Schaffer Online Library of Drug Policy Sign the Resolution
Contents | Feedback | Search
DRCNet Home
| Join DRCNet
DRCNet Library | Schaffer Library | Drugs and Driving 


Drinking and Driving in Great Britain: Which Way Forward?

Andrew B Clayton

British Institute of Traffic Education Research, Kent House, Kent Street, Birmingham B5 6QF, UK


In many countries, a steady decline in the incidence of drink driving fatalities has occurred over the past decade. Recently presented data from Britain (Clayton and Everest, 1994), Australia (McLean, 1994) and the United States (Stewart and Voas, 1994), suggests that the trend is similar in all three countries, despite very different countermeasures being employed.

To an extent, countermeasures against drink drivers have made the implicit assumption that the target group are homogenous in terms of their attitudes, beliefs and behaviour. If, however, there exists within drink drivers a hard core of anti-social drivers, as suggested by Simpson and Mayhew (1991), it is possible that much of the decrease in drink driving can be attributed to changes in attitudes and behaviour amongst essentially socially responsible people. The hard core may have been largely unaffected by such countermeasures as have so far been employed.

In Britain, relatively little detail is known about the characteristics of drink driving offenders. A recent study (The Portman Group, 1994) provided detailed information on a sample of men who had been convicted of a drink drive offence in 1992 in a large metropolitan area. By investigating the circumstances which led up to the offence and the attitudes and behaviour of offenders, it was hoped to develop a strategy for deterring people from offending in the future. It was also intended that the results should provide a comparison with an earlier study conducted in 1976-77 (Clayton et al 1980).

The research comprised two group discussions; 293 personal interviews with convicted male offenders in their home and follow-up depth interviews with a sub-sample of 15 offenders. The sample was skewed towards middle-aged males of social classes C2DE and also had higher proportions of unemployed respondents than would be expected in the catchment area. This profile is a reflection of the type of people who drink and drive, rather than any bias in the sample.

Cluster analysis was carried out on the data to identify groups of respondents with similar behavioural and demographic characteristics and attitudes in order to help in the development of strategies which might be targeted at different groups of offenders. Five clusters were developed and given descriptive labels. A summary of the characteristics of each group is given below.


This group comprised 23% of the sample, aged 25 to 44, typically drinking beer (81%) or strong beer (16%) in a pub or at home before offending. Members of this group were often unemployed C2DE men with previous convictions for drink driving and other crimes.


This group comprised 19% of the sample, aged 25 to 54, typically drinking beer (80%) or strong beer (13%) in pubs before offending. Members of this group were largely C2DEs who denied that drink driving is wrong and felt that the chances of being stopped were remote. Generally, they believed that they were good drivers who were unaffected by drinking.

Devastated Professionals

This group comprised 19% of the sample, aged 25 to 44, of whom 72% drank beer, 16% wine and 16% spirits, in either a pub or at home before offending. Members of the group were predominantly middle-aged, ABs or C1s who were shocked at being treated as criminals. They felt that they were capable of driving after drinking.

Young Irresponsible

This group comprised 17% of the sample, aged 25 to 35 who typically drank strong beer at home at friends' homes before offending. Members of this group were C1s and C2s with a carefree attitude to life, easily influenced in a group, and inexperienced drivers. They generally knew that their driving was impaired but took the risk.


This group comprised 7% of the sample, aged 35 to 54, C1C2DEs who typically drank beer at a friend's or relative's home before offending. Often unusual circumstances such as a celebration, argument or depression led to the offence being committed. Members of this group were severely affected by their conviction and claimed to be reformed as a result.

The results also revealed what offenders considered to be the worst consequences of their convictions. These included:

Comparison with Birmingham Study

Table 1 Summarises the main differences between the Birmingham and Portman Group studies

Table 1
Main Differences between the Birmingham and Portman Group Studies

  Birmingham Study Portman Group Study
Date of offences 1976-77 1992
Total sample 1,032 2,115
Court cases attended 284 -
No of interviews (No in-depth) 102 (102) 293 (15)
Control Group 384 -

In terms of BAC distribution, the Birmingham sample had a higher proportion of offenders with BAC levels in excess of 200mg/100ml. However, in comparison with the national data for 1976-77, the Birmingham study also had a higher median alcohol level.

In terms of time of day, the pattern has hardly changed with nearly half the offences being committed late at night. The major shift over the past decade in the place of drinking away from the pub to the home (noted by Goddard and Iken (1988)) is also reflected in these data. In the Birmingham study, 55% of the drinking occurred in the pub and only 9% in a private house; the corresponding figures for the Portman Group study were 38% and 36% respectively. The difference between the two studies in terms of the claimed amount of alcohol drunk (9 Units and 12.5 Units) does not match the difference in the alcohol levels. In both samples, the interviewees were probably being economical with the truth when asked about how much they had consumed prior to the offence.

