The Canadian Underage Substance Use Prevention Trial (CUSP)

University of British Columbia, Center for Addiction and Mental Health, Dalhousie University, Canadian Institutes of Health Research (CIHR)

Despite having made some strides with respect to reducing adolescent drinking rates, illicit substance use and substance use disorders (SUDs) remain significantly above national targets for health promotion and disease prevention in Canada and the United States. Now, more than ever, there is a pressing need for effective substance abuse prevention in Canada, particularly for those most at risk of developing substance use problems including prescription drug misuse. Clearly, new approaches to prevention (with lower numbers needed to treat) are needed and which translate new research on addiction vulnerability to personalised prevention and early intervention. The PreVenture Program involves brief cognitive-behavioural interventions targeting personality traits from a neurocognitive perspective. While the personality-targeted approach has been shown to be effective in reducing most substance use behaviors, it has yet to be evaluated for its impact on uptake of prescription drug misuse in adolescents. The Canadian Underage Substance use Prevention (CUSP) Trial aims to evaluate the long-term effects of a personality-targeted school-based prevention program on delaying the onset of drug and alcohol use in adolescence over three years across Canada. This is a hybrid effectiveness [E] and implementation-facilitation [IF] trial on delaying the onset of drug and alcohol use in adolescence. In the [E] part, the effects of a personalized prevention program will be tested against usual school-based prevention curricula. PreVenture is delivered through a TtT implementation model with or without [IF], e.g. with ongoing supervision and web-based support. The [IF] package is designed to support long-term sustainability of PreVenture after a community accesses PreVenture training.

