Motor vehicle accidents are the leading cause of death among youths and young adults aged 16-25 years in the United States (1). The prevalence of drinking and driving among high school students aged 16-19 years has declined by 54%, from 22.3% in 1991 to 10.3% in 2011 (2). However, the prevalence of weekend nighttime driving under the influence of marijuana (based on biochemical assays) among drivers aged ≥16 years has increased by 48%, from 8.6% in 2007 to 12.6% in 2013-2014 (3). Use of marijuana alone and in combination with alcohol has been shown to impair driving abilities (4-9). This report provides the most recent self-reported national estimates of driving under the influence of alcohol, marijuana, and alcohol and marijuana combined among persons aged 16-25 years, using data from the Substance Abuse and Mental Health Services Administration (SAMHSA) National Survey on Drug Use and Health (NSDUH) from 2002-2014. Prevalence data on driving under the influence of both substances were examined for two age groups (16-20 years and 21-25 years) and by sex and race/ethnicity. During 2002-2014, the prevalence of driving under the influence of alcohol alone significantly declined by 59% among persons aged 16-20 years (from 16.2% in 2002 to 6.6% in 2014; p<0.001) and 38% among persons 21-25 years (from 29.1% in 2002 to 18.1% in 2014; p<0.001). In addition, the prevalence of driving under the influence of alcohol and marijuana combined significantly declined by 39%, from 2.3% in 2002 to 1.4% in 2014 (p<0.001) among persons aged 16-20 years and from 3.1% in 2002 to 1.9% in 2014 (p<0.001) among persons aged 21-25 years. The prevalence of driving under the influence of marijuana alone declined 18%, from 3.8% in 2002 to 3.1% in 2014 (p = 0.05) only among persons aged 16-20 years. Effective public safety interventions,* such as minimum legal drinking age laws, prohibition of driving with any alcohol level >0 for persons aged <21 years, targeted mass media campaigns, roadside testing (e.g., sobriety checkpoints), and graduated driver licensing programs (10) have contributed to the decline in driving under the influence of alcohol in this population. These or similar interventions might be useful to prevent driving under the influence of other substances, such as marijuana alone or combined with other substances.