Research report No. 17

Chapter 6. Crash involvement, speeding and fatigued driving (PDF 143 KB)

This chapter follows on from the previous chapter, which looked at links between drink-driving and other types of risky driving, to examine inter-connections between three other problematic driving outcomes - crash involvement, high-level speeding, and fatigued driving.

It is well established that young drivers are particularly prone to involvement in crashes. For example, rates of crash-related injuries or deaths are considerably higher among young drivers than older drivers (Cavallo & Triggs, 1996; Dunsire & Baldwin, 1999; Palamara & Stevenson, 2003; Transport Accident Commission, 2009). Recent road safety statistics show that while 18 to 25 year olds represented only 14% of licenced drivers in Victoria in 2007, they accounted for 28% of drivers killed in road traffic accidents during that year (TAC, 2009b). Additionally, transport accidents were the leading cause of death among Australians aged 15 to 25 years in 2004 to 2006 (Australian Bureau of Statistics, 2008).

Two risky driving practices place young drivers at particularly high risk of serious crashes - speeding and driving when fatigued (Dobbie, 2002; Palamara & Stevenson, 2003). Both practices influence an individual's ability to drive safely, by affecting their ability to react to unexpected road events and take evasive action if required. For instance, a driver who speeds has less time to respond to changes in their environment, experiences a reduction in car manoeuvrability, and requires a longer stopping distance than a driver who observes the speed limit (Aarts & Van Schagen, 2006; Liu, Chen, Subramanian, & Utter, 2005; Williams, Kyrychenko, & Retting, 2006). A fatigued driver may be less attentive, experience impairments in judgement, and take longer to react to dangers in their driving environment than a more alert driver, or in some cases, may fall asleep and crash (Dobbie, 2002; Stutts, Wilkins, Osberg, & Vaughn, 2003). For these reasons, speed- and fatigue-related crashes tend to be among the most severe types of crashes, more often resulting in injury or death (Aarts & Van Schagen, 2006; Dobbie, 2002).

Both risky driving practices appear to be quite common among young drivers. For instance, looking at the prevalence of these behaviours among young adults participating in the Australian Temperament Project, most had engaged in low-level speeding (i.e., up to 10 km/h over the limit) on at least one of their ten most recent driving trips (84% at 19-20, 83% at 23-24 years) and about half were involved in moderate-level speeding (i.e., 11-25 km/h over the limit: 49% at 19-20, 45% at 23-24 years). Fatigued driving was also very common, with about two-thirds of young people having recently driven when very tired (64% at 19-20 and 63% at 23-24) (see Smart & Vassallo, 2005, and Chapter 3 of this report for further details).While there has been considerable interest in whether risky driving co-occurs with other problematic behaviours such as substance use and antisocial behaviour (Beirness & Simpson, 1988; Shope & Bingham, 2002; Smart & Vassallo, 2005; Vassallo et al., 2008), less is known about the degree to which different road safety behaviours themselves are inter-related. Such information could assist those working in the road safety field to appropriately target their intervention efforts. For instance, knowing whether the same group of young people who engage in one form of risky driving (e.g., speeding) are also more likely to engage in other risky driving practices (e.g., driving when fatigued) may be useful in determining whether generalist or more targeted intervention approaches are needed to combat particular unsafe driving practices.

Several studies have shown that drivers who disregard speed limits are more likely to drink and drive, to have a history of speeding and driving-related offences and to speed at night (Harrison, Fitzgerald, Pronk, & Fildes, 1998; Palamara & Stevenson, 2003; Steptoe et al., 2004). Fatigued drivers are more likely to spend longer periods behind the wheel, and to more often drive late at night or during the early hours of the morning than other drivers (Connor et al., 2002; Stutts et al., 2003). Furthermore, those involved in fatigue-related crashes are more likely to have engaged in other risky driving practices such as drink-driving and excessive speeding (Dobbie, 2002; Engström et al., 2003; Pack et al., 1995). However, many of these studies have been conducted overseas and have focused on drivers of varying ages rather than young adults. Australian research examining the co-occurrence of different unsafe driving behaviours among young drivers remains limited.

This chapter will examine the degree of overlap between three different problematic driving outcomes at 23-24 years. Two steps will be taken to address this issue. First, we examine whether young people who have been involved in multiple crashes, engaged in high-level speeding, or driven when tired, differ from other young drivers on their driving history, on-the-road experiences, and involvement in other forms of risky driving. Second, we will compare the multiple crash, high-level speeding and fatigued driving groups to explore their similarity on these characteristics, as well as the degree to which the same individuals are involved in these three types of problem driving outcomes.

The following aspects will be investigated:

• driving history (as assessed by the length of time licences have been held);
• number of hours currently spent driving each week, and the times at which driving occurs;
• rates of apprehension for driving-related offences; and
• engagement in other risky driving behaviours.

