Alarming Facts About Smoke Detectors

Editor's note: This article is an edited transcript of the live seminar Alarming Facts About Smoke Detectors. To view the recorded course, click here.

Residential Fire Statistics

According to the Federal Emergency Management Agency, fires represent the fourth largest accidental killer in the United States behind motor vehicle accidents, falls, and drownings. Each year nearly 3,000 Americans die in a residential fire, which averages out to approximately eight deaths per day. Data show that the majority of victims are children and/or the elderly. While kitchen fires are the number one cause of residential fires, most residential fire fatalities start somewhere other than the kitchen. Additionally, the majority of residential fire fatalities occur between 11 p.m. and 7 a.m., when most people are asleep.

There is an assumption that the majority of residential fire fatalities occur when there are not any smoke detectors installed in the home. However, data compiled by the U.S. Fire Administration show that when someone perishes in a home fire, 40% of the time a working smoke detector was actually present. While there are most likely multiple factors contributing to a 40% residential fire fatality rate in the presence of a working smoke detector, one issue that needs to be addressed involves the apparent limitations of current smoke detectors in effectively arousing individuals with hearing loss.

Growth of Residential Fire

It is important to examine how quickly a fire can develop in a home.

According to a timeline established by the National Fire Protection Agency (NFPA), within the first 30 seconds that a fire ignites in a home, it grows fairly rapidly. After more than just one minute from the initial development, flames of the fire spread, consuming oxygen in the air while sending increasing concentrations of carbon monoxide in the atmosphere. As a result, smoke starts to fill the room of origination. Thirty seconds later, at approximately one minute and 35 seconds, smoke layers descend rapidly and the room temperature rises to 190 degrees Fahrenheit. At this point, occupants can still exit the home safely. At the two minute and three second mark, the temperature in the originating room approaches approximately 400 degrees Fahrenheit. A few seconds later, smoke begins to pour quickly into other areas of the house. This is a major problem since smoke accumulation cuts off access to oxygen, increasing the chances of asphyxiation or suffocation. In addition, smoke can fill the lungs with carbon monoxide and other toxic gases that develop as a byproduct of different synthetics, plastics, and paints that are now burning in the house.

At approximately three minutes and three seconds, the temperature in the room of origin is more than 500 degrees Fahrenheit. At three minutes and 20 seconds, a safe escape is categorized as difficult. Just short of four minutes, flashover occurs whereby the energy in the room of origin suddenly ignites everything resulting in temperatures of 1400 degrees Fahrenheit. Thirty seconds after flashover, flames are visible from the exterior of the house, which generates the first visible evidence of a residential fire from the outside. At this point, rescue is very unlikely.

Most experts say that you should plan on having no more than about two minutes to escape a residence once a smoke detector has been triggered. Once the smoke detector triggers, at approximately the one minute mark, the premises must be vacated before the three minutes and three seconds mark, especially since escaping exponentially decreases at about three and a half minutes. This is a very important timeline to keep in mind, particularly when reviewing research that assesses arousal success in response to various smoke detector alarms.

Smoke Detectors

A smoke detector is designed to warn occupants of the presence of fire by detecting the presence of flame or smoke. In turn, it generates a distinct audible alarm so that occupants immediately recognize the need to evacuate the premises. Smoke detectors do not necessarily provide an early warning of fire.

Recall that experts agree that home dwellers have about two minutes to escape a residential fire once the smoke detector is triggered. Since most residential fires occur during the night when most people are asleep, it may be safer to assume that there is less than two minutes to exit since someone won't really know how quickly it actually took them to wake up to the smoke detector's alarm.

A smoke detector has two basic parts: the smoke detector component and an audible alarm component. The smoke detector component is responsible for reliably detecting the presence of smoke or flame. Once smoke or flame is sensed by the smoke detector, it must trigger the audible alarm. This alarm must also be audible and loud enough to be heard and recognized as an urgent alarm.

How effective a smoke detector will be in achieving these two functions is influenced by the specific smoke detector technology integrated in a product as well as the characteristics of the sound emitted by the alarm.

