A school-based prevention program increased cannabis knowledge for 2 years but had inconclusive effects on actual cannabis use | Research

The takeaway

A cluster RCT of 2,190 Australian students found that both the universal Climate program and the combined Climate and Preventure (CAP) program significantly increased cannabis-related knowledge for up to 24 months, but evidence was inconclusive regarding whether they reduced cannabis use or harms over 3 years.

School-based prevention program developers; education policymakers; public health researchers evaluating drug prevention strategies.

Cannabis knowledge increased significantly for 2 years, but use reduction was inconclusive

What the researchers found

Researchers randomized 26 Australian high schools (2,190 students, mean age 13.3) to four conditions: universal prevention (Climate), selective prevention for high-risk students (Preventure), combined (CAP), or health education as usual.

Both Climate and CAP groups showed significantly greater increases in cannabis-related knowledge compared to controls (p < 0.001), with higher knowledge maintained at 6, 12, and 24 months.

There was no significant difference between Climate and CAP groups, suggesting the combined approach did not add knowledge benefit beyond the universal program alone.

No significant differences were detected between intervention and control groups on cannabis use or cannabis-related harms. However, the prevalence of these outcomes was lower than anticipated, which may have limited statistical power.

Bayesian analyses were inconclusive: there was insufficient evidence to conclude either that the interventions had no effect or that they had a meaningfully large effect on cannabis use.

Why it matters

Increasing knowledge is an important first step in prevention, but knowledge alone does not always change behavior. This study shows that well-designed school programs can sustain cannabis knowledge gains for years, while the behavioral impact remains uncertain during early adolescence.

The numbers in context

2,190 students from 26 schools. Knowledge significantly higher in Climate and CAP groups at 6, 12, and 24 months (p < 0.001). No significant differences in cannabis use or harms. Cannabis use prevalence lower than anticipated, limiting power.

How the study worked

Cluster RCT. 26 Australian high schools, 2,190 students (mean age 13.3). Four conditions: Climate, Preventure, CAP, control. Assessments at baseline, 6, 12, 24, and 36 months. Multilevel mixed linear models. Bayesian sensitivity analyses.

What this study cannot tell us

Underpowered for cannabis use outcomes due to lower-than-expected prevalence. 3-year follow-up may be insufficient to see behavioral effects in a population starting at age 13. Australian context may not generalize to other countries. No blinding of students or teachers to condition.

How to read the evidence

Moderate. Well-designed cluster RCT with long follow-up, but underpowered for behavioral outcomes and lack of significant use reduction limits practical conclusions.

When this study was published

Published in 2018. The Climate Schools program has continued to be evaluated and adapted in multiple countries.

The bigger picture

School-based drug prevention faces the fundamental challenge that knowledge does not automatically translate to behavior change. The longer-term follow-up may reveal whether knowledge gains eventually protect against cannabis use as these students transition into early adulthood when use rates typically increase.

Questions still open

Will knowledge gains translate to reduced use as students enter peak cannabis initiation years? Would the program work differently in populations with higher baseline cannabis use? Is selective prevention (Preventure) more effective than universal approaches for behavioral outcomes?

Common questions

Does drug education in schools actually work?

This study shows it can effectively increase knowledge for up to 2 years. Whether that knowledge prevents cannabis use is less clear. The most effective prevention programs tend to combine knowledge with skills training, social influence approaches, and in this case, selective targeting of high-risk students.

Why did not the program reduce cannabis use?

Several factors may explain this: the students were young (13) with low baseline cannabis use, making it hard to detect prevention effects. The behavioral impact may emerge later when cannabis use typically increases. The Bayesian analysis could not rule out either no effect or a meaningful effect, indicating uncertainty rather than failure.

Read the original research

Universal cannabis outcomes from the Climate and Preventure (CAP) study: a cluster randomised controlled trial.

Substance abuse treatment, prevention, and policy, 13(1), 34

View on PubMed ↗Publisher’s paper ↗

Citation

Newton, Nicola C; Teesson, Maree; Mather, Marius; Champion, Katrina E; Barrett, Emma L; Stapinski, Lexine; Carragher, Natacha; Kelly, Erin; Conrod, Patricia J; Slade, Tim. (2018). Universal cannabis outcomes from the Climate and Preventure (CAP) study: a cluster randomised controlled trial.. Substance abuse treatment, prevention, and policy, 13(1), 34. https://doi.org/10.1186/s13011-018-0171-4