Binge drinking: A review and research agenda

Binge drinking is a wicked social problem associated with poor health and safety outcomes. However, most research on binge drinking occurs outside consumer research and marketing. This paper presents a systematic review of the contributions of key disciplines (Health, Marketing, Psychology, Economics, Management, Social Science, and Sociology) to the understanding of binge drinking. Using the theory, context, characteristics and methods (TCCM) framework, we find the need for more theoretical investigation in binge drinking research and more research conducted outside western countries. A wide set of independent factors has been included in binge drinking research, but few studies consider mediating and moderating variables. Methods used vary across disciplines and those used are mainly descriptive research, regression, and experiments. We conclude that whilst progress in this field is evident much needs to be done for greater generalizability and robustness of findings.


| I NTRODUCTION
Excessive alcohol consumption has been the focus for many studies in consumer behavior (Banister & Piacentini, 2006;Gallage et al., 2018;Martin et al., 2013;Previte et al., 2015;Zhang & Shrum, 2009). Of particular concern to consumer behavior researchers in the area of excessive alcohol consumption is binge drinking (Banister & Piacentini, 2006;Gallage et al., 2018;Kubacki et al., 2011;Martin et al., 2013;Siemieniako et al., 2010;Wolburg, 2005;Zhang & Shrum, 2009). Binge drinking is excessive alcohol consumption in a short period of time. Clinically this is five standard drinks consumed within 2 h for men, or four for women in the same period, and excessive binge drinking being 10 standard drinks consumed for men and eight for women in the same period. (Alcohol Research: Current Reviews, 2018). Binge drinking has also been defined as "a pattern of heavy drinking that occurs over an extended period of time set aside for the purpose" (Herring et al., 2008, p. 476).
The World Health Organization views binge or excessive drinking as a significant health problem requiring worldwide attention. In its 2018 report (World Health Organization, 2020), it concluded, "Far too many people, their families and communities suffer the consequences of the harmful use of alcohol through violence, injuries, mental health problems and diseases like cancer and stroke". These concerns are mirrored in statistics on binge drinking effects. Deaths related to alcohol consumption globally in 2016 (the most recent figures) were estimated at 38.8 per 100,000, some 56.4 million people (World Health Organization, 2020, p. 85), or 5.3% of all deaths worldwide. Conditions associated with excessive alcohol consumption include cancer (4.7%), epilepsy (11.7%), cardiovascular diseases (4.1%), and digestive diseases (25.2%, World Health Organization, 2020, p. 85).
Binge drinking is associated with a myriad of health and social problems. In particular, it has been associated with greater risks of injury and death (Caamaño-Isorna et al., 2017), antisocial and illegal behavior such as drink-driving, assault, stealing and damaging property (Yang et al., 2016). There are long term health impacts on individuals who regularly binge drink (World Health Organization, 2020).
Making matter worse, binge drinking across generations has also been shown to predict problem gambling behavior (Zhai et al., 2017), and binge drinking among young women places them at a greater risk of sexual assault (Luke, 2009). Whilst systematic reviews on excessive drinking exist (Kubacki et al., 2015), the area of binge drinking from a multidisciplinary perspective has yet to be examined in detail.
Binge drinking produces substantial detrimental harm to individual consumers and society at large. The numerous studies in different disciplines using different approaches to understand and limit the extent of binge drinking indicate the importance of this complex issue in consumer behavior and for policy makers. Given that consumer behavior is a multi-disciplinary field (Inman et al., 2018) which can influence and is influenced by other fields of research and public policy (Prothero et al., 2011), there would seem a need to bring together this varied research to understand its contribution both as part of the discipline of marketing and to the important area of consumer misbehavior.
To the authors' knowledge, there is no systematic review that brings these various research streams together and identifies how they contribute to our understanding of the phenomenon. Arguably, this is because the research is spread across different fields and different methodological approaches. A systematic and coordinated approach to addressing the complex problem of binge drinking is difficult without understanding the contributions of different research fields.
To this end, this paper conducts a more inclusive systematic review by consolidating the binge drinking research across multiple disciplines. The aim is to take a structured approach following the TCCM framework (Chen et al., 2021;Paul & Rosado-Serrano, 2019) to summarize and highlight the important aspects of the research with reference to theory, context, constructs, and methodology. Through this approach, a more comprehensive understanding of the key features related to binge drinking is established and novel insights for future research identified. Our research shows a distinct lack of theoretical explanation for binge drinking across disciplines and the need for cross-disciplinary and methodological approaches to understand binge drinking better and offer practical advice for policy development and the design of interventions.

