The IEC 60601-1-8 standard has taken many years to harmonize. Now that it is finalized, device OEMs may need to reconsider how to design alarms into medical equipment.
Audible alarms serve multiple functions in medical equipment, not the least of which is that they protect manufacturers against liability. As device makers continue to integrate more functions into each piece of medical equipment, they must also incorporate more types of warning sounds.
Medical equipment manufacturers usually develop proprietary alarms for their products. Sometimes these alarms sound completely different, even on devices made by the same company. One of the drafts of IEC 60601-1-8 highlighted some of the user discontent with medical alarm signaling. Problems identified included difficulty in identifying the source of an alarm, alarms being too loud and distracting, and high rates of false-positive or negative alarm conditions. The confusing environment resulted in users disabling alarm systems or adjusting the alarm limits to extreme values, which essentially renders the alarms useless. The ultimate concern was that problems deciphering medical equipment audible warnings could lead to patient injury or even death.
Efforts to harmonize alarm systems in medical equipment had been moving slowly over the last decade.1 However, this changed in 2003 when international standard IEC 60601-1-8 was issued. Although compliance is voluntary, it is expected that many medical equipment manufacturers will eventually move toward adopting this standard. In not following its guidance, manufacturers risk liability issues, but even more, they risk missing out on sales to larger institutions that may soon begin to require compliance to IEC 60601-1-8.
What is IEC 60601-1-8?
IEC 60601-1-8 is a comprehensive international standard that specifies basic safety and essential performance requirements and tests for alarm systems in medical equipment. Its 71 pages, including four annexes, cover both visual and audible alarms, but the vast majority of the specifications are devoted to audible-alarm issues for medical equipment and applications.
The comprehensive document addresses several aspects of alarm issues, including what kind of medical condition should trigger an audible warning sound and what the specific frequency and shape of the sound's waveform should be. Many of its most important topics for audible alarms have direct bearing on how medical device designers should approach alarm systems.
Priority Condition. The equipment designer is responsible for gauging when an alarm should trigger. The designer determines the priority of the condition being monitored by the medical equipment. IEC 60601-1-8 gives guidance on whether a patient's condition should be assigned a high, medium, or low priority. This guidance is based on the potential result of a failure to respond to the cause of the alarm condition and how fast the potential harm could happen to the patient. For example, a high priority would be assigned if death or irreversible injury could happen quickly, but a low priority would be assigned if only minor injury or discomfort may happen after a period of time. A heart rate of 170 on a treadmill test may warrant a low-priority condition whereas this same heart rate at an intensive-care monitoring station may be assigned a high priority.
Audible Alarm Bursting. There are three burst requirements listed in IEC 60601-1-8. As this term is used here, a burst is essentially a pulse train of individual sounds (e.g., a three-burst train would work like this: beep, beep, beep, pause, beep, beep, beep). The burst requirements correspond to the three priority conditions discussed above. A high-priority-condition burst has 10 fast pulses that repeat, a medium-priority-condition burst has 3 slightly slower pulses that repeat, and a low-priority-condition burst has 1 or 2 even slower pulses that may optionally repeat. The standard details the number of pulses, the shape of the pulse train, and the spacing of the pulses for each of the priority conditions.
Figure 1. (click to enlarge) This example of an audible sound frequency shows how to meet the pulse requirements set forth in section 18.104.22.168 of IEC 60601-1-8.
Characteristics of the Individual Pulses. IEC 60601-1-8 requires that an individual sound pulse have a fundamental frequency (musically known as pitch) somewhere between 150 and 1000 Hz. It also specifies that the pulse contain at least four harmonic sounds from 300 to 4000 Hz. Figure 1 shows a visible example of an audible sound that meets these requirements. The peak to the far left in Figure 1 is the fundamental frequency of the sound, and the four peaks to the right of the fundamental peak represent the harmonic tones. To the ear, the resulting blended sound would be similar to hearing four different musical notes played at the same time. A final requirement for the sound pulse characteristic is that the sound level (measured in decibels) of the four harmonic tones be within ±15 dB of the fundamental frequency tone.
