Patterns of drug use in fatal crashes

To characterize drug prevalence among fatally injured drivers, identify significant associations (i.e. day of week, time of day, age, gender), and compare findings with those for alcohol.

Descriptive and logistic mixed-model regression analyses of Fatality Analysis Reporting System data.

US states with drug test results for >80% of fatally injured drivers, 1998-2010.

Drivers killed in single-vehicle crashes on public roads who died at the scene of the crash (n = 16 942).

Drug test results, blood alcohol concentration (BAC), gender, age and day and time of crash.

Overall, 45.1% of fatally injured drivers tested positive for alcohol (39.9% BAC ≥ 0.08) and 25.9% for drugs. The most common drugs present were stimulants (7.2%) and cannabinols (7.1%), followed by 'other' drugs (4.1%), multiple drugs (4.1%), narcotics (2.1%) and depressants (1.5%). Drug-involved crashes occurred with relative uniformity throughout the day while alcohol-involved crashes were more common at night (P < 0.01). The odds of testing positive for drugs varied depending upon drug class, driver characteristics, time of day and the presence of alcohol.

Fatal single-vehicle crashes involving drugs are less common than those involving alcohol and the characteristics of drug-involved crashes differ, depending upon drug class and whether alcohol is present. Concerns about drug-impaired driving should not detract from the current law enforcement focus on alcohol-impaired driving.