Smoke Detector Technology

There are two general types of residential smoke detector technology: photoelectric technology and ionization censor technology, with some smoke detectors incorporating a combination of both technologies. Photoelectric technology relies on the use of a light beam to detect the presence of smoke or flame. This type of technology is similar to the technology used by businesses that alerts employees when someone enters or exits a store. In that scenario, an infrared light beam illuminates when a customer crosses the threshold of the entrance. A sensor detects the interruption in the light beam and an audible signal such as a ding or doorbell sound is generated.

Smoke detectors equipped with photoelectric technology operate in the same manner. One of the components inside the smoke detector generates a beam of light. Another component, called the photo detector, constantly monitors the light beam. When smoke enters the chamber of the smoke detector, the light beam will become partially blocked and the photo detector senses that blockage and triggers the audible alarm. In contrast, ionization sensor technology incorporates the use of a small amount of radioactive material that is enclosed in a metal chamber. The chamber contains two metal plates that are positioned a small distance apart from one another. One plate carries a positive charge while the other plate carries a negative charge. The radioactive material in the chamber emits particles between the two plates, which causes a steady flow of ions resulting in the presence of a small but steady current. When smoke or even hot air enters the chamber, the property of the ionic flow will change and interrupt the current flow which triggers an audible alarm.

According to the NFPA, photoelectric smoke detectors comprise approximately 10 percent of the smoke detectors sold in the United States whereas the ionization smoke detectors comprise 80 to 90 percent of smoke detectors sold in the United States. A photoelectric smoke detector will generally have a greater sensitivity, meaning a faster response time to a slow smoldering fire that produces a lot of smoke but not necessarily a lot of flame. In contrast, smoke detectors equipped with ionization sensor technology will respond more quickly to fast burning fires that generate a lot of flame and not necessarily a lot of smoke. Fire fatalities will occur as a result of both smoldering types of fires as well as flaming fires. It is impossible to predict the type of fire that may occur in a home. According to the NFPA, either smoke detector type will provide sufficient time for escape for most people in most fire situations. However, for the best protection the NFPA recommends incorporating both technologies in the home, either by installing multiple smoke detectors whereby some are photoelectric and some use ionization sensors. Or, using smoke detectors that incorporate both type of technologies in the same unit.

Smoke Detectors - Alarm Characteristics

With a better understanding of smoke detector technology, the second area to explore involves alarm characteristics of smoke detectors. It is important to note that NFPA is a nonprofit agency whose mission is to reduce the burden of fire on quality of life through research, training, and education. The NFPA develops and issues consensus, codes, and standards intended to minimize the possibility and effects of fire. The NFPA does not test, label, or approve smoke detectors, rather it issues codes and standards related to fire safety. One such well known standard is the National Fire Alarm Code (NFCA), also referred to as NFPA 72. This standard has been adopted by federal, state, and local municipalities across the United States and includes performance requirements of residential smoke detectors.

According to the NFPA 72, smoke detectors must generate a distinct audible alarm for a minimum amount of time at a defined intensity level. That specific alarm may not be used for any purpose other than alerting individuals of a fire and the subsequent need to evacuate.

As of July 1996, all smoke detectors were required to generate a very specific audible signal known as the audible emergency evacuation signal as described in the ANSI standard S3.42 1990. More commonly referred to as the temporal 3 or the T3 pattern, smoke detectors must generate an audible signal that consists of a three pulse tonal sequence of .5 second on and .5 second off with an additional one second of silence following the third and final .5 second off portion of the cycle. One full T3 pattern takes four seconds to complete.

As outlined by the NFPA, once activated, smoke detectors must be able to emit this T3 signal pattern continuously for no less than 180 seconds which equates to a duration of three minutes. Furthermore, the intensity of such an alarm must be reliably heard. In the United States, this currently translates to the following: Residential smoke detectors must emit an alarm that measures 75 dBA at pillow level. Since fire code requires the installation of a smoke detector immediately outside of a bedroom, the assumption is that installation of a smoke detector in that location will result in an alarm intensity of 75 dBA at pillow level.

Unless someone physically checks each house and actually measures the intensity of a smoke detector at the pillow level with a sound level meter, whether or not the standard is actually being achieved is difficult to quantify. The NFPA has outlined additional performance requirements whereby the measured intensity of the signal emitted by the smoke detector must minimally measure 85 dBA at a distance of 10 feet (approximately 3 meters). Residential smoke detectors are tested by independent objective agencies that determine whether or not a product is compliant with NFPA's current performance standards.

