This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.
Early initiation of alcohol and cannabis use markedly increases the risk of harms associated with use, including the development of substance use and mental health disorders. To interrupt this trajectory, effective prevention during the adolescent period is critical. Despite evidence showing that parents can play a critical role in delaying substance use initiation, the majority of prevention programs focus on adolescents only. Accordingly, the
This paper outlines the protocol for a cluster randomized controlled trial (RCT) of the CSP program, a novel internet-based program for parents and students to prevent adolescent substance use and related harms. The CSP program builds on the success of the
A cluster RCT is being conducted with year 8 students (aged 12-14 years) and their parents from 12 Australian secondary schools between 2018 and 2020. Using blocked randomization, schools are assigned to one of the two groups to receive either the CSP program (intervention) or health education as usual (control). The primary outcomes of the trial will be any student alcohol use (≥1 standard alcoholic drink/s) and any student drinking to excess (≥5 standard alcoholic drinks). Secondary outcomes will include alcohol- and cannabis-related knowledge, alcohol use-related harms, frequency of alcohol consumption, frequency of drinking to excess, student cannabis use, parents’ self-efficacy to stop their children using alcohol, parental supply of alcohol, and parent-adolescent communication. All students and their parents will complete assessments on three occasions—baseline and 12 and 24 months postbaseline. In addition, students and parents in the intervention group will be asked to complete program evaluations on two occasions—immediately following the year 8 program and immediately following the year 9 program.
Analyses will be conducted using multilevel, mixed-effects models within an intention-to-treat framework. It is expected that students in the intervention group will have less uptake and excessive use of alcohol compared with the students in the control group.
This study will provide the first evaluation of a combined internet-based program for students and their parents to prevent alcohol and cannabis use.
Australian New Zealand Clinical Trials Registry ACTRN12618000153213; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=374178 (Archived by WebCite at http://www.webcitation.org/71E0prqfQ)
RR1-10.2196/10849
Alcohol and cannabis are the most commonly used licit and illicit drugs in Australia and are associated with substantial socioeconomic costs [
Despite the importance of including parents in prevention efforts, relatively few substance use prevention programs have involved both students and parents, and the programs developed have faced numerous challenges during their implementation (eg, high attrition rates, lack of engagement, and lack of sustainability [
The course is an internet-based universal prevention program delivered to all students regardless of their level of risk and is based on a social influence approach to prevention [
The parent component for the CSP program is based on the successful Dutch Prevention of Alcohol Use in Students (PAS) program [
Currently, we are seeking to evaluate the effectiveness of the CSP program in preventing alcohol and cannabis use and increasing parents’ self-efficacy to prevent their child from using these substances. Incorporating an internet-based parent component into an effective school-based student program has the potential to remarkably enhance prevention outcomes and reduce alcohol- and cannabis-related harms among adolescents.
We will determine the effectiveness of the CSP intervention by running a cluster RCT in Australian Independent and Catholic secondary schools (ACTRN12618000153213). Cluster randomization will be used to avoid contamination of the control group with the intervention groups through student and staff communication. Schools will be randomly allocated to the CSP condition or the “control” condition. Students and parents in the CSP condition will be provided access to the
The
This study is set in Independent and Catholic secondary schools in Australia. We obtained Human Research Ethics Committee (HREC) approval to approach Independent schools (HREC 17852) and Catholic schools in two dioceses in the Greater Sydney area (reference: 201731 and 060318) and invite them to participate. In October 2017, 156 schools were approached using a variety of methods, including advertisement during the Personal Development Health and Physical Education (PDHPE) Teachers’ Association Annual Conference, emails sent directly to schools or school principals, newsletter entries (ie, the Independent Secondary Schools Association and
For cluster randomization, sample size calculations were based on sample size requirements developed by Heo and Leon [
Eligible participants are students attending participating schools and are enrolled in year 8 in 2018 and these students’ parents. These students will be 12-14 years of age at baseline and 14-16 years of age at the final assessment point. Furthermore, to be eligible to participate, students and parents must have at least intermittent internet access and basic proficiency in English.
After school principals agree to participate in the study, active consent will be sought individually from parents, students, and teachers. Parents will be asked to provide active consent for (1) their child’s participation in the study and (2) their own participation. Eligible students with parental consent will be directed to the participant information statement and consent form when registering on the CSP website, and all parents will similarly be directed to their participant information statements and consent forms when registering for the first time. Conversely, students and parents in the CSP group who do not consent to participate in the research trial will still be offered access to the content of the program; however, they will not be prompted to complete assessment surveys, and no data will be collected from those individuals.
The student component of the intervention consists of the effective and validated
Screenshot of example student cartoon. Source: Netfront Pty Ltd.
The parent component was developed in consultation with parents of Australian secondary school students and relevant experts through a large scoping survey, beta-testing, and pilot-testing of the developed program [
Screenshot of the year 8 webinar. Source: Netfront Pty Ltd.
Screenshot of the rule-ranking component (when top 3 rules have been finalized). Source: Netfront Pty Ltd.
Parents will also have access to 6 brief internet-based modules (under 10 minutes each, 4 in year 8 and 2 in year 9) covering a range of topics about alcohol and cannabis use, as well as parenting strategies and communication tips (
Students and their parents will follow the same registration and consent process in both the control group and the CSP group and will be asked to complete the baseline questionnaire. Then, students will receive their regular drug and alcohol education during their PDHPE classes (ie, they will cover topics such as drug use, health and well-being, sources of support, personal safety, exploring risk, etc, in accordance with the student outcomes defined by the Australian National Health and Physical Education syllabus) during the study. These students will be asked to complete the 12-month follow-up questionnaire in 2019 and the final follow-up questionnaire in 2020. Similarly, parents in the control group will be asked to complete the baseline questionnaire and 12- and 24-month follow-up questionnaires over the same period.
Students and parents in the CSP group will be invited to register for the CSP program online during the first year of the study (for the majority of schools, this will be during term 1 of 2018; however, for some, it will be term 2 or 3 of 2018). Students will complete their registration and baseline survey during their regularly scheduled PDHPE classes, and their teachers will facilitate their progression through the
Upon conclusion of the trial, all schools in the control group will receive access to the same materials offered to the CSP group, free of charge.
Module 1: Prevalence, patterns, and harms of adolescent alcohol use
Module 2: Parental attitudes and rule-setting
Module 3: Parental supply and use
Module 4: Communication and involvement
Module 5: Prevalence, patterns, and harms of adolescent cannabis use
Module 6: What parents can do to prevent adolescent cannabis use
Screenshot of one of the parent modules (Parent Module 4, in the year 8 program). Source: Netfront Pty Ltd.
We selected the following primary outcomes to reflect the overall target of the intervention (ie, to prevent alcohol use and related harms in adolescents). All primary outcomes will be measured at baseline and 12- (prior to the delivery of the year 9 intervention) and 24-month follow-up, with the primary endpoint being 24 months.
Any alcohol use is defined as the consumption of at least 1 full serve or standard drink of alcohol (ie, any drink containing 10 grams of alcohol). To measure this outcome, students would be provided with a chart used to illustrate a standard drink (as used in the NHMRC Australian Guidelines [
Any drinking to excess is defined as the consumption of ≥5 standard drinks on a single occasion, in line with the NHMRC Australian Guidelines for risky drinking on a single occasion [
All secondary outcomes will be assessed at baseline and 12- (prior to the delivery of the year 9 intervention) and 24-month follow-up, except for cannabis-related knowledge and patterns of cannabis use, which will only be measured at 12- and 24-month follow-up.