6.1 Crash involvement

Young people were divided into three groups on the basis of the number of crashes they had been involved in as drivers since starting to drive.⁷⁰ These were:

• a zero crash group (n = 392, 42% of drivers);
• a single crash group (n = 284, 30%); and
• a multiple crash group (2 to 6 crashes; n = 259, 28%).

The groups did not significantly differ on gender composition, with a similar ratio of males to females in each group.

The zero, single and multiple crash groups were compared on the length of time they had held their licences; the amount of time they spent driving each week and the times at which this typically occurred; their rate of committing driving-related offences; and their engagement in a range of risky driving behaviours. The results of these comparisons follow.⁷¹

While most of the ATP cohort had held a licence for close to 6 years, individuals in the multiple crash group tended to have held their licences longer than those in the zero crash group.⁷² Thus, the multiple crash group had held their licences for an average of 6 years and 2 months, compared to an average of 5 years and 10 months for the single crash group, and 5 years and 9 months for the zero crash group.

Young people in the three crash groups did not significantly differ in the total amount of time they spent driving each week, or in the amount of time that they spent driving at particular times of the day or week.

On average, young people in the multiple crash group had been detected speeding on significantly more occasions (M = 2.3, SD = 2.5) than those in the single (M = 1.3, SD = 1.6) and zero crash groups (M = 1.1, SD = 2.5).⁷³ Multiple crash group members were also significantly more likely than individuals in the other two groups to have had police contact for a driving-related offence in the past year (22% vs 13% of the single and 9% of the zero crash groups).⁷⁴ Furthermore, young adults in the multiple crash group had significantly more often had their licence suspended or cancelled than those in the single or zero crash groups (14% of the multiple crash group, compared to 4% of the single and zero crash groups).⁷⁵

The multiple crash group had experienced more crashes resulting in property damage than the single crash group, but not more crashes resulting in injury or death.⁷⁶

As can be seen from Table 20, the multiple crash group was significantly more likely than the other two groups (but particularly the zero crash group) to have engaged in most forms of risky driving. Thus, a higher percentage of the multiple crash group had exceeded the speed limit, driven when fatigued, driven when affected by illegal drugs, or used a mobile phone when driving on at least one of their ten most recent driving trips. There were also trends for differences on driving when affected by alcohol and having nearly fallen asleep or fallen asleep when driving.

Differences were particularly evident for speeding and mobile phone use when driving. For instance, more than half of the multiple crash group (55%) had recently exceeded the speed limit by 11-25 km/h, compared with 43% of the single crash group and 38% of the zero crash group. Similarly, while talking on a handheld mobile phone when driving was relatively common overall, the proportion of young people in the multiple crash group who had engaged in this behaviour (66%) was noticeably higher than among the single and zero crash groups (about 50% in each group).

Table 20. Engagement in risky driving during their past ten trips, zero, single and multiple crash groups, at 23-24 years

	Crash group						χ2 results (df = 2)
	Zero		Single		Multiple
	n	%	n	%	n	%	χ2	p
Speeding
Up to 10 km/h over limit	303	78.7	233	82.3	234	90.7	16.1	< .001
11-25 km/h over limit	147	38.3	121	42.8	143	55.4	18.8	< .001
> 25 km/h over limit	48	12.5	24	8.5	52	20.2	16.4	< .001
Fatigued driving
Drove when very tired	215	55.7	184	65.0	177	68.3	12.0	.002
Nearly fell asleep/fell asleep	34	8.8	28	9.9	40	15.4	7.4	[.025]
Driving under the influence of alcohol or illegal drugs
Affected by alcohol	72	18.7	62	21.9	71	27.4	6.9	[.031]
Affected by marijuana	17	4.4	7	2.5	25	9.7	15.0	.001
Affected by ecstasy	2	0.5	6	2.1	10	3.9	9.2	.010
Affected by amphetamines	5	1.3	4	1.4	12	4.6	9.1	.010
Mobile phone use when driving
Talked on hands-free mobile	145	37.6	129	45.7	132	51.2	12.2	.002
Talked on handheld mobile	190	49.4	141	50.0	172	66.4	21.2	< .001
Used mobile phone function	228	59.2	183	64.7	195	75.3	17.8	< .001

6.2 Speeding

Young people who had exceeded the speed limit by more than 25 km/h on at least one of their ten most recent trips (high-level speeders; n = 131, 14% of drivers) were compared to those who had not (other drivers; n = 828, 86%). The majority of high-level speeders were male (70%), while females were more likely than males to be classified as other drivers (66% of this group was female).⁷⁷

The two speeding groups did not differ on the average length of time they had held their driver's licences.