Optional Melodies. IEC 60601-1-8 requires that at least one set of audible warning sounds used in the medical equipment meet the specified frequency, pitch, and priority conditions. However, it also allows the option of providing more than one set of audible warning sounds that vary the fundamental frequency of the audible warning sound.
Essentially that means that the equipment could play a musical melody instead of just beeping over and over with the same-frequency sound. The sequence of musical notes for each melody is spelled out for several different kinds of medical applications. For example, a ventilation alarm is assigned one melody while a cardiac alarm has a different melody. By assigning a unique melody for each kind of medical equipment, the hope is that medical personnel can become familiar with each different kind of melody and more quickly identify the source of an audible warning sound.
It would be a significant investment of time and money to design a system that produces such melodies. Nevertheless, some medical equipment designers may want to consider incorporating that type of design because it would provide a competitive advantage. This is especially true if large medical institutions decide that the efficiencies gained by having all medical equipment play such melodies outweighs additional equipment cost.
Sound-Level Requirements. IEC 60601-1-8 contains only basic requirements for the volume of audible warning sounds. The only requirement listed is that a high-priority-condition warning sound be louder than a medium-priority-condition warning sound, which in turn must be louder than a low-priority-condition sound. One practical recommendation is that there should be a 3–6-dB difference between each of the three warning sounds. It takes at least a 3-dB difference for the human ear to discern that one sound is louder or softer than another sound. Conversely, you don't want the high-priority-condition sound to be too much louder than the low-priority-condition sound, or either the lower sound will be too soft to be heard or the higher sound will be too loud and distract those nearby.
IEC 60601-1-8 does not address exactly how loud the sound levels should be. The standard's authors are relying on the experience of the medical equipment designers, the audible-alarm designers, and the equipment users to determine the volume of the warning sounds. It is a fine line to make a warning sound loud enough to be heard in all situations, but not so loud that the people near the warning sound are annoyed or distracted.
Technical Alarm Condition. Besides the high-, medium-, and low-priority conditions that require a warning, there may be other reasons for medical equipment to trigger an audible sound. These conditions are called technical alarm conditions. Examples include issuing a sound because of a keyboard input error, a power failure, or an equipment error.
The requirement in IEC 60601-1-8 states that sounds issued for technical alarm conditions should not be easily mistaken for the sounds issued for the three priority conditions. Although this seems to be common sense, it does not hurt to remind equipment designers of this fact. For example, nurses don't like paging doctors only to find out that an unplugged blood pressure monitor power cord caused an equipment warning sound.
There are several ways to make technical-alarm-condition sounds distinguishable from the priority-condition sounds. Examples include using very-high-frequency pitches, very short staccato or clicking sounds, or sounds that are much longer than priority-condition ones.
Reminder Signals. It is common for equipment designers to allow the users of medical equipment to temporarily silence the audible alarms so that medical personnel can concentrate on resolving the condition of the patient that caused the alarm in the first place. IEC 60601-1-8 allows for an alarm to be temporarily silenced but states that if the alarm condition is still present, it may be appropriate for the equipment to continue to issue a periodic beep or other indicating sound to remind the medical personnel that the alarm condition still persists even though the alarm sound itself has been muted.
Verbal Alarm Option. IEC 60601-1-8 does provide the option of using verbal messages for warning signals, and this choice may at first seem appealing. However, verbal messages do present some unique challenges. For example, in a medical emergency, verbal messages from equipment could quickly become distracting to responding personnel who are trying to talk to each other. In addition, verbal messages are language dependent. Language is a significant issue considering the needs of the increasingly multilingual workforce in the United States and abroad. A device maker would have to provide and support a number of verbal message language sets.
Because of these complex issues, the standard adds a requirement that states a verbal message must be validated through clinical testing or by simulating clinical usability testing. Essentially, equipment that uses verbal messaging must work in a clinical environment, which is a significant step over the usability testing normally required for other alarm types. Nonverbal priority conditioning requires the designer only to document that all standards are met; it is assumed that such equipment will function well in the field.
Of course, a designer may decide that verbal messages offer a competitive advantage. However, the advantage offered by such messaging must offset the added complexity and cost inherent in verbal alarms.