Notice that the current standard does not mention or stipulate any frequency requirements for the alarm. As long as these three requirements are met, a smoke detector may generate an alarm of any frequency per the discretion of the manufacturer.

To understand why this is the case, consider the fact that when residential smoke detectors were first introduced and widely distributed during the 1970s the buzz about these devices focused on their ability to detect the presence of early stages of smoke or fire. Very little attention was put on the effectiveness of the smoke detector alarm to arouse sleeping individuals. Research in the past decade, however, has specifically addressed this issue. Nonetheless, one of the most alarming facts about smoke detectors is that even today the standard signal emitted by the majority of residential smoke detectors is 3100 Hertz (Hz) or higher in frequency.

Taking into consideration hearing loss demographics and traditional audiometric configurations related to presbycusis, the ability for individuals with hearing loss to effectively hear a standard smoke detector signal of 3100 Hz or higher should become an immediate concern. Even if an individual is an active hearing instrument wearer, recall that most residential fire fatalities occur between 11 p.m. and 7 a.m., a time when individuals are asleep and, therefore, will not be wearing hearing instruments. Furthermore, when a person is asleep, awareness thresholds to auditory signals are elevated. It has been estimated that an auditory signal may need to be as much as 40 dB louder for a person to detect that signal while asleep compared to detecting the same signal when that same individual is awake.

Smoke Alarm Effectiveness - Research Review

Over the past decade there has been a relatively significant amount of research published on the effectiveness of smoke detectors in alerting sleeping individuals. Most of the data show that signal type, i.e. - auditory versus tactile versus visual - as well as the specific frequency of the audible alarm greatly influence the probability that adults, particularly those with hearing loss, will wake up in direct response to that generated alarm.

Based on a number of studies published over the past decade, adults with normal hearing wake up fairly quickly to an activated smoke detector, even when the intensity of the signal is softer at pillow level than the current fire or code requirements of 75 dBA.

In one study, Bruck (1998) documented that 100 percent of normal hearing adults woke to a standard smoke alarm signal of 3100 Hz with a measured intensity of 60 dBA at pillow level within 32 seconds of alarm initiation. Since the intended purpose of a smoke detector is to effectively warn sleeping occupants of a fire so that they can evacuate the premises in a timely fashion, the mission is accomplished for adults with normal hearing.

Bruck also looked at how well children with normal hearing performed in the same situation. She found that only 17 percent of children with normal hearing awoke to that same audible 3100 Hz alarm during that same time period of 32 seconds.

Since smoke detectors must continually emit an audible signal for at least 180 seconds, Bruck continued to document how many additional subjects woke up during the course of the entire three minute period that the audible alarm was actually activated.

When examining the cumulative percentage of adult subjects that woke up to the smoke detector during that entire three minute period, since all adult subjects woke up by the 32 second mark, the number of awake adult subjects at the subsequent time intervals remained at 100 percent. With regard to children, 17 percent of children with normal hearing successfully awoke to the standard audible 3100 Hz alarm within the first 32 seconds of alarm activation.
During the entire three minute period, only an additional 6 percent of children were aroused by the alarm. This resulted in a 23 percent wake up rate for children. In other words, 77 percent of children with normal hearing did not respond to the smoke alarm after a full three minutes despite having normal hearing.

Which Signal Is Best?

Turning back to the adult population, the question arises as to what is the best signal for waking adults who have hearing loss? Du Bois, Klassen & Roby (2005) investigated the waking effectiveness of different types of auditory, vibrotactile, and visual alarms in three specific groups of subjects with different degrees of hearing loss. Group one consisted of 32 adults with normal hearing; group two consisted of 45 individuals with hearing impairment with audiometric thresholds that ranged from 20 to 90 dB HL; and the last group was comprised of 32 deaf individuals whose audiometric thresholds were greater than 90 dB HL at 500 Hz and beyond. Each group was exposed to five different signals over the course of this study. These signals included: the standard 3100 Hz alarm generated by most commercially available smoke detectors; a low frequency 450 Hz tone alarm; two different bed shakers generating a vibrotactile alarm, one emitting a continuous signal and the other emitted a pulsating signal; and a visual alarm utilizing a strobe light generating a 110 candela light, which is the minimum light intensity outlined in the NFAC.

