Sleepiness at the wheel

Abstract

Prevention of road crashes is a major priority in most countries. The present thesis focuses on driver sleepiness and road crashes. The general aim of the thesis was to explore the relation between driver sleepiness and driver impairment and the changes that precede a crash or a similar safety-critical event, but also what constitutes good reliable behavioural / physiological measures of sleepiness. A second question concerned particular groups and situations is post-night shift driving characterized by increased sleepiness, is more complex driving also affected by sleepiness, and are younger drivers affected by sleepiness behind the wheel more than older drivers? A third question concerned countermeasures. What are the preferred self-administered countermeasures and what are the effects of the structural countermeasure like rumble strips?

The results showed that driving home from the night shift was associated with an increased number of incidents (2 wheels outside the lane marking, from 2.4 to 7.6 times). There were a decreased time to first crash, increased lateral deviation (from 18 to 43cm), increased eye closure duration (0.102 to 0.143sec), and increased subjective sleepiness. Moreover, a night of prior sleep loss increased levels of established indicators of sleepiness at the wheel even if the driving situation required frequent interactions with other cars on the road. However, blink duration (mean and variability) was shorter during overtaking, compared to other situations, even during the drive home after a night shift. Young drivers showed a higher risk than older drivers. Sleepiness increased with time on task, with higher levels for young drivers than for older ones, and the results indicate that younger age is associated with greater susceptibility to sleepiness at the wheel. In another study it was shown that a sleepy driver leaving the lane and hitting the rumble strip will be awakened and show an improved driving behaviour. However, the sleepiness signs (KDS, lateral deviation, eye blink duration) returned 5 minutes after the rumble strip hit. It was concluded that various aspects of sleepiness are increased before a rumble strip is hit and that the effect is very short-lived.

In a methodological study a combined scoring of Electrooculogram and Electroencephalogram, the so called Karolinska Drowsiness (KDS) scoring, is a promising method for estimating physiological sleepiness under conditions of driving. At KDS level 30 % (meaning sleepiness signs 30% of the time within a given time frame) the risk of lane departure is 2.6 times higher; at KDS 40% the risk of lane departure is more than 6 times higher. The relation between KDS and variability of blink duration shows that at KDS 30% the blink duration has changed from 0.16 seconds (level 0) to 0.18 seconds.

The thesis has also shown the importance of taking into account driver group when working with countermeasures against sleepiness related crashes. The most common countermeasures among drivers were to stop to take a walk, turn on the radio/stereo, open a window, drink coffee and to ask passengers to engage in conversation. None of them has so far been proven to be effective. Counteracting sleepiness with a nap (a presumably efficient method) was practised by those with experience of sleep related crashes or of driving during severe sleepiness, as well as by professional drivers, males and drivers aged 46-64 years. The most endorsed means of information to the driver about sleepiness was in-car monitoring of driving performance. This preference was related to the experience of sleepy driving, not being a professional driver and male gender.

In conclusion, the studies show that sleepiness is a major determinant of impaired driving performance, and many drivers need to be educated of alertness enhancing strategies that can prevent the negative consequences of driving when wakefulness is reduced.