DRINKING AND DRIVING

The automobile age brought with it unprecedented prosperity and freedom of movement, but motor vehicles have also caused the deaths and injuries of millions of people. From the beginning the abuse of alcohol has been universally viewed as one of the major causes of vehicular carnage, with severe punishments being deemed the best way of dealing with the self-indulgent reprobates responsible.

According to the sociologist Joseph Gusfield, noted for his work on alcohol in American society, behind all legislation aimed at curtailing drinking and driving is the image of "the killer drunk," the morally flawed character who has committed more than an ordinary traffic violation. Unlike the social drinker, who knows his limits and respects the law, the drinking driver is a villain who threatens the lives of the innocent through indulgence in his own pleasure. In this legislation, unlike other kinds of traffic law, it is the behavior itself, the hostile, antisocial menace, which is singled out for special disapproval. From this perspective, the enforcement of drinking-driving legislation is as much a matter of public morality as it is of public convenience and safety (Gusfield).

The specter of the killer drunk is the key image that animates "the dominant paradigm," to use the term coined by H. Laurence Ross, another American sociologist who has done more than any other scholar to elucidate, from an international perspective, the causes and prevention of drinking and driving (Ross, 1982, 1992). The dominant paradigm understands that there is a safe drinking level for the great mass of responsible drivers, differentiated from the levels regularly achieved by the small minority of reckless "drunken drivers." The problem, in fact, is not "drinking and driving" at all, but "drunken driving." The dominance of this paradigm in the United States is one reason why the term drunken driving is used so often there, in contrast to most European nations and Australia, where "drinking and driving" or "drink-driving" are the more popular terms.

How one defines the problem is fundamentally important in determining how one thinks about responses. The dominant paradigm calls for severe punishments administered through the criminal justice system. Not only are such punishments fitting, they are capable of deterring further offending, especially if they are backed by rigorous police enforcement. To the extent that the problem is construed in terms of the pathetic drunk rather than the cold-blooded killer, proponents of the dominant paradigm are also comfortable with offering treatment to offenders, provided such programs are not used to evade punishment.

Another way of viewing the problem is through what Laurence Ross calls "the challenging paradigm." Those who think within this framework are uncomfortable about drawing a rigid line between dangerous drunks and social drinkers, although they recognize that heavy drinkers are a critical part of the problem. Their inspiration is the public health perspective, which is not primarily concerned with righting the moral balance of the world but with minimizing alcohol-related harms. Adherents of the challenging paradigm view alcohol-related accidents as the product of the conjunction of the social institutions of transportation and recreation, rather than as a manifestation of moral dereliction. All developed societies rely, to an increasing extent, on private vehicles for all daily functions including recreation, while the consumption of alcohol is accorded an honored place in after-work camaraderie, weekend leisure, and business lunches. Large taverns with even larger car parks are built in the suburbs, and drinking to intoxication remains a core recreational activity for large numbers of people.

If the problem is institutions, perhaps the solutions lie in modifying the way these institutions operate. The challenging paradigm has a place for the criminal justice system, especially if the emphasis is on the general deterrence of the whole driving population. However, they also look beyond the criminal justice system to alcohol and transportation policy, exploring the utility of such measures as reducing alcohol availability or making vehicles or roadside hazards more "forgiving" of the errors of the drinking driver.

In the remainder of this discussion we explore many of the issues raised by the dominant and challenging paradigms, and assess the scientific evidence for the claims made.

The role of alcohol in road accidents

Around the middle of the twentieth century the technical means became available to measure the quantity of alcohol in a person's blood (the blood alcohol concentration, or BAC, usually measured in terms of grams of alcohol per milliliter of blood). Laboratory research using this technology showed that at BAC levels much lower than those normally associated with intoxication, tasks related to driving performance (such as divided attention tasks) were noticeably affected. Although the effects of BAC depend on such factors as an individual's weight, rate of drinking, and presence of food in the stomach, deterioration in performance becomes quite marked between BACs of .05 and .08. As a guide, the average man would attain a BAC of .05 or higher if he drank three "standard drinks" (e.g., three mid-size glasses of mid-strength beer) within one hour, without eating.

