BACKGROUND AND AIMS: Although the co-occurrence of cannabis and depression is well established, less is known about the temporal sequence of cannabis use and depression. The present study had three main aims: to test a symptom-driven pathway in which depression may drive increases in cannabis use, to test a substance-induced pathway in which cannabis use may drive increases in depression and to assess a shared vulnerability model assessing associations between individuals who have (and have not) experienced adverse childhood experiences (ACEs).

DESIGN: Data are from an ongoing, longitudinal, cohort study (n = 2234). Data were set up in an accelerated longitudinal design from age 17 to 24 years. Initial sample was recruited from Southern California, USA. The majority of participants still live in Southern California. On average, participants were aged 18 years at wave 8, with more than half identifying as female (54.3%; n = 1350). Most participants identified as Hispanic (1127; 45.4%), followed by non-Hispanic white (510; 20.5%), Asian (503; 20.2%), multi-racial/other (284; 11.4%) and non-Hispanic black (60; 2.2%).

MEASUREMENTS: Primary outcomes were past-month days of cannabis use and depression symptoms [patient health questionnaire (PHQ)-8]. The Adverse Childhood Experiences scale was used as our main grouping measure.

FINDINGS: In the full sample, we showed that prior levels of depression symptoms were associated with a decrease in cannabis use [opposite to the proposed symptom driven model; B = -0.33 (-0.58, -0.09)]. Dynamic coupling parameters noted individuals who evidenced greater increases in cannabis use between two prior ages reported greater increases in depressive symptoms between subsequent ages [support for a substance-induced pathway; B = 0.53 (0.18, 0.89)]. Similar to the overall sample, for those who had not experienced ACEs, as cannabis use increased we saw a steady increase in depression [support for a substance induced pathway; B = 0.14 (0.04, 0.29)]. However, for those who experienced ACEs, as cannabis use increased we saw a consistent decrease in depression [opposite to the proposed substance-induced pathway; B = -0.18 (-0.28, -0.08)].

CONCLUSION: There is mixed support for both symptom-driven and substance-induced pathways between cannabis use and depression.