Young people who engaged in high-level speeding spent significantly more hours behind the wheel than other drivers (on average, 13 hours per week, compared to 10 hours for other drivers).⁷⁸ Looking more closely at when driving occurred, high-level speeders spent significantly more hours driving at night on weekdays than other drivers (2.8 hours vs 2 hours) and there were trends for differences on rates of weekend driving.⁷⁹

As might be expected, high-level speeders had been detected speeding significantly more often than other drivers,⁸⁰ having been detected on average 3 times during their driving careers (M = 3.2, SD = 4.2), compared with an average of just over one instance of being detected among other young drivers (M = 1.3, SD = 1.7). Similarly, high-level speeders were significantly more likely to have had contact with police regarding a driving-related offence in the past 12 months (31% vs 11%).⁸¹ High-level speeders had also significantly more often had their licence cancelled or suspended at some time during their driving careers (19% vs 5%).

While similar proportions of high-level speeders and other drivers had been involved in a crash at some stage during their driving career, focusing only on those who had been involved in a crash, high-level speeders were found to have been involved in more crashes than other drivers (1.4 crashes compared to 0.9 crashes).⁸² Looking at the circumstances of these crashes, high-level speeders had been involved in more property damage crashes when driving alone than other drivers (high-level speeders: M = 1.7, SD = 1.1; other drivers: M = 1.3, SD = 0.8),⁸³ but did not differ in their rates of involvement in other crash types (property damage crashes when carrying passengers, injury crashes when driving alone or carrying passengers).

High-level speeders were compared to other young drivers on their engagement in risky driving. As can be seen from Table 21, high-level speeders were significantly more likely to have engaged in almost all other types of risky driving except talking on a hands-free mobile phone. Hence, a higher percentage of high-level speeders had exceeded the speed limit by low and moderate levels; not worn a seatbelt or helmet for all or part of a recent driving trip; had fallen asleep or nearly fallen asleep when driving; driven when affected by alcohol or marijuana; and/or talked on a handheld mobile phone or used a mobile phone function when driving. A similar trend was found for driving when very tired, but did not reach the adjusted significance level.

Table 21. Engagement in risky driving during their past ten trips, high-level speeders and other drivers, at 23-24 years

	High-level speeders		Other drivers		χ2 results (df = 1)
	n	%	n	%	χ2	p
Speedinga
Up to 10 km/h over limit	130	99.2	668	80.8	27.7	< .001
11-25 km/h over limit	129	99.2	299	36.1	181.1	< .001
Failure to wear seatbelt/helmet
For part of trip	33	25.2	95	11.5	18.3	< .001
For all of trip	15	11.5	35	4.2	12.1	< .001
Fatigued driving
Drove when very tired	94	71.8	505	61.0	5.59	[.018]
Nearly fell asleep/fell asleep	35	26.7	69	8.4	39.4	< .001
Driving under the influence of alcohol or illegal drugsb
Affected by alcohol	70	53.4	146	17.6	83.1	< .001
Affected by marijuana	18	13.7	31	3.7	23.3	< .001
Mobile phone use when driving
Talked on handheld mobile	102	77.9	420	50.8	33.3	< .001
Used mobile phone function	106	80.9	521	63.0	16.1	< .001

Note: ^a Groups were not compared on speeding by more than 25 km/h, as this was the criterion used to identify high-level speeders and other drivers. ^b Groups were not compared on driving when affected by ecstasy or amphetamines, as the number of individuals in the high-level speeding group who had engaged in these behaviours was too small for reliable statistical comparisons.

Rates of these behaviours differed markedly across the two groups. For instance, not surprisingly, virtually all of the high-level speeding group (99%) had engaged in moderate-level speeding (between 11 and 25 km/h) during at least one of their ten past trips compared with just over a third of other drivers (36%). The percentage of high-level speeders who had engaged in highly dangerous behaviours such as driving when affected by alcohol (53%) or falling asleep or nearly falling asleep when driving (27%) was also much higher than among other drivers (18% and 8% respectively).

6.3 Fatigued driving

To determine whether fatigued drivers significantly differed from other drivers, ATP study members were divided into two groups on the basis of how often they had driven when very tired during their past ten trips. Two groups were identified:

• those who had driven when very tired on two or more of their ten most recent driving trips (fatigued drivers; n = 364, 38% of drivers); and
• those who had not (other drivers; n = 599, 62%).

While the groups differed on gender composition, with more females than males being in the fatigued drivers group (66%), these differences did not reach the adjusted significance level.⁸⁴

The two groups did not significantly differ on the average length of time they had held their driver's licences.