92 percent of adults with normal hearing, 57 percent of the hard of hearing group and 0 percent of the deaf subjects were aroused to the 3100 Hz signal alarm. The fact that none of the deaf subjects woke up to the 3100 Hz alarm is not surprising. What is quite compelling, however, is the fact that almost half of the adults with mild to severe hearing loss did not wake up to this signal.

In the case of the low frequency 450 Hz signal, 11 percent of deaf subjects woke up to this alarm whereas all the normal hearing subjects along with 92 percent of the hard of hearing subjects woke to this signal. In this situation, the 450 Hz signal enabled the hard of hearing group to perform as well as the normal hearing group did to the traditional 3100 Hz signal. With regard to the bed shaker type alarms, the presence of a continuous vibrotactile alarm resulted in both the normal hearing and the deaf groups yielding essentially identical wake up rates of 93 percent and the hard of hearing group performing slightly poorer with an 82 percent wake up rate.

In contrast, the investigators reported that every single subject across all three groups woke up to a bed shaker alarm that generated a pulsating or intermittent signal. In terms of the effectiveness of the visual alarm, nearly 60 percent of deaf subjects responded to this type of alarm, although only approximately one third of normal hearing and one-third of the hard of hearing subjects were aroused by this type of stimulus.

The findings from this study initially suggest that if you have normal hearing, any alarm is sufficient as long as the alarm is not a strobe light-only alarm. For those who are hard of hearing, a low frequency signal and/or a pulsating vibrotactile alarm seems to be most effective. Finally, for deaf subjects, a vibrotactile signal, particularly a pulsating signal, represents the most effective alarm signal.

Sleep Cycle

Interestingly, some of these findings are not consistent with previously published research. Earlier research strongly suggested that deaf individuals are very responsive to visual alarms, reporting a 92 to 100 percent wake up rate in response to strobe lights. Other earlier research showed that adults with hearing loss showed a 70 percent wake up rate to vibrotactile stimuli rather than the 80 to 100 percent reported by Du Bois and colleagues.

A lot of the discrepancy between the Du Bois et al. study and earlier research stems from sleep cycles of subjects as well as alarm presentation factors utilized across different studies. With regard to the sleep cycle, there are two main categories of sleep: slow wave sleep and rapid eye movement sleep (REM). Slow wave sleep consists of four distinct stages whereas REM sleep is only one stage. Each stage of slow wave sleep is associated with a specific type of brain activity that corresponds to a unique pattern of brain waves. Therefore, when a subject is hooked up to an EEG machine, it is fairly simple for a technician to recognize what stage of sleep a subject is in based on the type of brain wave that is generated. In general, the higher the sleep stage number, the deeper the level of sleep. With regard to slow sway sleep, stage one represents the lightest stage of sleep whereas stage four represents the deepest level of sleep. It is much easier to wake someone up during stage one and stage two sleep as compared to stages three and four. Stage four is the most difficult to interrupt.

Stage five or REM sleep is technically considered the deepest stage of sleep. However, sleep research has consistently shown that individuals are more likely to spontaneously awake from REM sleep than any other stage of sleep. With that in mind, the most difficult sleep stages to wake up individuals from are the slow wave sleep stages of stages three and four. These are collectively known as delta sleep stages. The reason that the Du Bois and colleagues (2005) findings showed poorer performance by subjects to the strobe light alarm compared to earlier studies, is that earlier studies did not control for the subjects' sleep stages at all whereas the Du Bois study did. However, one of the limitations of the Du Bois at al. (2005) study is that it only controlled for this factor to a small degree. For the Du Bois study, as long as a subject was not in stage one sleep, data could be collected during any of the other sleep stages. This is potentially a very large variable, which is why additional research must be reviewed in this area to determine the best alerting signal for the hearing impaired population. Also, the alarms in the Du Bois et al. (2005) study were triggered at out of the box intensities and they were presented for a full three minutes at maximum levels. As long as the subject woke up to the alarm at any time during the three minutes, it was considered a successful awakening. It was unclear as to exactly when each subject woke up to the alarm. If most of the subjects woke up to the alarms at the end of the three minute alarm cycle, and you add an additional 30 to 60 seconds of time that it would have taken for the smoke alarm to trigger after the start of the fire, it is possible that subjects in this study actually woke up at a point in time when flashover already occurred. Recall, that experts indicate that successful escape from a fire at this point in time is considered highly unlikely.