This outcome will be assessed using the 23-item Rutgers Alcohol Problem Index, which measures the consequences of alcohol use. Students would be asked to report the consequences of their alcohol use over the past 12 months, in which higher scores indicate greater harms [
Data regarding parental self-efficacy will be collected by a 3-item scale measuring parents’ confidence in their ability to prevent their adolescent from drinking alcohol, in which higher scores indicate a greater sense of self-efficacy [
This will be measured by a 2-question scale from the Australian Parental Supply of Alcohol Longitudinal Study (APSALS) [
Theory of Change Logic Model for
The timeline of student and parent or guardian participation in the study.
Timeline | Control | ||||
Parent | Student | Parent | Student | ||
Assessment | Parent survey 1: baseline | Student survey 1: baseline | Parent survey 1: baseline | Student survey 1: baseline | |
Intervention | Parent modules: alcohol | Year 8 student lessons: alcohol | — | — | |
Evaluation | Evaluation of alcohol modules | Evaluation of alcohol lessons | — | — | |
Assessment | Parent survey 2 (12-month follow-up) | Student survey 2 (12-month follow-up) | Parent survey 2 (12-month follow-up) | Student survey 2 (12-month follow-up) | |
Intervention | Parent modules: alcohol and cannabis | Year 9 student lessons: alcohol and cannabis | — | — | |
Evaluation | Evaluation of alcohol and cannabis modules | Evaluation of alcohol and cannabis lessons | — | — | |
Assessment | Parent survey 3 (24-month follow-up) | Student survey 3 (24-month follow-up) | Parent survey 3 (24-month follow-up) | Student survey 3 (24-month follow-up) |
Cannabis and other drug use (tobacco or cigarettes, amphetamines, ecstasy, hallucinogens, sedatives, inhalants, or “other”) will be measured by 5 questions adapted from the National Drug Strategy Household Survey [
This outcome will be measured by a 6-question scale from the APSALS, examining parental knowledge and child disclosure of activities, friends, and whereabouts (eg,
Alcohol-related knowledge will be assessed using a 16-item “Knowledge of Alcohol” scale originally adapted from the School Health and Alcohol Harm Reduction Project (SHAHRP) questionnaire and used in previous trials of the
Parents’ and students’ knowledge about cannabis will be assessed by 16 items of the “Knowledge about Cannabis” scale adapted from the SHAHRP questionnaire, as used in previous trials of the
Students will be asked to report how frequently they consumed alcohol, in terms of standard drinks; this outcome will be measured by questions that were originally adapted from the SHAHRP “Patterns of Alcohol” index and have been used in previous
Students will be asked to report the frequency of drinking to excess (defined as having ≥5 standard drinks on a single occasion). These questions were originally adapted from the SHAHRP “Patterns of Alcohol” index and reflect those used in previous
Additional measures include demographic information, such as gender, age, country of birth, truancy rates, and academic performance (to determine the baseline equivalence of groups), which will be assessed using questions that have been included in previous
Following school consent, schools have been randomly allocated to the CSP or control group by an external researcher; this process involved stratified random allocation, in which schools were divided into three mutually exclusive strata: (1) coeducational (mixed males and females) school, (2) single-sex or predominantly girls school, and (3) single-sex or predominantly boys school. The allocation was random within strata to achieve balance across intervention and control groups with respect to the number of males and females participating in the trial. Randomization was achieved using a randomization table created in StataSE, version 14, using the Stratarand procedure. The randomization sequence was computer-generated by an external biostatistician, who then informed the research team of which group each school had been allocated to.
It is not possible for the research team to remain blinded to the group allocation of schools during the study, given the nature of the intervention and the need to manage schools (particularly teachers as they progress through the program). However, the research team were not involved in the
Students and parents will complete a self-assessment questionnaire online through the CSP website at baseline and 12- and 24-months postbaseline. The questionnaires contain scales used in previous
All measures are self-reported, as it has been found to be the most favored method of assessment for young people and has excellent discriminant [
Internet-based tracking will be used to monitor the extent to which students and parents engage with the intervention (ie, tracking the number of webinar views, the number of parents who ranked the rules, the number of parents who completed each module, the number of students who completed each lesson, etc). While teachers in the intervention group will be asked to complete a logbook as a record of how they delivered the intervention, teachers in the control group will complete a logbook detailing the type of drug and alcohol education they offered to their students over the same period.