As shown in Figure 32, fatigued drivers spent significantly more hours driving in total per week, and driving at different times of the weekend, with trends for driving more often during the week as well.⁸⁵

Figure 32. Number of hours spent driving, at weekdays and weekends and during daylight and night-time hours, fatigued drivers and other drivers, at 23-24 years

Fatigued drivers had not been detected speeding more often than other drivers,⁸⁶ but were significantly more likely to have had police contact for a driving-related offence in the past year (18% vs 12%).⁸⁷ However, they did not significantly differ on rates of licence cancellation or suspension.

Fatigued drivers were significantly more likely than other drivers to have been involved in a crash while driving (67% versus 56%),⁸⁸ and had also been involved in more crashes on average (fatigued drivers: M = 1.2 crashes, SD = 1.1; other drivers: M = 0.9 crashes, SD = 1.1).⁸⁹ Looking at the characteristics of these crashes, young people who had driven when very tired had been involved in more crashes resulting in property damage when they had passengers in the car (fatigued drivers: M = 0.9, SD = 0.7; other drivers: M = 0.7, SD = 0.6).⁹⁰ There were no significant differences on the circumstances and outcomes of other crash types (property damage when driving alone; injury or death when driving alone or when with passengers).

In terms of involvement in other forms of risky driving, Table 22 shows that fatigued drivers had significantly more often engaged in most types of risky driving. Not surprisingly, the most powerful differences between fatigued drivers and other drivers were on the percentage who had fallen asleep or nearly fallen asleep while driving (25% versus 3%). Additionally, large differences were also found on speeding and mobile phone use when driving. For instance, more than half of fatigued drivers (57%) had engaged in moderate-level speeding, compared to 37% of other drivers, while two-thirds (66%) of fatigued drivers had talked on a handheld mobile when driving, compared to 48% of other drivers. There were also trends for differences on failure to wear a seatbelt for the duration of a trip and driving when affected by marijuana or amphetamines.

Table 22. Engagement in risky driving during their past ten trips, fatigued drivers and other drivers, at 23-24 years

	Fatigued drivers		Other drivers		χ2 results (df = 1)
	n	%	n	%	χ2	p
Speeding
Up to 10 km/h over limit	327	90.1	474	79.3	19.1	< .001
11-25 km/h over limit	209	57.4	221	37.1	37.8	< .001
> 25 km/h over limit	66	18.2	65	10.9	10.1	.001
Failure to wear seatbelt/helmet
For part of trip	70	19.2	60	10.0	16.4	< .001
For all of trip	27	7.4	23	3.8	5.9	[.015]
Fatigued drivinga
Nearly fell asleep/fell asleep	89	24.5	15	2.5	112.8	< .001
Driving under the influence of alcohol or illegal drugs
Affected by alcohol	104	28.6	114	19.0	11.8	.001
Affected by marijuana	27	7.4	23	3.8	5.9	[.015]
Affected by ecstasy	13	3.6	6	1.0	7.7	.005
Affected by amphetamines	14	3.8	8	1.3	6.4	[.011]
Mobile phone use when driving
Talked on handheld mobile	241	66.2	285	47.7	31.1	< .001
Used mobile phone function	270	74.4	360	60.1	20.4	< .001

Note: ^a Groups were not compared on "Drove when very tired", as this was the criterion used to identify fatigued drivers and other drivers.

Summary: Crash involvement, speeding and fatigued driving

Crash involvement
Young people who had experienced zero, one, or two or more crashes since starting to drive differed significantly on several aspects of their road safety behaviour.

Individuals who had been involved in multiple crashes had held their licences for slightly longer than those who had never crashed (6 years 2 months vs 5 years 9 months).

The groups did not significantly differ in gender composition, nor the amount of time they spent driving each week (both in total, and at different times of the day and week).

Rates of driving offences were higher, with the multiple crash group having been detected speeding on significantly more occasions than the zero crash group. They were also more likely to have come to police attention for driving-related offences and to have had their licence cancelled or suspended.

The multiple crash group also showed a greater propensity for risky driving than other drivers (but particularly the zero crash group), with more individuals from this group having engaged in most forms of risky driving. Differences were particularly evident in rates of moderate-level speeding (between 11 and 25 km/h above the limit) and handheld mobile phone use while driving.

High-level speeding
Young people who had exceeded the speed limit by more than 25 km/h on at least one of their ten most recent driving trips (high-level speeders) did not differ from other drivers in the length of time they had held their licence.

However, the high-level speeding group (who was predominantly male) tended to spend more time in total driving each week, and more time driving at night during the week than other drivers.

High-level speeders had more often been detected speeding, were more likely to have come into contact with police for driving-related offences, and were more likely to have had their licence cancelled or suspended than other young drivers. Furthermore, they had engaged more often in almost all types of risky driving practices. The most powerful differences emerged in moderate-level speeding, falling asleep or nearly falling asleep when driving, and driving when affected by alcohol. However, it should be noted that group differences on other risky driving practices were also generally very strong.

