Alcohol and fire: a bad mix

Most people know that smoking in bed is a fire hazard. But not many know that having a few beers or glasses of wine before bed can increase their risk of becoming a fire fatality - even with a smoke alarm in the house.

"Danger, danger there is fire! Get up now! You must get up and investigate, there is fire! Get up now! Danger, danger there is fire! Get up now! You must get up and investigate, there is fire! Get up now!"

When I first hear a woman reiterating this urgent message, the hairs on the back of my neck rise. What I find even more disturbing is that a young man is soundly sleeping nearby, despite the fact that her cry is coming from speakers just a metre away from his head.

The warning, which started at a whisper, or at 35 decibels, is getting uncomfortably loud for research assistant Michelle Barnett and myself. We are in the dining room of a suburban home, separated from the young man's bedroom by a hallway and two sets of closed doors. He is still fast asleep.

It is the early hours of the morning and I am observing research being undertaken by Victoria University that is examining how alcohol affects one's ability to wake up to alarms. A world first, the research is being carried out by PhD candidate Michelle Ball, who is working under the supervision of Professor Dorothy Bruck, head of the University's School of Psychology.

It seems an eternity before Barnett - who is monitoring the young man's brain waves on a laptop for changes in his sleep patterns - tells me that he has woken up. It has taken the young volunteer, Thomas, a full nine minutes to wake. He 'reacts' by pressing a button on his bedside table three times. During the last three minutes of his sleep, the warning was at 95 decibels, a noise level that by occupational health and safety standards would require the wearing of ear protection. Another unsettling fact was that Thomas had only had a few standard drinks before he went to bed. His blood-alcohol concentration (BAC) was only .04.

Thomas soon goes back to sleep and Barnett will wait until he is at stage four of sleep - the deepest state - signified by long, 'ropey' green waves on the laptop screen. Before midnight, Barnett had spent 45 minutes wiring Thomas to the Electroencephalogram on her laptop by placing electrodes on his scalp and head.

"Thomas says he's a fairly deep sleeper, but he is not as deep as some of the others in the trial," says Barnett, as she watches the tight green waves on her computer become looser and looser, indicating that he is once again sinking into deep sleep. One of the volunteers in the trial slept through 95 decibels.

Barnett, who is studying a Master of Psychology in Clinical Neuropsychology, will stay up in her dimly lit Keilor Downs home to repeat the exercise two more times during the night. Volunteers are tested on three separate occasions - first with no alcohol in their systems, then with a BAC of approximately .05, and then with a BAC of about .08. Ball and Professor Bruck have been conducting the sleep trials since the beginning of the year. They were also keen to find out whether deep sleeping young adults under the influence of alcohol would wake to three different alarm signals, including the shrill, high-pitched sound of the smoke alarms in most Australian homes. As well as the lower-pitched female actor's voice, they also tested a low frequency sound - not unlike the sound of a truck in reverse gear. This sound is now the standard smoke alarm signal across North America.

The researchers found that alcohol had an effect on the amount of time it took volunteers to wake up in response to all three alarm sounds. Even low to moderate levels of alcohol could seriously affect a person's ability to respond. "The biggest jump in the time it takes to respond to the alarms is between being sober and .05," Ball says. "The results so far also show that the more the young people had to drink the longer it took them to respond to the alarms."

But perhaps the most important finding was that the lower-pitched signals were significantly more successful than the high-pitched signals of Australian smoke detectors found in homes. Professor Bruck has had similar findings in her trials with children aged six to 10 - high frequency signals were also less effective in waking them.

"What we think is an urgent sound when we are awake is different when we are asleep," Ball says. "When we are asleep our brains monitor the environment for a range of sounds close to human pitch. That is why we suspect the T3 [low-pitched, beeping signal] and human voice signals have been more successful in our trials so far."

Ball and Professor Bruck have been working closely with Victoria University's Centre for Environmental Safety and Risk Engineering (CESARE), a world leader in developing a comprehensive risk model to determine the probability of risk in fire under a wide variety of residential circumstances. CESARE director, Professor Ian Thomas, says the centre has been involved in Professor Bruck and Ball's research because it recognises that the conventional wisdom of a few years ago may not be correct.

Professor Thomas was part of a Victoria University team,which included Professor Bruck and Ball, that presented six papers at the Third International Symposium on Human Behaviour in Fire in Belfast last September.

"It is well understood by the engineering and fire safety community that quite a few people who die in fires are alcohol affected," Thomas says. "The longer it takes for a person to respond to a fire or a smoke alarm, the more likely that the fire will grow and the person will be affected by the smoke and/or heat from the fire."

So what are the implications of Ball and Professor Bruck's preliminary findings? Does the Australian public need to be warned that the type of fire alarms they have at home could be designed to be more effective? According to Ball, more work needs to be done. "First our results need to be replicated," she says. "We in Australia need to look closely at the type of [smoke alarm] signal we are selling, pending the conclusion of our ongoing research. In the Australian standards for smoke alarms, no requirements are laid down regarding pitch, just volume. But our findings so far need to be investigated further."

And do we need to be warned about the effects of even one drink on our ability to wake up in response to a signal during the night? "It is of vital importance that the general public be provided with this information in the hope that awareness will lower the number of alcohol-implicated fire fatalities," Ball says.

Printed with permission from Victoria University's Connections magazine.

"She watches the tight green waves on her computer become looser and looser, indicating that he is once again sinking into deeper sleep."

"Lower-pitched signals were significantly more successful than the high-pitched signals of Australian smoke detectors found in homes."

"Does the Australian public need to be warned that the type of fire alarms they have at home could be designed to be more effective?"