How do people drink alcohol at a low-risk level?

ABSTRACT Reducing the risks associated with drinking is an ongoing public health goal. Approximately two-fifths of Australian adults consume alcohol within low-risk guidelines, yet little is known about their drinking patterns or practices. In this paper, we use social practice theory to consider low-risk drinking at home as a routinised social practice with material, meaning and competence dimensions. We analysed open-text survey responses from 252 Australian adults (30–65, 89% female) who were considered low-risk drinkers. A low-risk drinking occasion was typically closely linked to other practices such as eating dinner or connecting with family or friends. Drinking alcohol, even in small amounts, was associated with enjoyment. Being attuned to bodily sensations and applying some self-imposed rules were competencies that allowed low-risk drinkers to avoid intoxication. Low-risk drinking practices entail some elements that can inform health promotion, including encouraging efforts to limit drinking to times of the day (e.g. during meals) and to attend to bodily feelings of sufficiency. The study also shows how low-risk drinking is entangled with gendered and age-related norms about drinking, and facilitated by rarely being in ‘intoxigenic’ environments. These factors are imbricated with individual decisions in our respondents’ capacity to consume alcohol moderately.


Introduction
The potential harmful consequences of episodic and regular heavy alcohol use are well established (Rehm et al., 2010).Regardless, alcohol's cultural currency remains intact.In Australia, alcohol is the most prevalent psychoactive drug consumed, with 79.1% of adults (i.e.18 years and older) consuming alcohol in the past year (Australian Institute of Health and Welfare, 2020).Recent population statistics also report that a majority of the drinkers, constituting 41% of Australian adults did not consume five or more Australian standard drinks (one ASD contains 10 g ethanol) on any occasion in the past year (Australian Institute of Health and Welfare, 2020).Twice as many women as men (56% vs. 28%) were in this category, which we defined here as 'low-risk' drinkers, as it is within the 2009 national drinking guidelines for reducing the risk of short-term harm arising from a single drinking occasion (e.g.injury) (National Health Medical Research Council, 2009).
Drinking guidelines, generally referred to as 'low-risk guidelines' by the public health sector (e.g.Stockwell et al., 2012), are a clear example of neoliberal interests shaping public policy (Maddison & Denniss, 2013).As a health intervention, low-risk drinking guidelines operate on the assumption that if individuals are aware of, and accept the evidence-based expert advice, they will in turn make informed and healthier choices about their alcohol consumption.This places the onus on individuals to modify their alcohol consumption (Lindsay, 2010).Aligned with this, the alcohol industry's pervasive 'drink responsibly' narrative is deliberately vague as to how responsible drinking might be done.
Rather than seeing alcohol use as the outcome of a set of unique decisions by individuals, this paper aims to better understand drinking from the viewpoint that alcohol use, even lower-level alcohol use, is a cultural or social practice that is enacted by groups of people.It is therefore sustained through its integration into social life, rather than simply by individual decisions to drink in this way.Social practice theory (SPT), with its focus on materials, meaning and competence (Meier et al., 2017;Shove et al., 2012) provides a useful framework to show how restrictions on alcohol use are knitted into each of these social practice elements.SPT is increasingly used in studies examining health behaviours such as alcohol use (Hennell et al., 2021;Supski et al., 2017), and smoking and vaping (Keane et al., 2017;McQuoid et al., 2020).While many SPT studies draw their data from interviews, responses to open ended survey questions have also been used (e.g.Keane et al., 2017;MacLean et al., 2019).
Practice frameworks differ from the approach that is most frequently taken to analyse drinking, which is strongly influenced by social cognition and rational choice approaches (Meier et al., 2017).This tendency is evident in a review by Muhlack et al. (2018) which describes 13 qualitative studies that included data on middle-aged adults (30-65) whose drinking was considered 'non-problematised'.The authors operationalised this based on participant's narratives of their day-to-day experiences rather than asking about actual amounts of drinking (Muhlack et al., 2018).None of the reviewed studies explored alcohol consumption only among adults who drank within country-specific drinking guidelines; rather, the cohorts included a mix of low-risk and heavier drinkers.Except for two studies (Brierley-Jones et al., 2014;Holloway et al., 2008), alcohol consumption was predominately discussed using a social cognition and rational choice lens, with less attention to the routinised social practices embedded in and surrounding alcohol consumption.
Recent SPT studies of alcohol use have focused on heavy drinking occasions among atrisk groups such as young people (Hennell et al., 2020;Supski et al., 2017), sexual minorities (MacLean et al., 2019), and groups of men (Roberts et al., 2019) and women (Wright et al., 2022).A recent study employed a practice approach to explore drinking at home for heavier and lighter consumers (MacLean et al., 2022), finding that being at home afforded a sense of cosiness and safety, and reduced barriers to intoxication, for heavier drinkers.Each of these studies articulated how social, cultural, material, and embodied forces shape and perpetuate drinking, extending the analysis beyond individual choices and behaviour.Comparatively few studies, and none informed by SPT, have focused on adults who drink within national endorsed guidelines about alcohol consumption.
We build on these studies by applying a SPT lens to examine the elements of a low-risk drinking occasion among Australian adult drinkers who consumed no more than four standards drinks on an occasion in the last year.Our intention in this paper is to develop knowledge on the social dynamics of low-risk drinking occasions and factors that sustain and reproduce this drinking style.These factors include both individual decisions about drinking that, as we show, are entangled with the elements of this social practice.This is important because insights into the social dynamics of low-risk drinking occasions may help shape public policy and health messages encouraging heavier drinking Australians to reduce their consumption.It also illuminates the social practice of alcohol use for a sizeable proportion of Australian drinkers who have been largely ignored by researchers.

Data collection
This study is based on data drawn from a convenience sample of Australians who completed an online survey about drinking alcohol at home.The survey was advertised on Facebook in late 2018.Eligible respondents were Australian residents aged 30 to 65 years who reported drinking at least weekly.Respondents who completed the survey were eligible to enter a prize draw to win a $50 grocery voucher.In total, 2101 met the eligibility criteria and completed the survey.Of these, 252 reported that they had not consumed five or more ASD on an occasion in the previous year and answered a series of open-ended questions.Most of these respondents were women (89%) and the mean age of respondents was 52 years.The low-risk drinkers were mostly fairly affluent, with 43% with a household income over $110,000 AUD and a further 29% over $65,000 AUD.Ethical approval for the survey was granted by La Trobe University's Ethics Committee.
We report here on this sub-group's written responses to five open-ended questions designed to elicit discussion about the place of alcohol use in everyday life for low-risk drinkers.These entailed describing the context and alcohol consumed during the last occasion when they drank alcohol, why and how they drink at a low-risk level and difficulties and advantages associated with this.
Respondents' quotes are used to illustrate the elements of practices identified across the sample.In doing so we indicate whether the respondent is male or female (M or F), their age range (e.g.40s) and the respondent number ascribed to them in the dataset.

Social practice theory
As noted above, because we regard alcohol consumption as a socially embedded practice rather than a set of individual decisions, we turned to SPT (Shove et al., 2012) to analyse our data.In SPT, the unit of analysis is the practice itself, and particularly how it is reproduced and has changed over time.Individuals are considered as carriers or hosts of a practice, rather than being the main focus.
SPT entails a 'focus on the configuration of elements that establish [a practice] as a normal or necessary thing to do' (Blue et al., 2016, p. 45).SPT theorists (e.g.Shove et al., 2012) provide specific advice on analysing data concerning social practices, including identifying meaning, material and competence aspects.'Meaning' refers to the social and symbolic significance attached to a practice; 'material' to the settings where it occurs and the objects or consumer goods that are used, and 'competence' to the sets of skills and practical know-how that are part of enacting it.In this paper, competence emerges as particularly important, as such, we consider the skills and strategies that people use to limit (whether deliberately or not) their alcohol consumption.
Central also to SPT is the notion that practices such as low-risk drinking are not isolated or discrete (Shove et al., 2012).For example, low-risk drinking is practiced by people who also drink at higher levels, although in this paper we focus on people who consistently drink at a low level, rather than those who shift between high-and lowlevel consumption.But even for consistent low-risk drinkers, low-level drinking is part of a network of other practices, including, as we shall explore here, eating dinner, watching television with a partner or housemate, or meeting up with a friend.
The elements of social practicemeaning, material and competenceguided the coding framework we used to analyse the data.We explain in the sections below the data that were coded to each.Sub-elements were identified as JM and SM read the open-text responses from 25 respondents together.JM then coded the remaining cases; however, discussions took place during this stage of the coding to ensure interpretations were consistent.NVivo 12 was used to organise the data into the elements listed above, with child nodes developed to group data in accordance with broad themes that arose within each (Willis, 2012).

Findings
In the sections below we synthesise material, meaning and competence elements of their drinking as described by survey respondents.Consistent with a SPT approach, we also identify the proximate practices that accompany (and can be seen as containing or supporting) low-risk drinking.

Material
We coded comments to the element 'material' when they were concerned with physical quantities of alcohol consumed, the quality or financial cost of the product and the settings such as a person's home where drinking occurred.The embodied pleasure in drinking a consumable product, in this case alcohol, is also discussed in this section.
Respondents' accounts suggested that most consumed 1 to 2 drinks in their most recent drinking occasion.Less common was an occasion with 3-4 drinks and, at the other end of the range, a few respondents had half a glass of alcohol.The predominant beverage was wine, and less often beer or spiritsstraight or with a mixer.Within the same response, most wrote about who they were with, if anyone, and the timing and context of the drinking occasion.These aspects will be discussed in a later section of the paper, but for now the information has been retained in the quotes below as they overlap with the material element of the practice.
We note here that the sample of respondents were drawn from a study of home drinking.As is evident in the quotes below, home is frequently the setting in which low-risk drinking occurs.This is at least in part a function of the older age of our respondents, who are less likely than younger people to find themselves in 'intoxigenic' (McCreanor et al., 2008) drinking places such as nightclubs and bars.For this group, low-level drinking was routinised alongside other home-based practices and engagements with others: A glass of wine mainly with dinner every night.(M 50s #26) Two gin and tonics when my friend popped around for a chat whilst I prepared dinner.(F 40s #859) I usually share a bottle (half each) [which is approximately 3.5-4 ASD] of white wine with my husband on a Friday and Saturday night while we watch a movie.Usually half that on a Sunday.(F 60s #2601) The financial cost of alcohol constrained how much, and how often, respondents drank.Alcohol consumption for these low-risk drinkers was also about aesthetics in the form of the sensory appealmainly taste, but also the texture and aromaof different beverages, rather than quantity.Wine, boutique beer or top-shelf spirits were materials to be appreciated by respondents, who as we noted earlier had an average age in their 50s and were well off compared with the general population.This cost in itself reduced consumption.
Have reached the stage in my life where quality is a meaningful factor in my choice of drink, and that often comes at a much higher price point.(F 60s #2599) I like $$$ champagne so one glass savoured is better than a bottle of cheap stuff.(F 40s # 593) We noted a tendency among respondents to seek a relaxed or mild alcohol-affected state rather than the heavy intoxication that has been associated with more youthful drinking styles (Szmigin et al., 2008) but which is also sought by some older drinkers (MacLean et al., 2022).Most respondents did not elaborate on this sensation, but those who did described modest effects such as a 'gentle buzz' (F 50s #2601), feeling 'slightly tipsy' (F 50s #362) or a 'mellow' feeling (F 60s #1852).This desired state was often reached with one or two drinks.
Two drinks is enough.I don't like the 'drunk' feeling, but like the slightly relaxed feeling it brings.(F 50s #2178) I enjoy the taste of the wine and the little rush it gives me.(F 50s #584) As a practice, drinking alcohol can been seen as a sensuous experience, and an expression of taste, or cultural capital (Bourdieu, 1984).These embodied affects tell us something about the social fieldthat our respondents, in the main, enjoy consuming quality liquor, afforded by savouring a glass of wine (rather than cheap alcohol products).This is a key dimension of the way drinking practices are shaped socially by expectations of appropriate tastes and pleasures for people in middle age, rather than simply being expressions of individual desires.

