Using an extended theory of planned behaviour to investigate sleep impaired driving

Abstract

Background: A significant proportion of vehicle accidents in the UK are caused by drivers falling asleep at the wheel and these accidents often occur at high speeds and result in fatalities (Home & Reyner, 1995b). Research has shown that driving after 15 or more hours of wakefulness increases the risk of being involved in a sleep-related vehicle accident (Stutts et al., 2003). Further, young and elderly adults have been found to have more sleep-related accidents between midnight and 6am and between 3pm and 6pm, respectively (Summala & Mikkola, 1994). An extended theory of planned behaviour (TPB; Ajzen, 1988,1991) was used to understand the reasons why young, middle-aged and elderly adults drive while sleep impaired with a view to reducing the incidence of this behaviour. Specifically, the determinants of driving after 15 or more hours of wakefulness, between midnight and 6am and between 3pm and 6pm were explored. Methods: A preliminary study in which sleep duration and quality were measured in young, middle-aged and elderly adults established the validity of actigraphy to reliably differentiate periods of sleep and wake. After identifying the accessible beliefs underlying each specific behaviour and constructing questionnaires tailored to each age group and behaviour, the main study was conducted. Two hundred and ten young, middle-aged and elderly adults completed TPB questionnaires in relation to driving after being awake for 15 or more hours in the following week. Measures of behaviour (using both subjective and objective methods, i.. e., actigraphy) were obtained eight days later from the young adults. Additionally, the young and elderly adults completed TPB measures regarding driving between midnight and 6am and between 3pm and 6pm, respectively. The young adults provided behavioural data prospectively. The ability of anticipated regret, impulsive sensation seeking, past behaviour and gender to account for additional proportions of variance over and above the TPB was also explored. Results: The TPB explained significant proportions of variance in intention to drive after 15 or more hours of wakefulness in all three age groups. Injunctive norm was a consistent predictor in all groups. The model predicted a subjective measure of driving after 15 or more hours of wakefulness but failed to predict a more objective measure, obtained from the young adults. Both the young adults' intentions to drive between midnight and 6am and their self-reported behaviour were successfully predicted by the TPB, as were the intentions of elderly adults to drive between 3pm and 6pm. Several beliefs which discriminated those who did and who did not perform (or intend to perform) the behaviours were identified. Anticipated regret independently predicted the intentions of young adults to drive after 15 or more hours of wakefulness and between midnight and 6am, as well as the intentions of elderly adults to drive between 3pm and 6pm. The TPB failed to mediate the effect of gender on driving between midnight and 6am in the young adults. Conclusions: The TPD provided useful, but not sufficient, accounts of the determinants of intentions to drive while sleep impaired. The model was able to predict subjective measures of sleep impaired driving but failed to predict a measure based on more objective methods. This research can be used to inform interventions attempting to reduce sleep impaired driving. A number of possible behavioural change strategies are proposed based on the empirical, theory-based results obtained here.