Students, parents, and teachers in the CSP group will also be asked to complete two additional questionnaires, asking them to evaluate the intervention content that they received (ie, they will be asked how acceptable, appropriate, and enjoyable they found the program content). Teachers will be asked to indicate the likelihood that they would recommend the program to others, whereas students and parents will be asked to rate the likelihood that they would use the information and skills they learned in their own lives. Students and parents will be invited to complete these evaluation questionnaires once in 2018 after completing the
To facilitate the retention of students and parents in the trial, frequent communication will be maintained through email prompts generated from the CSP website, which reiterates the requirements of the study. Participants will also be offered reimbursement after completing each study questionnaire; specifically, students will be entered into a draw to win one of three Aus $500 JB Hi-Fi or Apple store vouchers at each assessment occasion, and parents will enter the draw to win one of three Aus $500 Westfield or Coles gift cards. Moreover, teachers will receive a one-off reimbursement of Aus $50 for the extra administration involved during the trial. When completing the baseline questionnaire at the beginning of the study, students and parents will be asked to enter their contact details (ie, name, address, home number, mobile number, and secondary email), which will be used solely for contacting students who are absent from school on the day of a questionnaire (absent students will be identified by their teacher, who will inform the research team of the absent students) and inviting them to complete subsequent follow-up assessments. Any data collected from students and parents who have consented but discontinue the study will be used in the analysis, in accordance with the intention-to-treat principle. The chosen statistical analysis techniques consider missing data resulting from participant discontinuation.
In this study, primary and secondary outcomes will be analyzed in longitudinal analyses using multilevel mixed-effects regression models. The effects of highest interest are intervention × time interactions that reflect differences between intervention groups in the growth of each outcome over time. The multilevel modeling can account for the expected correlations between different observations of the same individual and between individuals in the same school [
Mixed-effects logistic regression with a logit link function will be applied when analyzing binary outcomes. A range of potential fixed effects and random effects structures will be compared using likelihood ratio tests and model fit statistics, such as the Akaike information criterion, to determine the best fitting model for each outcome. For all outcomes, between-condition effect sizes (eg, Cohen
The primary aim of this study is to evaluate the efficacy of the CSP program in comparison with the standard health education received by the control group. Therefore, planned comparisons for each outcome will compare students and parents in the CSP condition with students and parents in the control condition at baseline and 12- and 24-month follow-up.
This study will be overseen by an external biostatistician, and any adverse events will be reported to the University of New South Wales Sydney HREC to maintain the integrity of the study including the data collected, trial progress, and ethical compliance. However, given that the intervention reflects normative alcohol and cannabis education provided as part of the PDHPE curriculum, no serious adverse events are anticipated to occur during the study; therefore, a formal steering committee is not required.
Ethics approval was obtained by the University of New South Wales Sydney HREC (HC17852), the Sydney Catholic Education Office (Ref: 201731), and Catholic Education Parramatta (Ref: 060318).
Confidentiality of the collected information will be strictly maintained, and participants’ data will remain anonymous. To access internet-based questionnaires and materials, students and parents will be required to register on the CSP website and once logged in, all data collected will be automatically deidentified, and the database will generate a unique ID code for each participant and the individual’s data files across sessions will be linked with this unique ID code. All data collected will be in a computerized format and stored in password-protected files on university servers, accessible only to the research staff and stored separately to questionnaire data. These procedures mirror those used in previous and existing school-based prevention trials conducted by the research team (eg, The CAP Study, HREC 11274 [
The results of this study will be presented to academic peers at conferences and published in health and education peer-reviewed journals. The feedback will be provided to participating schools in the form of a deidentified report of the study’s findings. This report will also be available to students and their parents at the end of the study. When publishing results of this study, no information will be published on the basis of individual cases, and all published data will reflect group data.
