Author ORCID Identifier

Defense Date


Document Type


Degree Name

Doctor of Philosophy


Clinical Psychology

First Advisor

Ananda B. Amstadter


The college years encompass a period of increased risk recreational cannabis use (RCU), as well as a time of increased risk for trauma exposure and developing posttraumatic stress disorder (PTSD). Given the high co-occurrence between RCU and PTSD, and the potentially negative consequences of the two (e.g., worse academic outcomes), there is a need to understand the etiologic mechanisms of these commonly co-occurring conditions. Two primary phenotypic models exist: self-medication model (i.e., PTSD to RCU) and the high-risk model (i.e., RCU to PTSD). To date, there are two existing studies longitudinally examining the etiologic models proposed to explain co-occurring RCU and PTSD in a college sample, but they are limited to only investigating the first two years of college. Thus, Aim 1 of this study examined these models of co-occurrence in a large, ongoing longitudinal study of college students (Spit for Science [S4S]; NIAAA-R37 AA011408, PIs Kenneth Kendler & Danielle Dick) throughout the first three years of college. Cannabis use and PTSD have been shown to be moderately heritable in twin studies. Thus, Aim 2 conducted aggregate genome-wide analyses (i.e., genome-wide complex trait analysis [GCTA], polygenic risk scores [PRS]) of RCU and PTSD to examine their molecular heritability, as well as the association of aggregate genetic risk with RCU and PTSD. Given evidence of latent heritability, as well as overlapping latent heritability of lifetime cannabis use and PTSD, examination of molecular genetic risk is also needed. Thus, Aim 3 further examined the self-medication and high-risk models by incorporating PRS for lifetime cannabis use and lifetime PTSD as potential influences of same- and cross-phenotype prediction (e.g., PRS for lifetime cannabis use predicting RCU and PTSD in S4S). To limit genetic heterogeneity, study participants were limited to individuals in S4S with European- (n = 3721) and African- (n = 1469) ancestry based off of their genomic super-population assignment. Aim 1 results supported both the self-medication and high-risk model. Aim 2 results did not provide support for significant molecular heritability of RCU or TRD in individuals with European or African ancestry in S4S likely due to low statistical power. Aim 2 results did provide evidence of same-trait prediction of PRS for lifetime cannabis use predicting non-experimental (i.e., use ≥ 6 times) cannabis use in individuals with European ancestry in S4S. Aim 3 results did not provide support for significant moderation of PRS for lifetime cannabis use or PRS for lifetime PTSD in the self-medication or high-risk models, respectively. However, Aim 3 results did provide evidence of same-trait prediction of non-experimental cannabis use based on PRS for lifetime cannabis use. Given the relatively small sample size, genotypic results should be interpreted with caution. However, as a whole, these findings provide support for the self-medication and high-risk models explaining the development of co-occurring PTSD and cannabis use. Implications of these findings, in light of study limitations, are discussed.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission