The Canadian Underage Substance Use Prevention Trial

Aim:

To evaluate the effectiveness and implementation of the PreVenture program on reducing substance-related problems and substance use behaviours in adolescence when delivered in real-world school settings using a train-the-trainer (TtT) model, with and without implementation facilitation.

Questions:

This hybrid Type II effectiveness-implementation trial examines whether a personality-targeted prevention program, delivered through a TtT model, reduces substance-related problems and substance use behaviours among high-risk adolescents in routine school settings. In the effectiveness [E] part, the effects of PreVenture will be compared with usual school-based prevention curricula. PreVenture is delivered through a TtT implementation model with or without implementation facilitation [IF], consisting of ongoing coaching, supervision, and web-based support. The [IF] package is designed to support long-term sustainability of PreVenture after a community accesses PreVenture training.

[E] questions: Does the PreVenture intervention, when delivered through a TtT model (with or without IF), reduce the severity of substance-related problems among high-risk secondary school students compared with treatment as usual (primary)? Does the intervention reduce substance-related problems, binge drinking, cannabis use, illicit drug use, and non-medical prescription drug use at approximately 12 and 24 months post-baseline (secondary)? Does the intervention demonstrate sustained effects across available follow-up assessments (secondary)?

[IF] questions: Does the addition of the implementation facilitation [IF] package improve implementation outcomes (feasibility, reach, fidelity, sustainability) compared with the standard PreVenture TtT model?

Design:

Three sites (Nova Scotia, Ontario, British Columbia) participated in the trial. Each site recruited up to 9-10 high schools, randomised to one of three intervention conditions. Twenty-nine schools were randomised in total. Schools that withdrew prior to the baseline survey were replaced with another school assuming the same randomised intervention condition.

1. Treatment-As-Usual (TAU) control: standard curriculum and mental health care.
2. TtT PreVenture delivered using a train-the-trainer model (PTtT): Locally certified trainers provide structured PreVenture training to school-based professionals, who subsequently deliver the intervention to students.
3. TtT PreVenture with Implementation Facilitation (PTtT+IF): In addition to PTtT, schools receive enhanced implementation support, including: a) involvement of youth in promoting the program, b) ongoing coaching and supervision sessions for group facilitators, and c) access to easy-to-use performance metrics.

Each school screened all consenting Grade 10 students using the Substance Use Risk Profile Scale (SURPS). A small number of schools also screened Grade 9 students as part of local implementation decisions. Students were enrolled across three successive cohorts. Cohort 1 completed follow-up assessments at approximately 12 and 24 months post-baseline; Cohort 2 completed a single follow-up at approximately 12 months post-baseline; Cohort 3 contributed baseline screening and implementation data only and did not complete post-baseline effectiveness assessments.

The primary analysis population includes high-risk students - defined as scoring at least 0.9 standard deviations above school-based norms on one or more SURPS subscales - from Cohorts 1 and 2 attending randomised schools that contributed at least one post-baseline assessment. Students will be analysed according to their school's randomised assignment, regardless of intervention participation or level of follow-up completion.

The PreVenture program is a personality-targeted prevention program designed to help youth understand their target personality trait and develop adaptive coping strategies using motivational and cognitive restructuring techniques. Four personality-specific motivational pathways to substance misuse are targeted: hopelessness, anxiety sensitivity, impulsivity, and sensation seeking. PreVenture involves two 90-minute group sessions, with one week separating each session. Sessions use structured manuals incorporating psychoeducational, motivational enhancement, and cognitive-behavioural therapy components.

Main Hypotheses:

[E] Primary hypothesis: Among adolescents identified as high risk based on SURPS personality profiles, assignment to PreVenture delivered via a TtT model (with or without IF) will reduce the probability of screening positive on the CRAFFT tool (score ≥2) at follow-up compared with assignment to treatment as usual.

[E] Secondary hypotheses:

1. Assignment to PreVenture (PTtT or PTtT+IF) will reduce substance-use behaviours at follow-up, including binge drinking, cannabis use, illicit drug use, and non-medical use of prescription drugs, compared with treatment as usual.
2. Intervention effects on substance-related problem severity and substance-use behaviours will be sustained across available follow-up assessments.