Meaning
As others have shown, alcohol has many meanings and serves diverse functions.It is a social lubricant, a reward, and a sign of social reciprocity (e.g.Sulkunen, 2002).Discourse around appropriate use of alcohol also invokes moralising views on personal accountability, self-control and social acceptability (Järvinen, 2012).As noted above, middle-aged women were well-represented in our sample of low-level drinkers, and seem often to share values and expectations about what constitutes acceptable alcohol consumption.These discourses shape the meaning of alcohol consumption and contribute to the constitution of low-risk drinking practice.Thus, they were coded to the 'meaning' element.
Consuming alcohol marked the physical or cognitive transition between employment and household responsibilities to times for leisure or self-care.An alcohol drink symbolised a 'treat' or 'reward' for completing an everyday task, or a source of motivation to undertake a task.It also represented a release from dull or taxing activities as suggested by terms such as 'down time', 'wind down' and 'time out' to characterise the low level drinking occasion (Lyons et al., 2014).This temporal framing together with the work-reward dynamic seems to have the effect of constraining the alcohol consumption of respondents.
For many, enjoyment was bundled with the sense of relaxation attributed to alcohol, both figuratively and physiologically, and the pleasant taste of alcohol.It also symbolised a shared moment with family, friends, neighbours or colleagues; a moment to connect and relax, and was generally considered to be largely unproblematic if kept within limits.For those who lived alone, drinking with another person, or in the presence of others, offered temporary companionship.In a formal setting or party context, consuming one or two drinks triggered an increased sense of sociability.Even holding a glass containing alcohol eased small talk and quelled social awkwardness for some respondents.The sense of solidarity engendered through the performance of low-level drinking maintains this practice.
It promotes quality time with my husband and, on occasion, my son.They both have stressful jobs and need opportunities to debrief.(F 60s #268) Enjoying the flavours of a special wine or beer with friends and the good social times associated with it.(F 50s #760) On a social occasion I will sometimes just hold a glass and barely have any of it.No-one notices.(F 40s #1941) The importance evident in these quotes of adhering to the prevailing social norm of consuming alcohol at celebrations or events such as 'after work drinks' was particularly salient for some of the younger respondents.
At times I feel that people talk more freely if they are aware that I am having a drink with them.(F 40s #2116) At work functions everyone has 1 or 2 drinks.You are in the great minority if you are not drinking.(F 30s #2202) Drinking has integrated itself with Australian culture and it's really just expected by a lot of people.Why would they invite you to go for a social drink if you don't actually drink?(F 30s #2227) Respondents also took the time to explain that consuming alcohol provided enjoyment but that becoming drunk was neither the intended nor desired endpoint.Although, as is evident above, many respondents used alcohol to promote sociability, others insisted that alcohol was not needed to authenticate new or existing connections.More explicitly, consuming non-alcoholic beverages with others carried equal social meaning for a few, particularly for some women.
… I can have just as much fun socially without drinking.(F 40s #755) Drinking with my neighbour is a very social thing to do.We sometimes do the same thing but with a cup of tea or coffee.(F 60s #193) As much as low-risk drinking was framed around enjoymentthe taste, relaxed state, and positive personal connectionsequal if not more weight was placed on what it didn't represent: loss of control, drunkenness, intoxication, impairment, addiction.These states, and motivations to avoid them, were largely rationalised from a health perspective; however, some comments moralised alcohol use and heavier drinking.This concern with not drinking to become drunk was clearly also influenced by norms about appropriate performances of gender by the predominantly female low-risk drinkers, cognisant perhaps of the heavy weight of judgement that women who drink immoderately incur (MacLean et al., 2018).In our study, respondents contrasted their drinking with heavier drinking patterns and drunkenness, with this legitimising and affirming drinking at low levels.
