Introduction
Overview
Gambling is a relatively common recreational activity for older adults across a range of cultures (Subramaniam et al., Reference Subramaniam, Wang, Soh, Vaingankar, Chong, Browning and Thomas2015; Luo and Ferguson, Reference Luo and Ferguson2017). Researchers have demonstrated that older adults may experience a range of benefits from gambling, including that gambling is a mechanism for socialisation and increased cognitive functioning (Desai et al., Reference Desai, Maciejewski, Dausey, Caldarone and Potenza2004). However, researchers have also demonstrated that gambling can pose significant risks to older adults' health and wellbeing (Guillou Landreat et al., Reference Guillou Landreat, Cholet, Grail Bronnec, Lalande and Le Reste2019; McCarthy et al., Reference McCarthy, Thomas, Pitt, Daube and Cassidy2020). van der Maas et al. (Reference van der Maas, Nower, Matheson, Turner and Mann2021: 211) argue that while the international literature generally indicates that older adults have lower rates of gambling disorders as compared to younger age groups, ‘aging populations in the global north, possible cohort effects as a result of the “baby boom” generation, and overall increases in gambling availability are all trends that may increase the number of older adults with Gambling Disorder’. A systematic review of the international literature by Tse et al. (Reference Tse, Hong, Wang and Cunningham-Williams2012) demonstrated that while older adults' participation in gambling is increasing, there are clear differences across different countries and contexts, ranging from 26.6 per cent (Lai, Reference Lai2006) to 85.6 per cent (Rönnberg et al., Reference Rönnberg, Volberg, Abbott, Moore, Andrén, Munck and Svensson1999). Estimates of the life-long prevalence of ‘problem gambling’ in older adults varies between countries, ranging from 0.01 to 10.6 per cent (Subramaniam et al., Reference Subramaniam, Wang, Soh, Vaingankar, Chong, Browning and Thomas2015). In Australia, the Household Income and Labour Dynamics (HILDA) survey reported that 1.1 per cent of the Australian adult population are classified as ‘problem gamblers’, with 29.1 and 11.1 per cent of those classified as ‘problem gamblers’ aged 50–64 years old and 65 years and over, respectively (Armstrong and Carroll, Reference Armstrong and Carroll2017). Jurisdictional differences exist in the prevalence of ‘problem gambling’. In 2019, 0.7 per cent of Victorian adults were classified as experiencing problems with gambling, with rates of 0.8 per cent for 55–64 year olds, 0.5 per cent for 64–75 year olds, and 0.1 per cent for those aged 75 years and over (Rockloff et al., Reference Rockloff, Browne, Hing, Thorne, Russell, Greer, Tran, Brook and Sproston2020). In the same year in New South Wales, 1 per cent of the adults were classified as experiencing a problem with gambling, with rates of 1.5 per cent for 55–64 year olds and 1.0 per cent for those aged 65 and over (Browne et al., Reference Browne, Rockloff, Hing, Russell, Boyle, Rawat, Tran, Brook and Sproston2020). Additionally, the HILDA survey reported that about one-third (30.4%) of 50–64 year olds and about one-quarter (23.8%) of adults aged 65 years and over were considered regular gamblers – gambling at least once in a typical month (Armstrong and Carroll, Reference Armstrong and Carroll2017). The survey also found that older adults in the 54–65 age category had a higher probability of ‘problem gambling’ as compared to other age groups (Wooden and Wilkins, Reference Wooden and Wilkins2017).
While it is important to understand the factors that may contribute to gambling disorders, public health researchers have increasingly focused on the range of factors that may contribute to normalised and regular participation in gambling, including the broader negative experiences with gambling. John et al. (Reference John, Holloway, Davies, May, Buhociu, Cousins, Thomas and Roderique-Davies2020: 2) argue that understanding a spectrum of gambling harms is important in moving away from a ‘false dichotomy of “safe, social, or responsible” gambling on one side and “problem, dependent or pathological” gambling on the other’. Gambling-related harms are not only associated with those classified as ‘problem gamblers’, but are experienced across all levels of risk classification (Miller, Reference Miller2017). van Schalkwyk et al. (Reference van Schalkwyk, Petticrew, Cassidy, Adams, McKee, Reynolds and Orford2021) argue that gambling harms occur on a continuum, take a range of forms, are constantly changing, and impact not only on the individual but also on their families and communities. However, research syntheses of gambling in older adults have found an overemphasis on the characteristics and experiences of older adults who have been classified as ‘problem’ or ‘pathological gamblers’, rather than exploring the beliefs and behaviours of those who may be at low or moderate risk of gambling-related harm (Ariyabuddhiphongs, Reference Ariyabuddhiphongs2012; Tse et al., Reference Tse, Hong, Wang and Cunningham-Williams2012; Subramaniam et al., Reference Subramaniam, Wang, Soh, Vaingankar, Chong, Browning and Thomas2015; Guillou Landreat et al., Reference Guillou Landreat, Cholet, Grail Bronnec, Lalande and Le Reste2019).
