An Online School Program Improved Alcohol and Cannabis Knowledge but Only Changed Alcohol Behavior | Research

The takeaway

An internet-based school prevention program significantly improved alcohol and cannabis knowledge and reduced alcohol use and intentions, but had no effect on cannabis use or cannabis intentions, in 1,103 Australian students.

Educators and policymakers interested in evidence-based school prevention programs.

Cannabis knowledge improved (d=0.72) but cannabis behavior was unchanged

What the researchers found

Researchers conducted a cluster randomized controlled trial of the Climate Schools: Alcohol and Cannabis course, an internet-based prevention program, among 1,103 students (average age 13.25) from 13 Australian schools.

Immediately after the intervention, students in the program showed significantly greater alcohol knowledge (effect size d=0.67) and cannabis knowledge (d=0.72) compared to controls. They were also less likely to have consumed any alcohol in the past 6 months (OR=0.69) and less likely to intend to use alcohol in the future (OR=0.62).

However, there were no significant effects on binge drinking, cannabis use, or intentions to use cannabis. This cross-validation trial replicated the knowledge gains from the original trial but showed mixed behavioral results.

Why it matters

Internet-based prevention programs are scalable and can reach large numbers of students. This trial confirmed that the program effectively increases knowledge, but the disconnect between knowledge gains and behavioral change for cannabis highlights the challenge of translating education into prevention.

The numbers in context

1,103 students from 13 schools. Age: mean 13.25 years. Alcohol knowledge: d=0.67. Cannabis knowledge: d=0.72. Reduced any alcohol use: OR=0.69. Reduced alcohol intentions: OR=0.62. No effects on binge drinking, cannabis use, or cannabis intentions.

How the study worked

Cluster randomized controlled trial with 6 intervention schools and 7 control schools. Students (N=1,103, mean age 13.25) completed surveys at baseline and immediately post-intervention. Mixed-effects regressions were used to account for school-level clustering.

What this study cannot tell us

Only immediate post-intervention effects were assessed; longer-term follow-up is needed. The program was tested in Australia and may not generalize to other countries. The lack of effect on cannabis behavior may reflect the younger age of participants (many had not yet initiated cannabis use). School-level clustering limits effective sample size.

How to read the evidence

This is a cluster RCT providing moderate evidence for knowledge effects but no behavioral change for cannabis, with only short-term follow-up.

When this study was published

Published in 2016. The Climate Schools program has continued to be evaluated with longer follow-up periods.

The bigger picture

The finding that a program can effectively increase knowledge without changing behavior is common in prevention science. It underscores that knowledge alone is often insufficient to change substance use behavior, and that programs may need additional components (skills training, normative feedback) to affect cannabis use specifically.

Questions still open

Would longer-term follow-up show delayed effects on cannabis behavior? What program components are needed to change cannabis use behavior beyond knowledge?

Common questions

Does drug education actually prevent drug use?

This study found that education improved knowledge significantly but did not change cannabis behavior. Research consistently shows that knowledge-based approaches alone are often insufficient, and more comprehensive programs that include skills training and normative feedback tend to be more effective.

Why did the program work for alcohol but not cannabis?

Several possibilities exist: alcohol use was more common at this age, so there was more behavior to change; the content may have been more compelling for alcohol; or cannabis use among 13-year-olds may have been too low to detect effects. Longer follow-up is needed to assess delayed cannabis effects.

Read the original research

A cross-validation trial of an Internet-based prevention program for alcohol and cannabis: Preliminary results from a cluster randomised controlled trial.

The Australian and New Zealand journal of psychiatry, 50(1), 64-73

View on PubMed ↗Publisher’s paper ↗

Citation

Champion, Katrina E; Newton, Nicola C; Stapinski, Lexine; Slade, Tim; Barrett, Emma L; Teesson, Maree. (2016). A cross-validation trial of an Internet-based prevention program for alcohol and cannabis: Preliminary results from a cluster randomised controlled trial.. The Australian and New Zealand journal of psychiatry, 50(1), 64-73. https://doi.org/10.1177/0004867415577435