The alcohol-crash link was confirmed in a series of case-control studies that compared the BACs of drivers experiencing crashes with those of matched non-crash-involved drivers. These studies found that relative crash risks increase exponentially with BAC: at .05 the risk is double that for a zero-BAC driver, at .08 the risk is multiplied by ten, while at .15 or higher (the levels typically attained by drivers arrested for drinking and driving) the relative risk is in the hundreds. The curve is even steeper for serious and fatal crashes, for single-vehicle crashes, and for young people.

While it is likely that factors other than alcohol, such as a propensity to take risks, contribute both to the levels of drinking and to crash involvement, there is a near universal consensus that there is a direct and causal link between alcohol consumption and crashes, especially serious crashes. For example, eliminating alcohol would probably have prevented about 47 percent of fatal crashes in the United States in 1987 (Evans).

Prevalence and patterns of drinking and driving

The most direct way of measuring the prevalence of drinking and driving is to take breath tests from a random sample of motorists. A number of countries carry out these surveys periodically, usually at nights and at weekends when drinking drivers are more numerous. Two groups of nations emerge in these studies. One group includes Scandinavia and Australia, where there are relatively few drinking drivers on the roads. Moderate to high BACs are found among less than 1 percent of drivers in these countries, even at peak leisure times. The second group includes the United States, Canada, France, and the Netherlands, where between 5 and 10 percent of drivers during nighttime leisure hours have moderate to high BACs. These patterns are broadly consistent with overall road fatality rates for different countries, and also with analyses of the BACs of drivers killed. However, in these latter studies even the Scandinavian countries have found that more than a quarter of drivers have positive BACs, despite the low numbers overall of drinking drivers on the road.

A second main way of estimating the prevalence of drinking and driving is to ask random samples of drivers about their behaviors in the recent past. For example, a 1988 study comparing Norwegian, Australian, and American drivers found that 28 percent of Australians, 24 percent of Americans, but only 2 percent of Norwegians admitted to driving in the past year after four or more drinks (Berger et al.). Despite their poor behaviors, 78 percent of the Australians agreed that it was morally wrong to drive after so many drinks, a higher figure than in the United States, but (again) lower than for the Norwegians, who scored a very high 98 percent. Overall, "general prevention," defined as the influence of moral inhibitions and of social pressures, had taken greater hold in Norway than in the English-speaking countries, but general deterrence (behavior change in response to fear of the threat of legal sanctions) was a more potent force in Australia than in the other countries.

Using intoxication among drivers in fatal crashes as an indicator, dramatic reductions in drinking and driving were experienced in most developed countries in the 1980s. However, the indicators reversed direction in the early 1990s, but then continued in modest decline in the second half of the decade. Formal and informal controls on drinking and driving differ markedly from country to country, but nevertheless there appear to be some common influences. Levels of police enforcement (not the severity of penalties) stand out in all countries as an influence, together with a reduction in per capita alcohol consumption. Attention paid to the problem by political leaders, and the visibility of drinking and driving in the press, appear to be critical factors.

Deterrence

The deterrence of drinking and driving depends primarily on increasing the perceived probability of apprehension in the target population. One way of accomplishing this is to introduce laws that replace the vague offense of "driving under the influence" with the offence of driving with a BAC above a prescribed level (usually .08 or .05). Another way is to initiate a police crackdown on drinking and driving for a period of time. The experience of the United Kingdom in 1967, when it introduced for the first time a .08 BAC limit, illustrates well the usual impact of such interventions. The law was extremely controversial at the time, with the result that most drivers were aware of it and believed they would be caught if they drove after drinking. There was a marked decline in serious accidents at nights and weekends, but not at times when drinking and driving would not be expected. However, the deterrent impact wore off within a few years as drivers gradually became used to the new law, and realized that their chances of detection were in fact not very high.

This pattern of a sharp decline in drinking and driving coincident with a new law or with intensified police enforcement, followed by a gradual decline to pre-intervention levels, is commonly found. Deterrence is an unstable psychological process dependent on continuous publicity and on the perception of a credible police threat. However, random breath testing (RBT) is a major exception to the rule that enforcement effects are invariably temporary.