| Selection of articles
The articles were selected via a library mega-search across 53 EBSCO databases, Web of Science and Scopus containing the keywords "Binge drinking" or "Heavy Episodic Drinking" (often a term also used to describe binge drinking) and "Research" in the title, abstract or keywords. The systematic review was conducted between May 2021 and February 2022.
The EBSCO mega-search covered databases across social sciences (e.g., EconLit, ERIC, Humanities Source Ultimate, and Sociology Source Ultimate), health/psychology (Medline, Psychology, and Behavioral Science Collection) and business (Business Source Ultimate), plus others. The search did not include the terms alcohol abuse and heavy drinking as this systematic review aimed to examine the distinctive phenomena of binge drinking.
The initial search resulted in some 24,521 identified papers. The research was narrowed down by including the term "Consumer" and selecting academic/peer-reviewed articles and then removing duplicate entries, which reduced the number to 391 papers.
We next followed a well-established criterion for inclusion of papers according to scholarly quality that is an impact factor greater than or equal to 1.0 for systematic reviews (Chatterjee & Sahasranamam, 2018;Dabi c et al., 2020;L opez-Duarte et al., 2016;Paul & Rosado-Serrano, 2019). We must draw the line somewhere in binge drinking research, and the papers selected were well cited and in respectable journals. Also having an impact factor of greater than one, means the journal has significant impact in that its papers are cited on average at least once.
Next, we selected articles with an impact factor (2018) of 1.0 or above, as suggested by (Chatterjee & Sahasranamam, 2018;Dabi c et al., 2020;L opez-Duarte et al., 2016;Paul & Rosado-Serrano, 2019 Papers were classified as belonging to a discipline on the basis of accepted disciplinary lists (e.g., ABDC list, https://abdc.edu.au/ research/abdc-journal-quality-list/), Scimago journal and country rank (www.scimgor.com) or on the basis of the aims and description of the journal homepage. Figure 1 shows the PRISMA flowchart (Page et al., 2021) for the selection of articles.
The references for the review, were then imported into excel and coded according to the dimensions of the TCCM systematic review.
This included theories used, countries studied, independent, mediating, moderating and dependent variables and type of method used.
Coding was done on the basis of committee by agreement method.
Citations for each article were also recorded so as to assess further the scholarly impact of research. First though we examine papers with the greatest impact in the field.

| TCCM FRAMEWORK
Paul and Criado (2020) provide a thorough overview of several ways to conduct systematic literature reviews. As the aim here is to bring together research across multiple disciplines to provide a richer understanding of binge drinking in relation to consumer behavior, an inclusive framework-based approach is used to structure the review.
Specifically, this systematic review follows the TCCM framework (Paul & Rosado-Serrano, 2019). The TCCM framework is appropriate as it involves a careful consideration of the various guiding theories used in binge drinking research, a discussion of the different contexts in which the research was conducted (i.e., countries), a summary of the characteristics used which includes independent, mediating, moderating, or dependent variables, and a review of the types of methodological approaches taken. From this systematic review, the research gaps in binge drinking are exposed and future directions for consumer behavior research identified.

