Construct validity of DSM-5 cannabis use disorder diagnosis and severity levels in adults with problematic substance use.

The DSM-5 definition of cannabis use disorder (CUD) differs from DSM-IV by combining abuse and dependence criteria (without the legal criterion) and including withdrawal and craving criteria. Information on construct validity of the DSM-5 CUD diagnosis and severity levels is lacking. This study examines the associations between DSM-5 CUD and severity classification and a set of concurrent validators. Adults with problematic substance use were recruited from two settings: a research setting in an urban medical center and a suburban inpatient addiction treatment program. Participants who reported past-year cannabis use (n = 392) were included in this study and completed a semi-structured, clinician-administered diagnostic interview. Regression models estimated the associations between binary DSM-5 CUD and severity levels with a set of validators, including cannabis use variables, psychopathology, and functional impairment. DSM-5 CUD and all severity levels were associated with cannabis use validators, including number of days used, self-reporting that cannabis use was a major problem, and greater cannabis craving. DSM-5 CUD and severe CUD were associated with other psychiatric disorders and social impairment. Findings add information about the validity of DSM-5 CUD diagnosis and severity levels, with severe CUD receiving the strongest support from its association with validators across all domains, as distinct from the mild and moderate CUD measures that were associated with cannabis-specific validators alone. Severe CUD is likely to require more intensive treatment to bolster physical, psychiatric, and social functioning, whereas the mild and moderate severity thresholds provide useful information for identifying less severe disorders for prevention and brief intervention.