Background: Whereas the role of externalizing disorders is relatively well established in predicting the onset of cannabis use (CU) or cannabis use disorder (CUD), the status of anxiety and mood disorders in predicting CU and CUD remains controversial. Objective: (1) To examine cross-sectional and prospective associations of CU and CUD with a range of mental disorders and whether anxiety and mood disorders are associated with CU/CUD after adjusting for externalizing disorders. Methods: N = 1395 community subjects aged 14–17 at baseline were followed-up at three waves prospectively over 10 years. Substance use, substance disorders and mental disorders were assessed using the DSM-IV/M-CIDI. Results: (1) The baseline prevalence rates where 19.3% at t0 for CU and 2.6% for CUD. Cumulative incidence rates at t3 were 54.3% for CU and 13.7% for CUD. (2) In cross-sectional and prospective analyses other substance use disorders, mood and anxiety disorders were associated with CU and CUD. (3) Associations of panic-anxiety with CU and of depressive and bipolar disorders with CU and CUD were significant after controlling for externalizing disorders. Conclusion: A range of psychopathological conditions, including depressive, bipolar and less consistently anxiety disorders as well as the degree of their comorbidity are significantly associated with incident CU and progression to CUD, even when controlling for externalising disorders. A better understanding of this complex interplay may result in better aetiological models and intervention strategies.

Background: Whereas the role of externalizing disorders is relatively well established in predicting the onset of cannabis use (CU) or cannabis use disorder (CUD), the status of anxiety and mood disorders in predicting CU and CUD remains controversial. Objective: (1) To examine cross-sectional and prospective associations of CU and CUD with a range of mental disorders and whether anxiety and mood disorders are associated with CU/CUD after adjusting for externalizing disorders. Methods: N = 1395 community subjects aged 14–17 at baseline were followed-up at three waves prospectively over 10 years. Substance use, substance disorders and mental disorders were assessed using the DSM-IV/M-CIDI. Results: (1) The baseline prevalence rates where 19.3% at t0 for CU and 2.6% for CUD. Cumulative incidence rates at t3 were 54.3% for CU and 13.7% for CUD. (2) In cross-sectional and prospective analyses other substance use disorders, mood and anxiety disorders were associated with CU and CUD. (3) Associations of panic-anxiety with CU and of depressive and bipolar disorders with CU and CUD were significant after controlling for externalizing disorders. Conclusion: A range of psychopathological conditions, including depressive, bipolar and less consistently anxiety disorders as well as the degree of their comorbidity are significantly associated with incident CU and progression to CUD, even when controlling for externalising disorders. A better understanding of this complex interplay may result in better aetiological models and intervention strategies.