Perhaps our most important finding was that substantial numbers of America’s high school seniors continue to put themselves and others at risk for harm. More than a quarter (28%) reported driving under the influence or riding in a vehicle with a driver who had used drugs or alcohol in just the past 2 weeks. The 2011 figure of 28% was down significantly from 2001, when it was 32%. Alcohol use also decreased in this interval, and that may account for the decline in driving after substance use and riding with drivers who had used alcohol or drugs. Driving or riding after marijuana use was slightly lower in 2011 (23%) than in 2001 (25%), but this behavior increased in each of the last 3 years of the study period. Similarly, driving after smoking marijuana increased in each of the last 3 years in our data, from 10% in 2008 to 12% in 2011. This increase is particularly concerning, in light of evidence that marijuana has been implicated in dangerous driving.11,22–24 A recent meta-analysis of 9 studies that met criteria suggested that driving under the influence of marijuana was associated with a significantly increased risk of a motor vehicle crash.15
The data in Tables 2 and and33 show that these risky behaviors are quite pervasive, occurring in all sociodemographic groups. Another indication that the behaviors are pervasive is that the intraclass correlations by school for driving after using alcohol or drugs were all between 2% and 3% (for 2011); in other words, 97% or more of the variation in the behaviors was within schools, and 3% or less was between schools, indicating that this behavior was not found in only a few schools.
Male students were much more likely than female students to report driving after smoking marijuana or drinking heavily, but we observed no significant differences by gender in driving or riding after use of marijuana or other illicit drugs or after a driver smoked marijuana, used other illicit drugs, or consumed 5 or more alcoholic drinks. This presumably is attributable to the likelihood that young women ride as passengers in male-driven vehicles. Both genders are putting themselves at a similar risk of harm.
Driving after marijuana use or heavy drinking did not vary much by levels of SEP, as indicated by parental education, but we found some slight curvilinear association between SEP and driving or riding after substance use: students with low and high SEPs had lower rates. The relatively few students (7%) who reported living with neither a father nor a mother were significantly more likely to report all 4 behaviors.
After we controlled for all other demographic and lifestyle factors, African American students were significantly more likely to report driving after smoking marijuana. The factors that revealed this association were religious commitment and miles driven. African Americans reported higher-than-average religious commitment and lower mileage; adjustment for these variables produced a significant association with driving after marijuana use.
Although demographic factors generally did not relate strongly to driving under the influence of drugs, lifestyle factors certainly did. Students with strong religious commitment and good grades were much less likely than average to drive after using drugs or drinking alcohol. Students who engaged in more than an average amount of truancy, spent more evenings out for fun and recreation, worked more hours per week, or drove more miles were all more likely than average to report driving after drugs or alcohol.