Background
Motor vehicle crashes are the leading cause of death among teenagers in the United States of America (U.S.), resulting in the deaths of approximately 2,400 teens each year (CDC, 2023). These crashes represent about one-third of all fatalities in this age group, highlighting the urgent need for enhanced safety protocols and educational programs to reduce these preventable deaths (Yellman, 2020). In the United States, teen driving safety is addressed through a combination of educational programs, regulatory measures, and community initiatives, reflecting a deep-seated concern for the well-being of young drivers (NHTSA, 2021). In the U.S the Graduated Driver Licensing (GDL) system is designed to reduce risk by gradually introducing driving privileges to new drivers. (CDC, 2022). Typically, the program starts with a supervised learner’s phase, followed by an intermediate licence that limits high-risk activities (e.g., nighttime driving, carrying peer teenage passengers), culminating in full licensure once these foundational safe driving practices are well-established (CDC, 2022).
There are several important differences between teenage drivers in the United States and Australasia. In Virginia and most U.S. states, teenagers can begin supervised driving as early as 15 or 16 years, with provisional or unsupervised driving often permitted by 16 or 17 years. This is considerably younger than in most parts of Australasia, where the minimum age for unsupervised driving is typically 17 years, and 18 years in Victoria.
The legal drinking age in the United States is 21 years, meaning that all alcohol consumption by teenage drivers is both underage and illegal. In this study, questions about alcohol use and driving were handled confidentially and responses were de-identified to encourage honesty, given the sensitive and unlawful nature of underage drinking. By comparison, the legal drinking age in Australasia is 18 years and coincides with the minimum age for unsupervised driving.
Blood alcohol concentration (BAC) limits also differ. In Virginia, there is a zero-tolerance BAC policy for drivers under 21, meaning any detectable alcohol (0.02 grams per decilitre or above) is considered a violation. For fully licensed adult drivers, the limit is 0.08. Whereas in Australia, the BAC limit is zero for learner and probationary licence holders, and 0.05 for fully licensed drivers.
Parental involvement is emphasised as a decisive influencer in these driving safety programs. Initiatives like the CDC’s “Parents Are the Key” campaign equip parents and carers with the tools and knowledge to guide and monitor their teens’ driving habits effectively (CDC, 2022). This proactive involvement is supported by research that underscores the significant influence parents and carers have over their teens’ driving behaviours, particularly in the initial stages of learning to drive (CDC, 2022; Curry et al., 2015).
Moreover, the rise of mobile technology has introduced new challenges such as distracted driving, a significant concern among teen drivers (Delgado et al., 2016; Madden & Lenhart, 2009; Ortega et al., 2021). To combat this, educational campaigns and federal initiatives, including legislation from the Bipartisan Infrastructure Law, have been implemented to integrate distracted driving awareness into driver’s licence exams and public safety messaging (NHTSA, n.d.; US Government, 2024). These efforts aim to make teens aware of the dangers of texting and using smartphones while driving (US Department of Transportation, 2015).
Historically, adolescents have displayed varied attitudes toward drink driving, influenced by social norms, peer pressures, and individual risk perceptions (Carter et al., 2014). Studies consistently show that targeted educational interventions can shift these attitudes, highlighting the dangers of drink driving and promoting safer behaviours (National Academies of Sciences et al., 2018). For example, research indicates that teens often underestimate their impairment by alcohol, a misperception that can lead to higher risk-taking when driving on the road (François et al., 2017). Addressing these attitudes through programs like IMPACT is vital, as they help reshape understanding and behaviour regarding the intersection of alcohol consumption and driving.
The United States adopts a comprehensive strategy for teen driver safety that integrates educational efforts, strict regulations, and technology to foster a safe driving culture. This holistic approach mitigates immediate driving risks and also encourages lifelong safe driving habits, maintaining its importance in national efforts to reduce traffic-related injuries and fatalities among teens.
