will also continue to make use of the literature and other published information. FDA scientific, medical, nursing and engineering staff are made aware of the survey results as needed. If FDA believes there is a significant risk of adverse events as noted from the survey, it will combine those results with data gained from the other sources. FDA will work with the manufacturers and health care professional organizations to make important information known to the clinical community. Additionally, FDA continues to work with manufacturers to ensure the development, testing and promulgation of methods for reducing the risk associated with these devices and to minimize the complications from adverse events that may occur in the course of normal usage. If the results of any survey raise serious concerns about the safety of these devices, FDA may convene an Ad Hoc group of clinical and manufacturing representatives to discuss further actions.