Comparison of a Personalized Parent Voice Smoke Alarm with a Conventional Residential Smoke Alarm: Can Children be Effectively Awakened from Slow Wave.

Comparison of a Personalized Parent Voice Smoke Alarm with a Conventional Residential

Smoke Alarm: Can Children be Effectively Awakened from Slow Wave Sleep?

Gary A. Smith, MD, DrPHMark Splaingard, MD

Huiyun Xiang, MD, PhD, MPHJohn Hayes, PhD

Center for Injury Research and PolicyThe Research Institute at Nationwide Children’s Hospital

Background

• Approximately 800 children <15 years old die each year in the U.S. due to a residential fire

• Approximately half of residential fires occur at night with victims asleep

• To be effective, smoke alarms must be capable of consistently awakening individuals and prompting rapid escape

• However, conventional residential smoke alarms fail to awaken the majority of children

Sleep Stages Overview

• The sleep cycle begins with 3 stages of non-REM sleep with progressive slowing of the brain and deeper sleep

• Stage 3 is referred to as slow wave sleep (SWS)

• Sleep stage can be identified by EEG/PSG

Stage 2 Sleep

Slow Wave Sleep

Several factors associated with sleep place children at greater risk for residential fire-related injury and death

• Children sleep more than adults

• Auditory arousal thresholds are much higher in children than adults for each sleep stage

• Children have disproportionately more SWS than adults, which has a higher arousal threshold than other sleep stages

• Most SWS occurs during early sleep cycles, and house fires are also more common during the early hours of the night

The Problem

• Children are more likely than adults to be in a stage of sleep that is refractory to arousal at the time of a nocturnal residential fire

• In addition, once awake, children must be able to perform escape behaviors that require decision-making and action

– however, children are particularly susceptible to the effects of “sleep inertia”

Study by Busby and Pivik, 1985

• 50% during Rapid Eye Movement (REM) sleep

• 34% during stage 2 sleep

• 4.5% during SWS

Frequency of awakenings among boys 8 to 12 years of age with tone sound intensities up to 123dB

Study by Bruck, 1999

• Evaluated arousals from sleep to smoke alarms among children

– only other investigator to do this

• 20 children 6-17 years old compared with their parents age 30-59 years

• 60dB alarms for 3 minutes

Study by Bruck, 1999 (Continued)

• All adults awakened

• 31% of children awakened at least once

• Only 15% of children awakened consistently to the alarm

• However, EEG/PSG monitoring was not performed, and therefore, effect of sleep stage was unknown

• Performance of escape not evaluated

Response to One’s Own First Name

• “Cocktail party” phenomenon (Moray, 1959; Howarth and Ellis, 1961)

• Numerous studies demonstrate that the sound of one’s own name

– is an intrinsically significant stimulus

– elicits a differential human response as early as 4-5 months of age

– response persists during sleep, even when reactivity to other stimuli has disappeared

Study Objective

To compare a personalized parent voice alarm with a conventional residential smoke alarm, both presented at 100dB, with respect to their ability to awaken children 6-12 years old from slow wave sleep and prompt their performance of a simulated self-rescue escape procedure

Study Subject Eligibility Criteria

• 6-12 years old

   No diagnosis or current medication that may affect sleep, arousal or ability to perform escape procedure

  No hearing impairment

  No acute illness at the time of study

  Child and child’s caretaker speak English

  Family able to be contacted by telephone

Methods• Hearing and health screening was done

• Children were trained how to perform a simulated self-rescue escape procedure when they heard a smoke alarm

• Each child’s mother recorded a voice alarm message “First Name! First Name! Wake up! Get out of bed! Leave the room!”

• Sleep stage was monitored by PSG

• Children were awakened twice during night

– during first and second cycles of slow wave sleep

• Children received either the parent voice alarm or conventional tone alarm in each cycle

– each type of alarm was used once in random order

• Children were manually awakened if they did not wake up within 5 minutes of the alarm onset

Results

• Twenty-four children were enrolled

• One-half of children received the parent voice alarm first, and one-half received the tone alarm first

– the order of alarm presentation was not statistically associated with awakening or escaping

Median time to awaken

Parent voice 20 seconds

Tone 3 minutes

Wilcoxon signed rank test, p<0.001

Median time to escape

Parent voice 38 seconds

Tone 5 minutes (max)

Wilcoxon signed rank test p<0.001

Study Limitations

• Tested at 100dB

• Used large speakers

• Unclear which factors were responsible for success, for example:

– alarm stimulus intensity

– alarm stimulus frequency (Hz)

– speaker size

– mother’s voice

– use of child’s first name

– other message content

Conclusion

The personalized parent voice smoke alarm significantly out-performed the conventional residential smoke alarm with respect to awakening children from slow wave sleep and prompting their performance of a simulated self-rescue escape procedure.