The conceptual framework of this study is based on behavioural prediction models, which identify attitudes as fundamental factors that shape and determine behaviour. Although intentions and willingness are considered more proximal indicators of behaviour and behaviour change, attitudes represent the fundamental cognitive structures that shape these factors (Ajzen, 1991). Numerous studies have demonstrated that altering attitudes is often a prerequisite for influencing more immediate behavioural outcomes. For instance, interventions designed to change adolescents’ attitudes toward drinking and driving may serve as a critical first step in modifying their driving behaviours over time (Ajzen, 1991; Hingson et al., 2002). Therefore, while attitude changes are not direct indicators of behaviour change, they are a necessary precursor that can eventually lead to modifications in intentions and behaviours.
This study evaluates adolescent driving history, focusing on four key categories: attitudes towards alcohol, attitudes towards alcohol-related driving risks, attitudes towards driving a short distance after drinking, and driving knowledge. It assesses demographic factors (e.g., age, gender), driving licence status, car crash experiences, seat belt use, and the influence of parents and peers on driving behaviours.
Methods
Study design
A cross-sectional study evaluated data collected from 2019 to 2023 on a cohort of 1,234 matched records, who were recruited from various high schools in Richmond, Virginia, as part of the IMPACT Program. Both pre- and post-intervention surveys were collected prospectively from participants, and responses were matched at the individual level to allow for within-subject analysis of changes over time.
Target population and recruitment
The study targeted adolescents aged 14-19 years residing in the Richmond area. Participants were recruited from local high schools using a convenience sampling method. Schools were responsible for informing parents and carers about the event and obtaining permission for their child’s participation, with the option for students to opt-out if desired.
The program coordinator provided comprehensive details about Program IMPACT to the school administration of selected high schools in Richmond, Virginia. This included the objectives of the program, planned activities, and the importance of student participation in promoting safer driving behaviours. Following this, school administrators disseminated the information to students and families. Communication channels included school assemblies, official school communication platforms (i.e., emails, newsletters), and classroom announcements. To facilitate informed participation, the school administration distributed consent forms provided by the program coordinator to parents and guardians through the school’s digital platforms. These forms detailed the study’s procedures, the voluntary nature of participation, and the confidentiality measures in place. While individual written consent was not collected from students, participation was authorised through the established agreement between the program and the school administration, which governed the engagement of students in the survey activities. Once school-level agreement was finalised, teachers provided students with a secure survey link during class periods. Teachers played a critical role in guiding students on accessing and completing the surveys responsibly, thereby ensuring a controlled environment for data collection and maximising the quality and reliability of the data gathered. Importantly, both pre- and post-intervention surveys were administered directly to students using this process.
The pre-intervention survey was distributed one week before the program, and the post-intervention survey was distributed immediately after the program, with a one-week window for completion to capture the immediate impacts of the intervention. Survey responses were matched at the individual level using unique participant identifiers. Throughout the recruitment and data collection phases, the program coordinator remained actively engaged, addressing any questions from school staff, students, or parents. Regular check-ins with school administrators and teachers were conducted to monitor progress, maintain adherence to the data collection schedule, and resolve any emerging issues promptly.
Program description (Intervention)
The Program IMPACT utilised a comprehensive educational framework designed to engage high school students in understanding the severe consequences of risky behaviours, especially those linked to driving. This 90-minute intervention strategically integrated interactive educational sessions with live demonstrations, facilitated by a multidisciplinary team of professionals that consisted of the local Fire Department, Emergency Medical Services (EMS), local law enforcement, trauma survivors, and trauma centre nurses from our institute. This approach ensured that students received a well-rounded perspective on the immediate and long-term impacts of unsafe driving behaviours.
The program began with an EMS presentation that illustrated the critical roles of first responders during emergency situations, providing students with a realistic perspective on the immediate medical responses to crashes caused by risky driving behaviours. This was complemented by trauma experience simulations, where students witnessed realistic scenarios of emergency responses to trauma, enhancing their grasp of the physical and immediate dangers of such actions. These activities engaged students and provided hands-on learning experiences that underlined the key messages of the program.
