Updated IEC 60601-1-8 Breaks New Ground in Development of Alarm Sounds

September 7, 2021

Categories: AAMI News

The International Electrotechnical Commission’s (IEC’s) 60601 series of technical standards is concerned with the safety of medical electrical equipment and represents a benchmark for the commercialization of medical equipment in many countries. IEC 60601-1-8 is concerned with safety, as indicated by its title (Medical electrical equipment—Part 1-8: General requirements for basic safety and essential performance—Collateral standard: General requirements, tests and guidance for alarm systems in medical electrical equipment and medical electrical systems).

IEC 60601-1-8 recently underwent considerable development and updating, particularly in relation to the auditory alarm signals now recommended for use. The amendment to the standard was published in July 2020.

Our ears, rather than our eyes, are our primary “warning” sense, as we are unable to “shut” them or turn away from the source. Auditory alarm signals therefore are seen as essential safety measures and are used throughout the medical world. The downside of this is that auditory alarm signals are used for everything from immediately life-threatening cardiac events, to infusion pumps running low, and to batteries running low.

As a result, in certain clinical areas where considerable monitoring is taking place (e.g., intensive care unit, operating room), an overabundance of alarm signals can lead to the phenomenon of “alarm fatigue,” whereby listeners are overwhelmed by alarm signals to the extent that they tune them out, which of course leads to the possibility of overlooking important alarm conditions.

In 2011, a summit hosted by AAMI, The Joint Commission, ECRI, and the Food and Drug Administration was held in Washington, DC, to address the “alarm problem.” Ambitious targets were identified by AAMI’s then CEO, Mary Logan, who set the goal that “by 2017 no one will be harmed by an adverse alarm event.” AAMI set up a National Alarms Coalition and has been at the forefront of alarm fatigue reduction through proper rationalization of alarm system use.

Alongside this work, the alarm signals themselves have been the topic of considerable effort, and the amendment of 60601-1-8 includes a substantial update of the alarm signals themselves (i.e., the sounds that are generated when an alarm condition is deemed necessary). This represents a much-needed and long-overdue effort that will make a positive impact on alarm fatigue and safety.

An audio waveform flows in and out of an ear The updated version of IEC 60601-1-8 includes substantial content on alarm signals themselves—or the sounds that are generated when an alarm condition is deemed necessary. This long-overdue effort is expected to make a positive impact on alarm fatigue and safety.

The design, development, testing, and adoption of new auditory alarm signals is a challenging one for a number of reasons:

  1. In the past, the technology used to produce sounds was limited to either the output of a mechanical process (e.g., a bell or horn) or the output of a relatively simple piezo-type device producing a low-quality sound that is both harsh on the ears and of limited effectiveness in terms of localizability and the type of sounds possible.
  2. Few, if any, agreed-upon criteria exist for determining whether an auditory alarm signal can be deemed effective or acceptable.
  3. People appear to be very nervous about changing an alarm signal from one sound to a different one—a reasonable consequence given that in the past, few alternatives were possible and listeners are, by and large, unaware of the range of acoustic, psychoacoustic and psychological issues that can influence the effectiveness of an alarm signal.

The first challenge has been resolved in large part as a result of many current-day medical devices having music-quality sound generators and speakers integrated into them, making it possible in principle to use almost any type of sound as an alarm signal. The second two issues have been tackled head-on in the amendment of 60601-1-8.

Before the new auditory alarm signals were developed, it was recognized within the clinical alarms community that the sounds supporting previous versions of the standard (2012 and 2006) were less than optimal. They had been shown to be difficult to learn and discriminate, which is not surprising because each of the eight high-priority auditory alarm signals indicated in the standard had the same acoustic structure and same temporal pattern. The only difference could be found in their pitch patterns. Essentially, the alarm signals were a set of eight five-pulse melodies with the same rhythms.

The studies carried out among nurse and other clinical listeners showed that even after many exposures, the listeners were unable to identify each of the alarm signals to a safe criterion. It was even demonstrated that alarm signals designed for a previous incarnation of the standard in the 1980s were easier to learn and discriminate, with the suggestion that the standard should revert to these signals. However, enormous technology changes have occurred since the 1980s and our understanding of how listeners experience sound also has advanced; therefore, a new approach was devised. Although the current alarm signals could quite easily be improved upon, the challenge was to design and test sounds that represent the best that can be achieved, given that the process of updating a standard is slow and the results remain in place for many years.

An AAMI-funded project was undertaken to develop new alarm sounds for the eight categories of risk indicated in 60601-1-8. This project involved many of the world’s best-known alarms researchers and was a combination of research studies and collaboration with the relevant standards committees.

In the first studies, several sets of auditory alarm signals of different types were developed and compared for learnability and localizability (i.e., the ability to determine the direction from which a sound is coming, which is important when clinical staff are not at the patient’s bedside). All of the new designs outperformed the existing sounds on both tests, and the types of sounds that performed best of all were auditory icons.

Auditory icons are real, or close to real, sounds that are representative of their function. For example, the obvious auditory icon for the cardiovascular category is the sound of heartbeats. For other functions, the most appropriate icon is not necessarily as clear, so the selection of suitable icons went through an iterative design-and-test procedure during development of the auditory alarm signals. The final auditory icons appear in Table G4 of the amendment to the 60601-1-8 standard.

Because they are indicative of their function, auditory icons are both easy to learn and hard to forget. In the learning studies, participants were 80% to 100% correct in identifying the meaning of the alarm signals after a single exposure to the sound, which was much better compared with all other types of sound and far better than the existing sounds. Because the sounds are real-world sounds, they are much richer acoustically, meaning that they also are easier to localize.

The previous version of the standard and the amendment also differentiate between high-, medium-, and low-priority alarm conditions. In the amendment, the priority is indicated by combining the auditory icon with a “pointer”—a short burst of sound integrated into the icon that varies in urgency depending on whether the high- or medium-priority alarm condition is intended. The low-priority alarm signal is a single or double tone, similar to the previous version of the standard.

These results were presented to the relevant committees, and it was agreed that auditory icon alarm signals should be developed for the updated version of the standard.

Subsequent testing was carried out in simulations, from which it was determined that trained clinicians could recognize and respond to the auditory icons faster and more accurately compared with current alarm signals. This study also found the auditory icon alarm signals to be less irritating and fatiguing than traditional alarm signals.

Other tests showed that the auditory icon alarm signals are audible at low signal-to-noise levels, to the extent that some of them are still detectable in noise when that noise is three or four times higher than the alarm signal sound.

The series of studies carried out not only demonstrated the superiority of auditory icons in a number of meaningful areas but also provided a set of procedures that can be used in benchmarking alarm signals in the future, as all relevant research has been published in peer-reviewed journals and therefore is publicly accessible (see Edworthy et al. [2017], Edworthy et al. [2018], Mcneer et al. [2018], and Bennett et al. [2019]).

The auditory alarm sounds therefore are much improved in the updated 60601-1-8. In addition, the revised standard addresses another key issue. Considerable debate exists as to whether the eight categories of risk named in the standard are appropriate, and a couple of studies have suggested that a different way of categorizing risk might be more appropriate or effective. The standard now recognizes this and allows for the development of both new sounds and new categories by manufacturers and others that may wish to develop their own auditory alarm signals.

In support of this, the updated 60601-1-8 provides a data table showing how the new alarm signals in the standard perform in terms of learnability, localizability, simulation, and audibility. This table can be used by anyone wishing to develop new sounds, thereby allowing for meaningful comparisons between proprietary alarm signals and the signals indicated in the standard.

The new auditory alarm signals are freely available as WAV files on the ISO and IEC websites.