Under RBT as it is practiced in Australia and some Scandinavian countries, large numbers of motorists are pulled over at random by police and required to take a preliminary breath test, even if they are in no way suspected of having committed an offense or been involved in an accident. Thus RBT should be sharply distinguished from the U.S. practice of sobriety checkpoints, in which police must have reasonable suspicion of alcohol consumption before they can require a test. The RBT law has been very extensively advertised and vigorously enforced in Australia, with the result that 82 percent of motorists reported in 1999 having been stopped at some time (compared with 16 percent in the United Kingdom and 29 percent in the United States).

Time series analyses of accidents show that in Australia RBT had an immediate, substantial, and permanent impact, with every extra one thousand tests conducted each day by police resulting in a 6 percent decline in daily serious accidents (Henstridge et al.). The direct deterrent impact was enhanced by the fact that RBT gave heavy drinkers a legitimate excuse to drink less when drinking with friends. This is a good example of how formal sanctions can reinforce informal sanctions.

The same time series analyses show that a reduction in the legal BAC in some states from .08 to .05 resulted in an average 10 percent decline in serious accidents. This is consistent with experience in other countries where the BAC level has been reduced.

RBT and lower BAC levels concern certainty of detection. Administrative license revocation, the practice in some U.S. states where drivers who drink have their licenses revoked almost as soon as they fail a breath test, concerns swiftness of punishment. Research supports the potential of this procedure to reduce the recidivism of sanctioned drivers and to deter others. As a general rule, the only sanction applied to drivers who drink that reduces recidivism is loss of license. Although many drivers continue to drive while unlicensed, they tend to be more cautious and hence safer. Thus it seems that license loss has (to some extent) a physically incapacitating effect.

Other countermeasures

License loss is effective for both alcohol-related and non-alcohol-related accidents, but its impact on drinking and driving can be enhanced if combined with alcohol treatment. While treatment without license suspension is generally ineffective, suspension plus education, psychotherapy counseling, or follow-up contact probation (preferably in combination) produce an additional 7 to 9 percent reduction in recidivism and accidents (Wells-Parker et al.). Ignition interlock devices, which prevent a vehicle being started until the driver passes a breath test, have been shown to be very effective for many high-risk offenders. However, the effects tend to be limited to the period of the court order unless combined with treatment within a case management framework to deal with the underlying problems.

The problem with all countermeasures focused on apprehended offenders is that most serious alcohol-related crashes involve drivers with no prior drinking and driving convictions. Hard-core drivers who drink comprise about 1 percent of drivers on the road, but more than a quarter of drivers killed. Many of these drivers have a history of violence and serious antisocial behavior including crime, with alcohol abuse simply one facet of their deviant careers. It is likely that for this group a radically different approach is needed, involving early childhood interventions (Farrington).

Most accidents do not involve hard-core offenders, and there is therefore a continuing need for countermeasures directed at the general population. Promising measures include promotion of responsible beverage service for bar staff and managers of on-premise alcohol outlets combined with deterrence of drinking and driving through local enforcement; reduction in retail availability of alcohol to minors; and reductions in the number and density of alcohol outlets to limit general access to alcohol. Any measure that reduces per capita alcohol consumption, such as increases in price through taxation, will reduce alcohol-related accidents.

Reducing dependence on driving has similar promise. Successful measures include designated driver programs (someone in a group stays sober so that that person can drive home), safe rides programs, and increasing the age of driver licensing or restricting licenses to daytime use for young drivers. Promoting public transport would certainly be effective if it were ever evaluated for its impact on drinking and driving. Contrary to expectations, there is no evidence that driver education for young people reduces crash involvement. Indeed, the evidence suggests the reverse: by encouraging young people to gain their license at an earlier age, such training increases exposure to risk, and hence accidents.

Finally, making the vehicle and roadside environment more forgiving of the errors of drinking drivers will reduce deaths and injuries. Frangible poles that minimize damage to vehicles; improved response times and skills of emergency medical teams; more use of seatbelts and airbags; and brighter reflective road signs (so impaired drivers notice them) are but a few examples of effective environmental interventions.

