Development and Evaluation of ‘High Alert’, A Brief Smartphone Intervention to Reduce Cannabis-Impaired Driving Among Canadian Youth
Abstract
Young drivers in Canada report the highest rates of cannabis use and impaired driving while expressing the least concern toward driving under the influence of cannabis (DUIC). Since cannabis increases crash risk, preventing youth from DUIC remains a top public health priority. This dissertation aimed to apply the six-step Intervention Mapping framework to develop, implement, and evaluate a brief smartphone intervention called High Alert, designed to reduce DUIC among high-risk youth in Ontario. In Study 1, the first four steps of Intervention Mapping were applied to systematically develop High Alert while integrating evidence, theory, and user feedback. This process involved conducting a needs assessment (Step 1), identifying intervention objectives (Step 2), selecting theory-based methods and practical strategies (Step 3), and creating intervention components and materials (Step 4). The program was pre-tested with five youth and five adult cannabis educators, and their feedback informed refinements before pilot testing. Study 2 applied Step 5 to pilot test High Alert’s implementation and preliminary efficacy through a three-arm pilot randomized controlled trial involving 102 Ontario youth with recent DUIC (≥3 instances in the past 3 months). Findings reveal that High Alert participants exhibited the greatest mean reductions in DUIC compared to both the active and passive control groups. These reductions were statistically significant when comparing all DUIC instances to the passive control, but not for the active control. No significant differences were observed in risky cannabis use. These findings suggest High Alert’s potential to reduce DUIC, warranting further investigation. Finally, Study 3 evaluated High Alert using Step 6, incorporating formative, process, outcome, and acceptability evaluations. While High Alert significantly increased DUIC knowledge and expectations of reducing DUIC compared to the active control, and received better engagement and acceptability scores, no significant differences emerged in other behavioral determinants. Process evaluations highlighted challenges in recruitment, retention, adherence, and measurement, offering insights to inform future trials. Overall, this dissertation highlights High Alert’s potential as a youth DUIC intervention and provides crucial guidance for advancing DUIC research to improve youth road safety.