A range of factors may contribute to older adults' increased risk of regular and routine engagement in gambling, and gambling-related harm (van der Maas et al., Reference van der Maas, Nower, Matheson, Turner and Mann2021). These include significant life changes such as retiring from paid work and seeking alternative recreational activities, the loss of friends and family networks (either through death or the inability to engage in social activities), social isolation and loneliness, low and often fixed incomes, and their subsequent inability to recover financially from gambling losses (Subramaniam et al., Reference Subramaniam, Wang, Soh, Vaingankar, Chong, Browning and Thomas2015; Guillou Landreat et al., Reference Guillou Landreat, Cholet, Grail Bronnec, Lalande and Le Reste2019). It is these very factors that may also make older adults an attractive target for the gambling industry. Researchers argue that older adults may attend gambling venues due to free transport, affordable meals and social activity incentives, as well as the perception that gambling venues are a ‘safe’ environment (van der Maas et al., Reference van der Maas, Mann, Matheson, Turner, Hamilton and McCready2017; Thomas et al., Reference Thomas, Pitt, Randle, Balandin, Cowlishaw, McCarthy, Bestman and Daube2020; McCarthy et al., Reference McCarthy, Thomas, Pitt and Bellringer2021). In some cases, social associations and agencies responsible for the health and socialisation of older adults may also facilitate and encourage gambling activities for older adults, e.g. through excursions of retirement residents to gambling venues (McNeilly and Burke, Reference McNeilly and Burke2001; Thomas et al., Reference Thomas, Pitt, Randle, Balandin, Cowlishaw, McCarthy, Bestman and Daube2020). The above studies provide evidence for ensuring that a range of individual, socio-cultural, environmental and commercial influences are considered when investigating older adults' gambling behaviours and motivations.
Despite these well-documented vulnerabilities for older adults, few studies have sought to understand why older adults gamble and how a range of factors may contribute to regular or routine gambling practices. Researchers have called for detailed qualitative studies of the gambling experiences of older adults to understand the complex range of factors that influence their gambling behaviours and practices (Tse et al., Reference Tse, Hong, Wang and Cunningham-Williams2012). Qualitative approaches seek to understand the lived experiences of different health and social phenomena, and enable researchers to develop a detailed understanding of people's experiences and the meanings they make of these phenomena (Hansen, Reference Hansen2006; Green and Thorogood, Reference Green and Thorogood2018). We would take this a step further and argue that, in particular, critical qualitative inquiries (Denzin, Reference Denzin2017) are needed to understand not only the gambling practices associated with different risk categories, but also the range of power structures (including from the gambling industry) that may influence these practices. These types of qualitative inquiries seek to understand how power, inequity and justice may influence different health and social phenomenon (Charmaz, Reference Charmaz2017). In the case of gambling, this includes understanding how strategies of the gambling industry, social contexts, the accessibility and availability of gambling, and gambling environments, may influence the normalisation of gambling, and gambling practices that are potentially risky for different population subgroups (Goyder et al., Reference Goyder, Blank, Baxter and van Schalkwyk2020; Thomas et al., Reference Thomas, Pitt, Randle, Balandin, Cowlishaw, McCarthy, Bestman and Daube2020; McCarthy et al., Reference McCarthy, Pitt, Bellringer and Thomas2022). These types of studies are also important in developing prevention initiatives. As Venuleo et al. (Reference Venuleo, Marinaci and Mossi2021) argue, understanding why gambling harm may occur in older adult populations cannot be understood without investigations into social and cultural contexts, including the meanings that are associated with gambling.
Theoretical lens
To explore the range of factors that may influence how older Australian adults participate in gambling, and the meanings they offer about these experiences, this paper draws upon Social Practice Theory (Reckwitz, Reference Reckwitz2002; Shove et al., Reference Shove, Pantzar and Watson2012; Supski et al., Reference Supski, Lindsay and Tanner2017). Social Practice Theory shifts the focus from the individual and their behaviour, towards an understanding of the social and collective organisation of practices (Hargreaves, Reference Hargreaves2011). Theorists such as Reckwitz (Reference Reckwitz2002), Shove et al. (Reference Shove, Pantzar and Watson2012) and Warde (Reference Warde2005) pursue a middle ground between agency and structure which ‘has been found in the everyday and routine performances of social practices’ (Hargreaves, Reference Hargreaves2011: 82). The performance of social practices is viewed as the ‘routine accomplishment of what people take to be normal ways of life’ (Shove, Reference Shove2004: 1054). Reckwitz (Reference Reckwitz2002) and Shove et al. (Reference Shove, Pantzar and Watson2012) assert that social practices, such as playing sport or smoking, consist of a range of elements that are interconnected – thus fitting well with an examination of the range of factors that may contribute to gambling. Importantly, social practices are dynamic, shaping perceptions, interpretations and actions, and participation in social practices leads to the reproduction and transformation of them (Shove et al., Reference Shove, Pantzar and Watson2012; Blue et al., Reference Blue, Shove, Carmona and Kelly2016). People are practitioners or carriers of the social practice and are recruited to the social practice when they enact it (Reckwitz, Reference Reckwitz2002; Shove et al., Reference Shove, Pantzar and Watson2012). For example, when a person gambles on electronic gambling machines (EGMs, pokies or poker machines), they have been recruited to a social practice which influences not only the person participating (such as their perceptions of EGMs) and those around them but also the way the social practice of EGM participation may occur.
