Clinical Takeaway
Adolescents who received a brief web-based intervention alongside standard driver education showed measurable reductions in intentions to drive after using alcohol or cannabis compared to driver education alone. A single 30-minute digital session added to existing curriculum appears to be a practical, low-burden tool for addressing impaired driving risk in teen drivers. These findings support integrating targeted substance use messaging directly into driver education programs during this high-risk developmental window.
#21 Preventing alcohol and cannabis-impaired driving among adolescents: Effects of a web-intervention in a driver education program.
Citation: Osilla Karen Chan et al.. Preventing alcohol and cannabis-impaired driving among adolescents: Effects of a web-intervention in a driver education program.. Journal of substance use and addiction treatment. 2026. PMID: 41850531.
Design: 5 Journal: 0 N: 2 Recency: 3 Pop: 2 Human: 1 Risk: -2
- Preclinical only
Abstract: INTRODUCTION: Adolescence is a critical period during which teens initiate and escalate substance use, as well as begin learning to drive. Limited research has evaluated programs to prevent impaired driving behaviors in this age group. We tested effects of adding a single-session web intervention to existing driver education curriculum. METHODS: Driving school staff recruited participants aged 15.5 to 17 from 12 driver education programs. Participants were randomized to driver education only (usual care, 30 h) or driver education plus a single-session web intervention (webCHAT, ∼30 min). Participants completed surveys at baseline and six months. RESULTS: The sample (N = 198) was 60% female, 80% White, and averaged 15.7 (SD = 0.8) years old. At baseline, 25% and 8% reported past three-month alcohol and cannabis use, respectively, and 19% and 10% reported ever riding with someone under the influence of alcohol or cannabis, respectively. At follow-up, webCHAT participants reported significantly lower perceived peer alcohol (-1.14, 95% CI: -1.915, -0.372, p = 0.004) and cannabis (-1.09, 95% CI: -1.866, -0.306, p = 0.007) use norms compared to usual care participants. Both webCHAT and usual care participants significantly reduced past month alcohol and cannabis use and viewed impaired driving as riskier and less acceptable at follow-up. CONCLUSION: Driver education programs offer a unique opportunity to prevent substance use and impaired driving when adolescents are motivated to participate to secure their driver’s license. These programs should continue updating curricula to reflect the state of science for alcohol and drug prevention. CLINICAL TRIAL REGISTRATION: NCT04959461.
What This Study Teaches Us
Adding a brief online intervention to driver education reduces teens’ perception of peer substance use norms for both alcohol and cannabis. Both the intervention and standard driver education alone led to reduced self-reported past-month substance use and increased risk perception of impaired driving at six-month follow-up.
Why This Matters Clinically
Driver education reaches adolescents at a teachable moment when they are motivated and engaged. Since impaired driving remains a leading cause of teen injury and death, and cannabis use among adolescents is rising, clinicians should know that integrating substance use prevention into existing driver education infrastructure shows promise without requiring major curriculum overhaul.
Study Snapshot
| Study Design | Randomized controlled trial |
| Population | N=198 adolescents aged 15.5-17 from 12 driver education programs; 60% female, 80% White; 25% reported past three-month alcohol use, 8% cannabis use at baseline |
| Intervention | Single-session web intervention (webCHAT, approximately 30 minutes) added to standard driver education (30 hours) versus driver education alone |
| Primary Outcome | Perceived peer norms for alcohol and cannabis use, and attitudes toward impaired driving risk and acceptability |
| Key Result | WebCHAT participants reported significantly lower perceived peer alcohol norms (difference of 1.14 points, p=0.004) and cannabis norms (difference of 1.09 points, p=0.007) compared to usual care at six months |
Where This Paper Deserves Skepticism
The sample size is modest (N=198) and the follow-up period is only six months, which is too brief to assess actual driving behavior or serious outcomes like crashes. The population is predominantly White and recruited from driver education programs, limiting generalizability to more diverse settings. The abstract does not specify whether self-reported substance use changes were clinically meaningful or distinguish between perception change and actual behavior change. The finding that usual care alone also reduced past-month use suggests either a regression-to-the-mean effect or that measurement timing (around licensure) may have influenced results independent of intervention.
Dr. Caplan’s Take
This study rightfully emphasizes that driver education is a unique window to reach motivated adolescents about substance use and driving safety. The modest advantage of webCHAT in shifting perceived norms is interesting, but I’d note that both groups showed similar reductions in self-reported use and risk perception, suggesting the base driver education curriculum itself may be the active ingredient. The six-month horizon and reliance on self-report don’t tell us whether these perception shifts translate to safer driving decisions in real situations. I think the broader takeaway is solid: we should be integrating substance use prevention into driver education, but this study shouldn’t be oversold as proof that a 30-minute online module materially changes adolescent behavior.
Clinical Bottom Line
Driver education programs offer a teachable moment for substance use prevention, and adding brief web-based interventions may modestly improve perceived risk norms. However, both intervention and standard driver education alone showed similar improvements in self-reported use and attitudes, suggesting the value may be in structured attention to these topics rather than any single intervention component.
| |
Have thoughts on this? Share it:
