Long-term impact of online school-based mental health and substance abuse program: A 72-month study in Australia

In a recent study published in The Lancet Digital Health, a group of researchers assessed the long-term effectiveness of four school-based, online interventions for preventing mental health and substance use disorders in adolescents, measured at a 72-month follow-up.

Study: Effectiveness of a universal, school-based, online programme for the prevention of anxiety, depression, and substance misuse among adolescents in Australia: 72-month outcomes from a cluster-randomised controlled trial. Image Credit: Matej Kastelic/Shutterstock.com

Background 

Mental health and substance use disorders cost over the United States (US)$2.4 trillion globally and are expected to double by 2030. Trends show increasing mental health issues among youth, aggravated by coronavirus disease 2019 (COVID-19).

Although alcohol initiation is delayed in younger adolescents worldwide, consumption increases significantly between ages 15 and 18.

Further research is needed to determine prevention efforts’ long-term sustainability and to adapt interventions to evolving adolescent and societal needs.

About the study 

The present study tracked 6,386 adolescents from a four-arm, multicentre, cluster-randomized trial across 71 schools in New South Wales, Queensland, and Western Australia, which started between September 2013 and December 2016.

Schools were randomly assigned to one of four interventions: Climate Schools Combined (CSC), Climate Schools Substance Use (CSSU) alone, Climate Schools Mental Health (CSMH) alone, or standard health education. With their consent, participants were re-contacted for follow-up at 60 and 72 months.

The interventions involved 18 classroom sessions employing social learning and cognitive behavioral therapy principles, focusing on mental health and substance use. CSSU and CSMH interventions focused specifically on substance use and mental health, respectively, with a reduced number of targeted lessons.

Due to high attrition, the 84-month follow-up was canceled, focusing resources on earlier assessments. Follow-up included web-based surveys, with retention efforts including financial incentives and multiple communication strategies.

Primary outcomes assessed were alcohol and cannabis use and symptoms of anxiety and depression, using validated self-report measures.

Analysis utilized multilevel mixed-effect regression models, accounting for the clustered nature of the data, and analyzed by intention to treat to provide insights into the long-term efficacy of these interventions.

Study results 

Between September 2013 and December 2016, the study enrolled 6,386 students from 71 schools spanning New South Wales, Queensland, and Western Australia. Participants were allocated to four groups: 1,556 to standard education, 1,739 to CSSU, 1,594 to CSMH, and 1,497 to CSC.

The study followed up with these participants for up to 72 months post-baseline, focusing on those who had not declined further contact at the 30-month follow-up. Intention-to-treat analysis included all original participants, with a mean age of 13.5 years at baseline, over half of whom were female.

The follow-up assessments at 60 and 72 months revealed varied participation rates across the groups, with the CSC group showing substantial retention. The study observed no adverse events.

In terms of alcohol consumption, while weekly drinking and heavy episodic drinking increased over time among the control group, these increases were significantly slower in the CSC group.

However, there were no notable differences in the probability of weekly drinking or heavy episodic drinking between the CSC group and other groups at the 72-month follow-up.

Similarly, monthly cannabis use increased in the control group but changes over time were not significantly different among the intervention groups.

Mental health outcomes, including depression and anxiety, showed slight increases over time in the control group, with minimal evidence of significant changes between groups throughout the follow-up period.

Sensitivity analyses adjusted for baseline covariates related to outcomes and attrition indicated similar trends, though with increased uncertainty in effect estimates. 

Conclusions 

To summarize, the study assessed the effectiveness of a school-based prevention program targeting mental health and substance use disorders over 72 months among middle school students. It found that the CSC intervention significantly slowed the increase in weekly drinking and heavy episodic drinking.

However, these findings are tempered by baseline differences between the CSC and control groups, which could affect the results.

Sensitivity analyses further diminished the certainty of these outcomes, showing minimal long-term differences between groups for alcohol use disorder, cannabis use, and mental health symptoms.

The study suggests that while such programs can initially mitigate certain behaviors, their long-term effectiveness may require ongoing intervention.