I like a beer or wine with a meal, to accelerate relaxation after work and occasionally like to feel a bit silly with friends.Hate being drunk though.(F 50s #2241) I don't like being drunk, think drinking alcohol is damaging to your health; and I definitely don't like drunks.(F 60s #193) Perhaps expectedly due to their social positioning as mostly women and middle-aged people, our respondents were concerned to avoid drunkenness.At the same time, most respondents appreciated alcohol's widely-recognised capacity to signify time out and to heighten social engagement (MacLean et al., 2019).
Competence Following Twine's analysis of snacking, we considered 'social competencies', as enabling people to comply with expectations about when, where and with whom it is normatively appropriate to perform a specific practice.And as Twine explained, these aspects of social competency 'form an important link' with meaning (2015, p. 1276).We coded comments to the element 'competence' when they were concerned with skills and configurations of factors that enable low-level drinking, and particularly with capacity to consume only a few drinks and stop after this has occurred.Other elements of 'competence' such as knowledge of fine wines are not discussed here.
Much of what respondents described as enabling their low-level drinking was described by respondents as entailing deliberate decisions informed by technical knowledge.Even though respondents were directly asked how many standard drinks they consumed on the most recent drinking occasion, the precision of responses conveyed respondents' awareness of the drinking guidelines and an eagerness to demonstrate their level of competency with the 'standard drink' metric: My beer was a small but strong IPA, so maybe 1.5 standard drinks.(F 40s #638) 2 wines, equals 3 standard drinks.(F 50s #1605) Participants measured alcohol to enact this technical knowledge.This took the form of using a shot glass to measure the amount of alcohol they poured, or selecting a specific glass to drink from.This information was only offered by a small number of respondents, and those who did indicated that such actions helped them to adhere to self-imposed rules around alcohol use.
I'm concerned about drinking too much if I increase the size of the shots.I measure [the] shot to ensure I don't overindulge.I feel that by just having 2 shots per day, I can enjoy a regular drink without fearing that I'll slip into pouring larger drinks, as has happened in the past.(F 50s #927) Others capped the number of drinks consumed in a low-risk drinking occasion or the number of days they drank, including scheduling zero alcohol use days: 'I'd like to have a glass of wine with dinner most nights but restrict it to 3 or 4 nights per week' (F 50s #220).
However, decisions around when to drink, and how much (or little) to drink, especially before driving, were built on ideas about appropriate comportment for people of their age and gender.In a legal sense this meant having a blood alcohol limit under 0.05%, and in a familial sense, fulfilling parental duties at any hour of the day or night, or modelling 'responsible' drinking behaviour to children.For a few parents, demonstrating a healthy attitude towards drinking and modelling low-risk drinking provided adolescent children with a positive reference point for alcohol.This motivated respondents to count their drinks and/or keep to a limit of one or two drinks: Other self-imposed rules included only consuming alcohol with food, swapping from an alcoholic drink to water when eating dinner, limiting the amount of alcohol in the house and not drinking alone.The last two strategies were reported by respondents who disclosed their own or a family member's heavier drinking: My mother is an alcoholic and I find if I have alcohol in the house I drink it rather quickly.So it's best if I don't have it in the house and only drink when I go to my friend's house.(F 50s #2517) I want to only drink one wine with a meal when feeling good or happy.I avoid the company of heavy drinkers because this encourages me to drink more than I should.I enjoy a drink but I don't keep it in my home nor do I keep any junk food.(F 50s #362) Respondents described a range of factors that supported their competence in moderating drinking.While deliberate actions were taken by some respondents, as described above, it was evident that other dynamics were at play which facilitated this competence for members of our sample, most notably that a small number of drinks felt sufficient to them.
For example, in the sense of sufficiency that meant they were able to moderate drinking without conscious effort or a desire to consume more.For respondents in our study, phrases such as 'one glass is enough', 'I don't want [or need] more' and 'I don't feel like more' signalled the point where they were feeling relaxed but not too alcohol affected.This embodied sense of sufficiency enabled respondents to mitigate or avoid unwanted physiological effects of alcohol such as headaches, disturbed sleep, feeling nauseous, or weight gain.As many respondents explained, being attuned to bodily signs of intoxication helped them navigate their point of sufficiency with little deliberate effort.