Burke et al. (Reference Burke, Joesph, Pasick and Barker2009: 62S) describe that a social practice lens is important in investigating health and social behaviours because it recognises that behaviours do not occur in isolation but are a product of (and contribute to and alter) the social contexts within which those behaviours occur. Blue et al. (Reference Blue, Shove, Carmona and Kelly2016) recommended that a social practice lens is useful in developing public health policy by focusing interventions on the collective rather than on individual behaviour change. Social Practice Theory has been utilised to understand public health issues via health practices which emerge from the actions and interactions of people (Cohn, Reference Cohn2014), and has been utilised to identify public health responses to social practices such as substance use (Keane et al., Reference Keane, Weier, Fraser and Gartner2017; Supski et al., Reference Supski, Lindsay and Tanner2017; MacLean et al., Reference MacLean, Savic, Pennay, Dwyer, Stanesby and Wilkinson2018).
To use Social Practice Theory to help to understand older adults' gambling practices, consider the gambling activity, bingo. To play bingo the following are required: a venue, number cards, and a way to draw or mark off numbers (materials); a knowledge of how to play the game (competence); and an expectation of some form of benefit, such as financial or social gain (meaning). The social practice is the analytical unit (Hargreaves, Reference Hargreaves2011), which redirects the focus from individual decision making to the reproduction and transformation of social practices (Shove and Warde, Reference Shove and Warde1998). Social practices may be a site for both excessive (Supski et al., Reference Supski, Lindsay and Tanner2017) and inconspicuous consumption (Shove and Warde, Reference Shove and Warde1998). By examining social practices, researchers may be able to understand how best to change patterns of consumption, and understand the implications of patterns of consumption and the institutions and infrastructures associated with consumption (Hargreaves, Reference Hargreaves2011; Shove et al., Reference Shove, Pantzar and Watson2012).
To understand older adults' gambling as a social practice it is necessary to establish how the gambling practices of older adults are socially constructed, and how practitioners are recruited to and influenced by these practices. In addition, it is necessary to determine if social practices have their own normative behaviours that influence the practices (Supski et al., Reference Supski, Lindsay and Tanner2017). Investigating the gambling practices of older adults requires a focus on the practice as the unit of analysis, rather than any particular behaviours. This moves away from the individualised focus on gambling behaviours that has dominated gambling research (Miller et al., Reference Miller, Thomas, Robinson and Daube2014; Hancock and Smith, Reference Hancock and Smith2017).
The present paper aims to understand older adults' gambling by developing a model of the social practices that may influence their participation in gambling activities. Focusing on older adults with no, low or moderate risks associated with gambling, this study aimed to answer the following research questions:
(1) How do the social practices of older adults influence their participation in gambling?
(2) What is the relationship between agency and structure in the development of older adults' gambling practices?
(3) How do social practices interact with broader factors to contribute to routine gambling behaviours?
(4) What are the key lessons for initiatives aimed at preventing and disrupting routine or regular gambling participation among older adults?
Methods
Approach
The data analysed for this paper were part of a broader qualitative study which aimed to investigate the normalisation of gambling in adults aged 55 and over in Australia. The broader study investigated four key themes: gambling attitudes and behaviours; the normalisation of gambling, attitudes towards gambling harm; and strategies to reduce the normalisation of gambling.
A Constructivist Grounded Theory (CGT) approach was used to guide the study (Charmaz, Reference Charmaz2014). This approach was used because CGT aligns with a critical qualitative approach to inquiry (Charmaz, Reference Charmaz2017), and assumes there are multiple ways to understand reality, and that this reality is shaped by the influence of social contexts and the social interactions associated with the everyday experiences of individuals or groups (Charmaz, Reference Charmaz2014). For the constructivist researcher, there is no one objective truth to understanding the research issue and there are many different views of the reality presented (Houser, Reference Houser2015; Liamputtong, Reference Liamputtong2017). CGT provides a position from which to learn about the world, and a method to develop theory to understand the focus of investigation (Charmaz, Reference Charmaz2014). This theory is constructed by a systematic approach that is inductive, iterative and involves a constant comparative method (Maykut and Morehouse, Reference Maykut and Morehouse1994). Throughout this process, the researchers were flexible and responsive to the data as it emerged throughout the investigation (Charmaz, Reference Charmaz and Charmaz2006). Utilising CGT allowed the researchers to develop an understanding of the context of gambling for older adults. It also allowed the researchers to articulate and integrate the perspectives of participants, and to examine their views in detail as required (Charmaz, Reference Charmaz2014).
Approval for this study was obtained from the Deakin University Human Research Ethics Committee (2019-354) prior to fieldwork commencing.
Sampling and recruitment
Participants were invited to participate in the study if they were aged 55 years and over, lived in Australia, spoke English and had participated in any form of gambling within the last 12 months. Adults aged 55 years and over were considered to be older adults to ensure consistency with the age ranges associated with previous studies on older adults (Tse et al., Reference Tse, Hong and Ng2013; Thomas et al., Reference Thomas, Pitt, Randle, Balandin, Cowlishaw, McCarthy, Bestman and Daube2020). This age range also allowed the research team to explore the gambling of older adults pre- and post-retirement. A range of sampling approaches were used. First, convenience sampling was used to gain access to participants via community and online settings across different jurisdictions in Australia. Once interviews commenced, snowball sampling was used whereby participants were asked to refer potential participants who met the criteria to join the study. Commensurate with CGT approaches, data interpretation occurred throughout the study. This contributed to purposive sampling of participants, to ensure that the sample reflected a diversity of ages and gambling experiences.