The Program IMPACT utilises interactive components such as live demonstrations and trauma simulations to engage students and provide them with a hands-on understanding of the consequences of risky driving behaviours. These demonstrations simulate real-life emergency situations, such as car crash rescues by first responders, to enhance students’ awareness of the immediate impacts of unsafe driving. These interactive demonstrations are designed to be age-appropriate and are tailored to high school students’ developmental stage. Research has shown that such real-life simulations can significantly improve retention of information and encourage attitude shifts, particularly when students are actively engaged in the learning process (LaVelle & McLaughlin, 2008; Wolfe et al., 2009). The objective was to create an impactful educational experience that emphasised the dangers of distracted and impaired driving without inducing fear or anxiety in the participants.
Further deepening the educational impact, the program included focused discussions on substance use and distracted driving. These sessions highlighted the dire consequences of driving under the influence of alcohol or drugs and the dangers associated with texting or other distractions while driving. Adding a forensic angle, the program also covered the judicial consequences of risky behaviours through a presentation from the Commonwealth Attorney. This segment educated students on the legal repercussions and the long-term effects of their actions, reinforced by a forensic presentation on evidence collection.
Data collection
Data were collected using a self-report survey method from the IMPACT event-based intervention program in Richmond, Virginia. The data are captured through Research Electronic Data Capture (REDCap), a secure web application designed specifically for building and managing online surveys and databases (REDCap, 2013). Participants completed the survey in classrooms via cell phones or school computers. The survey duration was approximately 30 minutes. The survey included various sections: Demographic Information (4 Items), Driving History (7 Items: Binary responses), Driving Behavioural Questions (5 Items: Likert scale, 1=Never, 2=Rarely, 3=Sometimes, 4=Most of the time, 5=Always), Alcohol Attitudes Scale (7 Items: Likert scale, 1=Disagree strongly, 2=Disagree somewhat, 3=Uncertain, 4=Agree somewhat, 5=Agree strongly), Drinking and Driving Scale (12 Items: Likert scale, 1=Disagree, 2=Disagree somewhat, 3=Unsure, 4=Agree somewhat, 5=Agree), and Alcohol-Related Driving Risk Assessment Scale (15 Items: Likert scale, 1=Very Unlikely, 2=Unlikely, 3=Somewhat Likely, 4=Likely). Additionally, there was a section on Driving Knowledge consisting of 9 Items with a mix of multiple-choice and binary questions, scored out of 9 (Namoos et al., 2025).
Survey administration timeline
The pre-intervention survey was distributed one week prior to the intervention day, giving participants adequate time to complete it. This approach ensured that we captured baseline data of the participants’ attitudes and behaviours before they were exposed to the intervention. Following the intervention, the post-intervention survey was distributed in class, and participants were given another week to fill it out. This allowed us to assess the immediate effects of the intervention while providing participants sufficient time to reflect on their experiences and provide thoughtful responses.
Data cleanup
Data cleaning was conducted using R programming (Appelhans et al., 2015), incorporating several essential steps to ensure the accuracy and reliability of the dataset from the IMPACT program. Missing or incomplete data points were managed through imputation techniques for continuous variables and exclusion for significant cases of categorical missingness, thereby preserving the integrity of the dataset. Variable formats were standardised to ensure consistency across the dataset (e.g., date formats were unified, categorical variables were encoded uniformly). Any inconsistencies or discrepancies between similar variables or across different data entries were corrected. Owing to the scale-based nature of the data, the incidence of outliers was minimal. Therefore, no specific outliers were addressed, as the structured response scales inherently limit extreme values that could potentially skew the results. These comprehensive data cleaning procedures significantly enhanced the dataset’s quality and integrity, providing a robust foundation for subsequent analysis and interpretation.