While high-level speeders and other drivers did not differ on rates of crashes when driving, of those who had been involved in a crash, high-level speeders had experienced more crashes than other drivers, particularly more crashes resulting in property damage when driving alone.

Fatigued driving
Young people who had driven when very tired on two or more recent driving trips (fatigued drivers) significantly differed from other young drivers on numerous aspects.

While this group did not differ from other drivers in the length of time they had held their driver's licences, they tended to spend more time behind the wheel in total and on weekends than other drivers.

A higher proportion had been in police contact for driving-related offences. However, the percentage of fatigued drivers and other drivers who had had their licences cancelled or suspended did not significantly differ.

Differences were also evident on most types of risky driving. The most powerful group differences emerged on rates of falling asleep or nearly falling asleep when driving, moderate-level speeding and use of a handheld mobile while driving.

Fatigued drivers were more likely to have been involved in a crash when driving and had experienced a higher number of crashes on average than other drivers (particularly property damage crashes when passengers were present).

6.4 Similarities and differences between problematic driving outcome groups

A summary of the characteristics that were significantly related to each of the three problematic driving outcomes - crash involvement, high-level speeding and fatigued driving - is provided in Table 23. A cross (X) indicates that a particular road safety behaviour or characteristic (e.g., length of time licences have been held) was found to be associated with a specific outcome (e.g., crash involvement), while the absence of a tick indicates that a significant association was not found.

Table 23. Road safety characteristics associated with problematic driving outcomes, at 23-24 years

	Multiple crashes	High-level speeding	Fatigued driving
Driver history
Had licence longer	X
Time spent driving
More hours driving per week		X	X
More weekday daylight driving			T
More weekday night-time driving		X	T
More weekend daylight driving		T	X
More weekend night-time driving		T	X
Driving offences
Detected speeding on more occasions	X	X	X
More police contact for driving-related offences	X	X	X
Had licence cancelled/suspended	X	X
Risky driving (past ten trips)
Drove up to 10 km/h over limit	X	X	X
Drove 11-25 km/h over limit	X	X	X
Drove more than 25 km/h over limit	X	-  b	X
Failure to wear seatbelt/helmet for all of trip		X	X
Failure to wear seatbelt/helmet for part of trip		X	X
Drove when very tired	X	X	-  b
Nearly fell asleep/fell asleep when driving	X	X	X
Drove when affected by alcohol	X	X	X
Drove when affected by marijuana	X	X	X
Drove when affected by ecstasy	X	-  a	X
Drove when affected by amphetamines	X	-  a	X
Talked on hands-free mobile when driving	X
Talked on handheld mobile when driving	X	X	X
Used mobile phone function when driving	X	X	X
Crash involvement
Involved in a crash when driving	-  b		X
Higher number of crashes	-  b	X	X
More property damage crashes (driving alone)	X	X
More property damage crashes (passengers)	X		X
More injury crashes (driving alone)	T
More injury crashes (passengers)

Note: ^a Group comparisons were not undertaken for these aspects due to low prevalence of these behaviours among one or more groups. ^b As the groups were defined on these variables, they were not compared on this aspect. T = trend.

There was considerable overlap in the factors associated with each outcome. Overall, 39% of the characteristics on which the groups were compared were common to all three problem driving outcomes. Overlap was especially evident on apprehension for driving offences and engagement in risky driving practices. Further, high-level speeding and fatigued driving shared 52% of characteristics in common, while crash involvement and high-level speeding, and crash involvement and fatigued driving, shared 50% of characteristics in common.

Some specific associations were also apparent. Length of time licences had been held was only associated with crash involvement. Type and amount of time spent driving was associated with high-level speeding and fatigued driving, but not with crash involvement. However, overall, there was a very high degree of overlap.

Next, the degree to which the same individuals were involved in multiple crashes, high-level speeding and fatigued driving was investigated. Table 24 shows that co-occurrence was very common, with 57% of drivers in the multiple crash group, 72% in the high-level speeding group, and 45% in the fatigued group also found to engage in another problem driving outcome. The highest degree of overlap was found between the multiple crash and fatigued driving groups, while for the high-level speeding group, the greatest level of overlap was with the fatigued driving group. Of concern is that one-fifth of high-level speeders had engaged in all types of problem driving outcomes, a rate much higher than among the multiple crash and fatigued driving groups, for whom rates were 11% and 8% respectively.

Table 24. Co-occurrence of multiple crashes, high-level speeding and fatigued driving, at 23-24 years

	N	% of multiple crash group	% of speeding group	% of fatigued driving group	% of all three groups
Multiple crashes only	109	42.4	-	-	20.9
Speeding only	35	-	28.2	-	6.7
Fatigued driving only	192	-	-	54.5	36.9
Multiple crashes and speeding	25	9.7	20.2	-	4.8
Multiple crashes and fatigued driving	96	37.3	-	27.3	18.4
Speeding and fatigued driving	37	-	29.8	10.5	7.1
All three	27	10.5	21.8	7.7	5.2

Note: A small number of individuals did not have complete data over all three outcomes and were excluded from this analysis.

