Tracking Why People Use Cannabis Over Time Could Help Personalize Addiction Interventions

Latent transition analysis across four studies identified that transitions into or remaining in latent classes characterized by multiple cannabis use motives predicted adverse outcomes including cannabis use disorder. This pattern held across different data collection frequencies (daily, monthly, yearly, biennial).

Secondary analysis of four studies with varying data collection frequencies: Medical Cannabis Certification Cohort (n=801, biannual), Cannabis Health and Young Adults Project (n=359, annual), Monitoring the Future Panel (n=7,851, biennial), and Text Messaging Study (n=87, daily). Latent transition analysis with random intercepts was applied across all datasets.

Most cannabis interventions are static, offering the same approach regardless of how someone's relationship with cannabis evolves. This study shows that monitoring changes in why people use cannabis -- not just how much -- could identify when someone is shifting toward riskier patterns, enabling interventions to adapt in real time.

Adaptive interventions -- treatments that change based on how a patient is doing -- represent the next frontier in addiction medicine. This study provides the motivational framework needed to trigger those adaptations: when someone starts using cannabis for more reasons, it signals escalating risk.

Secondary analysis of existing datasets that were not specifically designed for this purpose. The four studies used different populations, measures, and timeframes, making direct comparisons challenging. The daily data study had a very small sample (n=87).