CHAPTER EIGHT: CONCLUSIONS AND RECOMMENDATIONS
Overview conclusions
8.1 One quarter of drivers (25%) drove within a few hours of consuming any alcohol, which is a legal activity (provided they were under the legal limit). This prevalence has declined since 2001, as has the incidence rate, of 154 incidents per 100 drivers. The prevalence of driving while over the limit in the past 12 months was low, engaged in by only one in 20 drivers (5%), and has remained constant since 2001. The incidence rate for drink-driving has decreased and was 11 incidents per 100 drivers in 2007. The reason for the broad drop in prevalence and incidence is likely to be because of the dominant view among all drivers that drink-driving was socially unacceptable, indicated by the fact the prevalence of ever driving after drinking and drink-driving has fallen among all age groups. The one measure that has not changed, driving while over the limit in the past 12 months, indicates that there is a small group of drivers who persist in drink-driving; the drop in incidence among this group suggests the prevailing view of society is affecting their behaviour as well, but to a lesser extent.
8.2 Given the strength of collective view expressed in the survey that drink-driving is anti-social, it is not surprising that attitudes were not linked to behaviour. Where there were differences in view, these tended to be small and were shared by those who drove after drinking and those who did not. Instead behaviour was linked to circumstance. Drivers who drank before driving suggested that the situations in which they did drink and drive were often unconsidered and circumstantial, a by-product of the mixing of drinking for pleasure and driving for convenience. There was no consensus among all drivers over whether it was easy to end up over the limit without realising it, or not.
8.3 The concept of drink-driving, and indeed the phrase drinking and driving, was associated with being significantly, rather than marginally, over the limit. This was, therefore, often seen by drivers as something that other people, rather than they, do. As a result, most drivers who drank alcohol before driving did not identify with that behaviour; the majority was convinced that their drinking before driving was measured and controlled and rarely resulted in them being more than a little over the limit.
8.4 Driving at, or marginally over, the limit was seen by those who drink and drive as low risk. This risk was perceived to be low both in terms of the effects of the quantity of alcohol consumed, as drivers only drive when they feel in control, and also in terms of the likelihood getting caught, a perception which was backed up by experience for some drivers. It is possible that the study underestimated the prevalence of drink-driving, as doing so in this marginal way was not always recognised as drink-driving. Driving the morning after drinking alcohol was also under-reported in the survey. This is because it was not thought of as drink-driving; drivers find it difficult to estimate the extent of alcohol left in the system the morning after drinking and as a result it is viewed as accidental and, therefore, excusable.
8.5 There was confusion about the drink-drive limit and drivers who did consume alcohol used what they considered the accepted rule of thumb as to what is a 'safe' amount to drink before driving, which is 2 drinks. This is often applied regardless of the type of drink, size of glass, strength of alcohol content or circumstance in which the drink was consumed. Drivers were aware of these issues, but were uncertain about how they could measure alcohol levels practically. On occasion, usually portrayed as unplanned and outwith the driver's control, their accepted limit of 2 drinks (what they perceived to be 'safe', even though it is potentially over the limit) could be exceeded.
8.6 Only half of those who did not drink alcohol before driving thought that drinking before driving should not be allowed. Drivers who did drink alcohol were also split as to whether the legal limit should be zero; some considered this would be impractical.
8.7 Not everyone who drove after drinking alcohol knew the penalties that would apply should they be caught drink-driving, and this caused them to underestimate the risks. It was clear that drivers who did drink and drive, especially at or marginally above the limit, had given little thought to, and severely underestimated, the consequences of being caught and convicted of drink-driving.
8.8 The study demonstrated that current drivers (i.e. those who had driven a motor vehicle on a public road in the past 12 months) tended to drive often. Virtually all used the car for social and personal reasons and the majority drank alcohol fairly often so it is inevitable that, at times, the 2 things coincide. The more a driver is behind the wheel, and the more alcohol they consume, the more likely it is that the 2 activities do coincide, however unintentionally. Having said that, the number of drivers who drove whilst over the limit was small. They were spread across age-groups, gender and social grade and, as a result, are difficult to target. Drivers did have perceptions of who does drink and drive, for example those who live in rural areas or younger drivers, most of which appear to have some foundation in reality, though none of which represent the full picture.
Examining specific drivers
8.9 The study specifically set out to examine the drinking and driving practices of some different groups.
Drivers in rural areas
8.10 There was a strong perception that drivers living in rural areas drink-drive more often as 'they can get away with it', it is more socially acceptable, or practically, there are few alternatives. At the same time, drivers in rural areas pointed out that they have more to lose, as they tend to have a greater dependency on their car, which increases the risk. In common with previous research on the subject, the research found no evidence that drink-driving, or indeed driving after consuming any alcohol, was more prevalent in rural than urban areas. The incidence of driving after consuming alcohol was actually lower in rural areas, so drivers living in those areas drove after drinking alcohol less often than other drivers.
Women aged 25 and over
8.11 Statistics on convictions for drink-driving between 1998 and 2002 showed that the rate at which convictions among women, especially those over 25, increased was higher than the overall rate and may indicate an increase in prevalence over that period. This research set out to identify whether such an increase could be detected since the previous study in 2001.
8.12 Compared to men, women more often did not drink alcohol before driving, expressed stronger views on driving as being anti-social and assessed the behaviour as being riskier. The proportion of women who drove after drinking alcohol, and who suspected they had driven while over the limit, was smaller than the proportion of men who had done either of those things. Since the previous study in 2001, there has been a larger decrease (from a higher base) in the prevalence of driving after drinking alcohol among men than among women, so the gap between the 2 is reducing.
Other groups of interest
8.13 Young people, particularly males, were sometimes considered to drink and drive more than other groups. They tended to socialise more, were seen as being more inclined to take risks and take more risks, or perhaps have less to lose, than other age groups. An opposite view was that for many, their cars were important for the independence they provided and status they conferred. The qualitative research demonstrated that for some, the pressure felt from peers did encourage drink-driving. Reported prevalence of driving after drinking any alcohol among young people was similar to that of older people (60+) and lower than other age groups. Incidence was lower for young people than among any other age group. A higher proportion of young drivers (17-29 year-olds) had driven while over the limit in the past 12 months than other ages. The fact that young people are a relatively small proportion of all drivers means it would be misleading to suggest that they are the biggest group among drink-drivers - only one in 5 drink-drivers (driven while over the limit in the past 12 months) was aged 17-29.
8.14 Older people (60+) were perceived to drink-drive more than others, mainly because they started driving when the behaviour was regarded as acceptable rather than anti-social, and they have been drinking and driving for a long time so are unlikely to change. The evidence from this research is mixed. Prevalence among older people was lower than for those in their middle years (30 to 59), although their incidence of driving after drinking alcohol was higher. While fewer older people said they drank alcohol before driving, and admitted to drink-driving, those who did drive after drinking alcohol appeared to do so more frequently than other age groups.
8.15 Finally, incidence of driving after drinking any amount of alcohol was higher for those in AB social grade than any other demographic group, and for men. Prevalence of driving after drinking alcohol was also high among these 2 groups and among those aged 30 to 59.
Future drink-drive strategies and initiatives
8.16 The research has shown that the accepted social view is that drinking and driving is anti-social. This has no doubt contributed to the reduction in reported prevalence and incidence of that activity on most measures. It is important that those views are maintained and ideally strengthened further, so that it becomes even less acceptable to drink and drive. There is a need to target according to situation, characteristics and attitude to risk. These are considered below.
Targeting Drivers
8.17 As part of developing the deterrence strategy, it is important to identify the focus of future communications and initiatives. Is the aim to target those who drive when well over the limit, who do so repeatedly and are aware of what they are doing? Alternatively, is the target those who drive after drinking at or just above the legal limit who do not recognise their behaviour as drink-driving? A further target might be those who drive the morning after drinking who might not identify themselves as drink-drivers. The tactical approaches for the different groups need to be tailored. Those who do not recognise their behaviour by the language currently used to describe drink-driving will not respond to initiatives and communication designed to modify that behaviour.
8.18 Communications and initiatives also need to be targeted towards specific subgroups; for example, younger drivers may respond to the message that drinking and driving can result in the loss of status, independence and freedom; older drivers may be influenced by the message that drink-driving can lead to being branded a criminal and living with a criminal record. Communications should tactically target the different key audiences, through both media selection and the messages themselves. Targeting is made harder in this case because the numbers drink-driving are relatively low and are spread across most types of drivers; there is no one clear target that will make the most impact on reducing this activity. To some extent, targeting might be determined by which groups, if any, have the most, or most serious, drink-drive related accidents. Further investigation would be required as to whether relevant statistics are currently available or could be gathered to assist targeting.
Using the risk audit
8.19 The risk of drink-driving, especially when marginally over the limit, is considered low by those that do it. Drivers believe they are still in control of the car (because they would not drive if they were not): they are not going far and know the road and they will drive carefully. Although the impact of having an accident would be high, particularly if someone was killed or injured, the likelihood of having an accident is judged as negligible, especially if a driver has driven after drinking on various occasions before, so this risk is discounted. The likelihood of getting caught is also presumed to be low especially if they look after their cars and drive carefully, not drawing attention to themselves. A driver would be considered unlucky if he or she got caught. The more a person drives after drinking, the more this view is confirmed. Even if a driver is 'unlucky' enough to get stopped by the police and breathalysed, the fact they have consumed their 'personal safe limit' means they believe they may not be over the legal limit at all. The impact of getting caught is also misjudged due to a lack of knowledge about penalties and lack of thought about the emotional and practical consequences of being arrested, charged and convicted.
