Poster #30 - Associations Between Cannabis use, Major Depressive Disorder, and Suicidality: A Clinical Longitudinal Sibling Study

• In Event: Poster Session 09

Abstract

Background and Aim: Understanding causal factors underlying suicidality is of central importance. Cannabis use disorder, depression, and suicidality co-occur more frequently than would be expected by chance. Although direct causality remains unclear, cannabis has been proposed as a plausible causal mechanism for mood maladjustment in both animal and clinical models (Chen et al., 2002; Degenhardt et al., 2001; Smolkina et al., 2017). Thus far, research addressing these questions has not been conducted in clinical samples where cannabis use occurs at notably higher frequencies than in the general population. The current study seeks to extend the present literature by examining the effects of cannabis use on suicidality using a family-controlled, sibling comparison analysis in an adolescent clinical sample with earlier initiation and heavier use patterns than commonly reported (M=13.6 years-old at onset of regular use; M=7-9 days per month at age 17).

Method: We analyzed data from Denver, Colorado- and San Diego, California-based clinical probands and their siblings, recruited from substance use treatment programs, alternative schools, and juvenile probation departments. Participants completed repeated, prospective measures of cannabis use, major depression, suicide attempts, and suicidal ideation at two waves during adolescence (M age=17 ) and early adulthood (M age=24). We leveraged these measures to conduct both longitudinal and family-controlled analyses testing a) cross-sectional and prospective associations between cannabis use and suicidality, b) whether these associations persisted after adjusting for shared environmental and family confounding.

Analysis Plan: Analyses will be conducted using linear mixed-effects models for dichotomous outcomes that account for the nested family structure. Family-controlled models will be used to estimate the association between cannabis use and suicidality within families (i.e., differential sibling exposure in each family). This multilevel model approach accounts for any unmeasured familial factors that make siblings alike by decomposing each siblings’ cannabis exposure into between-pair and within-pair effects (Begg & Parides, 2003; McGue et al., 2014). First, analyses will examine the association between cannabis use and depression and suicidality. Second, multilevel models will partition phenotypic associations into between-family and within-family effects. A final multilevel model will include alcohol use and depression as covariates. We will address associations between adolescent cannabis use and subsequent suicidality and between adolescent suicidality and subsequent cannabis use to address prospective associations. If within-family, sibling comparisons suggest that varying levels of cannabis use are associated with greater depressive symptomology and suicidality, this would provide evidence consistent with, but not definitively prove, causality between cannabis use and later depression and/or suicidality. The lack of any significant finding would indicate that adolescent cannabis use, even with high-frequency use, may not have any significant effect on later mood disorder or adjustment.

Expected Results: Based on the extant literature, we expect significant within-wave associations between cannabis use and suicidality. Additionally, we expect significant prospective associations between cannabis use and subsequent suicidality between adolescence and early adulthood, even after controlling for familial confounds. Conversely, we do not expect significant prospective associations between suicidality on subsequent cannabis use between adolescence and adulthood, before or after controlling for familial confounds.

Authors

• Elisa Fumika Stern, University of Colorado Boulder
  Presenting Author

• Jarrod Ellingson

• Robin Corley, University of Colorado, Boulder

• Michael Stallings

• Christian Hopfer

• Tamara Wall

• Soo Rhee, University of Colorado Boulder