These findings may reflect differences in group sizes to some extent, as the high-level speeding group was considerably smaller than the multiple crash and fatigued driving groups. They also suggest that high-level speeding may be a more extreme form of risky driving. High-level speeding (defined in this study as exceeding the speed limit by more than 25 km/h) is more likely to reflect an act of conscious risk-taking than fatigued driving or crash involvement.

Summary: Similarities and differences between problematic driving outcome groups

Numerous aspects of driver history, current on-the-road experiences, and risky driving behaviours were in common across the crash involvement, high-level speeding and fatigued driving groups.. For instance, higher rates of driving offences and a greater propensity to engage in a range of risky driving practices were linked to all three problem driving outcomes.

The highest degree of concordance was found for high-level speeding and fatigued driving. However, many of the factors that were associated with crash involvement were also linked to high-level speeding and fatigued driving.

Nevertheless, some characteristics were more strongly associated with particular outcomes than others. For instance, length of time licences had been held was only associated with crash involvement, while the amount of time currently spent driving and the times at which this typically occurred, were linked to high-level speeding and fatigued driving, but not crash involvement. However, overall, there was a very high degree of association.

Individuals who engaged in one form of risky driving were very likely to engage in another type of problem driving. Thus, 57% of young people in the multiple crash group, 72% in the high-level speeding group, and 45% in the fatigued driving group had also engaged in one of the other two types of problem driving outcomes. The highest degree of overlap occurred between the fatigued driving and multiple crash groups. Concerningly, 20% of high-level speeders had engaged in all three problem driving outcomes, compared with 11% of those in the multiple crash group and 8% of those in the fatigued driving group.

6.5 Discussion and implications

This chapter examined the extent to which crash involvement, high-level speeding and fatigued driving were related to driver history, current on-the-road experiences, and other risky driving behaviours. This issue was addressed in two ways: first, by looking at the profile of each group separately and comparing these to other drivers; and second, by looking at the characteristics that were common or unique across the three problem driving outcomes, and the degree to which the same individuals were involved in the differing problem driving outcomes.

Looking first at each outcome, we found that young people who had been involved in multiple crashes tended to be slightly more experienced drivers, were more likely to have been apprehended for driving-related offences, and were more likely to have recently engaged in a wide range of risky driving practices than other drivers (particularly those who had never crashed).

Young drivers who engaged in high-level speeding and/or had driven when very tired on two or more recent trips also exhibited many of these characteristics. For instance, high-level speeders (who were predominantly male) had higher rates of apprehension for driving offences and were much more likely than other drivers to have engaged in a wide range of risky driving behaviours. In addition, this group spent more time on the road (particularly at night during the week) and had experienced higher rates of crashes than other drivers (especially property damage crashes when driving alone).

Similarly, fatigued drivers were more likely than others to have come to police attention for their behaviour on the road, and to have engaged in various types of risky driving. This group (the majority of whom were female) also tended to spend more time driving (in total, and at the weekend), and had experienced more crashes than other drivers (particularly property damage crashes when passengers were present).

There was considerable similarity in the road safety experiences and behaviours of those who had been involved in multiple crashes, engaged in high-level speeding, or driven when fatigued. However, some unique differences were found. For instance, while higher levels of driving exposure (time on the road) were linked with high-level speeding and fatigued driving, they were not associated with crash involvement. Similarly, length of time that licences had been held was associated with crash involvement but not the other two outcomes.

Individuals in each of the three problem driving outcome groups were often involved in another of the types of problem driving outcomes examined in this chapter, with overlap especially evident for multiple crash involvement and fatigued driving. High-level speeders were particularly likely to engage in the two other problem driving outcomes.

Some conclusions and implications that can be drawn from these findings are next presented.

Problematic driving did not occur in isolation

It was notable that problematic driving did not occur in isolation. Thus, young drivers who had been involved in multiple crashes, engaged in excessive speeding, or had driven when very tired on multiple occasions were more likely than other young drivers to have engaged in a wide range of risky driving behaviours and to have come to the attention of police for unsafe or illegal driving. Risky driving practices common to all three groups included low- and moderate-level speeding, falling asleep or nearly falling asleep when driving, driving when affected by alcohol and/or marijuana, and using a handheld mobile when driving. In addition, high-level speeders and fatigued drivers had been in more crashes than other drivers.

