Background
Gambling for money is a popular leisure time activity in most countries, which has major social and economic impacts not only affecting the gambler, but his/her significant others, and the society. Gambling impact studies can help researchers and policymakers compare the health and social costs and benefits of different gambling policies and can be used when considering which gambling policies will reduce or increase costs or benefits the most. In a public health approach, the impacts of gambling, negative and positive, are assessed across the entire severity spectrum of the activity. Although some studies have created basic principles for conducting impact studies, a theoretical model is currently lacking. The aim of this debate is to review complementing and contrasting views on the effects of gambling to create a conceptual model, where a public health perspective is applied.
Main text
The effects of gambling can be structuralized using a conceptual model, where impacts are divided into negative and positive; costs and benefits. Costs and benefits are categorized into three classes: financial, labor and health, and well-being. These classes manifest in personal, interpersonal, and societal levels. Individual impacts cause effects on a personal level to gamblers themselves. External impacts influence the interpersonal and society/community levels and concern other people. The temporal level refers to the development, severity and scope of the gambling impact. These include general impacts, impacts of problem gambling and long-term impacts of gambling.
Conclusions
The conceptual model offers a base on which to start building common methodology for assessing the impact of gambling on the society. While measuring monetary impacts is not always straightforward, the main issue is how to measure the social impacts, which are typically ignored in calculations, as are personal and interpersonal impacts. The reviewed empirical work largely concentrated on the costs of gambling, especially costs on the community level. The Model can be used to identify areas where research is scarce. Filling the gaps in knowledge is essential in forming a balanced evidence base on the impacts of gambling. Ideally, this evidence could be the starting point in formulating public policies on gambling.
Peer Review reports
Background
Gambling can be defined as betting money on an outcome of uncertain results to win money. All forms of gambling, even those typically considered to be more skill-based, like poker and sports betting, contain an element of luck [1]. Another common characteristic of gambling is that it is a zero-sum game: when one player wins, the other must lose [2]. Gambling is a popular leisure time activity in most countries, and the vast majority of adults have engaged in some gambling activity at least once in their life, and between 40 and 80% have participated in some form of gambling in the last 12 months [3]. For most individuals, gambling is a form of entertainment [4, 5]. For some consumers, the motivation for gambling is influenced by social interactions because gambling venues offer social settings to meet people [6, 7], whereas others are mainly motivated by the dream of winning money [8]. By contrast, some use gambling to escape their problems, and this is especially common among problem gamblers [9].
Gambling is typically viewed as a continuum, with most people gambling only occasionally or not at all and some gambling more frequently. Along this continuum, people can experience negative financial and social consequences, although harms tend to be more common among frequent gamblers [10]. Based on harms experienced because of gambling, gamblers are usually divided to recreational, at-risk, and problem and pathological gamblers [11, 12]. Problem and pathological gamblers are usually called problematic gamblers. Pathological gambling is a disorder included in both diagnostic manuals: International Classification of Disorders [13] and Diagnostic and Statistical Manual [14]. Prevalence of problem and pathological gambling varies between countries, but it is estimated that among adult population 1 to 4% are problem gamblers [15], whereas prevalence estimates of pathological gambling range from 0.1 to 0.8% [16]. There are, however, much more people suffering from gambling-related harms.
Harms caused by gambling can co-occur with other difficult situations in life, usually intensifying along with crises and continuing even after the problematic behavior comes to an end [17, 18]. Gambling-related harm can affect multiple domains of life [17], including financial [1\and health problems\], psychological and emotional distress [\, and impaired social and cultural relationships \]. They have an influence on multiple levels: gambling-related harms restrict the gambler and their family, friends, workplace, community, and society [17, 18, 28, 29]. Because of these significant influences on society and the population’s overall health, gambling is a critical public health issue [30, 31].
The impacts of gambling on societies is positive and negative and depends on a number of factors, including what type of gambling environments and games are available, how long gambling has been possible, whether gambling revenues are derived locally or outside the jurisdiction, and the effectiveness of gambling policy [32,33,34,35]. Overall, there are several main purposes for conducting impact studies on gambling. First, to demonstrate that gambling has major social and economic impacts. Impact studies can also help researchers and policymakers compare the impact of different health and social problems and benefits; for example, gambling impacts can be weighed against alcohol impacts. Additionally, impact studies can be used when considering which gambling policies will reduce or increase costs or benefits the most [36].
Different approaches have been used to study the impacts of gambling. Research into the socioeconomic impacts of gambling can be conducted from a cost of illness perspective, commonly used in alcohol and drug research; however, this approach neglects the benefit side [37]. Economic cost–benefit analysis (CBA) measures changes in well-being in common units (dollars) [38] and attempts to discover whether increased gambling opportunities are positive for society [39]. In this approach, monetary value is also assigned to intangible harms (harms not necessarily monetary in nature, e.g., the pain and suffering of problem gambler), and harms are known to affect others in addition to the gamblers themselves. This approach, however, has been criticized because an arbitrary monetary value is applied to these intangible harms [37, 40]. Anielski and Braaten [39] also examined the impacts of gambling by using an approach they called full cost–benefit accounting, which attempts to overcome the obstacles of CBA. However, like Williams, Rehm and Stevens [32] stated, figures obtained by this approach are not reliable and somewhat arbitrary, and it is not clear how the monetary values for some variables are created. Anielski and Braaten [39] also described many other approaches to study gambling impacts.
