Markham, Francis2018-09-062018-09-06b53532016http://hdl.handle.net/1885/147215Gambling is an important public health issue in Australia. According to recent estimates, gambling-related harm is the third largest contributor to the burden of disability in the state of Victoria, measured in terms of disability-adjusted life years. The gambling product most associated with gambling-related harm in Australia is the electronic gaming machine (EGM), which accounts for over half of all Australian gambling expenditure. Around 30 per cent of weekly EGM gamblers experience moderate or severe adverse impacts from their gambling. This thesis consists of six studies on the spatial distribution of the impacts of electronic gaming machines (EGMs) and the relationship between EGM losses and problem gambling. All have been published or were accepted for publication in peer-reviewed academic journals at the time of submission. Jointly, these studies developed theoretical and methodological tools to advance the production of small area estimates of gambling-related harm, as well as beginning the exploration of its consequences. The six studies in this thesis can be grouped into three inter-linked themes that contribute to this aim in different ways. Two studies are concerned with developing the applied and methodological tools for investigating the spatial distribution of problem gambling. The first of these studies presents a calibrated Huff model of the spatial behaviour of gamblers. The second of these uses the Huff model to refine spatial microsimulation derived small area estimates of the prevalence of problem gambling. Together, they provide a toolkit for estimating the local impacts of EGMs. Three studies provide the theoretical underpinning of the thesis by investigating the relationship between gambling losses and problem gambling at the scales of the individual, the EGM venue and state or territory. In order to develop the methods for investigating the spatial distribution of problem gambling, a sustained engagement was required with Total Consumption Theory in the context of gambling. These studies find a consistent relationship between EGM losses and the risk of harm at all spatial scales. At the scale of the individual, there is no evidence to support a J-shaped dose-response relationship, meaning that risk of gambling problems increases monotonically with money lost. A final study estimates the spatio-temporal correlation between EGM accessibility and a single gambling-related harm, domestic violence. Whereas research in the earlier phases of this project sought to estimate the distribution of ‘problem gambling’ as an outcome measure, phase four seeks to measure the relationship between EGM accessibility and specific gambling-related harms directly. In this instance, the spatial association between EGMs and police-recorded domestic violence incidents is investigated in Victorian postcodes over a ten-year period. A significant spatio-temporal association between these two variables is found, providing evidence of a link between EGM gambling and violence. This study concludes that future research might usefully explore the spatio-temporal co-occurrence of EGM gambling and specific gambling-related harms to better understand the social and health impacts of EGM gambling. The research developed in this thesis has contributed toward bringing knowledge of the geography of the impacts of EGMs closer to that of cognate public health issues. While Total Consumption Theory was developed in the context of gambling to underpin the production of local area estimates that incorporate gambling consumption as a risk factor, the findings in this section have broader implications for gambling regulation. More broadly, the approaches developed in this thesis and the research findings have the potential to contribute to improving the regulation of EGMs and thereby reduce the incidence of gambling-related harms.en-AUgamblingelectronic gaming machineshuman geographyhealth geographytotal consumption theorypublic health201710.25911/5d63c0f29f3da