Recruitment flyers were distributed to the research team's existing networks, through local councils and community organisations (e.g. Men's Shed, community centres, libraries), online via Twitter and Facebook, and through the team's database of individuals who had agreed to be contacted about research studies on gambling. Potential participants contacted the research team via email or phone, were provided with a copy of a Plain Language Statement and Consent Form, and were given the opportunity to ask questions before they gave consent to participate. Prior to being invited to participate, individuals were asked screening questions about their age and gambling behaviours to ensure that they met the inclusion criteria.
Data collection
Data were collected via semi-structured telephone interviews between July 2020 and January 2021. Interviews lasted from 45 to 90 minutes. To ensure consistency in the interview process, all interviews were conducted by RHJ. Most of the participants were interviewed alone, while two couples were interviewed together. The interviews were audio-recorded with the participants' permission and transcribed verbatim by a professional transcription company.
During the data collection process, a semi-structured interview guide was used to assist the data collection.
A range of demographic and gambling data were collected, including age, occupation, education and postcode, to determine participants' socio-economic status (Australian Bureau of Statistics, 2016). The Problem Gambling Severity Index (PGSI) was used to measure at-risk behaviour in problem gambling (Ferris and Wynne, Reference Ferris and Wynne2001). The PGSI is a standardised measure that includes nine questions about individuals' gambling behaviour over the past 12 months, including questions such as ‘Have you bet more than you could afford to lose?’ and ‘Have you felt guilty about the way you gamble or what happens when you gamble?’ The response options were never = 0, sometimes = 1, most of the time = 2 and always = 3, with each option attracting the score indicated. The scores were summed, and individuals were categorised as either non-problem (score 0), low risk (1–2), moderate risk (3–7) or problem gamblers (8–27).
A range of open-ended questions were used to explore gambling practices and experiences. These included asking older adults to describe their gambling experiences, the social practices that surrounded these experiences, the range of influences on their gambling and how their gambling had changed over time. RHJ engaged in the co-creation of data with the participants, allowing participants to raise issues that they felt were important to them. This allowed for an in-depth examination of the participants' experiences, beliefs and opinions about gambling. As such, the data collection process itself was a construction where the research reality was co-constructed by both the participant and the researcher (Charmaz, Reference Charmaz2014).
After completion of the interview, participants were sent a Aus $50 grocery voucher as a token of appreciation for their time. All identifying data were removed from transcripts prior to analysis to maintain the confidentiality of participant identities.
Data analysis
The interpretation of qualitative data was managed in NVivo 11 (QSR International, Melbourne). An inductive and reflexive thematic analysis approach was taken to analyse the data (Braun and Clarke, Reference Braun and Clarke2020). This involved reviewing the content of the interview transcripts to construct themes from the data (Braun and Clarke, Reference Braun and Clarke2022). The data analysis process is outlined in Table 1. The aim of the interpretation was to provide a thick and detailed description of participants' experiences (Denzin, Reference Denzin1989).
Data analysis process (Braun and Clarke, Reference Braun and Clarke2022)
Reflexivity process
Reflexivity was the main process used to ensure the rigour and trustworthiness of the data. To enable reflexivity to occur during the data analysis, weekly team meetings were held. During these meetings, the researchers reflected critically on the research, the data being generated and the role of the researchers in the data interpretation (Green and Thorogood, Reference Green and Thorogood2018). The research team discussed and reflected upon the ideas that were generated from the data. This involved discussions about the interpretation of the data, creation of categories and eventually themes. The processes that were followed during the team meetings allowed for the development of alternative interpretations and ensured that the interpretations of the data were credible (Green and Thorogood, Reference Green and Thorogood2018). Social Practice Theory was selected as an appropriate theory to help understand and explain the social processes that were occurring for older adults during their gambling participation. Social Practice Theory enabled the researchers to consider a wider context of factors that influence older adults' gambling, and how older adults' participation in gambling influenced the social practice. Sampling concluded when the data were able to demonstrate a range of concepts relating to older adults and their gambling behaviour, attitudes, perceptions of the normalisation of gambling and strategies to respond to this.
Results
Sample characteristics
The sample characteristics are shown in Table 2. The sample consisted of 40 participants, with an age range of 55–78 years (mean = 65.7 years, standard deviation = 6.7). The sample was skewed slightly towards female participants (N = 23, 57%), with about two-thirds currently retired (N = 26, 65%). The sample was highly educated, with just under half of the participants having completed an undergraduate or postgraduate degree at university (N = 19, 48%). Most participants lived in one of Australia's two largest states, Victoria (N = 21, 53%) or New South Wales (N = 15, 37%). Most (N = 35, 87%) lived in areas of high or moderate socio-economic advantage (SEIFA (Socio-Economic Indexes for Areas) score of 4 or above). Nearly all (N = 38, 95%) of the participants had gambled on at least two products in the previous 12 months, and about two-thirds (N = 25, 63%) scored from 1 to 4 on the PGSI, indicating that they were at low to moderate risk of problem gambling. Figure 1 shows the number of participants who engaged with the different types of gambling products.
Number of participants who gambled monthly or more, per gambling product.
Note: N = 40. * Total more than sample size as participants gambled on multiple products. EGMs: electronic gambling machines.
Socio-demographic and gambling characteristics of the sample
Note: 1. Total more than sample size as participants gambled on multiple products.