Data analysis
A descriptive analysis was conducted to summarise key demographic characteristics of the cohort, including gender, race, and age. Frequencies and percentages were calculated for categorical variables to provide a detailed breakdown of the sample composition, while measures of central tendency and variability, such as means, medians, standard deviations, and ranges, were computed for continuous variables.
For comparative analysis, T-tests were performed to evaluate differences in continuous variables across demographic groups, with a confidence interval (CI) of 95 percent to determine the precision of the estimates. Additionally, logistic regression analyses were conducted to calculate odds ratios (ORs) for categorical outcomes, assessing the strength of associations between demographic factors and attitudes towards alcohol and driving behaviours.
Results
Baseline (Pretest) Analysis
Demographic characteristics
The study sample included 1,243 individuals, predominantly aged 15 years (59.7%), with a significant drop in representation as age increases. The gender distribution consisted of females (53.1%), males (42.1%), and a small percentage of teenagers who identified as transgender/non-binary (3.2%), other (1.1%), or missing (0.6%). The racial composition of the sample included White (49.0%), Black/African American (22.5%), Hispanic/Latinx (10.7%), alongside smaller percentages from other racial groups (Table 1).
Adolescent driving behaviours, seat belt usage, and alcohol-related risks (driving history)
Table 2 presents the driving history of the study sample revealed diverse behaviours and attitudes toward driving, seat belt usage, and exposure to alcohol-related risks. The majority of participants did not have a license or learner’s permit, accounting for 59.2 percent of the sample, while 34.7 percent held a learner’s permit and only 5.6 percent reported having a full driver’s license. Most participants were actively engaged in driver’s education classes, with 78.4 percent currently enrolled and 21.6 percent not participating. Car crash experiences were reported by 22.5 percent of participants, whereas 77.5 percent indicated they had never been involved in a crash. Seat belt use was generally high: 72.7 percent reported always wearing a seat belt, 19.2 percent wore it most of the time, and 5.9 percent sometimes. Only a small proportion reported rarely (1.2 percent) or never (0.5 percent) using a seat belt.
When asked about exposure to alcohol-related risks, 92.9 percent of participants stated they had never ridden as a passenger with a driver who had been drinking alcohol. Smaller groups reported one instance (2.7 percent), two or three times (2.3 percent), four or five times (0.2 percent), and six or more times (1.1 percent) in the past 30 days. Regarding their own driving after drinking in the last 30 days, 53.6 percent reported not driving at all, 45.9 percent reported zero instances of drinking and driving, and only a very small number reported one time (0.2 percent), two or three times (0.2 percent), or six or more times (0.1 percent).
Participants also reported on their parents’ driving behaviours. Most participants (92.8 percent) indicated that their parents never engaged in drinking and driving, while 5.5 percent reported it occurred sometimes, 0.9 percent often, and 0.6 percent always. In terms of digital distraction, 68.2 percent reported their parents never drove distracted, while 31.6 percent observed their parents driving distracted.
Peer driving behaviors were similarly assessed. The vast majority of participants (95.7 percent) stated their friends never drank and drove, while 3.5 percent reported this happened occasionally, 0.5 percent frequently, and 0.4 percent constantly. Regarding digital distraction among peers, 79.8 percent reported their friends did not drive distracted, whereas 20.2 percent indicated that their friends did.
Pre and Post Analysis
Changes in attitudes towards alcohol post-intervention
Participants showed a decreased likelihood of agreeing that alcohol contributes to happiness or improves mood (OR 1.42, 95 percent CI 1.18 to 1.70). There was also a shift in attitudes regarding alcohol’s impact on social interactions (OR 2.02, 95 percent CI 1.60 to 2.55). The perception that alcohol improves sexual experiences also decreased (OR 1.41, 95 percent CI 1.12 to 1.77). The belief that alcohol aids cognition and coordination was largely unchanged (OR 1.03, 95 percent CI 0.87 to 1.22). The intervention resulted in a decrease in the belief that alcohol impairs thinking and coordination (OR 1.41, 95 percent CI 1.10 to 1.82). The perception that alcohol makes one feel stronger or more powerful showed no significant change (OR 1.03, 95 percent CI 0.88 to 1.22). Finally, there was a decline in viewing alcohol as a coping mechanism for relaxation and distraction from problems (OR 1.40, 95 percent CI 1.09 to 1.79).
