Identifying Subtypes of Cannabis Use Disorder: A Latent Class Analysis Among Daily Cannabis Consumers and Associated Sociodemographic and Cannabis Characteristics.

Struble C.; Alschuler D.; Wall M.; Hasin D.; Borodovsky J.; Aharonovich E.; Livne O.; Habib M.; Budney A. · Journal of addiction medicine · 2026

**Background & Methods** This cross-sectional study recruited 4,140 US daily cannabis consumers (aged ≥18 years) via social media (February-April 2022) to complete an online survey assessing sociodemographic characteristics, cannabis use behaviors, and DSM-5 Cannabis Use Disorder (CUD) criteria. Latent class analysis (LCA) was employed to identify distinct subtypes of CUD based on endorsement patterns of diagnostic criteria, with subsequent bivariate testing to examine differences in sociodemographic and cannabis use correlates across identified classes. **Key Findings** • Five distinct CUD subtypes were identified among daily cannabis consumers: (1) Minimal Problems (30%; 0-1 criteria, no CUD), (2) Common Problems (45%; 2-3 criteria, mild CUD), (3) Social, Physical, and Emotional Problems (6.6%; 2-5 criteria with specific psychosocial impacts), (4) Physical Dependence Problems (16%; 4+ criteria emphasizing tolerance/withdrawal), and (5) Severe Problems (3.1%; 6+ criteria, severe CUD). • Sociodemographic differences between classes were statistically significant for age and reasons for cannabis use (P≤0.001), with cannabis use frequency and quantity primarily distinguishing the Minimal Problems class from other subtypes. • The majority of daily consumers (75%) endorsed at least mild CUD criteria, and intervention recommendations were stratified by subtype severity, including behavioral strategies for limiting daytime use and coping with cravings, psychologically-tailored interventions for moderate CUD, and comprehensive treatment targeting multiple symptom domains for severe CUD. **Dosage & Administration** Not reported. **Safety & Adverse Effects** Not reported. **Evidence Quality** This study provides a descriptive classification of CUD subtypes with limited causal inference capacity. Key limitations include: social media-based recruitment introducing selection bias toward digitally-engaged individuals; cross-sectional design precluding temporal relationships; self-reported data subject to recall and social desirability bias; and lack of clinical validation in specialty treatment settings. The study represents exploratory evidence (Level 4) establishing CUD heterogeneity among daily consumers but requires prospective validation and clinical utility assessment before implementation in treatment algorithms.

More recent cannabis research