Psychosocial characteristics of adolescent problem gambling

Problem gambling among adolescents has emerged as a significant area of research interest. Youth gambling problems are associated with a range of interpersonal, familial, economic, psychological and legal problems. However, because not all adolescents who gamble will develop gambling problems, the research literature has begun to emphasise potential factors that may increase or ameliorate the risk of developing such difficulties. Those characteristics associated with higher levels of severity, earlier onset and longer duration of symptoms are described as risk factors, while those which serve to reduce the severity of problems or lessen the influence of risk factors are referred to as protective factors. Although a stronger focus on these characteristics has emerged, there is still much to be learned about factors that may be associated with adolescent problem gambling behaviour. In particular, given their potential as targets for intervention, potentially malleable psychosocial factors represent an important area of research. Presented as a thesis by publication, the first chapter of this thesis provides a narrative review of problem gambling among adolescents. The use of a biopsychosocial model as a framework within which to identify and organise relevant variables is presented, and this framework is used throughout the thesis. The second chapter provides a rationale and aims for the current thesis. Chapter 3 consists of a systematic review of the extant literature in relation to the psychosocial characteristics associated with problem gambling in high school students, while Chapter 4 contains an original empirical study of the psychosocial characteristics of problem gambling among a sample of high school students. The study in Chapter 5 further explores the relationships between attitudes to gambling, gambling involvement, coping strategies, parenting styles, and gambling problems in this sample. Finally, Chapter 6 provides a general discussion of the key findings contained within this thesis, and makes comment on their broader clinical and research implications. For the study presented in Chapter 3, a comprehensive search of the extant literature in relation to the psychosocial characteristics associated with adolescent problem gambling in high school students was performed. Strict inclusion criteria were applied, resulting in 19 studies being examined. Reflecting the diversity of the field, from these 19 studies, 46 individual psychosocial characteristics were identified. For the purposes of drawing conclusions, only the 11 characteristics for which a minimum of two studies existed were described further. Grouped using a biopsychosocial model, the review presented in Chapter 3 found evidence for a number of characteristics across various domains. Specifically, associations were found between problem gambling and: impulsivity and general risk propensity (temperament/personality domain); ineffective coping (cognitive domain); family problems (family environment domain), symptoms of ADHD, substance use, and delinquency (externalising problems domain), and emotional problems and anxiety (internalising problems domain). It was concluded that measurement issues and a lack of replication have an impact on the ability to determine the strength and direction of relationships between problem gambling and associated psychosocial characteristics. Implications of the findings for practice and research are also discussed. The study described in Chapter 4 is based on secondary data analysis from a larger study investigating the familial transmission of gambling problems. Participants (N = 612, 240 males, 371 females, 1 unreported) recruited from 17 secondary schools across Victoria, Australia, completed a self-report questionnaire assessing a number of relevant variables. Using a biopsychosocial conceptual model, the aim of the study was to examine the extent to which a number of previously identified demographic and psychosocial characteristics were predictive of gambling problems. In addition, the study sought to explore whether female gender acted as a protective factor by moderating associations between putative risk factors and gambling problems. Results from the study partially supported the hypothesised relationships between a number of risk factors and problem gambling. When controlling for the influence of other variables, perceived paternal problem gambling, perceived paternal problem drinking, the number of gambling friends, gambling attitudes and life stressors all emerged as unique predictors. Of note, female gender moderated the relationship between paternal problem drinking and problem gambling, such that this variable was significantly predictive of gambling problems only for females. The implications of these findings for intervention and treatment are discussed, and the importance of utilising multivariate approaches that can take into account the mutual influence of various biopsychosocial factors is emphasised. The final study, which is presented in Chapter 5, is based on the same data set as that described in Chapter 4, however employed a more sophisticated multivariate statistical technique to further examine the links between gambling attitudes, gambling involvement, coping strategies, parenting styles, and gambling problems. The aim of the study was to simultaneously explore predictors of problem gambling, while examining the extent to which coping skills and parenting styles may moderate the expected association between gambling involvement and gambling problems. Data from the 612 high school participants were analysed using a zero-inflated Poisson (ZIP) regression model, controlling for gender. Results provided mixed support for the hypotheses, with gambling involvement and inconsistent discipline both predicting gambling problems, and gambling attitudes predicting gambling involvement. In addition, significant interaction terms were found for gambling involvement by problem focussed coping, reference to others and inconsistent discipline. Finally, a significant indirect effect on gambling problems from gambling attitudes through gambling involvement was found, suggesting involvement fully mediated the relationship between gambling attitudes and gambling problems. In contrast, no relationships were found for the prediction of the inflated portion of the model, suggesting that the study variables were only related to the level of gambling problems, and not the probability of having gambling problems. The sixth chapter of the thesis provides a synopsis and general discussion of the key findings from Chapters 3, 4, and 5. The clinical and research implications are presented, and the strengths and limitations of the current thesis are discussed. Finally, recommendations for future research are provided.