The comparison of the age and social class of the two samples showed a very close correspondence with a continuing predominance of young (under 35s) manual workers.

Table 2 suggests that the proportion of offenders who thought that they were over the legal limit has increased only slightly. An extremely high proportion still claimed that they had been fit to drive as safely as normal. The major difference between the two samples was in the proportion who claimed that they had drunk more than they had intended. The observed increase in the Portman Group study may be partly a reflection of changes in public attitudes towards drinking and driving. If drinking and driving is now more unacceptable, then more offenders may claim to have drunk more than they had intended.

Table 2
Comparison of Drinking Attitudes and Behaviour

  Birmingham Study Portman Group Study
Drank more than intended 15% 39%
Thought was over legal limit 38% 43%
Fit to drive as safely as normal 83% 85%

This limited comparison of the two studies suggests more similarities than differences in the characteristics of the offenders. If the studies may be regarded as typical samples of the overall populations of offenders in the late 1970s and the early 1990s, it would appear that the characteristics of those unaffected by all the drink driving countermeasures have remained largely unchanged over a period of some fifteen years.

The Way Forward

A recent seminar hosted by The Portman Group examined some options for change in Britain (Clayton, 1994). By far the greatest initial effect is likely to come from changes in legislation or enforcement procedures. Two main options exist: to lower the legal limit or to increase the level of enforcement (through unfettered discretion or RBT). Either option is likely to create a major public debate which will in itself have positive effects upon the driving public's perception of drink driving. Although both measures are desirable in the long term, the order in which they are introduced may be of critical importance. An increase in the perceived risk of being breath tested caused by increased enforcement may be perceived by the public as tackling the problem of the existing offender and making it more likely that he will be caught. Lowering the legal limit will create a new offence from behaviour that is currently legal. If the aim is to reduce the number of offenders with high BACs, then increasing enforcement may be more appropriate. Its effects are also likely to cascade downwards and impinge upon those at or near the legal limit. It appears less likely that the converse would happen.

These two options are not mutually exclusive and either could be combined with changes in the penalties for drink driving. There is a case for increasing the penalties for the more serious offences by making greater use of vehicle confiscation, immediate licence withdrawals and longer periods of disqualification. For the least serious offences, the period of disqualification might be reduced (particularly if the legal limit were lowered) in favour of compulsory alcohol re-education.

Of the two enforcement options, unfettered discretion would appear to be the more cost-effective than RBT, but any change in legislation would inevitably take time. In the meantime, encouraging the police to make full use of their existing powers would appear appropriate. Any of the above changes would create a great deal of informal publicity as it would be perceived by the public as having a direct effect on their behaviour. A change to unfettered discretion would have a greater effect if the public sees it as a real change. Many drivers may not have a precise knowledge of the conditions under which the police may request a breath test. They may believe that police powers are already unfettered.

There is growing evidence that the majority of the public increasingly support the notion of not driving after drinking (Clayton and Everest, 1994). It is critical that this support is not lost through the introduction of measures that are perceived as draconian. Rather, the support should be built on through a measured campaign against the more serious, persistent offenders. Such a campaign is likely to cascade downwards and thus also have an effect upon those drivers at the less serious end of the problem.


The author gratefully acknowledges the assistance of The Portman Group and the Department of Transport in the preparation of this paper. The views expressed are those of the author and do not necessarily represent the views of either The Portman Group or the Department of Transport.


Clayton, A.B. (ed) (1994). Drink drive seminar. Proceedings of seminar. London: The Portman Group.

Clayton, A.B. and J.T. Everest (1994). Decline in drinking and driving crashes, fatalities and injuries in Great Britain. Linköping: Sweden: VTI, Proceedings of the Conference Strategic Highway Research Program and Traffic Safety on Two Continents, The Hague, 1993, 175 - 186.

Clayton, A.B., McCarthy, P.E., and J.M. Breen (1980). The male drinking driver: characteristics of the offender and his offence. Crowthorne: Transport and Road Research Laboratory, Report SR600.

Goddard, E. and C. Iken (1988). Drinking in England and Wales in 1987. London: Office of Population Censuses and Surveys.

McLean, A.J. (1994). Decline in drink driving and alcohol related fatal crashes in Australia. Linköping: Sweden: VTI, Proceedings of the Conference Strategic Highway Research Program and Traffic Safety on Two Continents, The Hague, 1993, 160 - 172.

Simpson, H.M. and D.R. Mayhew (1991). The hard core drinking driver. Ottawa: Traffic Injury Research Foundation.

Stewart, K. and R. Voas (1994). Decline in drinking and driving crashes, fatalities and injuries in the USA. Linköping: Sweden: VTI, Proceedings of the Conference Strategic Highway Research Program and Traffic Safety on Two Continents, The Hague, 1993, 202 - 218.

The Portman Group (1994). Drink driving: a survey of male offenders. London: TPG.