Did Access to the We Can Do This App Reduce Methamphetamine Use in the Trial?
Offering access to an app did not produce a demonstrated reduction in methamphetamine use in the We Can Do This trial. Reilly and colleagues randomized 210 Aboriginal and Torres Strait Islander participants to immediate app access or a waitlist. Low engagement and incomplete follow-up are central to understanding what that result does and does not establish.
The randomized comparison
Participants were aged 16 or older and reported at least weekly methamphetamine use in the preceding three months. The study assigned 115 participants to app access and 95 to the waitlist. Both groups could access a website containing harm-minimization information.
The primary outcome was days of methamphetamine use in the preceding four weeks, assessed at one, two and three months after randomization. Other outcomes included dependence severity, psychological distress, help-seeking and days out of role.
The trial did not demonstrate a group benefit
The authors reported no significant between-group differences in methamphetamine-use days at the follow-up points. They likewise did not demonstrate benefits on the listed secondary outcomes. This is the result of the randomized access comparison; favorable opinions about an app cannot replace those outcome findings.
The result should not be rewritten as proof that the groups were exactly equivalent or that all digital interventions are ineffective. The trial tested this intervention and delivery arrangement in its sampled population, with uncertainty and implementation limits.
Access was not sustained engagement
The paper reported poor adherence, with only a small proportion completing the app intervention. Forgetting, personal issues, motivation and practical access problems were among the barriers explored. Randomization offered an opportunity to use the program; it did not ensure that participants completed its therapeutic content.
A comparison restricted to people who completed the app would answer a different question. Completion was not randomly assigned, and completers could differ in motivation, stability or other factors related to outcomes. Selecting them after randomization does not preserve the original randomized comparison.
Follow-up and self-report remain important
Follow-up fell from 63% at one month to 54% at three months. Missing assessments can affect estimates, especially if nonrespondents differ in drug use or engagement. The study also relied on self-reported use without biological verification.
The anonymous format was intended to facilitate participation, but the investigators acknowledged that it limited verification of eligibility and reported information. These limitations qualify causal interpretation of the observed access comparison and confidence in selected subgroup analyses.
What can responsibly be concluded?
The study did not demonstrate that offering this self-directed app reduced methamphetamine use under the tested conditions. It also documented substantial implementation difficulties. It does not show how the same content would perform with consistently engaged participants or additional support, and it does not establish a treatment effect for an individual.
Source
Reilly R, McKetin R, Barzi F, and colleagues. Web-Based Application for Reducing Methamphetamine Use Among Aboriginal and Torres Strait Islander People: Randomized Waitlist Controlled Trial. Journal of Medical Internet Research. 2025;27:e58341. DOI: 10.2196/58341. Supplied PDF: PMC11909485.