Changes in attitudes towards alcohol-related driving risks post-intervention
Following the intervention, there was a significant reduction in acceptance of driving after consuming one drink with a meal (OR 1.97, 95 percent CI 1.62 to 2.40). Attitudes also shifted regarding the acceptability of being the least drunk driver among a group (OR 2.03, 95 percent CI 1.63 to 2.52). Participants reported significantly decreased agreement about several alcohol-affected driving scenarios, including driving with a blood-alcohol content within legal limits (OR 3.13, 95 percent CI 2.38 to 4.12), drinking and driving if all passengers were wearing seat belts (OR 2.06, 95 percent CI 1.62 to 2.62), and the belief that short trips justify drinking and driving (OR 2.09, 95 percent CI 1.65 to 2.66). Regarding scenarios involving driving alone, participants expressed stronger disapproval after the intervention (OR 1.98, 95 percent CI 1.53 to 2.57). Acceptance of impaired driving in emergency situations also showed a notable decline (OR 3.01, 95 percent CI 2.16 to 4.19). Participants were less likely to agree with driving after feeling sober following a few drinks (OR 2.01, 95 percent CI 1.56 to 2.60) or with the idea that drinking and driving is more acceptable during the daytime (OR 2.04, 95 percent CI 1.59 to 2.63). Increased disagreement was also seen regarding driving after drinking if participants were not alcoholics (OR 2.06, 95 percent CI 1.63 to 2.62), and acceptance of drinking and driving when no other transport option was available decreased (OR 2.09, 95 percent CI 1.65 to 2.66). The belief that it is permissible to drink and drive as a passenger also saw a substantial decline (OR 2.12, 95 percent CI 1.67 to 2.70).
Changes in attitudes towards driving a short distance after drinking post-intervention
The intervention led to a significant reduction in the likelihood of accepting driving a short distance after drinking. The odds of disagreeing increased with the suggested number of drinks: after one drink (OR 2.48, 95 percent CI 1.87 to 3.28), after two drinks (OR 2.86, 95 percent CI 2.10 to 3.88), after three to four drinks (OR 3.64, 95 percent CI 2.71 to 4.90), and after five to six drinks (OR 2.36, 95 percent CI 1.69 to 3.29). However, after six or more drinks, no significant change was observed (OR 1.13, 95 percent CI 0.85 to 1.49).
Changes in driving knowledge post-intervention
The results showed a statistically significant improvement in knowledge (p = 0.005). The mean score before the intervention was 7.594, compared to 7.707 after the intervention, indicating a positive effect on the participants’ understanding of the subject matter.
Discussion
The IMPACT program demonstrated significant effects on altering the attitudes and self-reported behaviours of adolescent drivers regarding alcohol consumption and driving. The intervention notably reduced participants’ acceptance of drinking and driving, with increased recognition of the potential dangers associated with such behaviours. The program’s emphasis on hands-on learning, real-life simulations, and interactive educational sessions may have been particularly effective in shifting perspectives among this age group, aligning with findings from similar educational interventions (Caetano & McGrath, 2005; LaBrie et al., 2007).
Changing attitudes is considered an essential first step in behaviour modification, consistent with established behavioural prediction models such as the Theory of Planned Behaviour (Ajzen, 1991). Although attitudes alone may not directly predict behaviour, shifts in attitudes often precede changes in intentions and eventually actual behaviours. Future studies could benefit from directly measuring behavioural intentions and willingness to enhance predictive validity regarding behaviour changes.