Aim: To evaluate the long-term effects of the PreVenture program on delaying the onset of drug and alcohol use in adolescence when delivered through a train-the-trainer (TtT) implementation model. Questions: This is a hybrid effectiveness [E] and implementation-facilitation [IF] trial on delaying the onset of drug and alcohol use in adolescence. In the [E] part, the effects of a personalized prevention program will be tested against usual school-based prevention curricula. PreVenture is delivered through a TtT implementation model with or without [IF], e.g. with ongoing 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 personalized intervention delay the onset and reduce the severity of problem substance use among secondary school students when schools are trained through a TtT model (primary)? Does the intervention impact the onset and frequency of binge drinking, cannabis and illicit drug use, and uptake of non-medical prescription drug use at (i) one year and (ii) two years post-intervention (secondary)? What are the factors which impact the delay of onset and reduction of severity? [IF] questions: What are the best conditions under which the program implementation is sustained to a high quality from one year to the next? What are the costs and potential economic benefits of the intervention with and without additional implementation facilitation? Design: Three sites (NS; ON; BC) will participate in the trial. Each site will recruit 9 high schools randomized to one of three intervention conditions: Treatment-As-Usual (TAU) control: standard curriculum and mental health care TtT PreVenture delivered through local trainers (PTtT): Each site will identify two Trainers to be trained to deliver high fidelity PreVenture training to 4 school-based professionals per school assigned to this condition and condition 3 (see below). Each school will identify the 4 staff members to receive training and administer practice sessions using structured manuals with assessment of quality of the interventions. TtT PreVenture with Implementation Facilitation (PTtT+IF): three new components designed to increase the likelihood that schools will continue to implement the program with high quality and satisfaction: a) Involvement of youth in promoting the program, b) ongoing coaching/supervision sessions for group facilitators, c) access to easy-to-use performance metrics. Each school will screen all assenting grade 10 students (N=150 per school) in 2019 (Cohort 2 followed up until 2021), 2020 (Cohort 3 followed up until 2021) and 2021 (Cohort 4 no follow-up), which will result in approximately 4,050 students screened across 27 schools in Canada each year totaling 12,150 students over 3 years. [Cohort 1 is training-only in 2018] All 4050 students will be invited to participate in the trial each year and will be followed annually to facilitate coordination of assessments at the school level and to reduce any stigma to targeted youth. Youth who assent to being randomized to an intervention condition and who score above the personality threshold will be considered the Intent-to-Treat [ITT] sample, regardless of whether or not the school managed to deliver the intervention. The whole grade will be followed to assess secondary outcomes at the whole grade level as reported in Conrod et al, 2013. The PreVenture program is a personality targeted prevention program designed to help youth understand the target personality trait and develop adaptive coping strategies for managing that trait using motivational and cognitive restructuring techniques. Four personality-specific motivational pathways to substance misuse are targeted hopelessness (HOP), anxiety sensitivity (AS), impulsivity (IMP) and sensation seeking (SS). PreVenture involves 2 90-minute group sessions, with one week separating each session. The interventions are conducted using structured manuals that incorporate psycho-educational, motivational enhancement (MET), and cognitive-behavioural (CBT) therapy components, and include real life 'scenarios' shared by local youth with similar personality profiles. Main Hypotheses: [E] primary outcome: PreVenture will result in less severity of substance use problems as measured by the CRAFFT score after 12- and 24-month follow-up (comparison of both arms with PreVenture against treatment as usual using Cohort 2 in all three conditions) [E] secondary outcome: PreVenture will result in lower prevalence of binge drinking, lowered frequency of cannabis use, other illicit drug use and non-medical prescription drug use after 12- and 24-month follow-ups (comparison of both arms with PreVenture against treatment as usual using Cohort 2 in all three conditions) [IF] primary outcome: In the PreVenture group with implementation-facilitation, the rate of successful interventions (less severity on CRAFFT score) after 12-month follow-up is higher than in the PreVenture group without implementation-facilitation (comparison of 12-month results of Cohort 3 between both conditions) [IF] secondary outcome: In the PreVenture group with implementation-facilitation, the rate of interventions in Year 4 after the start of PreVenture interventions is higher than in the PreVenture group without implementation-facilitation (comparison of Cohort 4 between both conditions) Data analyses strategy for [E]: The core data for exploring effectiveness will be in Cohort 2, followed up for 24 months. This allows for the use of growth models, which can model intra-individual change over time. The dependent variables for these models change, may be dichotomous (e.g., onset of substance use: yes/no), ordinal, expressing counts or have interval properties (for example, the CRAFFT scale can be used as an ordinal or count variable - i.e. this poses a number of problems - but has also been used as a scale with interval scale properties). Different scale types require different types of regression analyses to be used in growth modelling (i.e., linear regression, logistic regression, Poisson regression, etc.). For the CRAFFT score, three aspects require consideration: proportion of people with 0 values, i.e., no onset of substance use or no problems left-hand censoring, as values cannot be lower than 0 assumption that the change from 0 to 1 is a structurally different change than the changes involving values greater or equal to 1 The first problem can be dealt with by using specific models for zero inflation, which started with zero-inflated Poisson distributions, but by now has been extended to most forms of regression analyses. Censoring should not be a problem as the investigators are dealing with increasing slopes, and will treat the CRAFFT as a count variable, which by definition has 0 as starting values. As for structural changes between the transition from 0 to 1 vs. other changes, the investigators will explore via the Heckman model, which will model the transition from 0 to 1 separately, allowing for different influence factors. For the primary outcome, the following statistical procedures will be used (ITT analyses): Dependent variable: CRAFFT score Regression type used for growth modelling: zero-inflated Poisson regression Adjustment variables: sex, age, province Relevant statistics: adjusted CRAFFT score after combined 12- and 24-month follow-up of combined PreVenture groups vs. treatment as usual (TAU) group. For the secondary hypotheses, similar procedures will be used, with the specific regression type to be determined by the dependent variable. In addition to ITT analyses, sensitivity analyses with different assumptions on missing values will be conducted. The investigators will also test if the effect of PreVenture is significant in each of the two time points (i.e., not only for the combined time points). Other analyses will explore different influencing factors on the effectiveness of PreVenture and these analyses will be conducted in the same framework, adding different independent variables. The investigators will combine this framework with a latent framework (latent growth analyses); to separate groups with different trajectories, and will model influences on these separate trajectories. Data analysis strategy for [IF]: It is important for an intervention such as PreVenture, which has proven efficacy under certain conditions, to show effectiveness in the real world under less optimal conditions, and to measure the various impacts on effectiveness. Accordingly, one of the experimental conditions systematically varies implementation conditions by introducing three additional components: a) Involvement of local youth in promoting the program, b) ongoing coaching/supervision sessions for group facilitators, c) access to easy-to-use performance metrics. The impact of facilitation will be tested in the following primary implementation hypothesis: Dependent variable: rate of implemented interventions Regression type used for growth modelling: fractional response regression Adjustment variables: sex, age, province Relevant statistics: adjusted rates of implementations at the 12-month follow-up between the two PreVenture groups (with vs. without implementation-facilitation) The alternative hypotheses will be accepted, if the PreVenture group with implementation-facilitation has a higher rate. There will be more analyses to explore the implementation facilitation. One key will be measurement of the fidelity and quality of implementation, and the investigators will be using a specifically developed measure, the PreVenture Intervention Fidelity and Adherence (PIFA) scale, for the latter. Overall, the analyses will include the infrastructure, the number of interventions produced, and the success rates of these interventions. Economic considerations: For (IF), a cost-utility analysis of the two PreVenture conditions to determine the incremental cost-utility ratio per improvements of the Child Health Utility (CHU9D) scale (based on the primary hypothesis in [IF]) will be conducted. Analyses can be widened to include comparisons between TAU and PreVenture conditions. Alternatively, a cost-benefit analysis on the impact of PreVenture from a societal cost perspective could also be conducted. The costs of the implementation would then be compared to cost-savings in resources necessary to cope with the substance use problems, where the latter could either be derived from the literature, and/or from a subsample analysis for health and judicial service utilization.