It seems to be my natural amount.After one large glass of wine I don't want any more.Also, I think it is a good place to stop.Not too hard on my body.(F 40s #695) I usually only feel like drinking 1-2 standard drinks; I don't feel the need for more.(F 60s # 993) Thus, the sense of sufficiency that participants reported is not a simple biological mechanism, but instead emerges over time, reflecting the comportment seen as expected for our mainly female and middle-aged cohort, the domestic spaces they inhabit, responsibilities they fulfil, their experiences of intoxication and hangovers, physical and emotional health and the practical need to avoid intoxication.For example, respondents spoke of the adverse interaction between alcohol and their prescribed medication and the risk that alcohol consumption may exacerbate existing health conditions.The practice of low-risk drinking emerged in conjunction with or in response to changing bodily sensations, including lower tolerance of alcohol.I'm not a heavy drinker, my medication also impacts how much I can drink, so I keep to 2 drinks max at any one time.(F 30s #1089) Social occasions entailed a greater impetus to drink for some respondents.Yet many noted that their network of friends had similar drinking patterns to themselves, and thus they were rarely placed in a position where their low-risk drinking behaviour was questioned.Others noted that they disliked being in a social situation where people were drinking or were drunk, and a few tried to avoid socialising with heavier drinkers: 'I tend not to spend a lot of time with heavy drinkers, as I don't feel comfortable in such situations' (F 60s #1599).This is another example of the regulatory competence that is enabled by taking opportunities to situate oneself in low-alcohol consumption settings and by drinking cultures and expectations shared among older and frequently female low-level drinkers.
Low-level drinking is framed by temporal elements, as in the time of the day at which drinking occurs, its regularitydaily, weekly or less oftenas well as the ritualistic way alcohol consumption is paired with everyday activities, and the sequence of different activities.A strong tendency to restrict drinking to times when it could be done in tandem with other activitiessuch as cooking, eating dinner, cleaning up and postdinner activitiesdominated accounts of low-level drinking.This was both volitional and patterned into the elements of home life (meals, family, partners, watching tv), making it relatively easy to sustain.Today: brandy and soda.Probably 1 standard drink (1 cm in bottom of glass).I was preparing dinner and then eating dinner with the same drink.(F 50s #211) 2 glasses of wine on Sunday night with my partner. 1 before dinner, 1 with dinner.White wine as usual.(F 60s #1585) These pre-and post-meal activities, as well as the meal itself, coexist with drinking practices; each practice is repeated and constrained within the home life rhythm.

Discussion
This study used SPT to examine low-risk drinking, defined as always consuming fewer than five Australian standard drinks on a single occasion, in a middle-aged and predominantly female sample of respondents.Instead of focusing only on individual behaviours and decisions, we draw attention to practices that shape and reinforce low-risk drinking across a sample of Australian adults, mainly women, aged 30 to 65 years (average age was 52).Guided by the work of Shove et al. (2012), we have considered the social practice of low-risk drinking as three broad elements: materials, meanings and competency.Although these have been presented and discussed in turn, we acknowledge that these elements are intricately tied together.In this discussion we summarise some of the main features of these three elements of low-risk drinking practice, and then consider how these interact and mutually reinforce each other to produce and reproduce a particular drinking style in the everyday lives of middle-age-to-older adults who drink in a low-risk way.We also note some effects of gender in the enaction of low-risk drinking.

Materials, meaning and competence
In terms of the materiality of low-risk drinking, wine was the most common beverage consumed, however spirit-based alcohol or beer was preferred by others.As in other studies (Holloway et al., 2008;Ward et al., 2011), the practice of alcohol consumption was influenced by the availability of financial resources, and a more expensive beverage choice (e.g.boutique beer, expensive gin) symbolised actual or aspirational socio-cultural status (Thurnell-Read, 2018).Attending venues where high level drinking was common was neither regarded as desirable nor something that our participants, who were mainly middle-aged women, needed to do as part of everyday life.