These findings are consistent with past research that has demonstrated strong links between the three problematic driving outcomes examined here and a range of other risky driving practices (see Clarke et al., 2002; Engström et al., 2003). There is also consistency with research showing associations between speeding and higher rates of driving offences (Palamara & Stevenson, 2003; Steptoe et al., 2004; Williams et al., 2006). Earlier findings from this report (Chapter 5), which showed that drink-drivers also more often engaged in a wide range of other risky driving practices, corroborate the view that problematic driving behaviours are interrelated and may reflect a risk-taking approach to driving among a sub-group of young drivers.

The overlap found is somewhat at odds with popular perceptions. Crashes tend to be viewed as inadvertent "accidents", while driving when fatigued is often seen as an outcome of a busy lifestyle. While this may be true, the consistent relationships found between these problem driving outcomes and other risky driving practices also suggest a more risky, unsafe approach to driving may be present among some.

The inter-connectedness of many problem driving behaviours reported here supports the viability of interventions aimed at a range of unsafe and/or unlawful driving practices since they will target the same individuals to a large extent. For example, the strong overlap found between high-level speeding and the other problem driving outcomes suggests that road safety efforts targeting excessive speeding may decrease crash involvement and fatigued driving as well. Nevertheless, as the overlap is not complete, more targeted approaches aimed at specific road safety behaviours will still be vital.

Risky driving was linked to other driving experiences

Young people who drove when fatigued were found to spend longer periods on the road than other drivers, spending more hours each week driving at night and during the day, both on weekdays and weekends. Other Australian research on fatigued driving among young people reveals similar trends (Harrison, 2006). Based on a telephone survey of 400 young drivers aged 18 to 25 years and on focus groups undertaken with drivers identified as at-risk drivers through a telephone survey, Harrison concluded that fatigued driving appeared to be a consequence of lifestyle-related motivational factors that overrode young people's concerns about the potential negative consequences of driving when fatigued. Respondents believed that the effects of tiredness on driving were manageable except at the most extreme levels and that they could habituate to tiredness with experience. As a result of these findings, the Royal Automobile Club of Victoria developed a campaign to address the issue of fatigue among young drivers. Other research shows similar trends (Connor et al., 2002; Stutts et al., 2003), and reinforces the continuing need for road safety education programs aimed at raising awareness of the risks of driving for long periods without a break.

Similarly, 23-24 year olds who engaged in high-level speeding also spent more time driving, particularly at night (on weekdays and weekends) and on weekends (during daytime and night-time hours); the peak times at which speeding, and speed-related crashes typically occur (Harrison et al., 1998; Liu et al., 1997). These trends, and those of other research, support enforcement activities that take place at night or on weekends to detect and deter young drivers who speed excessively.

Turning now to the crash history of the ATP sample overall, 43% had experienced a crash at 19-20 years of age, and this had risen to 60% by 23-24 years, indicating that most crashes had occurred when the cohort were less experienced drivers. Likewise, the average number of crashes experienced had risen from 1.36 to 1.6 over this time period. These findings are consistent with past research, which has linked driver inexperience to heightened rates of crash involvement (Cavallo & Triggs, 1996; Engström et al., 2003; Triggs & Smith, 1996). Nevertheless, after almost 6 years of licensure, 40% of the ATP sample had not experienced a crash when driving, and a similar percentage had not been detected speeding.

There were gender differences in risky driving behaviours

The three problematic driving outcome groups differed in gender composition. While most high-level speeders were males (70%), two-thirds of fatigued drivers were females, and similar numbers of males and females had experienced multiple crashes.

These findings align with past research with regard to speeding, as a large body of research shows that males are more likely to speed, be detected speeding, and be involved in speed-related crashes (Engström et al., 2003; Harrison et al., 1998; Liu et al., 2005; Palamara & Stevenson, 2003; Smart & Vassallo, 2005). Further, as described earlier (Chapter 3), a higher percentage of the young men in the ATP sample had engaged in all levels of speeding at 23-24 years than young women. Young men had also been detected speeding approximately twice as often as their female counterparts. These findings reinforce the need for specific initiatives aimed at reducing speeding among young men.

One such initiative is the "anti-hoon" laws introduced by several Australian states, including the state of Victoria in 2006. These appear to have had success in reducing excessive speeding and similar unsafe driving practices. In Victoria, vehicles can be impounded if a driver is detected exceeding the speed limit by 45 km/h. The legislation also applies to offences such as organising or engaging in street racing, evading police, loss of traction (e.g., burnouts, donuts, fish tails), dangerous driving or the repeated occurrence of driving while disqualified. Figures show that young males aged 18-25 years were the most likely group to have their cars impounded. Three years after the introduction of this law, Victorian statistics show that just 5% of those whose cars had been impounded re-offend for a second time and fewer than 1% have had their cars permanently confiscated (Department of Justice, 2009).

