Introduction

Cover Page Technical Report Technical Summary Acknowledgements List of Tables Introduction Study Sites Observational Study Focus Group Research Conclusions References

Objectives General Approach Research spanning the last 20 years has consistently shown that the suspension or revocation of the driver's license is an effective specific and general deterrent for drinking and driving; the effects are both short-term and long-term (Jones and Lacey, 2000; Wagenaar et al., 2000; Peck, 1991; Klein, 1989; Preusser et al., 1988; Peck et al., 1985). Although vehicle-based sanctions (for example, ignition interlock devices) hold great promise as deterrent measures, states rely heavily on the removal of the driver's license as a primary penalty for alcohol-impaired drivers. Arguably, it has been the most cost-effective sanction available, particularly when applied to first-time offenders. In the U.S., license suspension or revocation traditionally follows conviction for alcohol-impaired driving. Implied consent laws provide that driving privileges also may be suspended or revoked for refusal to submit to a test of blood alcohol concentration (BAC). Administrative license revocation or suspension (ALR/ALS) laws allow police and driver licensing authorities to suspend or revoke a driver's license before conviction, based on the failure or refusal of a BAC test. Such laws have been found to have both general and specific deterrent effects (Wagenaar et al., 2000). As of January 2001, 40 states and the District of Columbia had some form of administrative license suspension/revocation laws applying to the first offense (BAC ³ .08 or .10). An additional two states had an alternative method for removing the license prior to criminal adjudication by the courts. In most jurisdictions, offenders are permitted to obtain an occupational or hardship license during part or all of the administrative/pre-adjudication and the court-imposed suspension periods (NHTSA, 2001). Despite the demonstrated deterrent effects of license removal, there is evidence that some convicted alcohol-impaired driving offenders continue driving after their license has been withdrawn (Griffin and DeLaZerda, 2000; Wiliszowski et al., 1996; Peck, 1991; Ross and Gonzales, 1988; Jones, 1987). For example, analyses of data on fatal crashes for the years 1993-1997 indicate that 6 percent of all fatal crashes in the U.S. involved at least one driver whose license was currently suspended or revoked for any reason; 27 percent of these drivers had had at least one conviction for alcohol-impaired driving during the three years prior to the crash (Griffin and DeLaZerda, 2000). Using induced exposure methods and data for California drivers, DeYoung et al. (1997) estimated that drivers with a suspended or revoked license (for alcohol-impaired driving or any other reason) comprised 8.8 percent of the driving population and were over-involved in fatal crashes by a factor of 3.7:1. Limited evidence, based on self-reported data from drivers convicted of alcohol-impaired driving, indicates that offenders who continue to drive after their license has been suspended may drive less often and more carefully (Williams et al., 1984; Ross, 1991). To date, evidence of continued driving by alcohol-impaired driving offenders has relied on drivers' self-report information or on rates of re-arrest for drinking and driving, traffic violations, or crash involvement. Studies using the former approach may be limited by small or non-representative samples and by biases associated with the self-report of illegal behaviors. The latter approaches are useful in documenting the extent to which suspended or revoked drivers continue to represent a highway safety risk, but are based only on those drivers whose behaviors result in crashes or traffic violations detected by enforcement agencies. What has been lacking is research to establish the prevalence of driving while suspended or revoked among a representative group of convicted alcohol-impaired offenders engaged in everyday travel. Prior to the current study, such research has been precluded by the considerable logistical challenges involved. This study seeks to fill this void. It represents the first systematic effort to gather objective, independent, and unobtrusive observational data on the driving patterns of drivers who are suspended for alcohol-impaired driving.