Many papers in alcohol studies allude to the particular power of alcohol to convey meaning (e.g.Holloway et al., 2008;Sulkunen, 2002).Consistent with other studies exploring alcohol use among middle to older aged adults, and frequently for women, (e.g.Emslie et al., 2015;MacLean et al., 2022;Wright et al., 2022), alcohol use symbolised a self-determined reward or treat, and for our sample this came in the form of one or two glasses of alcohol.Alcohol use also played a part in bringing people together, and the sense of connection and solidarity respondents experienced towards those they drank with, or in the company of, was not dissimilar to reports from heavier drinkers described in other studies (MacLean et al., 2019).While the social utility of alcohol use was recognised among our respondents, it was not considered essential to enjoying social events, such as 'after work drinks ' (MacLean et al., 2022).
From a public health perspective, competency in low-risk drinking is implicitly related to capacity to usually stop after one or two drinks when alcohol is consumed.Respondents were no doubt influenced by prevailing norms, where heavy drinking is generally not considered to be appropriate conduct for women in their 50s and 60s, as our sample predominantly were.Respondents noted some specific practices were also used to moderate alcohol use.Similar to other studies (Järvinen et al., 2014;Ward et al., 2011), some of the competency in managing drinking was effected through deliberate decisions.A 'one glass' rule when driving or role modelling acceptable drinking behaviour to children were rationales for moderate alcohol use, particularly for women (Cook et al., 2022).In these cases, situational and contextual elements called for drinks to be counted or capped; thus counting drinks and technical knowledge about the size of a standard drink illustrated a certain level of drinking know-how.Direct references to the Australian drinking guidelines were uncommon, however the drinking competency and specific practices our sample engaged in demonstrated a level of knowledge that was consistent with the public health messages about alcohol consumption.As we shall discuss below, however, these strategies were supported for many respondents by an embodied or affective response to drinking alcohol whereby one or two drinks felt sufficient, by social norms and tastes that encouraged only moderate alcohol use and where the situations in which they found themselves were rarely ones that fostered heavy drinking.

Interactions between elements to produce the practice of low-risk drinking
Pleasure in drinking a relatively small quantity of alcohol is one experience that cuts across the three elements of material, meaning and competence.The many manifestations of pleasureenjoyment, sensory experience of different tastes and aromas, relaxing effect, connectedness to othersacted as different layers of meaning, with each reinforcing the enactment of drinking, as has also been described in a cohort of young people who abstain or drink at a low-risk level (Caluzzi et al., 2020).For many of our respondents, pleasure in drinking was accompanied by a strong sense of embodied sufficiency.They attended to bodily sensations in order to drink to the point of enjoyment but stop before they experienced adverse effects of alcohol.Not wanting to be drunk or overly alcohol affected at the time or the next day was a strong deterrent.Graber and colleagues proposed that among young adults, 'enjoyment acted as an incentive to avoid excess' (2016, p. 85), and this was the same for the middle-aged to older adults in our sample.
The life stage, lifestyle and predominant gender of the adults in this study must be acknowledged again cutting across the three elements of low-risk drinking we have discussed in this paper.The majority lived with familya partner, children or both.Family and professional responsibilities, or a combination of both, were ever-present.Parents in the sample wanted to set an example for both 'enjoying' alcohol but also drinking moderately.There was a degree of affluence both in terms of social standing and financial means among our respondents.Alcohol was affordable to them, with some expressing a preference for more expensive alcohol.Social groups mainly consisted of friends and family members who drank in a similar way, and drunkenness was not valorised, which may reflect the lesser prevalence of short term risky drinking among middle-aged and older women than younger women and also men (Callinan et al., 2018).Both the feeling of being drunk and the experience of hangovers were avoided.Few were swayed to drink more due to social pressure or to atone for their low alcohol consumption compared with heavier drinking peers.There was a forthrightness about their preference to drink at a low-risk level, regardless of others' perceptions.Together with the acute sense of embodied sufficiency when drinking alcohol described above and the acquired competencies during early and later adulthood, the elements mentioned here reinforced and sustained low-risk drinking among our sample.
