Modelling experiences of self-medicating behaviours in the context of low mood: psychosocial impairment, hazardous drinking and alcohol-related consequences

Abstract

The main focus of this thesis is on alcohol use and depression, and the behaviour of self-medication. This will be an exploration of these experiences at the symptom level for those who do or do not self-medicate with alcohol. The aim is to go beyond comorbidity to distinguish the symptoms expected to be more prominent in people with depressive experiences who do report self-medicating behaviours, verses those who do not, right down to their experiences, and not at the disorder level. There have been multiple studies that have attempted to unpick the very complex relationship between problematic alcohol use and mood difficulties, but the novel aspect of this thesis is to focus on the finer details of the practices, contexts, and ‘symptoms’ involved in these two common human behaviours/experiences.

Aims: The main aims will be to explore the measurement of depression, if differences in low mood can be explored at symptom level, and the investigation of low mood in relation to impairment. Also, the exploration of alcohol use in relation to drinking patterns and drinking consequences, while trying to explain why people self-medicate their low mood with alcohol.

Method: The National Epidemiologic Survey on Alcohol and Related Conditions(NESARC) Wave 1, was the main data source used in this thesis which is a nationally representative survey of 43,093 participants aged over 18 years old living in all of the50 states of America. This is a two-wave survey, with the second wave conducted 3-4years post the first wave. Both waves of this survey were utilised in this thesis and were conducted by the US Bureau of the Census in aid of the National Institute on Alcohol Abuse and Alcoholism (NIAAA; Grant et al., 2003). The NESARC is a comprehensive nationally representative survey which explores the occurrence of more than one psychological disorder or substance use disorder experienced by the one individual, using definitions from the DSM-IV. A series of advanced statistical models including factor analysis, structural equational modelling, and latent class analysis were employed to test the research questions.

Results: A one factor structure for the latent variable of depression provided the best fit to the data. Compared to females, males were less likely to have higher estimates for depression. Compared to Whites, American Indians were more likely to have higher estimates for depression. In contrast, Blacks, Asians and Hispanics had lower estimates for depression compared to Whites. All individuals under 65 years old were less likely to have higher estimates for depression than those aged over 65 years. Those who did self-medicate were more likely to have higher estimates for depression than those who did not self-medicate. In regards to impairment, a two-factor structure measuring distress and dysfunction was identified as the best fit and this structure of distress and dysfunction was similar across the self-medicating and non-self-medicating group. Compared to Whites, Hispanics were more likely to have lower estimates for dysfunction items. Latent Class Analysis identified a four-class model of clinically relevant homogeneous groups of drinkers that were identified as the best fit across the two groups of self-medicators/non-self-medicators. These four classes were named based on their probabilities on each item as ‘seldom drinkers’ and ‘hazardous drinkers’ in the self-medication group and ‘normal drinkers’ and ‘very seldom drinkers’ in the non-self-medication group. Using these four classes, drinking related consequences were explored and the results indicated that the very seldom drinkers, seldom drinkers and normal drinkers did not experience most of the alcohol related consequences, whereas, the hazardous drinking group experienced these consequences more commonly. The most prominent consequences experienced were interpersonal difficulties with family and friends, tolerance and withdrawal issues and a substantial time spent procuring, using or recovering from drinking. The hazardous drinking group also showed the highest estimates for alcohol use disorder. A one factor structure for the latent variable of impulsivity provided the best fit to the data. Hazardous drinkers, compared to very seldom drinkers, had higher latent levels of impulsivity. Compared to 18-25 year olds, adults aged 45-64 and 65 and older were less likely to be seldom drinkers. Compared to those who were married, those who were never married were more likely to be normal drinkers and also compared to those who were married, widowed/separated/divorced individuals were more likely to be hazardous drinkers. Also, compared to males, females were less likely to be hazardous drinkers. Compared to very seldom drinkers, hazardous drinkers were more likely to endorse the item measuring: “Have you often done things impulsively?”

Conclusions: The importance of the investigation of the behaviour of self-medication is key as research has shown that individuals who self-medicate tend to have poorer functioning and lower mental health related quality of life and are at a greater risk for developing a substance use disorder. It is important to identify which sub-populations are at most risk of higher estimates of depression to improve clinical interventions and treatment. If individuals who self-medicate experience more distressing and debilitating depressive experiences, this is important to identify and prevent. A key gap in research was assessed for overall functional impairment in relation to self-medication and depression to provide a valuable insight into the specific types of problems experienced by those who self-medicate. Another important implication was the valuable insight provided into the specific types of drinking patterns experienced by those who self-medicate, and how alcohol is used in regard to type of alcohol consumed, frequency and context of drinking depending on the behaviour of self-medication, with the important consequences of using alcohol in this way. Finally, another key implication was the determination of what may cause these individuals to self-medicate, and the importance of impulsivity. Most research focuses on comorbidity(i.e. when two disorders occur at the same time), which is mostly based on age of onset and which disorder came first, but the intention for this thesis was to move away from comorbidity and to focus in more detail on attempting to explain the reasons why individuals may experience both ‘conditions.’

Date of Award	Jan 2019
Original language	English
Supervisor	Jamie Murphy (Supervisor) & Orla McBride (Supervisor)