How to Meet Alarm Requirements of IEC 60601-1-8
It will take time for medical equipment designers to digest all the requirements in IEC 60601-1-8 and decide how to best meet these requirements. For some critical or large types of medical equipment, the increased time and cost of the development cycle and the increased cost of the equipment itself is not a large concern. However, other equipment providers may have a hard time communicating the increased component and development cost of their equipment to customers. For example, many pieces of medical equipment currently use low-cost piezoelectric audible alarms for their signaling. These are the same kind of alarms used in smoke detectors or at checkout counters at grocery stores. These low-cost alarms can no longer be used because they will not meet the complex frequency requirements of IEC 60601-1-8. Although a higher-priced audible alarm for a large x-ray system may be absorbed easily, a high-cost audible alarm could require makers of low-cost medical equipment to raise their prices significantly to cover the increased development and component costs.
If piezoelectric alarms cannot meet the requirements of IEC 60601-1-8, what kind of alarm system can?
Microcontrollers and Microprocessors. Microcontrollers or microprocessors can be used to ensure compliance with the standard. Using sophisticated programming techniques, such components can electrically generate the complex frequency signal (consisting of the fundamental frequency and four harmonic sounds) shown in Figure 1. Once the complex signal is generated, it must be amplified using appropriate power components and audio processors. The signal is then sent to a speaker to generate the audible warning sound. One advantage of such designs is that they can accommodate the various melodies and verbal messages discussed above.
Designers should be aware, however, that this type of system takes time to develop. It is not easy to write software that integrates five different frequency signals into one electrical signal. In addition, audio processors, power components, and speakers for the system can be two to three times the cost of piezoelectric alarm components. Designers must be able to justify the price increase.
Acoustic Transducers. Another option is to create alarm systems that use acoustic properties. Instead of producing the four harmonic sounds by electronic means, it is possible to produce these sounds acoustically. To accomplish this, a transducer or speaker can be mounted in a specially designed acoustic chamber. The user applies a simple sine wave to the part, and the harmonic sounds are automatically generated inside the acoustic chamber. One example of how multiple-pitch sounds are generated acoustically is a wooden train whistle. When someone blows into the whistle, several different sound frequencies are heard at one time. Such a sound is very different from the single-frequency sound produced by a referee's whistle, for example.
Besides the relatively simple setup, another advantage of mounting a transducer or speaker in an acoustic chamber is that the sound level can be increased significantly by acoustic means. This means that high-power components and audio processors are unnecessary.
Figure 2. An internally mounted miniature speaker such as the one contained within this octagonal housing may be an appropriate choice for manufacturers looking to integrate an alarm system that is small and requires few electronic parts.
Figure 2 shows an octagonal plastic housing that contains an internally mounted miniature speaker. The acoustic chamber inside the octagonal housing shapes sound waves. The sound is emitted through the small hole in the front. The resulting sound meets the frequency requirements of IEC 60601-1-8. The part measures 22 mm in diameter and is 13 mm high. It uses a 0.25-W speaker but is capable of producing a 95-dB sound at 10 cm. It should be noted that this type of alarm assembly cannot produce different melodies. The fundamental frequency cannot by varied without losing significant sound levels.
Medical equipment providers with products already on the market may not believe that they should adopt the IEC 60601-1-8 standard, or at least not yet. There is risk involved in adopting the requirement now. Doing so may mean an extended design cycle and increased equipment costs. And if customers do not perceive the added value of the design, it could result in lost sales. However, not adopting the standard as soon as possible also represents a risk. If the market is hungry for medical equipment that meets the harmonized standard, companies that wait may also experience lost market share and slow-moving inventory.
Standardizing the alarms also means a company may save time on new products because they will not have to reinvent alarm systems for each device. Companies that implement IEC 60601-1-8 as efficiently as possible stand to gain a competitive advantage over slower moving manufacturers and will be better prepared for the global market. One way to gain such an advantage is to work closely with knowledgeable suppliers who can help interpret the complex requirements in IEC 60601-1-8 and offer solutions that may not be obvious at first glance.
Dan O'Brien is outside sales engineer for Mallory Sonalert Products Inc. based in Indianapolis. He can be reached at firstname.lastname@example.org.