Correlates of Engagement and Associations With Outcomes in a Cannabis Harm-Reduction Mobile App for Youth With First-Episode Psychosis: Exploratory Analysis of the CHAMPS Pilot Randomized Controlled Trial.

Background: Continued cannabis use among young people with first-episode psychosis (FEP) has been linked to poorer clinical and functional outcomes (eg, increased symptom severity and higher relapse rates). Digital harm-reduction interventions may represent a promising, person-centered approach to reduce at-risk cannabis use behaviors in this population. However, evidence remains limited regarding which subgroups are more likely to engage with these interventions and whether specific levels of engagement are required to achieve more favorable cannabis-related outcomes. Objective: This exploratory, hypothesis-generating analysis of the CHAMPS (Cannabis Harm-Reducing App to Manage Practices Safely) pilot randomized controlled trial (RCT) evaluated a cannabis harm-reduction mobile app for youth with FEP in early intervention services (EIS). The objectives of this study were to assess engagement by examining associations between selected sociodemographic factors and module completion and to determine whether achieving specific completion thresholds was associated with improvements in cannabis-related outcomes. Methods: Cannabis-related outcomes (Marijuana Problems Scale [MPS], Protective Behavioral Strategies for Marijuana [PBSM] scores, and days of cannabis use) were self-assessed at baseline and at week 6 (primary end point), week 12, and week 18 (postrandomization). Participants were categorized into low to moderate (0-4 modules) and high (5-6 modules) completion groups, and selected sociodemographic factors were compared between groups using bivariate analyses. Mixed-effects models, adjusted for baseline values and the covariates sex and cannabis use disorder (CUD) status, were fitted to evaluate whether module thresholds were associated with improvements in cannabis-related outcomes. Results: Data from 96 participants, including 46 in the CHAMPS+EIS arm and 50 in the EIS-only arm (1:1 ratio), were analyzed under a modified intention-to-treat principle. Individuals in the high-engagement group reported higher baseline social support and higher educational level (P=.005 and P=.01). No statistically significant associations were observed between specific module completion thresholds and cannabis-related outcomes in adjusted mixed-effects models. Conclusions: Participants with higher social support and higher education were more likely to engage with CHAMPS. Although descriptive analyses suggested a potential gradient of improvement with increasing module completion, no specific completion threshold was robustly associated with improvements in outcomes. Given the multifactorial nature of engagement, supporting subgroups at risk of lower app use and examining metrics beyond module completion may enhance the impact of CHAMPS. These findings are hypothesis-generating, given their exploratory nature, and require replication in a future efficacy trial.