Abstract
CUD refers to a pattern of cannabis use that causes clinically significant impairment or distress. It is estimated to occur in 22% of people who use cannabis, with an estimated 22 million individuals affected globally, comparable to the prevalence of opioid use disorder. CUD is commonly comorbid with other mental health and substance use disorders, and affected individuals may also have reduced cognitive function. Prevalence of cannabis use and CUD is higher in younger populations and experiencing CUD during adolescence might lead to negative consequences into adulthood. Reducing the global burden of disease caused by CUD requires a multifaceted approach. Research has focused on both identifying efficacious treatments for reducing CUD, as well as determining risk factors to identify and educate those at risk of developing CUD. The work in this thesis addresses both research priorities, with a focus on the role of individual profile of comorbidities and cannabis use in the prediction and treatment of CUD.The aims of this thesis were 1) to identify and estimate the role of mental health, cognitive function, and cannabis use profile on risk for CUD and 2) to evaluate and develop evidence on treatments for CUD, considering comorbidities of mental health and cognitive function. The analyses underpinning this thesis were conducted using existing, high-quality, longitudinal cohort and clinical trial datasets. Specific analyses were chosen based on consideration of the available data and attempts to reduce bias and increase the rigour of the findings. This included the use of mixed effects models, longitudinal latent class analysis and multiple imputation.
Firstly, I conducted analyses of two longitudinal datasets: ALSPAC, a prospective birth cohort dataset, and CannTeen, a 12-month observational study of adolescents and adults. Using the ALSPAC dataset, I identified participant patterns of cannabis problems over adolescence using longitudinal latent class analysis. I then estimated the role of childhood mental health and cognition prospectively assessed at ages 8-10 on later patterns of cannabis problems. I found that externalising disorders at age 10 were particularly associated with later patterns of problems characterised by early-onset, high levels of problems. Internalising disorders and cognitive function were not reliably associated with later problem classes.
Secondly, I conducted two analyses using the CannTeen observational study. The first was a longitudinal investigation of CUD symptoms, comparing adults and adolescents using mixed effects models. I found that adolescents scored on average 3.7 points higher on a measure of CUD symptoms than adults, with no indication that the pattern of symptoms over time was different in adolescents compared to adults. This association remained after adjustment for a comprehensive measure of cannabis use (standard THC units). I expanded this work in a further analysis of standard THC units by testing their ability to discriminate between people who use cannabis with and without CUD. Using receiver operating characteristic curve analysis, THC units demonstrated a good ability to discriminate people with and without CUD, and I was able to determine preliminary thresholds of weekly THC consumption that correspond to risk for CUD.
To address the second aim of the thesis, I conducted a secondary analysis from a randomised clinical trial of daily oral cannabidiol at 400mg and 800mg doses for treatment of CUD. Focusing on previous evidence that cannabis has a detrimental effect on cognition, and indications that cannabidiol may be beneficial for cognitive function, I analysed change in cognitive function over the 4-week trial. Findings did not indicate a strong effect of either cannabidiol doses compared to placebo on verbal learning and memory (the primary outcome), or most secondary cognitive outcomes. However, there was some evidence that cannabidiol at 800mg per day improved working memory manipulation when compared to placebo.
Finally, I reviewed and evaluated clinical trial literature on psychosocial and pharmacological treatments for CUD, including its comorbidity with anxiety, depression, psychosis and externalising disorders. Findings supported a role of psychosocial treatments on increasing abstinence from cannabis use, including when CUD was comorbid with anxiety and depression, but not psychosis. Treatments that incorporate the individual’s family might be particularly useful for adolescents with CUD and externalising disorders. Many pharmacological treatments for CUD have been investigated in clinical trials, but few have shown promise for reducing cannabis use. Small scale trials are emerging that act on the endocannabinoid system (including cannabidiol), however these are in the early phase of testing and require replication.
The studies described suggest that there are several efficacious treatment methods for CUD, however mental health comorbidity may impact on treatment efficacy, particularly psychosis. Cannabidiol shows emerging promise for reducing cannabis use but did not improve cognition across a 4-week randomised placebo-controlled clinical trial. Several risk factors including childhood externalising disorders, adolescent cannabis use, and amount of THC consumed might increase risk of CUD, and it was possible to preliminarily determine thresholds for risk of CUD by THC consumption.
Taken together, this thesis highlights the important role of comorbidities, especially mental health, on risk for and treatment of CUD. The findings also indicate the utility of using a detailed assessment of cannabis use when considering risk of CUD, including the ability to quantify this risk and prevent it by determining lower use thresholds. This has significance for both clinicians as well as researchers conducting future investigations of CUD.
| Date of Award | 23 Nov 2023 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Tom Freeman (Supervisor), Lindsey Hines (Supervisor) & George Stothart (Supervisor) |
Cite this
- Standard