Fecha de publicación: -- Fuente: PubMed "Cannabis" Addict Behav. 2026 Sep 21;184:108867. doi: 10.1016/j.addbeh.2026.108867. Online ahead of print.ABSTRACTBACKGROUND: The Diagnostic and Statistical Manual of Mental Disorders - Fifth Edition (DSM-5) introduced cannabis use disorder (CUD) severity indicators along a unidimensional spectrum. Limited research has explored how individuals compare in severity and risk for related problems (e.g., psychiatric symptoms) across CUD severity groupings.METHOD: Participants were adults with chronic cannabis use (N = 599, 57.6 % female) drawn from several studies conducted at large southeastern universities. Confirmatory factor and item response theory analyses were conducted on interviewer-assessed CUD symptoms to evaluate latent CUD construct and severity indicator validity. Multinomial regressions were used to compare severity groups on cannabis use behaviors (e.g., Marijuana Problems Scale [MPS], Marijuana Motives Measure [MMM]), psychiatric symptoms (e.g., Distress Intolerance Index [DII], Mood and Anxiety Symptom Questionnaire [MASQ]), and other substance use (e.g., Timeline Followback [TLFB]).RESULTS: DSM-5 CUD criteria demonstrated good discrimination/difficulty (>1.00) and reflected a unidimensional latent CUD construct (factor loadings = 0.439-0.779). Latent severity scores showed some overlap between individuals with adjacent severity indicators (1.34 - 31.78 % overlap). Moderate/severe CUD differed from mild/no CUD across multiple cannabis use indices, with greater comorbid problems (general distress, daily nicotine use) in severe CUD specifically. Clinically relevant differences between mild vs. no CUD were more limited.CONCLUSIONS: Findings provide support for the validity of the DSM-5 severity indicators of CUD. Results further suggest that moderate and severe CUD appear to be a better reflection of the theory-based conception of the disorder than mild CUD within adults with chronic cannabis use.PMID:42784880 | DOI:10.1016/j.addbeh.2026.108867