Our finding that the majority of fatigued drivers were female appears to be at odds with other research showing that male drivers are more often involved in fatigue-related crashes (Dobbie, 2002; Engström et al., 2003; Pack et al., 1995). However, this comparison is not straightforward, as the circumstances being compared are not the same. The current study looked at fatigued driving quite broadly and did not limit this to the occurrence of crashes, unlike many other studies. Studies that have found a link between male gender and driver fatigue have typically focused on individuals who have been injured or killed in fatigue-related crashes. The current study focused on young people who reported driving when very tired on several occasions during their ten most recent trips. A recent Australian study found that sleep-deprived women were more aware of the impact of fatigue on their driving performance in a simulated driving exercise than sleep-deprived men (Baulk, Biggs, van den Heuvel, Reid, & Dawson, 2006). This finding may partially explain why fatigue-related crashes are higher among men.

Additionally, previous research has shown links between male gender and crash involvement (Cavallo & Triggs, 1996; Clarke et al., 2002; Engström et al., 2003; Vick, 2003), but this trend did not emerge in the current study. However, the focus of previous research has generally been on serious injury crashes, whereas our study looked at the characteristics of drivers involved in a wide range of collisions (ranging from property damage crashes to injury crashes). Almost all of our study members who had been in a crash when driving had experienced relatively minor collisions resulting in property damage. Very few (n = 30) had been involved in more serious crashes in which a person had been injured or killed. It is possible that if we had restricted our focus to more serious crashes, a different pattern of findings may have emerged. However, the small number of participants in our study who had been involved in crashes of this type did not allow this approach.

Overall, our findings demonstrate that risky driving occurs among young men and women alike, although speeding is more common among young men and fatigued driving among young women. Intervention and prevention efforts should therefore be directed to both audiences, although could usefully highlight speeding for young men and driver fatigue for young women.

Conclusions

The driving histories, current on-the-road experiences and risky driving practices of young people who had been involved in multiple crashes, high-level speeding or fatigued driving were explored, as was the degree of overlap between these three problem driving outcomes. Considerable similarity was found in the road safety experiences and behaviours of those who had been involved in each problem driving outcome, with higher rates of apprehension for driving-related offences, and engagement in a wide range of other risky driving practices evident. Further, there was considerable overlap across the three problem driving outcomes and it was common for young people to be involved in more than one type.

• ⁷⁰ While 576 respondents reported they had been in a crash, 33 did not provide further details about the number of crashes experienced and therefore could not be included in these analyses.
• ⁷¹ Due to the number of statistical tests that were performed for each topic, an adjusted significance level of p < .01 is used to reduce the risk of Type 1 error (a finding of significant differences when there is no such difference). For reader interest, results that are significant at the conventional p < .05 level are reported as trends, but are not interpreted.
• ⁷² F(2,667) = 5.2, p < .01.
• ⁷³ F(2,919) = 22.5, p < .001.
• ⁷⁴ χ²(2) = 20.7, p < .001.
• ⁷⁵ χ²(2) = 29.0, p < .001.
• ⁷⁶ Driving alone, crash resulted in property damage: F(1,464) = 197.4, p < .001; carrying passengers, crash resulted in property damage: F(1, 329) = 29.5, p < .001; driving alone, crash resulted in injury or death: F(1,260) = 6.0, p < .05 (trend); carrying passengers, crash resulted in injury or death: F(1, 215) = 0.7, n.s.
• ⁷⁷ χ²(1) = 58.7, p < .001.
• ⁷⁸ t(936) = -2.8, p < .01.
• ⁷⁹ Weekday, night-time driving: t(148.8) = -2.4, p < .01; weekend, night-time driving: t(852) = -2.1, p < .05 (trend); weekend, daytime driving: t(907) = -2.3, p < .05 (trend).
• ⁸⁰ t(134.4) = -5.1, p < .001.
• ⁸¹ χ²(1) = 34.5, p < .001.
• ⁸² t(140.0) = -3.5, p < .01.
• ⁸³ t(87.0) = -3.4, p < .01.
• ⁸⁴ χ²(1) = 6.1, p < .05 (trend).
• ⁸⁵ Total time spent driving: t(940) = -3.8, p < .001; weekday, daytime driving: t(928) = -2.2, p < .05 (trend); weekday, night-time driving: t(530.9) = -2.3, p < .05 (trend); weekend, daytime driving: t(641.9) = -3.0, p < .01; weekend, night-time driving: t(584.0) = -3.7, p < .001.
• ⁸⁶ t(536.7) = -2.5, p < .05 (trend).
• ⁸⁷ χ²(1) = 7.5, p < .01.
• ⁸⁸ χ²(1) = 11.7, p < .01.
• ⁸⁹ t(926) = -3.6, p < .001.
• ⁹⁰ t(342) = -2.7, p < .01.