Theme 1: The role of gambling practices in meeting older adults' social needs
Participants' descriptions of gambling practices demonstrated that gambling was embedded in their social group activities. Participants who visited gambling venues reported engaging in a suite of activities while they were there. These included socialising while engaging in a range of gambling activities, such as EGMs, Keno and raffles, and during the consumption of food and beverages. While the frequency of participation in gambling varied, participants stated that their gambling patterns and practices were prompted by the gambling behaviours of their peer groups. For some participants, peer group-based gambling was ad hoc and incidental, e.g. gambling on Keno after a meal together. For others, gambling was a planned part of their outing with their friends:
we have one couple who are members of a country racing club and they've taken us along to the country horse races a few times. We have a very nice meal with the members and put a few dollars on the bookies. (Male, age 73, moderate-risk gambler (PGSI 3))
Peer influences ranged from encouraging occasional gambling to influencing individuals to become part of horse-racing syndicates. Some participants specifically engaged in gambling due to the influence of their partner. Women in particular described that if it were not for their husband's or partner's gambling interests they would rarely participate in gambling:
I guess the horse-racing thing now, that's only been in the last couple of years and that's because I've been with my husband now for 12 years. He comes from a really strong background of enjoying the horse racing. So it's only been in the last couple of years that I've really got into that. If he wasn't interested in it, I wouldn't be bothered with it. (Female, age 56, non-problem gambler (PGSI 0))
Other participants described that gambling was something they participated in as part of holidays with family members and friends. There were a range of unique behaviours associated with gambling during holiday trips. The following participant described how pooling money for horse racing on a recent holiday with family members created a sense of group belonging and bonding during their holiday. She used colloquial gambling terminology such as the word ‘quaddie’ to describe her engagement with gambling and reported feelings of guilt around gambling. She described how gambling created fun for the family group and created a point of discussion for them:
We've just been away for ten days and it was with my sister and her husband … I actually have gambled because my husband said, ‘do you want to take a quaddie each day for the next four days?’ So we all put $70 in each and that was to cover four days where we took quaddies between us. It was just a bit of a fun social thing to do to – so it creates a bit of discussion and things like that. Supposed to be a bit of fun. (Female, age 58, low-risk gambler (PGSI 1))
Infrequent gamblers stated that they gambled with family members or friends in order to maintain social connections and achieve a sense of belonging, being sociable or fitting in with their social group. For example, some participants indicated they were invited to attend horse-racing events with their friends and participated in gambling while they were there, but outside this they very rarely engaged in gambling. Another participant described gambling on EGMs or playing bingo while away visiting extended family members as she ‘was just trying to be sociable’, while another participant went to EGM venues to keep her mother company. Another participant described that she was not interested in playing the EGMs but still participated with her family after attending a club for lunch. The following participant described her gambling as ‘a flutter’, and sought to downplay any risks associated with her participation, particularly by describing that she was not enthusiastic about gambling on EGMs:
And then maybe once in a blue moon, I have family in the country, and we normally go to the local club for lunch to save everybody doing the dishes, and there are poker machines there. I'm not really wrapped in them but sometimes I have a go … So I'd probably say once or twice a year, have a flutter. (Female, age 73, non-problem gambler (PGSI 0))
The other social group that influenced gambling practices were work colleagues. Some participants explained that they participated in workplace gambling, including sweeps for the Melbourne Cup, workplace lottery syndicates and football tipping competitions. A few described that engagement in gambling within the workplace was part of demonstrating their commitment to their workplace and their colleagues, as well as their sense of belonging within the workplace. For example, participation in gambling activities enabled older adults to contribute to social interactions and discussions in the workplace, with some stating that gambling was something that created conversation and a sense of solidarity with colleagues. One participant described that syndicate gambling with work colleagues was an ‘intellectual exercise’, where individual participants pooled money together and gambled as a collective:
So, we would do it as an intellectual exercise and bit of social fun to see can we maybe win enough money to go out to dinner as a group, fun stuff. (Female, age 57, low-risk gambler (PGSI 2))
Finally, participants who gambled alone also perceived that their gambling created social connections with others, explaining that gambling was a social outing that led them to interact with other patrons and staff members at the venue. Participants stated that they valued the conversations with other people in the venue, creating a sense of social connection even if they never saw that individual ever again. Using gambling to combat loneliness or negative emotions was a repeated theme in the narratives of some participants:
Facilitator: And when you say it's social, could you tell me more about that, about the types of interactions you might have there?
Interviewee: Well usually people sitting beside you [on the pokies] tend to talk to, especially, I mean I'm 67, I was gonna say elderly people but especially ones that live on their own, go into the pokies as a real social outing and they talk to everyone and interact with everyone. Now that smoking's not allowed in there, they all have somewhere you can go out and have a cigarette and I live on coffee, so I make a coffee and go out and have a cigarette. People are always chatting, don't know them, never see them again but it's, and it's a chance I think for a lot of lonely people to talk to someone. (Female, age 67, moderate-risk gambler (PGSI 4))
However, not all participants perceived that they gambled for social connection, with some stating that they gambled as a skill-based leisure activity. For these participants, gambling was considered a form of entertainment that involved studying the information available (such as reading the form guide associated with horse racing) and making decisions about how to place their bets.
