Abstract To assess the importance of benzodiazepines as a risk factor for road accidents, the number of blood samples positive for these drugs was compared among injured drivers responsible for a traffic accident and among injured non-responsible drivers and pedestrians. Benzodiazepines were detected by qualitative immunoenzymatic assay. A total of 3147 subjects were registered as accident victims, of whom 2852 had a complete file, including levels of blood alcohol and benzodiazepines and degree of responsibility for the accident. Benzodiazepines were detected in the serum of 8% of the study population. Comparison of the responsible and non-responsible groups did not show any significant difference between them when the part played by alcohol was disregarded, even after adjustment for age and sex. In conclusion, this study showed, once again, that alcohol is clearly an accident risk factor, but found no statistically significant evidence that this also applies to benzodiazepines.