The findings reflect a clear shift in participants’ perceptions of alcohol-related driving risks. For instance, the observed decrease in acceptance of driving even after minimal alcohol consumption (one or two drinks) indicates that adolescents are increasingly aware of the dangers involved, which is consistent with research suggesting that targeted educational programs can significantly alter risk perceptions (Schwartz, 2008). Given that alcohol-related driving risks are more pronounced among younger drivers who often underestimate their impairment (Hingson et al., 2002), these results highlight the potential of preventive education to mitigate such risks early on. Future studies could benefit from directly measuring behavioural intentions and willingness to enhance predictive validity regarding behaviour changes. The observed attitude shifts could translate into long-term changes in behaviour, contributing to a broader cultural change in how young drivers view the acceptability of driving under the influence.
Furthermore, the study identified changes in attitudes towards alcohol’s broader social effects, such as reduced agreement with the idea that alcohol improves mood or enhances social interactions. These results underscore the importance of addressing misconceptions about alcohol, which is critical given that many adolescents report social motivations for drinking (Sjödin et al., 2021). By countering these beliefs, the IMPACT program not only addressed driving-related risks but also tackled underlying attitudes that contribute to alcohol use, a strategy supported by existing literature on adolescent risk behaviours (National Academies of Sciences, 2019).
While the study demonstrated positive shifts in attitudes and knowledge, certain beliefs remained resistant to change. For example, attitudes towards driving after consuming more than six drinks did not show significant alteration, possibly due to a pre-existing baseline of disapproval. This may indicate that participants already viewed this behaviour as particularly risky, consistent with literature suggesting that extreme behaviours like heavy drinking and driving are less socially acceptable among adolescents compared to more moderate drinking (Bonnie et al., 2004). Future programs could consider addressing nuanced scenarios and misconceptions about alcohol impairment at lower levels to reinforce understanding of the effects of alcohol across a range of consumption levels.
The program’s holistic approach, which combines educational events with interactive demonstrations, appears to be effective in increasing awareness among adolescent drivers about the dangers of impaired driving, as indicated by shifts in attitudes post-intervention. Given the high rates of motor vehicle fatalities among teenagers, particularly due to alcohol-related incidents, the findings from the IMPACT program suggest that multifaceted educational initiatives may play a role in promoting safer driving attitudes. These findings support the need for ongoing public health campaigns that address both the immediate and long-term consequences of risky driving behaviours. The IMPACT program has set a foundation for future interventions that aim to further shift cultural perceptions and reduce risky driving behaviours among adolescents.
Strengths and limitations
A major strength of this study was robust sample size of over 1,200 participants. Such a large cohort enhances the reliability of findings and provides a strong foundation for future research and public health initiatives.
This evaluation of the IMPACT program demonstrated several strengths that contribute to its effectiveness. Its multifaceted educational approach integrates interactive components, real-life simulations, and presentations from emergency responders and legal experts. This engaging format captured participants’ attention and contributed to a deeper understanding of the consequences of impaired driving. Additionally, the program’s comprehensive focus on both alcohol-related perceptions and broader social attitudes toward drinking addresses underlying beliefs contributing to risky behaviours. Significant positive attitude shifts observed among participants demonstrate the program’s meaningful potential to influence behaviour, aligning with previous findings on interactive educational interventions.
While the program incorporates interactive simulations, these demonstrations are intended to educate adolescents through realistic, age-appropriate scenarios, rather than to induce fear. Existing literature supports the educational efficacy of simulations, highlighting their ability to enhance engagement and learning without necessarily causing negative emotional reactions (Carter et al., 2014). However, as noted, we did not explicitly measure whether participants experienced fear or message rejection, which constitutes an acknowledged limitation. Although positive attitude shifts were observed, some participants might have rejected messages due to the simulations’ intense nature. Future research should explicitly measure message acceptance versus rejection, consistent with models like the Extended Parallel Process Model (EPPM) (Witte, 1992), to better evaluate potential unintended emotional responses that could affect program effectiveness.