Which Signal Really is Best?

Bruck and Thomas (2009) investigated the effectiveness of a variety of signals on waking adults with hearing loss, specifically, hard of hearing adults with binaural hearing thresholds consistent with mild to moderately-severe hearing loss. The auditory alarm signals used in this study were chosen based on previously published as well as pre experimental research performed by these investigators that showed that both the 400 Hz and the 520 Hz square wave signals were associated with the lowest arousal thresholds in adults with hearing loss. The 3100 Hz pure tone signal was also used in this study because it represents the standard alarm frequency emitted by most commercially available smoke detectors.

A square wave is different from a pure tone signal in that a square wave consists of a specific fundamental frequency and an infinite number of subsequent peaks at odd-numbered harmonics. All three auditory signals used in the study were presented in a T3 format to simulate the evacuation signal required of smoke alarms. How well subjects woke up to an intermittent vibrotactile signal that mimicked the T3 pattern was also assessed using both a bed shaker as well as a pillow shaker. Both devices were designed to vibrate; however, as the name suggests, the pillow shaker actually resides under the top most pillow at the level of the subject's head whereas a bed shaker is placed under the mattress at the level of the subject's navel. Finally the efficacy of a strobe light signal was also assessed and it, too, was designed to flash in a T3 pattern.

The waking effectiveness of subjects in this study was assessed while subjects exhibited delta sleep. In other words, the alarms were not presented unless the subject was exhibiting stage three or stage four slow wave sleep; the two stages where it would be most difficult for someone to wake up to a signal. Furthermore, alarm presentations were delivered in a discreet 30-second episode to determine the lowest threshold necessary to arouse subjects. For example, for the auditory alarms, a signal was initially delivered at 55 dBA at pillow level for 30 seconds and for 30 seconds only. If the subject did not wake up to the alarm after a specific period of pause time, the alarm intensity was then increased by 10 dBA to 65 dBA and delivered again for 30 seconds at pillow level. This process was repeated (pause, increase presentation level by 10 dBA, present signal for 30 seconds) until the subject was aroused, or until the maximum intensity of the signal was reached. At maximum intensity levels, the alarm was presented for a full three minutes rather than just the 30 second interval to ultimately determine whether or not the subject could be aroused by that signal at the maximum intensity.

This study found that the majority of subjects, 92 percent, woke up to a 520 Hz square wave signal delivered at benchmark, which represents the intensity level of the alarm consistent with current national fire code. That is, a cumulative 92 percent of hard of hearing adults woke up to a 520 Hz square wave signal measuring 75 dBA at pillow level within the first 30 seconds of its presentation.

A very close 87 percent of the same subjects woke up to the 400 Hz square wave at benchmark levels whereas only 56 percent woke up to the traditional 3100 Hz alarm. Thus, nearly half of the subjects with mild to moderately-severe hearing loss slept through a standard smoke detector alarm despite the fact that the alarm was being presented at an intensity level that met minimum performance standards of smoke detectors. At maximum alarm intensities, 100 percent woke up to the 520 Hz square wave signal, 95 percent woke to the 400 Hz square wave signal, and only 84 percent woke up to the traditional 3100 Hz signal. In other words, 16 percent of hard of hearing subjects slept through a standard 3,100 Hz smoke detector signal for the full three minutes it was presented at maximum intensity levels. In terms of vibrotactile alarm performance, 80 to 84 percent of hard of hearing subjects woke up to a bed shaker and pillow shaker respectively at benchmark, which in this context represents out of the box intensity levels of the product. When both vibrotactile alarms were manipulated to generate maximum intensity levels, a cumulative 87 percent woke up in response to the bed shaker whereas 97 percent woke up in response to the pillow shaker. In terms of visual alarm performance, only 27 percent of hard of hearing subjects woke up to the strobe light at benchmark. In this context, benchmark represented the lowest possible intensity of the strobe light signal,177 candela, which is a greater signal intensity than current fire code requirements of 110. Even at maximum levels, when researchers simultaneously exposed sleeping subjects to three strobe lights, the wake up rate only approached 70 percent.

