Vehicle accidents are the fourth most frequent cause of death in South Australia, and the primary cause of death among 15 to 24 year olds. In 30 percent of South Australia road fatalities, at least one driver was under the influence of alcohol. This report reviews two major categories of countermeasures: those which attempt to change the behavior of the drinking driver, and those which concentrate on modifying physical or social environments. In the first category, legal sanctions have a deterrent effect on the average motorist, but a high risk group of hard-core recidivists is impervious to all penalties. Behavior modification techniques have yet to prove useful, but the compulsory use of seat belts has substantially reduced rates of road deaths. Recently, South Australia has started administering random breath testing (RBT). The objective of this controversial technique is to enhance general deterrence by convincing drivers that there are high risks in detection. The results of research on RBT indicates that while it is effective, it may not serve as a long-term deterrent. RBT has led to altered drinking patterns: liquor sales are down, the taxi trade has boomed, city hotels have more lunch business, and residential lounges do more early evening business. In general, people are visiting bars within walking distance. Conflicting data make assessment of long-term RBT effectiveness difficult. Eleven tables accompany the text. Appendices include additional tables, a summary of penalties for drunken driving, a summary of opinion polls on RBT, random breath test samples, and a list of publications of the South Australia Office of Crime Statistics. A bibliography is included. (Author summary modified)