PTSD Symptom Severity, Cannabis, and Gender: A Zero-Inflated Negative Binomial Regression Model.

Introduction: In the general population, significant relations have been shown between experienced trauma and cannabis use. Approximately three out of four college students report experiencing at least one-lifetime traumatic event, putting them at greater risk for developing substance use disorder (SUD), and college students report using cannabis to cope with negative effect and stress. However, PTSD symptom severity predicting cannabis use in the college population has not been investigated. Methods: This cross-sectional survey study investigated the relation between PTSD symptom severity and cannabis use, and explored the moderating effect of gender on this relation, using a non-directional, exploratory analysis, in a sample of college students (N=536; 68% female). Due to excessive zero values in the primary outcome, a zero-inflated negative binomial regression model was used. Results: Gender predicted number of cannabis using days [incidence rate ratio (IRR) = 2.17; 95%CI = 1.41, 3.35; p < .001]. The probability of being a cannabis user was moderated by gender [odds ratio (OR) = 0.96; 95%CI= 0.93, 0.99; p=.026], such that for males, as PTSD symptom severity increased, likelihood of being a cannabis user increased. This relation was not supported in females, however. Conclusion/Implication: Discussion includes the potential role of the endocannabinoid system, social norms, and motives in gender differences in PTSD-related cannabis use, and explores the self-medication hypothesis in this context.