Sleep and Fire: Who is at risk and can the risk be reduced?

While sleep is clearly a major risk factor for dying in a fire, in this review it is argued that in most fire fatalities during the sleeping period additional risk factors are also operating. This view is consistent with the data from empirical sleep studies and smoke alarms where most (but not all) adults tested will awaken to a signal reaching the pillow at 75 dBA. However, research suggests that significant ‘staying asleep’ risk factors include high levels of background noise, being a heavy sleeper, sleep deprivation, being a child, sleeping tablets, alcohol intoxication, hearing impairment and being over age 60 (for high frequency signals). Recent sleep studies that have compared the waking effectiveness of different alarm signals (using children, sober adults and alcohol intoxicated adults) suggest that the high frequency beeping signal (usually around 3000 Hz) currently used in smoke alarms is significantly less effective in waking ‘at risk’ groups than either a voice alarm (300-2500 Hz) or a low pitched beeping signal (500-2500 Hz). Compared to the high pitched signal, the latter two signals required a 13 dBA lower volume to awaken sober adults and, when presented at 89 dBA to 6-10 year olds, were almost twice as likely to cause awakenings. The findings suggest that alarm signals of a frequency below 2500 Hz will reduce the likelihood of individuals sleeping through a smoke alarm and further research should seek to more narrowly define the most effective frequency band and signal for arousal from sleep.