[IF] Outcomes:

1. Fidelity: Schools randomised to receive IF will demonstrate higher implementation fidelity during delivery of PreVenture sessions compared with schools receiving PreVenture without IF.
2. Sustainability: Schools in the IF condition will be more likely to continue delivering PreVenture beyond the initial implementation period.
3. Reach: Schools in the IF condition will deliver both PreVenture sessions to a greater proportion of eligible high-risk students compared with schools without IF.
4. Uptake of IF components: Among IF schools, engagement with individual IF components (youth promotion activities, ongoing coaching and supervision, and use of performance metric tools) will be described to characterise interest in and utilisation of the different implementation support strategies.
5. Student feedback: Students in the IF condition will report higher perceived alignment between session content and core PreVenture learning goals, reflecting higher-quality implementation.

Data analyses strategy for [E]:

Primary and secondary effectiveness analyses will include all students identified as high risk at baseline using the SURPS who attended randomised schools contributing at least one post-baseline cluster-level assessment, and who contributed at least one post-baseline individual outcome assessment. Students will be analysed according to the randomised assignment of their school, regardless of intervention participation or level of follow-up completion. For effectiveness analyses, the two intervention arms (PTtT and PTtT+IF) will be pooled into a single arm, as the content of the PreVenture intervention is identical across delivery conditions.

The primary effectiveness outcome is CRAFFT-positive screening, defined as a CRAFFT score ≥2. The primary estimand is the average treatment effect of assignment to PreVenture across all post-baseline assessments, estimated using mixed-effects logistic regression models pooling all available follow-up waves. Models will include a school-level random intercept, an individual-level random intercept (dropped if variance is estimated at zero), and fixed effects for treatment condition, wave, cohort, baseline CRAFFT score, and additional baseline covariates (age, sex, province, SURPS subscale scores). Results will be reported as adjusted odds ratios with 95% confidence intervals, alongside marginal predicted probabilities and risk differences estimated via marginal standardisation. The primary hypothesis will be evaluated at a two-sided α = 0.05. Given the small number of contributing clusters, the Kenward-Roger correction will be applied to adjust degrees of freedom and standard errors.

Secondary behavioural outcomes - including binge drinking, cannabis use, illicit drug use, and non-medical prescription drug use - will be analysed using the same model structure. Binary outcomes will use mixed-effects logistic regression; ordinal frequency outcomes will use proportional odds mixed-effects regression.

Secondary analyses will examine: (1) wave-specific treatment effects in Cohort 1 to assess whether intervention effects are sustained over time; and (2) complier average causal effect (CACE) analyses using school-level randomisation as an instrumental variable for session attendance. Secondary analyses are exploratory and no multiplicity adjustment will be applied.

Missing outcome data will be handled using full maximum likelihood estimation under the missing at random (MAR) assumption, incorporating all available repeated measurements without requiring complete cases. A pre-specified inverse probability weighting (IPW) sensitivity analysis will assess the influence of attrition on study conclusions by upweighting students with observed outcomes whose baseline characteristics resemble those lost to follow-up.

Data analysis strategy for [IF]:

Analyses of implementation outcomes will be descriptive and will focus on fidelity, reach, uptake of IF components, sustainability, and student feedback across intervention arms. Implementation outcomes will be summarised for schools randomised to PTtT+IF and compared descriptively with schools receiving PTtT alone. These analyses will draw primarily on data from Cohorts 2 and 3. The trial was not powered for formal inferential testing of implementation outcomes.

The following implementation outcomes will be assessed:

1. Fidelity and adherence: quality of PreVenture delivery will be assessed using the PreVenture Intervention Fidelity and Adherence (PIFA) scale, summarised by arm and cohort, and reported separately by facilitator type where available.
2. Reach: the proportion of eligible high-risk students who received both PreVenture sessions will be calculated per school and summarised by arm and cohort.
3. Uptake of IF components: the proportion of IF schools engaging in youth-promotion activities, coaching and supervision sessions, and use of performance-metric tools will be reported.
4. Sustainability: the proportion of schools continuing to deliver PreVenture in Cohort 3 will be reported by arm and by whether delivery was school-led or externally supported.
5. Student feedback: descriptive summaries of student responses to open-ended and precoded questions about intervention content, likes, dislikes, and learning objectives will be reported by arm and cohort.

Structured reflections completed by research staff will be summarised narratively to contextualise quantitative implementation findings and characterise variation in delivery conditions across provinces and cohorts.