When looking at benchmark level performances, the 520 Hz square wave represents the most effective signal in waking adults with mild to moderately- severe hearing loss. Although not as effective as the 520 Hz square wave, the 400 Hz square wave as well as the bed and pillow shakers performed much better than the strobe light or the traditional 3100 Hz alarm. Clearly, the least effective alarm was the strobe light, which was effective in only waking 27 percent of sleeping adults at benchmark levels. At maximum intensity levels, the trend was much the same, with the 520 Hz square wave as the most effective signal in waking sleeping adults with mild to moderately-severe hearing loss.

Smoke Alarm Options for People With Hearing Loss

Based on the limitations related to current smoke detector technology and what research shows regarding the most effective alerting signal for this hard of hearing patient population, recommendations to patients with hearing loss can be made accordingly.

Several manufacturers offer or market smoke detectors specifically to the hard of hearing or deaf populations. It is important to appreciate the product performance for purposes of making the most informed decision about what may be a potentially appropriate option for a specific patient.

Gentex Corporation develops electronic devices that utilize photoelectric sensors for the automotive industry. Gentex also offers photoelectric smoke detectors and they offer a variety, including a hardwired version that mounts on the ceiling. Because it is hardwired, an electrician must install it. Gentex also offers a simpler plug-in version that mounts on the wall and plugs into a standard wall outlet via a 9 foot cord. Both models come equipped with audible and visual alarms.

The Gentex smoke detector works like any standard photoelectric smoke detector. When it detects the presence of smoke, the smoke detector is triggered to emit a 90 dBA signal, the standard 3100 Hz alarm, and it simultaneously activates a high intensity strobe light.

In the absence of research studies assessing the effectiveness of a combined auditory plus visual alarm presentation, there are apparent limitations of this specific product based on the most recent research. These include first and foremost, the use of a 3100 Hz auditory alarm. A 3100 Hz auditory alarm is not effective in waking people with mild to moderately-severe hearing loss. A strobe light used independently is not as effective of a signal for waking adults with varying degrees of hearing loss. The overall percentage of wake up rates for a strobe light alarm does not qualify this type of an alarm as a functional equivalent of alarms using other types of signals. Furthermore, the strobe light in the Gentex product is located at the base of the smoke detector. Therefore, if this device is installed in the hallway outside the bedroom, which is where most smoke detectors are installed, the chances of a sleeping adult reacting to the strobe light are going to low because of the distance between that adult and the device.

Silent Call is a company that provides personal communication products for people who are deaf, deaf and blind, or hard of hearing. They offer the Shake Up Smoke Detector in two different models: a bed shaker version and a strobe light version. It is comprised of several components, including a component that looks like a traditional smoke detector with a built in transmitter as well as a second piece, the bedside console, which includes a built-in receiver.

The bed shaker model includes a bed shaker unit whereas the strobe light version has a built in strobe light in the bedside console. The smoke detector portion of either version is mounted just like a traditional smoke detector, whereas the bedside console is placed at the bedside on a nightstand. The Shake Up Smoke Detector works as follows: In the presence of a fire, the smoke detector component detects the smoke and generates an audible 3100 Hz signal of 85 dBA, just like traditional smoke detectors. Simultaneously, the built in transmitter in the smoke detector wirelessly sends a signal to the receiver in the bedside console. In the case of the bed shaker version, the bedside receiver will receive the signal and cause the bed shaker to shake. In the strobe light version, a signal is sent to the bedside console which then causes the strobe light to emit a light.

This product is different from the Gentex in that the different models are available that pair the audible alarm with either a vibrotactile alarm or a visual alarm. Additionally, the secondary alarm signal, whether it is vibrotactile or visual, is delivered at bedside, which is closer to the person that needs to be alerted by the signal. Keep in mind, however, that this device generates a 3100 Hz alarm, a signal that has been found ineffective for those with varying degrees of hearing loss. While one version of this product generates a strobe light at bedside, current research suggests that a visual alarm is even less effective than a 3100 Hz alarm for those who are hard of hearing.