Theme 2: The role of social practices in the development of gambling routines
Routines were a particular feature of participants' gambling practices. Participants discussed the range of routines that they had in relation to gambling, including that they would gamble only on particular days of the week or during different times of the year. Others described that they would engage with a gambling product in a particular manner, such as choosing the same type of EGM every time or purchasing the same type of lottery ticket. Gambling routines were primarily associated with social group activities, but were also associated with the accessibility of gambling in everyday environments, including regular non-gambling behaviours such as going to the supermarket. Participants were often very specific about the days of the week that they would engage in gambling. They often recalled these days because they were linked with other routine social practices. The following participant described the routine social practices related to gambling for this social group, which involved playing golf on particular days of the week, and then ended up at the club to have a drink, watch the races and gamble:
We'll play golf on Wednesdays and Saturdays. And come into the bowling club in town, they have a TAB machine, and we use that … there's probably a group of about eight of us. We all come back there and have a drink and watch the races. (Male, age 70, non-problem gambler (PGSI 0))
Participants regularly visited gambling venues for a few hours at a time, often a number of times per week, and took part in a range of activities. The suite of gambling and non-gambling activities offered by venues across different days and nights of the week repeatedly drew participants back to the venue across the week. The suite of activities influenced the building of routines across the week. Participants were drawn to the venue for an initial non-gambling purpose, which was then followed by a gambling activity. For example, participants indicated that they would regularly attend the local club for one activity (to play trivia) which would then be followed by a gambling activity, such as gambling on EGMs. For example, the following participant described the local club was where her social interactions occurred multiple times per week, and often revolved around gambling:
There's bingo, there's a group of four or five of us that go together. I leave when bingo's finished. We go another night, there's a meat raffle on the Friday night. I used to go to Trivia there too on a Tuesday night, Trivia's not on and the bingo's not on, so and no meat raffle on at the moment, that's when we've been playing the Keno. We'll sit there and we'll play Keno while we sit there and have a talk and that. I usually get there about 5.30 and I usually leave about 7. (Female, age 73, low-risk gambler (PGSI 1))
Another participant described the regular visits of a family member, attending the bowling club for dinner and to gamble on EGMs:
My sister used to come over one night a week and stay over and then we'd go somewhere for dinner that night and then we'd do all the shops the next day. So one of our options was the local bowls club and they've got the pokies. So we'd have dinner and then we'd go and play the pokies. The other thing about the local bowls club is if you go down to where they've got the pokies the tea and coffee is free. (Female, age 56, low-risk gambler (PGSI 2))
Many of the routines associated with gambling were longstanding, and had developed over many years with friends, family members or on their own. For some participants, the routine nature of their gambling had developed since they retired due to having less to do, while for others gambling routines had been in place for a large proportion of their adult lives. For example, one participant had a routine of gambling with a friend every Saturday at a betting shop. This had evolved from having a social routine over the past 40 years with a friendship group. It began with attending football matches together every week and placing bets at the betting shop prior to the football, to the weekly routine of catching up, having coffee and gambling:
Every Saturday, I catch up with the same mate of about … 40 years. For 40 years, we've either been doing a mix of going to the footy on a Saturday via [the betting shop] or in more recent times just go to the [betting shop]. (Male, age 60, non-problem gambler (PGSI 0))
The most routinised gambling behaviour was associated with participation in the lottery. Participants regularly purchased lottery tickets as part of their everyday activities. Lotteriy tickets were most regularly purchased when going to the supermarket for groceries, where the newsagents (which sold lottery tickets) was readily accessible. For example, the following participant described shopping for her elderly mother and during that process walked past the newsagents and purchased the lottery ticket:
I do go get my mum's shopping because she's elderly. So it's a bit of a little routine type of thing, I've got to go to Aldi first and then walk past the newsagents and go to Woolworths, so it's real regular. So I'll drop into the newsagents [to buy a lottery ticket]. (Female, age 58, non-problem gambler (PGSI 0))
Most participants who engaged in the lottery reported regular routines associated with the purchasing of tickets. Some participants would play the same numbers every week and these numbers would often have a significant meaning to the individual, such as family members' birthdays. Participants often described ‘systems’ associated with their lottery practices:
I just buy the same tickets every two weeks and one has got – one is just a system 7 and it's got the kids ages and our ages on it. So that sort of changes because the kids get older and now I've got grandkids. But I usually buy a system 7 and a normal 12 game quick pick. (Female, age 58, low-risk gambler (PGSI 1))
Many participants explained that routines were disrupted when the lottery jackpot amount was higher. This was the only time that routines were altered, diverting participants from their normal purchasing behaviours to buying more tickets. Other participants would purchase tickets with family members or friends – another example of the role of social groups and connection in gambling practices for participants. For example, one participant mentioned that every six weeks her friend would buy their shared tickets, and that they would often have coffee together at this time:
So, we've been doing this for years. And hopefully we can win millions but we never do, so, what … she's the one that does it all. I give her enough money to buy six weeks of tickets, which is … we only have Saturday nights because they're the big ones. And she buys it. And then she'll either drop a copy of the ticket off or email me a copy of the ticket. Yeah, she'll just ring and say ‘I've got the ticket, I'll pop in and have a coffee’. (Female, age 78, non-problem gambler (PGSI 0))
Finally, gambling routines and rituals were associated with particular special events or seasons – especially related to sport or horse racing. For example, one participant described placing bets together with a family member throughout the football season and on the horses during the Spring Racing Carnival. Sometimes this gambling involved a social get-together with friends and family or communicating on the phone while watching the sport separately:
I'll bet in the football season. I'll put on bets with my brother just for something to do and pass the time and see how we go. For multi bets, you know, you don't need to lay out much and if you get them, you win quite a bit and we tend to win most of them anyway. I'll bet on the horses but only during the Spring Carnival. So usually from say September to the middle of November, yeah, we're doing multis and a few things and again, with my brother and friends and we get together and pool and even then, we still do pretty good most years as well. That's the extent of my gambling. (Male, age 62, non-problem gambler (PGSI 0))
The routines described by the older adults centred around the social aspect of gambling, whereby the gambling activity was seen as a way to bring established groups together. Many participants explained that they gambled on the Melbourne Cup but it was their only form of gambling on horse racing. Participants described social events that had become a regular activity for them each Melbourne Cup or around other local horse-racing events. These included going out for lunch with friends to watch the race or going to place their bets with their partner on the day of the Melbourne Cup. Other participants only gambled during particular special events or seasons, such as gambling on the horses during the Spring Racing Carnival or gambling on football during the football season.