The home context, particularly the way in which a routine evening unfolded, also sustained low-risk drinking.A recent study of home drinking found that light drinkers would generally consume alcohol as part of one or two domestic activities, usually eating dinner.In contrast, heavy drinkers regarded many things that they did at home (preparing food, eating, watching television, watering the garden etc) as best done with a glass on hand (MacLean et al., 2022).In the data reported on here, alcohol was also unquestionably consumed in unison with making and/or eating dinner, and this pairing of practicesdrinking and eatingdid not seem to waver if the household was a single person, couple or family group.As these proximate activities tend to follow a personalised schedule, the practice of drinking has a temporal mechanism.Alcohol-free days, which some respondents incorporated into their routine, allowed a disruption to the connections between time, homelife activities and drinking.The overlaying 'reward' or 'me time' attribution to drinking alcohol, albeit in small amounts, was a further reinforcing factor.

Limitations of the study
The data analysed in this paper was drawn from a convenience sample of adults who responded to a Facebook post to complete an online survey about their patterns of consuming alcohol in the home, not specifically low-risk drinking.Open-text responses were collected from a series of questions focused on low-risk drinking, only asked of those who drank fewer than five standard drinks on a single occasion.Nonetheless the large sample of 252 respondents gave thoughtful and sometimes extensive responses to our questions, so we are confident that our data is reflective (if not representative) of the drinking practices of middle-aged and older Australians who drink at lower risk levels.Most of the sample were women, so men's low-risk drinking practices are less well represented here.It is also acknowledged that the terminology used by respondents, such as reference to a standard drink, may reflect the language used in the survey questions more than their own voice and narrative around their alcohol use.Therefore, the findings are reflective of the data collection method as well as the specific topic.

Conclusion
Part of the purpose of studying the social practice of low-risk drinking is to learn how this is done, to inform efforts to encourage a shift towards this drinking style among those drinking at levels more likely to place them at risk of short-or long-term harms.Our respondents spoke of some elements of drinking practice that could be encouraged in heavier drinkers through health promotion.One response would be to encourage greater attendance to bodily feelings of sufficiency (for those individuals who experience these after only a few drinks) (Zajdow & MacLean, 2014).Campaigns positing slower drinking as a social good are worthy of further consideration as a strategy to support reduce heavier drinking patterns, as is shifting social expectations of temporal framing of drinking to limit it to particular times of the day or activities (e.g. during meals).
Scholars have devoted much attention to the term 'responsibility', and its socio-political usage.Attribution of 'moral blame' (Trnka & Trundle, 2014) is inherent in low-risk drinking guidelines which, as noted in the Introduction, take an individualistic and rationalist view of drinking behaviour.In this paper we outline features of the social practice of low-risk drinking.In doing so, we show how low-risk drinking practice is supported by deliberate and rational individual efforts to control drinking, but also emerges and is reinforced by meaning, material and competency elements.The emphasis on individual self-management through rational choice inherent in low-risk drinking guidelines has led to critiques about their efficacy in promoting behaviour change (e.g.Lindsay, 2010).Our study shows how substantially any individual's drinking practice is framed by the contexts of their life, and that a decision to drink at a level that is compliant with health advice is only one of many factors involved in low-risk drinking.

Disclosure statement
No potential conflict of interest was reported by the author(s).
It sends a signal to myselfif I'm having a drink, it's because I'm allowing myself that time to relax.(M 40s #344) It only happens when I'm off duty and all my chores etc are done, so I love the connection with being able to put my feet up.(F 50s #2317)

I
usually only drink at home after I've finished driving … If I'm out I count my drinks.(M 50s #65) I have a one drink rule if I have the car.(F 50s #2049)It's been recommended to only drink 2 standard drinks.I have teenagers so it's important for them to see a responsible parent.(F40s #2522)