Additional limitations include reliance on self-reported data, potentially introducing biases such as social desirability, where participants may underreport risky behaviours or overstate positive attitudes toward safe driving. Moreover, convenience sampling from specific Richmond, Virginia schools limit the generalisability of findings. The short follow-up period raises questions about the sustainability of observed attitude changes over time, highlighting a need for longitudinal studies to determine if positive shifts translate into lasting behaviour change and reductions in driving incidents. Lastly, while the program addressed alcohol-related driving risks, it did not fully explore other influential factors, such as peer pressure and social norms, which play critical roles in shaping adolescent driving behaviours.
Future research
Future research should explore the long-term retention of these attitude changes and behaviours and examine whether the observed shifts translate into reduced rates of impaired driving and crashes over time. Additionally, while the current study provided insights into the effectiveness of the IMPACT program, further investigation into the influence of parental and peer behaviours on adolescent driving attitudes could offer a more comprehensive understanding of the factors shaping young drivers’ decisions. This study contributes to the growing body of evidence supporting the effectiveness of targeted, interactive educational programs in shaping safer driving behaviours among adolescent populations, ultimately aiming to reduce the burden of alcohol related.
AI tools
AI tools were not used in this study nor in the preparation of this paper.
Acknowledgements
This study was supported by the Department of Surgery’s Injury and Violence Prevention Program at the School of Medicine, Virginia Commonwealth University, and conducted at the Medical Center West Hospital. We gratefully acknowledge the financial support from the Virginia Department of Motor Vehicles through a series of grants. Funding was provided under Grant #FM60T-2021-51374-21374 for fiscal year 2021, Grant #FDL*DE-2022-52047-22047 for fiscal year 2022, Grant #FDL*DE-2023-53163-23163 for fiscal year 2023, and Grant #BFDL*DE-2024-54138-24138 for fiscal year 2024. We appreciate the continued support and resources provided by these grants, which have been instrumental in enabling our comprehensive investigation into the pertinent issues addressed by this study.
Author contributions
Asmaa Namoos, MD, MPH, PhD: Conceptualised and designed the study, coordinated research, oversaw data analysis, interpreted findings, and prepared the final manuscript. Corresponding author. Nicholas Thomson, PhD: Contributed to study design, methodology, data interpretation, and manuscript revisions. Developed the theoretical framework for risky driving behaviour. Jerry Van Harris, MSW: Managed Program IMPACT implementation, data collection, and contributed insights from a social work perspective to the manuscript. Carol Olson, MA, LPC: Supported educational intervention, facilitated trauma-informed discussions, and assisted with manuscript refinement. Michel Aboutanos, MPH, MD: Provided clinical oversight, contributed to health outcome interpretation, and reviewed the manuscript critically.
Funding
Virginia Department of Motor Vehicles through a series of grants. Funding was provided under Grant #FM60T-2021-51374-21374 for fiscal year 2021, Grant #FDL*DE-2022-52047-22047 for fiscal year 2022, Grant #FDL*DE-2023-53163-23163 for fiscal year 2023, and Grant #BFDL*DE-2024-54138-24138 for fiscal year 2024.
Ethics approval
This study does not require IRB ethical approval as it is conducted under the scope of a public health program, in alignment with 45 CFR 164.512 (i). Since the data used are fully de-identified, according to the standards outlined in 45 CFR 164.514, the information no longer qualifies as protected health information (PHI) and does not require IRB review. Additionally, as a public health surveillance activity conducted or supported by a public health authority, this study qualifies for exemption from IRB review under 45 CFR 46.102 (l) (2).
Data availability
The data will be available upon request.
Conflicts of interest
The authors declare there are no conflicts of interest’.