Discussion
This paper aimed to provide in-depth qualitative data relating to the range of social practices that may lead to older adults' participation in gambling. Figure 2 shows a proposed descriptive model of factors that may influence the gambling practices of older adults, how these may lead to the development of regular gambling routines and the range of strategies that may be used to respond. This proposed model could be used to guide future studies which seek to explore the range of factors that may influence or prevent regular and routine gambling behaviours in different population subgroups of older adults. The model is useful in that it is developed from data relating to a range of recreational practices associated with different gambling products, and may be useful in helping to provide insights for public health and health promotion activities which seek to prevent or reduce regular or routine gambling behaviours.
A proposed model of older adults' gambling-related social practices and avenues for action to prevent risk.
This study found that a range of socio-cultural determinants impacted older adults' gambling and their motivations to gamble. While a few participants described that gambling was a skill-based leisure activity, most participants used gambling as way of bringing people together, developing and maintaining social connections, and creating a shared sense of group belonging. There was a shared meaning of social connection through participation in gambling across a variety of different social practices, with strong social drivers involved in these practices. These results support previous literature that has demonstrated that older adults engage in gambling or attended gambling venues to fulfil a need for social connection (Hagen et al., Reference Hagen, Nixon and Solowoniuk2006; Ohtsuka and Chan, Reference Ohtsuka and Chan2014; Thomas et al., Reference Thomas, Pitt, Randle, Balandin, Cowlishaw, McCarthy, Bestman and Daube2020). While some studies have suggested that the need for social connectedness and group belonging is a particular driver of women's gambling participation (Thomas et al., Reference Thomas, Pitt, Randle, Balandin, Cowlishaw, McCarthy, Bestman and Daube2020; McCarthy et al., Reference McCarthy, Thomas, Pitt and Bellringer2021), these findings suggest that the gambling practices of older men may also be influenced by broader social group dynamics which may normalise gambling as a routine social activity within their peer groups. Importantly gambling facilitated a sense of social belonging across a range of different social settings, including recreational settings and the workplace. While older adults may perceive that engaging in gambling is providing benefits for their health and wellbeing, we would argue that the normalisation of regular patterns of gambling could also be contributing to their risks of gambling-related harm. As such, it is important for future research to understand how social patterns of gambling may interact with broader environment and/or commercial determinants to influence transitions from social to harmful gambling (Clarke et al., Reference Clarke, Tse, Abbott, Townsend, Kingi and Manaia2006). The study of Venuleo et al. (Reference Venuleo, Marinaci and Mossi2021) of older adults in Italy also stressed the need to focus on socio-cultural and structural factors in understanding and responding to the potential for gambling harm in older adults, with moderate risk and problem gamblers in their study more likely to be lonelier, with lower levels of perceived social support. Previous research has demonstrated that diminishing social networks as people age may make older adults vulnerable to gambling-related harm (Strain et al., Reference Strain, Grabusic, Searle and Dunn2002). There might be some suggestion in this study that a shift of focus towards gambling activities may indicate a transition through the risk continuum. Future research should explore how the interplay between socio-cultural, environmental and commercial determinants may influence transitions through the risk continuum for older adults who gamble. Public health and health promotion interventions that seek to develop socially supportive activities in which older adults are able to connect away from gambling venues may be important in denormalising regular and routinised gambling behaviours.
This study also shows that gambling practices may be influenced by both perceptions of individual autonomy and control (agency) and the structures surrounding them. This includes the influence of social groups, the placement of gambling outlets in everyday spaces and factors that prompted regular social group attendance at gambling venues. There was a clear interplay between social drivers and structural influences, including the role of the availability and accessibility of gambling products in community settings. Previous research in Australia has suggested that co-location of non-gambling activities with gambling activities may lead to increased exposure and participation in gambling (Bestman et al., Reference Bestman, Thomas, Randle, Pitt, Daube and Pettigrew2016). This suggests that a combination of individual, social and environmental determinants of older adults' gambling may contribute to gambling becoming increasingly normalised for older adults, with some participants describing that gambling had become a normal part of everyday routines. As Burke et al. (Reference Burke, Joesph, Pasick and Barker2009: 62S) highlight, ‘no behaviour occurs in isolation from its immediate and broad context’. This study shows the importance of focusing on the broad contexts of social practices, including exploring how older adults' need for social connection may interact to influence regular gambling participation.
