Title Banner - Alcohol and Highway Safety 2001: A Review of the State of Knowledge

Conclusions and Recommendations

Our overall conclusions and recommendations flowing from our review of the scientific literature dealing with the alcohol-crash problem and societal responses to it are presented in this chapter. Detailed conclusions are contained in the summary and conclusions sections of prior chapters.


We conclude that currently available hard data on the nature of the alcohol-crash problem are adequate for defining the gross prevalence of alcohol-impaired drivers in fatal crashes. For example, it is known that some 12,500 persons are killed each year in crashes in which one or more drivers had a BAC of .10+, and about 16,000 persons are killed annually in crashes in which a driver had a BAC of .01+. (There is also evidence that drivers at BACs much higher than .10 account for a disproportionate share of the alcohol-crash problem.) Since virtually all drivers are impaired at .10+ (and recent research indicates impairment and high risk at even lower BACs), using a BAC of .10+ as a measure is reasonable for determining a lower bound to the current magnitude of the problem. Less is known about the role of alcohol in non-fatal crashes, since comprehensive data based on objective measures of impairment (such as driver BAC) do not exist at the national level.

Research also clearly indicates that the size of the alcohol-crash problem in general has declined significantly in recent years, to the point that it can be said that alcohol-related fatal crashes are a smaller societal problem at the millennium than they were 10 or 20 years ago.

The characteristics of persons who drink and drive are also generally better known than they were at the times of prior state-of-knowledge updates. Basic demographic data for such variables as age and sex exist in abundance, and data are starting to appear on ethnic and racial characteristics. From this knowledge it is more clear than ever that young drivers have especially high alcohol-crash risk and alcohol-crash involvement, and that, young, White males in particular account for a large share of the alcohol-crash problem . Other demographics are available for certain groups of drinking-drivers (e.g., DWIs), but, except in small studies, generally not for drivers in crashes. Also, the drinking patterns and drinking-driving patterns of drinking drivers are becoming better defined. The role of prior DWI convictions in drinking-driving, is now better understood, indicating that while multiple DWI offenders have higher recidivism rates than first offenders, persons with no priors at all may have the highest involvement in total crashes and in alcohol-related crashes of all degrees of severity. Further, research suggests that repeat DWI offenders and first offenders share many of the same characteristics.

Our review found that pedestrians and bicyclists account for a much smaller, but still highly significant, portion of the alcohol-crash (approximately 1,500 fatally injured pedestrians at .10+ BAC). Data from FARS indicate that fully 34% of fatal pedestrian crashes involved either a pedestrian or a driver whose BAC was .10 or higher, and that very high BACs were common among alcohol-positive pedestrians. The contribution of alcohol-impaired bicyclists to the problem is much lower than that pedestrians, probably of the order of a few hundred fatalities a year at the .10+ level.

In general, the literature suggests that data from existing research are sufficient for defining broad groups of alcohol-crash targets, but are still inadequate for identifying more narrowly defined target groups. For example, there are sufficient data to say that young male drivers should be a target group, but not enough data to say that young, unemployed males without a college diploma who drive light trucks are an important subgroup to be singled out for special countermeasure action. In a word, more research is needed on the characteristics of alcohol-crash involved drivers and their relative risk. Specific areas where significant knowledge gaps exist and where significant research efforts are recommended are:


Nearly all countermeasure programs that have been evaluated have focused on the pre-crash phase. Their objective has most often been to reduce driving after drinking, although there has been increasing attention given to reducing excessive drinking before driving. The great majority of programs have used strategies of deterrence and incapacitation carried out by elements of the criminal justice system.

Countermeasures with strong evidence favoring their effectiveness are:

Countermeasures that have shown promise but for which evaluations of alcohol-crash impact are as yet inconclusive are:

While the state of knowledge about ways of dealing with the alcohol-crash problem has grown enormously since the first comprehensive report on alcohol and traffic safety, significant knowledge gaps remain. The most glaring of these is the knowledge about the effect of countermeasures that do not rely on the Criminal Justice System. These other countermeasures include approaches focusing on technology, the vehicle, the highway environment, and the more effective control of alcohol consumption. To date, such approaches have either been insufficiently developed, insufficiently evaluated, or both. Two additional areas where significant new knowledge is needed are: countermeasures targeted at specific groups of drinking drivers, (e.g., groups defined by such variables as race / ethnicity and type of vehicle), and pedestrian countermeasures.

We recommend a coordinated program of countermeasure research and development to fill these gaps. For the short term, the major thrust of operational programs should be on maintaining the 20-year downward trend in alcohol-related crashes. This will require refining current deterrent / incapacitation programs and generating and evaluating new such programs. But concurrently, new approaches will have to be developed, evaluated, and refined for later widespread adoption as the marginal utility of deterrence-based programs becomes exhausted.

Such an effort across a such a broad front will be beyond the capability of any one governmental agency, advocacy group, or industry. Instead, it will require merging of the interests and resources of many organizations and individuals. NHTSA is planting the seeds of this approach in its Partners in Progress program, and is exploring ways of expanding this program into a collaborative effort involving many other partners outside its agency. Clearly, this effort must be cooperative rather than managed because of the diversity of its participating organizations and individuals. Establishing and operating such a program will no doubt require new organizational arrangements, and this may well be one of its most challenging aspects.