8.20 To counter this latter view, communications should continue to strongly emphasize the consequences of being caught and convicted of drink-driving, including the possibility of spending a night in a police cell, effects on current employment, inconvenience of travel and loss of car.
8.21 Drink-driving is not strongly associated with being a criminal act and the longer term effects of having a criminal record can also be emphasised. Such effects include: the need to disclose this on job application forms; the impact on car insurance; the impact on ability to hire a car or van after the ban finishes; and on potential difficulties entering certain countries with a criminal record. The communications should link these messages with the emotional effects of being caught, asking: Was it worth it? How are you feeling now? What happens next? Communication based on real testimony from real drivers who have been through the ordeal, with which different types of drivers can associate and identify, would be very powerful.
Areas for additional research
8.22 One objective of the research was to identify additional areas for research on this topic.
8.23 It will be useful to repeat this study in the future in order to monitor changes in prevalence and incidence over time. The low proportion of current drink-drivers means that it will be difficult to identify further reductions (or indeed increases unless they were very large) with any confidence, which will need to be recognised in the research design. Similarly the design will need to recognise that particular groups that are targeted through communications may only occur in low numbers in the sample. A further purpose of repeating the research is to monitor changes in attitudes of drivers in general. Any changes identified, for example reductions in the near universal agreement with a number of statements, could provide an early signal of change in driver behaviour.
8.24 We suspected in this research that the overall profile of drivers may have changed since 2001. No profile of drivers was available from 2001 16 so we were unable to verify this was the case. In the future a profile of current drivers, gathered cost-effectively using, for example, an Omnibus survey, would provide helpful background information for a repeat study.
8.25 There were two other areas where further research might be beneficial, principally to inform future communications. The first would be to investigate perceptions and attitudes towards driving the morning after drinking. There was some ambiguity around drivers' views on this as it appeared from the survey they did not consider it to be drink-driving, although in the qualitative research they also recognised there was the potential to be over the limit when driving the morning after.
8.26 The second area would be in relation to the issue of what people described as 'borderline' drink-driving, which was considered acceptable and actually was not really drink-driving at all. This would merit further investigation to establish how communications could be developed to address this perception.
8.27 In paragraph 8.18 we highlighted the difficulty of targeting persistent drink-drivers as they were spread thinly among drivers with different characteristics. We suggested that targeting could be based on groups of drivers who have the most, or most serious, accidents as a result of drinking and driving. To our knowledge there is no published data available. Further investigation might be useful to identify whether such information exists, or could be gathered.
Future communications and activities
8.28 Drink-drivers clearly do not take their conduct seriously enough. Communications and other activities need to be employed that increase the perceived level of risk and challenge behaviour. If perceptions of increased risk grow then this might change the attitude that those who are caught are just 'unlucky'. This can be done in a number of ways, including:
- Publicising the number of drivers who are breathalysed, rather than the number caught and charged by the police
8.29 The resulting higher figures, and potentially increasing trend, would act as a warning to the public that there is a heightened police presence and possibility of being caught. If these statistics are published at a local level, they would be more relevant. In addition, monthly reporting would reinforce the message as a constant threat, rather than through periodic campaigns when drivers may temporarily moderate behaviour in response to campaign activity.
- Communicating the numbers caught as a success rather than a failure of campaigns
8.30 Currently, the reporting of numbers caught drink-driving positions them as disheartening or a failure of campaigns to reach drink-drivers and alter their driving habits. Instead, the numbers caught could be presented as being due to the success of policing and the new strategy to reduce drink-driving. This would help raise the perceived level of risk associated with drink-driving.
- The introduction of random testing 17 would significantly increase the perceived risk and influence behaviours
8.31 Currently, there is a perception among drivers that, as long as they drive carefully, observing the rules of the road, and keep their cars in good working order, they are unlikely to be stopped by the police. If random testing were introduced, the likelihood of being stopped and breathalysed by the police, and as a result the perceived level of risk associated with drink-driving, would increase and may result in fewer drivers feeling able to take a risk and drive home when they think they may be at or just over the limit.
- Lowering the legal alcohol limit for driving
8.32 Some drivers indicated that they drink and drive because they are legally allowed to and that they would stop if such activity was not allowed. Others pointed out the impracticality of a zero limit due to issues of drinking the night before and alcohol in food. However, many respondents understood the effects of alcohol varied according to circumstances and characteristics, but did not know how to apply this understanding. They were confused by units and how these equated with their preferred measure (drink or glass); and did not understand how to assess whether or not they were within legal limits. As the limit stands, it gives scope for the '2 drinks are okay' belief, which may on occasion take people over the limit. There was a belief that if the limit was lowered slightly (e.g. to a level where 2 drinks were clearly over the limit) this would go some way to removing the ambiguities and, with it, the licence some clearly feel they have to push their drinking beyond the legal limit before driving.