While the social drivers of gambling varied for different participants, a range of social processes and contextual factors influenced the routine nature of their gambling. These were primarily linked to the behaviours and expectations of social group members. Older adults could clearly describe the social dynamics associated with their gambling, justified their gambling as being a social activity, and did not acknowledge that this influence may be problematic or potentially harmful – even when scoring at low or moderate risk of gambling-related harm. Understanding the interaction between perceived agency and the influence of structural factors may have a number of implications for gambling harm prevention interventions for older adults. While many participants in this study perceived that they were responsible or in control of their decision making relating to gambling, they also described the significant influences of social structures in the routines associated with their gambling. The findings highlight that individual responsibility paradigms (such as being a ‘responsible gambler’) may be less-effective prevention strategies than those that seek to address the social norms that may be constructed around gambling. This supports research which has recommended developing a range of alternative activities for older adults that do not inherently link to gambling (Thomas et al., Reference Thomas, Pitt, Randle, Balandin, Cowlishaw, McCarthy, Bestman and Daube2020).
Finally, this study demonstrated how participating in gambling leads to the reproduction and transformation of gambling practices. People are practitioners or carriers of social practices and are recruited to the social practice when they enact it (Reckwitz, Reference Reckwitz2002; Shove et al., Reference Shove, Pantzar and Watson2012). Older adults themselves (un)intentionally become the conduits for gambling in other older adults, particularly when other older adults were recruited to these practices. There were a range of social norms and expectations of those that participated in gambling activities, particularly when older adults participated in gambling social practices established by their family members, such as incidental gambling with their partners or routinised gambling with family members. For example, older adults described having a regular routine of attending gambling venues for non-gambling social activities, but there was also an unspoken expectation that this would be accompanied with engagement in gambling activities. As demonstrated in studies which have focused on the factors that may create a pathway to gambling venues (Tira et al., Reference Tira, Jackson and Tomnay2013; Bestman et al., Reference Bestman, Thomas, Randle, Pitt, Daube and Pettigrew2016), venues provided a convenient community space for recreation, social connection (even when older adults were by themselves) and meeting unmet needs, and drew some older adults back to these venues repeatedly throughout the week. This demonstrates how the interaction between a range of socio-cultural, environmental and commercial determinants may lead to regular and routinised participation in gambling for older adults. As has been highlighted by others in relation to the gambling practices of older adults (Venuleo et al., Reference Venuleo, Marinaci and Mossi2021; McCarthy et al., Reference McCarthy, Pitt, Bellringer and Thomas2022), this study suggests that public health prevention interventions should seek to understand and address the social practices of older adults' gambling, and the range of structural factors that may influence these practices.
Limitations
The majority of participants in this study lived in areas with moderate to high socio-economic advantage. Given that there are a range of well-recognised inequities in gambling-related harm (Goyder et al., Reference Goyder, Blank, Baxter and van Schalkwyk2020; Raybould et al., Reference Raybould, Larkin and Tunney2021), with economic and social factors linked to older adults' experiences of gambling-related harm (Guillou Landreat et al., Reference Guillou Landreat, Cholet, Grail Bronnec, Lalande and Le Reste2019; van der Maas et al., Reference van der Maas, Nower, Matheson, Turner and Mann2021), there is a need for critical qualitative inquiries that specifically investigate how older adults' experiences of social disadvantage may influence their gambling practices. It should also be noted that the participants from this study gambled in the Australian states that have EGMs within community venues. In one state, Western Australia, EGMs are limited to casino venues only. Consequently, older adults in these locations may have experiences that differ from the study participants due to the gambling environment.
Conclusion
This study demonstrates that the availability of gambling and how gambling is linked to broader social contexts may shape the gambling practices of older adults. For participants in this study, gambling was a way of consolidating social connections and strengthening their sense of belonging within social groups. Rather than focusing on individual determinants and behaviours, future research should seek to extend the model proposed in Figure 2 by continuing to develop an understanding of the interplay between the determinants of older adults' gambling and extending the model to specific groups of older adults, such as culturally and linguistically diverse groups. This may lead to developing interventions that disrupt normalised and routine gambling behaviours, provide alternative non-gambling spaces for social connection and belonging, and acknowledging the interplay between individual agency and social and commercial structures and how they may impact older adults' gambling. Ensuring these factors are addressed will be important in responding to the potential risks posed by gambling to older adults.
Acknowledgements
We would like to thank the older adults who participated in this study.
Author contributions
RHJ is a PhD student researcher who led the study conceptualisation, data collection, data analysis, manuscript drafting, and critical revisions of the manuscript. HP and MR are study investigators who contributed to the study design, data interpretation, manuscript drafting and critical revision of the manuscript. SLT is the study lead chief investigator who contributed to the conceptualisation of the research, and participated in the design of the project, data interpretation and critical revision of the paper. All authors have read and agreed to the published version of the paper.
Financial support
This work was supported by an Australian Research Council Discovery Grant (DP190100695), investigating the normalisation of gambling in Australia. The funders had no role in the design of the study; in the collection, analyses or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.
Conflict of interest
RHJ receives a PhD stipend funded by the Australian postgraduate research scheme. HP has received funding from the Australian Research Council Discovery Grant Scheme, VicHealth, Deakin University, the Victorian Responsible Gambling Foundation and the NSW Office of Responsible Gambling. MR has received funding for gambling research from the Victorian Responsible Gambling Foundation and the Australian Research Council Discovery Grant Scheme. SLT has received funding for gambling research from the Australian Research Council Discovery Grant Scheme, the Victorian Responsible Gambling Foundation, Healthway WA and the New South Wales Office of Responsible Gambling.
Ethical standards
Ethics approval was obtained from the Deakin University Human Research Ethics Committee (2019-354).
Open access