Prevention, Early intervention, Substance Use and Misuse, Mental Health, Personality-targeted Interventions, Adolescence

Schools randomized to this arm will identify 4 staff members to participate in a 2-day training workshop + 3 hours of supervised practice and will be provided with access to screening and PreVenture intervention materials through the local trainer. Local trainers will deliver 2-day workshops and then supervise school staff in the delivery of two 90-minute group sessions (for at least one personality profile).

Schools randomized this arm will receive the standard PreVenture TtT protocol plus an additional Implementation Facilitation package that will contain 3 new components designed to increase the likelihood that schools will continue to implement the program with high quality and satisfaction: 1) Youth Engagement, 2) ongoing coaching and supervision for Facilitators, and 3) access to easy-to-use performance metrics.

For schools randomized to this arm, students will have usual access to drug and alcohol prevention through the standard curriculum and mental health care provided through student counseling at the participating schools. The schools will be incentivized to participate in the study with the promise of free PreVenture training and materials in subsequent years of the trial. Information on other drug prevention efforts implemented at the school will be collected, but the randomized design should control for any potential differences between intervention conditions on this random factor.

PreVenture TtT model will involve identifying local expert trainers to be trained to train school-based professionals on the PreVenture program. Steps include: PreVenture Facilitator certification: Trainers participate in training workshops and deliver the program to eligible youth in Cohort 1 and are rated on quality of implementation using the PIFA. Reliable use of the PIFA: Each trainer demonstrates convergence with the Montreal Team's ratings on a set of previously-recorded PreVenture sessions. PreVenture Training Workshop Certification: Trainers participate in a one-day TtT workshop to review the main aims and exercises of PreVenture Training. Trainers then record mock TtT sessions in which key components of the PreVenture Training Workshop are delivered to an audience of 4-10 individuals (e.g., class of graduate students). These sessions are rated by the Montreal Team the PreVenture Implementation Fidelity and Adherence-Trainer Scale (PIFA-T).

PreVenture TtT model + Implementation Facilitation Package: Youth advisors identified the role of older youth to support program implementation. Youth will help promote PreVenture to students and families and help Facilitators optimize attendance at sessions. Trainer-led coaching sessions will focus on enhancing program fidelity and acceptability by Facilitators. 'Best practice' supervision elements from efficacy trials will be integrated. Fidelity and symptom monitoring will be conducted in real-time during program implementation. Performance metrics have been developed based on youth qualitative feedback to PreVenture sessions in previous trials. Analysis indicates key responses that are predictive of students most likely to benefit from PreVenture on 12-month substance use outcomes and indicates the extent to which Facilitator's interventions are leaving students with impressions consistent with positive long-term outcomes.

Compare change in severity of substance use problems as measured by CRAFFT score of both arms of PreVenture against TAU using Cohort 2 in all three conditions. The CRAFFT Screening Test is a short clinical assessment tool designed to screen for substance-related risks and problems in adolescents. CRAFFT stands for the key words of the 6 items in the second section of the assessment - Car, Relax, Alone, Forget, Friends, Trouble. CRAFFT employs a skip pattern: participants whose Part A score is "0" (no use) answer the Car question only of Part B, while those who report any use in Part A also answer Part B's five CRAFFT questions. Each "yes" answer is scored as "1" point and a CRAFFT total score of two or higher identifies "high risk" for a substance use disorder.

Compare change in severity of substance use problems as measured by CRAFFT score after the start of PreVenture intervention using Cohort 3 between both conditions. The CRAFFT Screening Test is a short clinical assessment tool designed to screen for substance-related risks and problems in adolescents. CRAFFT stands for the key words of the 6 items in the second section of the assessment - Car, Relax, Alone, Forget, Friends, Trouble. CRAFFT employs a skip pattern: participants whose Part A score is "0" (no use) answer the Car question only of Part B, while those who report any use in Part A also answer Part B's five CRAFFT questions. Each "yes" answer is scored as "1" point and a CRAFFT total score of two or higher identifies "high risk" for a substance use disorder.

