Traffic Tech
Technology Transfer Series
Number 217       March 2000


Repeat DWI (driving while impaired) offenders comprise a small but important percentage of drivers involved in traffic crashes. Unfortunately, there are very little data on the actual magnitude of the problem. Mid-America Research Institute recently completed a review of the literature on repeat DWI offenders for the National Highway Traffic Safety Administration (NHTSA) as part of a larger update of the alcohol literature.

The report summarizes a review of the scientific literature since 1990 addressing drivers who have been convicted more than once of a DWI offense. The main focus of the review was on the role that this category of drivers plays in crashes, their characteristics, and the nature and effectiveness of countermeasures intended to reduce their alcohol-crash involvement.

Data from NHTSA's Fatality Analysis Reporting System (FARS) suggest that 2-3 percent of those in all fatal crashes in America (including both alcohol-related and non-alcohol related crashes) were repeat offenders. Data from the State of California suggest that the percentage of repeat offenders in fatal alcohol-related crashes could be more in the 8 percent range.

The risk of a driver who has one or more DWI convictions becoming involved in a fatal crash is about 1.4 times the risk of a driver with no DWI convictions (1997 FARS data). The risk of being in an alcohol-related crash (using 1995 state data from California) increases steadily with number of prior DWIs, again amounting to about 1.4 for repeat offenders as a whole. By contrast, California's data showed that the risk of being in any crash (alcohol-related and non alcohol-related combined) decreased with the number of prior DWIs. One reason for this suggested by the literature is that repeat offenders who are sober may drive more carefully than first offenders, or may not drive at all because their license was suspended.

An unexpected finding about repeat offender characteristics was the relatively small practical difference in the average blood alcohol concentration (BAC) compared to that of first offenders (.18 vs .16, respectively). Other than this, repeat offenders share many of the same characteristics as first offenders. Some older studies have found that the two groups were indistinguishable.

Much of the recent research in this area has examined various sanctions that are imposed on repeat offenders. Traditional sanctions, such as license suspension or revocation combined with treatment, appears to be effective, and have the potential to reduce recidivism by as much as 50 percent. Alternative sanctions (for example, electronic monitoring and vehicle impoundment) also appear to be effective, offering potential reductions in recidivism between 15 to 90 percent.

The report concludes with several suggestions for additional research.

Additional research is needed to determine the exposure of repeat offenders to traffic crashes so that the risk of this group can be estimated more accurately.

More state and local level studies are needed using databases maintained by state motor vehicle departments. These studies should include information about the characteristics of persons in crashes. There is a particular need for multivariate studies that can identify high risk groups of repeat offenders. Where possible, personality and psychosocial variables should be merged with crash data.

There is a need for new evaluations of countermeasures such as sobriety checkpoints, jail, and license suspension. There is also a need to evaluate the preventative effect on the general public of countermeasures targeted at repeat offenders.

For a copy of State of Knowledge of Alcohol-Impaired Driving Research on Repeat DWI Offenders (44 pages), write to the Office of Research and Traffic Records, NHTSA, NTS-31, 400 Seventh Street, S.W., Washington, DC 20590 or send a fax to (202) 366-7096 or download from the web at Amy Berning was the contract manager for this project.

Summary of Attributes of Repeat Offenders

Variable Value Variable Value
Sex Mostly Male (over 90%) Personality & Psychosocial Problems Yes, probably more common and severe than those of first offenders
Age Usually under 40, mean about 35 Drinking Locations Multiple locations favoring bars, at home, parties. Often plan to drive after drinking
Race White Final Destination Home
Income Low Beverage Mostly beer, often distilled spirits
Marital Status Unmarried Recidivism ~10 %+ per year; increasing with number of prior DWIs
Education High School or less Implied Consent More than 50% are BAC test refusers
Employment Non-white collar Sentences Traditional, treatment often
BAC .18+ at arrest; higher in fatal crashes Reasons for DWI Thought he or she was fit to drive
Prior DWIs Typically 2 or 3, higher for some in treatment programs Perceived Detection Low for first offenders, increases with priors
Prior Other Traffic Infractions Several Alcohol Problems Often have problems; alcohol dependency common
Prior Criminal Offenses Yes, more than first offenders, include serious crimes against persons  

U.S. Department
of Transportation
National Highway
Traffic Safety

400 Seventh Street, S.W. NTS-31
Washington, DC 20590

Traffic Tech is a publication to disseminate
information about traffic safety programs,
including evaluations, innovative programs,
and new publications. Feel free to copy it as you wish.
If you would like to receive a copy contact:
Linda Cosgrove, Ph.D., Editor, Evaluation Staff
Traffic Safety Programs
(202) 366-2759, fax (202) 366-7096