This study is supported by funding from the Australian Government Department of Health and a Society for Mental Health Research Early Career Research Award to NCN. This study was also funded by the NHMRC through the NHMRC Centre of Research Excellence (APP1041129).
This study is funded by the Australian Government Department of Health from 2016 to 2020 and by Society for Mental Health Research Early Career Research from 2015 to 2017. Enrollment of schools began in January 2018, with 8 out of 12 schools enrolled at the time of submission (enrollment is expected to be complete by October 2018). Baseline assessments are currently underway, and the first results are expected to be submitted for publication in 2019.
This paper describes the design and protocol of the CSP study, the first international trial of an integrated internet-based intervention for students and parents to prevent alcohol and cannabis use. The effectiveness of the CSP program will be assessed through a cluster RCT, relative to health education as usual at 12- and 24-months postbaseline. Contamination will be avoided by the use of a cluster RCT design, in which each school forms its own cluster and is allocated to either the intervention or control group, preventing individuals at the same school being allocated to separate conditions and thus preventing contamination between staff and students at each school. We aim to ascertain whether a combined internet-based approach to drug prevention can be effective in preventing alcohol use among adolescents and improving parents’ self-efficacy to prevent their children from using substances.
To date, approaches to substance use prevention have traditionally focused on adolescents themselves, despite evidence suggesting that parents play a critical role in substance use initiation [
The CSP program utilizes cartoon storylines to engage students and maintain their interest and internet-based technology to engage parents and improve ease of access to substance use prevention information, based on a successful evidence-based intervention [
A potential limitation of this study is the use of self-report, which could be subject to the social desirability bias. However, previous research has demonstrated that self-report measures of substance use among adolescents have yielded excellent discriminant [
Attrition is another potential limitation to this study, which could result from students not being present on the day of assessments or not providing correct or complete contact details to allow the research team to link their responses over time. However, missing data are likely to be at random, and the chosen data analytic techniques (mixed-effects regression modeling) use all available data, thus reducing the bias brought about by participant attrition.
The school sample used in this study (ie, Independent and Catholic school types) may limit the generalizability of the findings to the broader population. However, we do not expect this factor to markedly impact the outcomes of the trial, as previous research has found that the consumption and frequency of cannabis use within independent (nongovernment) schools was comparable to the larger population of young people of the same age [
An additional limitation of this study involves the need to obtain active consent from participants, as this might introduce selection bias. We aim to overcome this risk of selection bias with a robust study design. As this is an RCT, both participants in the intervention condition and control condition will volunteer to participate; therefore, the impact of volunteering is likely to be spread evenly across the two conditions. Although active or voluntary consent procedures can introduce selection bias [
The CSP program was developed to address an unmet need for an integrated, internet-based program for students and parents to prevent alcohol and cannabis use. The CSP program fits within the school PDHPE curriculum and overcomes barriers to the implementation through online delivery, making it scalable to meet the needs of students and parents across Australia. If proven to be effective, this comprehensive program could be implemented widely, as part of a national strategy to significantly reduce the burden of disease, social costs, and disability associated with early substance use in adolescents.
Australian Parental Supply of Alcohol Longitudinal Study
Human Research Ethics Committee
National Health and Medical Research Council
Prevention of Alcohol Use in Students
Personal Development Health and Physical Education
randomized controlled trial
School Health and Alcohol Harm Reduction Project
This study is supported by funding from the Australian Government Department of Health and a Society for Mental Health Research Early Career Research Award to NCN. This study was also funded by the NHMRC through the NHMRC Centre of Research Excellence (APP1041129).
MT and NCN are two of the developers on the
Peer-reviewer report – Society for Mental Health Research.