Parents’ Cannabis-Related Attitudes and Emerging Adult Offspring Cannabis Use: Testing the Mediating Effect of Perceived Parental Approval

Abstract

Background: Contrary to parental alcohol use and expectancies work, little is known about how parent’s cannabis use (CU) and expectancies influence offspring CU. This is a notable gap in the literature given increasing acceptability and use of cannabis, especially among emerging adults (EA). Moreover, limited work has tested mechanisms of transmission of risk from these parent factors. This study addresses these gaps by testing prospective associations of parental CU and expectancies with offspring CU and CU problems, and perceived parental approval of offspring CU as a potential mediator. Method: A community sample of 314 EA and caregiver dyads completed three annual assessments (mean age = 19.13). The sample was 54% female and majority White/non-Hispanic (76%). Caregivers reported on their cannabis expectancies and use, and EA reported on their CU, CU-related problems, and perceived parental approval of CU. Results: Longitudinal structural equation modeling supported a mediated pathway such that high parental positive cannabis expectancies were associated with perceived parental approval of CU, which in turn, predicted increases in EA CU and CU problems. Parental negative expectancies had a significant indirect effect but in the opposite direction. Indirect effects were found above and beyond parental CU, which was not associated with offspring CU. Conclusions: This is the first study to test prospective indirect effects of parental cannabis expectancies on offspring CU. Findings suggest parents’ attitudes, even in the absence of parental use, confer risk for offspring use by shaping perceived acceptance of CU, suggesting parental expectancies as targets for parent-based CU interventions.

Keywords:

• Emerging adult
• cannabis
• marijuana
• parent expectancies
• parent approval

Acknowledgements

The content of this poster is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Drug Abuse.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We tested whether cannabis-related variables differed for those currently living at home with parents versus living away from home. No variables significantly differed by living status (Fs = 0.00–3.08, ps > .05) and effect sizes were small (all η² < .02). One exception was Wave 7 and 9 cannabis use-related problems did significantly differ (Fs > = 4.72, ps < .05) such that those living away from home experienced more problems but with small effect sizes (all η² < .02). We therefore did not expect the proposed pathways to differ substantially by living status.

2 We ran separate models, treating cannabis use and cannabis use problems as count variables. First, we specified negative binomial distributions for these variables (to accommodate zero inflation) and this model did not converge. Second, we specified them as a Poisson distribution (which does not accommodate zero inflation). This model did converge and results largely matched results obtained using MLR for continuous variables. We retained the model specifying continuous distributions and MLR estimation because count and ordinal variables with more than 5 response options adequately approximate a continuous distribution (Bürkner & Vuorre, 2019; Tabachnick & Fidell, 2013) and does not significantly bias estimates. MLR estimation also has the advantage of allowing estimation of within-time error covariances and simplifying testing mediation.

Additional information

Funding

Funding for this study was supported by a grant from the National Institute on Drug Abuse (NIDA) awarded to C. Colder (R01 DA020171).