The reviewed research did not specifically assess whether a combined auditory plus visual alarm generates some sort of complex interaction that would yield significant improvements in waking adults when pairing the two together. Given the features of this product and considering what current research shows, this is probably not the best product choice for individuals with hearing loss. Of the two models, the bedside shaker version probably holds the most promise as a potential option for individuals with hearing loss. Anecdotal evidence seems to suggest that the combination of an effective audible signal, with specifically the 520 Hz square wave as best audible signal, with a vibrotactile alarm is probably most effective. Note that research tends to specifically indicate that when pairing a vibrotactile stimulus to a low frequency 520 Hz square wave signal, maximum effectiveness will be achieved.

Since research currently supports a low frequency 520 Hz square wave alarm as the most effective signal for alerting sleeping adults with mild to moderately-severe hearing loss, there are two products worth noting. The LoudenLow model 520 is a smoke detector that emits a 520 Hz square wave signal. Unlike most commercially available smoke detectors, which tend to be round and relatively small, this particular product is square and substantially larger. These larger dimensions are necessary in order to accommodate components needed to generate such a low frequency signal. Most smoke detectors typically generate a high frequency alarm, possibly because a high frequency alarm can utilize smaller components, which provides the ability to market a product that is smaller in size and, therefore, associated with more inherent market appeal. Just like other smoke detectors, the LoudenLow mounts on the wall. In the presence of a fire, this smoke detector will detect the smoke and generate a low frequency 520 Hz square wave signal at 85 dBA. This product is unique because it is capable of generating the specific audible alarm that has been identified as the most effective in waking hard of hearing individuals. One disadvantage is that the smoke detector may be installed at a relatively far distance from the source. Whether or not the intensity of the signal will be sufficient in certain circumstances, where perhaps installation of the device is outside of the bedroom or in a very large bedroom remains unknown.

Finally, there is the Lifetone HL Bedside Fire Alarm Clock. It is essentially a bedside alarm clock. It is unique in that it is designed to work with existing smoke detectors already installed in the home. This product contains a built in patented sensor that is specifically tuned to recognize and react to a T3 acoustic pattern signal. When a standard smoke detector is activated in the home, it will detect the smoke and generate the standard 3100 Hz alarm. The Lifetone device, which is at bedside, will detect the alarm of the smoke detector and in turn generate its own 520 Hz square wave signal at 90 dBA at bedside. Even though a homeowner's regular smoke detector may emit a 3100 Hz tone, because that has to be emitted in a T3 pattern signal, the Lifetone will detect the activated smoke alarm and in response will then generate a 520 Hz square wave. The Lifetone product also comes equipped with a bed shaker that will also be triggered once the Lifetone detects the activated smoke detector. One of the benefits of the Lifetone HL Bedside Fire Alarm device is that it is designed to work with existing smoke detectors, so there is no need to replace current smoke detectors. Secondly, it resides at bedside which is in much closer proximity to the patient's ears than a signal generated outside of the bedroom. Finally, it also generates a tactile signal so the individual is getting the best of both worlds, a 520 Hz square wave as well as a vibrotactile signal.

Summary

In conclusion, individuals with mild to moderately-severe hearing loss have increased risk in terms of failing to hear traditionally high frequency alarms generated by most commercially available smoke detectors. Regardless of whether or not a patient is a current hearing instrument wearer, audiologists should inform patients who have hearing loss that most residential smoke detectors emit a high frequency signal, typically 3100 Hz or higher, which places individuals with hearing loss at risk for not waking to an alarm while sleeping. Recent research has found that nearly half of those with mild to moderately-severe hearing loss will not wake up to a standard auditory alarm. The signal that was found to be most effective in waking sleeping adults with hearing loss is the 520 Hz square wave.

References

Bruck, D. (1998). Arousal from sleep with a smoke detector in children and adults (Technical report FCRC-TR 98-04). Sydney, Australia: Fire Reform Center.

Bruck, D. & Thomas, R. (2009). Smoke alarms for sleeping adults who are hard-of-hearing: comparison of auditory, visual, and tactile signals. Ear & Hearing, 30(1),73-80.

Du Bois, J., Ashley, E., Klassen, M., & Roby, R. (2005, Nov). Waking effectiveness of audible, visual and vibratory emergency alarms on people of all hearing abilities. Proceedings of the Accessible Emergency Notification and Communication: State of the Science Conference, Gallaudet University, Washington D.C. Retrieved February 15, 2010 from https://tap.gallaudet.edu/emergency/nov05conference/Papers/Du Bois.asp