Conclusion

Overall, the picture is one of steady progress, with some setbacks. The challenging paradigm, based on the principles of population health, continues to score successes through such strategies as reducing the legal blood alcohol concentration. General deterrence, especially utilizing random enforcement methods, has achieved permanent reductions in alcohol-related crashes, as has administrative license revocation. Treatment combined with license suspension and ignition interlocks reduce recidivism and accidents. Tougher penalties, the major emphasis of the dominant paradigm, show no promise at all.

The challenges include maintaining the deterrent impact of random enforcement; finding long-term ways of dealing with hard-core offenders; optimizing the use of alcohol and driving controls in politically acceptable ways; and maintaining political and media interest in the drinking and driving problem in the face of stiff competition from other social issues. The fact that drinking and driving declined in most countries in the latter part of the twentieth century, despite wide variations in prevention strategies, suggests that within the challenging paradigm there are many pathways to a safer motoring environment.

Ross Homel

See also Alcohol and Crime: Behavioral Aspects; Alcohol and Crime: The Prohibition Experiment; Alcohol and Crime: Treatment and Rehabilitation; Police: Policing Complainantless Crimes.

BIBLIOGRAPHY

Berger, Dale E.; Snortum, John R.; Homel, Ross J.; Hauge, Ragnar; and Loxley, Wendy. "Deterrence and Prevention of Alcohol-Impaired Driving in Australia, the United States and Norway." Justice Quarterly 7 (3) (1990): 453–465.

Evans, Leonard. Traffic Safety and the Driver. New York: Van Nostrand Reinhold, 1991.

Farrington, David. "Early Developmental Prevention of Juvenile Delinquency." Criminal Behavior and Mental Health 4 (3) (1994): 209–227.

Gusfield, Joseph R. The Culture of Public Problems: Drinking-Driving and the Symbolic Order. Chicago, Ill.: The University of Chicago Press, 1981.

Hauge, R., ed. Scandinavian Studies in Criminology. Vol. 6, Drinking and Driving in Scandinavia. Oslo, Norway: Universitetsforlaget, 1978.

Henstridge, John; Homel, Ross; and Mackay, Peta. The Long-Term Effects of Random Breath Testing in Four Australian States: A Time Series Analysis. Canberra, Australia: Federal Office of Road Safety, 1997.

Hingson, Ralph. "Prevention of Drinking and Driving." Alcohol Health and Research World 20, no. 4 (1996): 219–226.

Homel, Ross. Policing and Punishing the Drinking Driver: A Study of General and Specific Deterrence. New York: Springer-Verlag, 1988.

——. "Drivers Who Drink and Rational Choice: Random Breath Testing and the Process of Deterrence." In Routine Activity and Rational Choice. Edited by Ronald V. Clarke and Marcus Felson. Vol. 5 Advances in Criminological Theory. New Brunswick, N.J.: Transaction Publishers, 1993. Pages 59–84.

Laurence, Michael D.; Snortum, John R.; and Zimring, Franklin E., eds. Social Control of the Drinking Driver. Chicago, Ill.: The University of Chicago Press, 1988.

McCord, Joan. "Drunken Drivers in Longitudinal Perspective." Journal of Studies on Alcohol 45 (1984): 316–320.

Ross, H. Laurence. Deterring the Drinking Driver: Legal Policy and Social Control. Rev. and updated ed. Lexington, Mass.: Lexington Books, 1984.

——. Confronting Drunk Driving: Social Policy for Saving Lives. New Haven, Conn.: Yale University Press, 1992.

——. "Prevalence of Alcohol-Impaired Driving: An International Comparison." Accident Analysis and Prevention 25(6) (1993): 777–779.

Wells-Parker, Elizabeth; Bangert-Drowns, Robert; McMillen, Robert; and Williams, Marsha. "Research Report: Final Results From a Meta-Analysis of Remedial Interventions with Drink/Drive Offenders." Addiction 90 (7) (1995) 907–926.

Wilson, R. Jean, and Mann, Robert E., eds. Drinking and Driving: Advances in Research and Prevention. New York: The Guilford Press, 1990.