In a public health approach, the impacts of gambling, negative and positive, are assessed across the entire severity spectrum of the activity [41]. According to the literature, harms can occur also among those classified as nonproblem gamblers [42]; however, examining only problem or pathological gambling and its impacts on society is still common in economic costing studies [43]. When concentrating solely on problematic gambling, only the tip of the iceberg is observed and gambling harms and its costs to society are underestimated [42]. Additionally, in a public health approach, the positive effects associated with gambling are recognized [17]. In the economic literature, gambling revenues and positive impacts on public services have been observed [32], but fewer studies have examined the positive impacts of gambling on gamblers or their significant others. In a public health approach, the negative impacts of gambling can be assessed by health-related quality of life (HRQL) weights, known as disability weights (DW), which measure the per-person burden of health state on quality of life [44, 45]. DWs have been used to measure intangible social costs of gamblers, but could be also used to discover gambling harms that affect a gambler’s social network. Some studies have attempted to quantify the benefits of gambling by “consumer surplus,” which is the difference between what people would be willing to pay for a product or service versus what they pay [32]. In Australia, the estimated consumer surplus for gambling is AUS$8–$11 billion per year [2]. However, using this arbitrary monetary amount to quantify something that is clearly nonmonetary creates similar problems when trying to place a monetary value on the “social” impacts of gambling [32].
Since the expansion of the gambling market, the question of gambling impacts has piqued researchers and policymakers interest [37]. Despite increased interest in gambling impacts, no consensus has been reached regarding the appropriate theoretical and methodological approach to studying them [32]. A theoretical model is still lacking, although some studies have created basic principles for conducting socioeconomic impact studies. Based on Anielski and Braatan’s socioeconomic impact of gambling (SEIG) framework [39], Williams et al. [32] proposed a simpler categorization of impacts. By doing this, Williams et al. ignored that impacts can be evaluated on different levels, like the individual, family, household, community, regional, and national levels.
Several limitations of earlier gambling impact studies have been highlighted [37, 40], but one major concern has been how to capture and quantify the social impacts [32, 46]. While quantifying the economic impacts is reasonably straightforward (e.g., costs of treating problem gamblers or of preventing problem gambling), this is not the case for social impacts (e.g., invisible costs like the impacts of emotional stress and relationship problems caused by gambling), which cover the major negative impacts from gambling and cannot be evaluated in monetary terms [42].
Thus, studies have mostly ignored social impacts, choosing to measure only the economic costs or benefits that are quite easily quantifiable. This approach, however, presents a very biased view of the situation. There are no established ways to define the social impacts of gambling. Based on Williams et al. [32] social impacts are costs or benefits that are nonmonetary in nature. Walker and Barnett [40] stated that social costs must aggregate societal real wealth, that is, cause harm to someone in the society and benefit no one. They also defined that social cost must be social, rather than personal.
According to these definitions, when a gambler becomes ill because of excessive gambling, their suffering should not be counted as a social cost as long as someone in society gains from this excessive gambling and gamblers do not demand any treatment that would cause costs to society. In our study, rather speaking of social impacts, we use the term nonmonetary impacts (i.e., nonmonetary costs and nonmonetary benefits). Costs and benefits refer to overall negative or positive gambling impacts and not only those with monetary value. We also state that impacts should be examined at the societal, individual, and interpersonal levels.
Compared with existing models, this model combines aspects from costing studies [32, 39] and from gambling harm literature [18, 33,34,35] making the present model more comprehensive and up to date. The Model emphasizes the public health perspective, which is somewhat different from the one in costing studies. It covers both positive and negative effects of gambling and examines costs and benefits on individual, interpersonal and community/society levels. The model includes a temporal dimension, which refers to the development and severity of gambling behavior. From the public health perspective, it is not presumed that costs and benefits result only from problem gambling; instead we are interested in the whole spectrum of gambling behavior. Costs and benefits can be general, come from problem gambling and/or can have long term effects. In summary, a common and comparable methodology for evaluating the impacts of gambling is necessary [32, 37], and none has been created. Studies have usually concentrated on impacts of problem gambling while ignoring the entire continuum of gambling. Additionally, the emphasis has been on economic costs, whereas most gambling costs are “social.” The benefits of gambling are usually examined at the societal level (e.g., government revenue), and the influence of gambling on gamblers and their significant others are ignored. This debate argues for a conceptual theoretical model based on the gambling impacts literature, where a public health perspective is applied.
Main text
Structure of the public health impacts of gambling (PHIGam) model
Gambling impacts can be observed at the personal, interpersonal, and community/society levels (Fig. 1). Personal level refers to the gamblers themselves and interpersonal level to people close to the gambler: friends, family and work colleagues. Impacts can be individual or external. Individual impacts induce effects on a personal level to the gambler. External impacts influence the interpersonal and society/community levels and concern those who are not necessarily gamblers themselves. Gambling creates costs and benefits that others must pay for or can exploit. For example, gambler’s increased debt and financial strain affects family members’ lives, and the effects escalating into bankruptcy and homelessness can also be observed on the community level. Thus, it is of utmost importance that impacts are examined on multiple levels. In the model, impacts can be divided into negative and positive. However, impacts can simultaneously be both negative and positive. For example, gambling is linked to increased criminality [44] but can also decrease illegal gambling [45]. Similarly, tourism revenues are positive [46] but on the other hand tourism can increase crime [47].
