JMIR Publications

ABSTRACT

Background:

Among young people experiencing a first episode of psychosis (FEP), abstinence-based approaches for comorbid substance use may not align with their personal goals and lived experience. Digital harm reduction interventions could offer a promising alternative, yet few have been tailored to this population; among those developed for other clinical groups, many have faced challenges with uptake and implementation. Moreover, research indicates that successful uptake of these could rely on factors such as individual characteristics, needs, and user engagement.

Objective:

This exploratory pilot study aimed to identify moderators such as baseline participant profiles and engagement factors associated with improved outcomes in CHAMPS (Cannabis Harm-reducing App to Manage Practices Safely) among young people with FEP in a context of Early-Intervention services (EIS), with the goal of informing the optimization of future iterations of CHAMPS

Methods:

Primary outcomes included Marijuana Problems Scale (MPS), Protective Behavioural Strategies for Marijuana (PBSM) scores, and days of cannabis use, measured at baseline, week 6 (primary endpoint), week 12, and week 18. Associations between baseline demographics, engagement (low, moderate, high) and participation status (completers vs non completers) were examined. Mixed-effects models assessed outcome changes at week 6, week 12 and week 18, and tested whether baseline demographics or, within CHAMPS+EIS, module completion thresholds moderated outcomes improvements at week 6.

Results:

Data from 96 participants including 46 in the CHAMPS+EIS arm and 50 from the EIS only arm were analyzed under a modified intention-to-treat principle. Engagement level differed by baseline social support (Kruskal-Wallis, χ²1 (2) =8.1, P=.02); post-hoc analysis indicated that the high engagement group had significantly greater social support than both low (W=103.5, P=.03) and moderate (W=25, P=.02) groups, while lower support was linked to non-completion. Secondary education or more was globally associated with engagement (Fischer’s exact, P=.01) while secondary education or less was linked to non-completion week 6 (Fischer’s exact test, P=.005), week 12: (χ²1= 14.5, P<.001), and week 18 (χ²1= 4.6, P=.03). Higher baseline social support moderated PBSM increase at week 6, no other moderators emerged (F1,68 = 2.3, P <.001). Module-threshold regression was non-significant, though visual trends suggested that completing at least five modules could be associated with more favorable outcomes for MPS and PBSM.

Conclusions:

Although exploratory and requiring replication in a larger trial, these findings highlight social support as a key factor of engagement, suggest CHAMPS performs similarly across a diverse population and indicate a possible dose-response relationship with user engagement for specific cannabis-related outcomes. These exploratory findings will be critical in enhancing our digital harm reduction tool CHAMPS Clinical Trial: This study was registered in ClinicalTrials.gov (NCT04968275)