The specific deterrence of administrative per se laws in reducing drunk driving recidivism

To determine if administrative per se laws are more effective than other forms of sanction against drunk drivers.

The overall goal of the search strategy was to identify all relevant research concerning the specific effects of administrative per se laws in reducing drunk driving recidivism, traffic crashes, and other alcohol-related driving offenses by those drivers with suspended licenses. Known review articles and MEDLINE reviews formed the reference bibliography; numerous databases were searched from 1966 to the present, using such terms as alcohol, driver's license, recidivism, deterrence, and legislation.

To be selected the study had to be designed to test the presence of an administrative per se license revocation or restriction in a defined cohort, have a suitable comparison cohort whose sanctions for drunk driving were not administrative per se, and provide relevant data that lead to an objective assessment of recidivism. Types of studies included were randomized controlled trials, nonrandomized controlled trials, other specialized cohort studies, and case-control studies. Three studies were identified; all met inclusion criteria.

One of the studies provided Kaplan-Meier survival curves for failure times defined as days to new conviction following the initial arrest. Odds ratios and 99% confidence intervals were extracted from two of the studies and additional information was supplied by the author of one of the studies.

One study found that one state in the United States experienced a reduction of about one third in repeat arrests for drunk driving over a 3-year period among those who were arrested under administrative per se, relative to recidivism seen in a comparison cohort of drivers prior to administrative per se. Two other states did not experience any change in recidivism. The second study found that drivers whose licenses were suspended under administrative per se were 39% less likely during the first year following suspension to be rearrested on the charge of driving while intoxicated compared with a comparison cohort. This differential persisted into the second year of follow-up, but disappeared by the third year. The third study found both first offenders and repeat offenders arrested under administrative per se were 34% less likely to be involved during the year following their arrest in a subsequent motor vehicle crash compared with those in the comparison cohort. Drivers with administrative per se suspensions were 21% less likely to be involved in additional drunk driving offenses, and 27% less likely to be involved in reckless driving offenses related to alcohol.

Administrative per se laws governing license restriction for drivers have been shown to be effective in some states but not others in decreasing the rates at which these same drivers are subsequently involved in a motor vehicle crash or in another alcohol-related offense, compared with drivers who were sanctioned through other conventional judicial processes. Replications are needed in other states or large driver populations using improved methodology.