Abstract
This article discusses the evidence for supporting a change in the legal drink drive limit from 17.4 mmol/L to 10.9 mmol/L. Results from the Grand Rapids case control study, the Medical Research Council's research on simulated driving, and data from case studies from the United States and Australia all indicate that a figure lower than 17.4 mmol/L is necessary if crashes are to be avoided. Behavioural studies also suggest that driving related skills are significantly impaired at blood alcohol concentration below 10.9 mmol/L. The effect of reducing the limit from 17.4 mmol/L to 10.9 mmol/L upon the proportion of drivers with blood alcohol concentrations above 17.4 mmol/L and upon accident rate is also discussed briefly.