Compare change in prevalence of binge drinking of both arms with PreVenture against TAU using Cohort 2 in all three conditions. Change in frequency of binge drinking assessed by a sub-item on the Alcohol Use Disorders Identification Test (AUDIT). Items asks "How often do you have five or more drinks on one occasion?" (5 or more drinks = binge drinking) and the options are [options: 1=Never/2=Less than monthly/3=Monthly/4=Weekly/5=Daily or almost daily]. Higher numbers indicate higher rates of dependence.

Compare change in frequency of cannabis use of both arms with PreVenture against TAU using Cohort 2 in all three conditions. Change in frequency of cannabis use assessed by asking: "have you ever used cannabis in the last 12 months?" [options: yes/no]. If a student answers "yes", then asked: "how often used in the last 12 months?" [options: 1=monthly or less/ 2=2-4 times a months/ 3=2-3 times per week/ 4=4-6 times per week/ 5=everyday]. Reporting "yes" to 12 month use and reporting more frequent use indicates higher rates of use.

Compare change in frequency of non-medical prescription drug (NMPD) use of both arms with PreVenture against TAU using Cohort 2 in all three conditions. Change in frequency of NMPD use assessed by asking: "have you ever used prescription drugs for non-medical use (without a doctors prescription) in the last 12 months?" [options: yes/no]. If a student answers "yes", then asked: "how often used in the last 12 months?" [options: 1=monthly or less/ 2=2-4 times a months/ 3=2-3 times per week/ 4=4-6 times per week/ 5=everyday]. Reporting "yes" to 12 month use and reporting more frequent use indicates higher rates of use.

Compare change in frequency of illicit substance use of both arms with PreVenture against TAU using Cohort 2 in all three conditions. Change in frequency of other illicit substance use assessed by asking: "have you ever used the following substances in the last 12 months?" [options: yes/no]. Substances: Solvents, Cocaine, methylenedioxymethamphetamine (MDMA), Methamphetamine, Heroin, Fentanyl, Psilocybin, lysergic acid diethylamide (LSD), Steroids, Phencyclidine (PCP). If a student answers "yes", then asked: "how often did you use this substance in the last 12 months?" for each substance [options: 1=monthly or less/ 2=2-4 times a months/ 3=2-3 times per week/ 4=4-6 times per week/ 5=everyday]. Reporting "yes" to 12 month use and reporting more frequent use indicates higher rates of use.

Compare change in the rate of interventions in Year 4 after the start of PreVenture intervention in Cohort 4 between both conditions. Rate of the intervention will be measured by assessing the number high risk students delivered the intervention.

Inclusion Criteria: Survey: All year 10 students with either active parental consent or youth assent (depending on local site requirements) will be eligible to participate in the survey PreVenture: Eligibility will include assenting to study protocol and scoring one or more Standard Deviation (SD) above the school's norm on the 23-item Substance Use Risk Profile Scale (SURPS). The SURPS is a 23-item validated tool that assesses four personality dimensions along the internalizing (hopelessness (HOP) and anxiety sensitivity (AS)) and externalizing (impulsivity (IMP) and sensation seeking (SS)) spectrum. It has been shown to predict specific substance use patterns, including cannabis use and non-medical use of prescription drugs. Of those screened in a given year, 1350 (45%) will report elevated personality profiles and will be eligible for intervention. Exclusion Criteria: - Students that do not have either active parental consent or youth assent (depending on local site requirements) will not be eligible to participate in the survey or PreVenture

Conrod PJ. Personality-Targeted Interventions for Substance Use and Misuse. Curr Addict Rep. 2016;3(4):426-436. doi: 10.1007/s40429-016-0127-6. Epub 2016 Nov 4.

Woicik PA, Stewart SH, Pihl RO, Conrod PJ. The Substance Use Risk Profile Scale: a scale measuring traits linked to reinforcement-specific substance use profiles. Addict Behav. 2009 Dec;34(12):1042-55. doi: 10.1016/j.addbeh.2009.07.001. Epub 2009 Jul 8.

Conrod PJ, O'Leary-Barrett M, Newton N, Topper L, Castellanos-Ryan N, Mackie C, Girard A. Effectiveness of a selective, personality-targeted prevention program for adolescent alcohol use and misuse: a cluster randomized controlled trial. JAMA Psychiatry. 2013 Mar;70(3):334-42. doi: 10.1001/jamapsychiatry.2013.651.

O'Leary-Barrett M, Mackie CJ, Castellanos-Ryan N, Al-Khudhairy N, Conrod PJ. Personality-targeted interventions delay uptake of drinking and decrease risk of alcohol-related problems when delivered by teachers. J Am Acad Child Adolesc Psychiatry. 2010 Sep;49(9):954-963.e1. doi: 10.1016/j.jaac.2010.04.011. Epub 2010 Jul 31.