| Theory
Theory provides a method of predicting the interactions of phenomena, and research enables the testing of theory and the discovery of new knowledge (Thomas, 2017). Binge drinking literature draws on a variety of theoretical frameworks and paradigms to explain consumption behaviors, attitudes, motivations, social influences, and many other general effects in regards to both positive and negative consequences. Interestingly, the use of theory fluctuates from year to year (see Figure 2) and 58% of the articles do not address any theory at all, which is a main weakness of binge drinking research. That is not say that papers can conceptualize and theorize about a problem without using an established theory, though we expressed concern about the lack of theoretical application in this field. Note papers could have multiple theories.
The theories identified within this literature review cover a broad range. Most of the research that incorporates theories focuses on individual characteristics and social aspects. While there are no determined categories or associations that have been consolidated within the research, Table 2 provides a summary of the types of theories used. As indicated in Table 2, social theories are drawn on the most, with social norms theory being the most common. While individual psychological theories are applied often, no primary theory is evident.
Of the behavior change theories, the theory of planned behavior/reasoned action dominates. In the following sections, we discuss four of the more prominent theories used, including: social norms theory, the theory of planned behavior, ritual behavior and parental style/ monitoring.
3.1.1 | Social theories: Social norms theory Social norms are important determinants of health-related behaviors (Ajzen, 1991;Gerrard et al., 2008;Perkins & Berkowitz, 1986). Social norms theory focuses on two types of (mis)perceived social norms in Misperceptions of these social norms can have important implications for consumers, such as overestimating peers' alcohol use, which may make higher consumption perceived to be a socially desirable behavior leading individuals to match what they perceive to be the social norm (Festinger, 1954). Such misperceptions have been associated with a range of behaviors, including increased alcohol and other substance abuse (Martens et al., 2006). Many studies in the review focus on the role of social norms misperceptions of alcohol use, and how misperceived social norms can influence perceptions of what is considered to be healthy .
While many studies develop interventions based solely on social norms theory, or in conjunction with other strategies, to influence positive health behavior changes (Trockel et al., 2008), not all scholars find the social norms approach effective. In Real and Rimmel's (2005), research perceived degree of alcohol use was not predictive of alcohol abuse if other normative influences were considered. Baer and Carney (1993) also report no relationship between misperceptions of drinking problems and personal alcohol consumption. More empirical research is needed to further delve into binge drinking and social norm influence to determine the circumstances in which binge drinking misperceptions influence behavior.
3.1.2 | Behavior change theories: Theory of reasoned action / planned behavior Social-psychological models of behavior change, such as the Ajzen's (1991) theory of planned behavior (TPB) which is an extension of Fishbein and Ajzen's (1975) theory of reasoned action (TRA), are used within 6 of the articles to study individual behavior. TPB is considered one of the most influential attitude-behavior relationship models as it has been shown to predict behavior effectively (Christian et al., 2007;Previte et al., 2015).
Many risk behaviors have been studied using TPB (Fishbein and Ajzen, 1975) which makes it appropriate for binge drinking research.
TPB has been applied to understand how attitudes, subjective norms, and perceived behavioral control influence binge drinking intentions and behavior. Some articles consider TPB in conjunction with parental theories, emotional theories, expectancy-value theory, and brand choice (Knox et al., 2019;Previte et al., 2015;Sharma, et al., 2013;Piper et al., 2021). Several articles suggests that a planned approach is appropriate for addressing binge drinking as TPB captures the key challenges faced by the individual when intending to drink moderately (Martin et al., 2013;Turrisi, 1999;Yan & Brocksen, 2013). TPB also captures individuals' beliefs about their ability to control drinking situations (Turrsi, 1999). Others suggest that persuasion, rational choice and sensation seeking can also be applied to risk taking, ultimately leading to behavior change for better or for worse (Blinn-Pike and Worthy, 2008;Meyer, 2010).
TBP approaches presume consumers can influence their binge drinking at least in part through a rational planning process. While much of the research focuses on reducing binge drinking, more work is needed to understand when consumer intentions to moderate their drinking break down. Established research in self-regulation failure (Baumeister and Heatherton, 1996;Robson et al., 2020) suggest inertia, transcendence and acquiescence may play a part. Additional research is also needed to look at consumer intentions to binge drink and how positive attitudes towards binge drinking are formed.

| Ritual behavior
The papers that examine binge drinking as a ritual consumption behavior mainly draw on Rook's (1985) seminal work on ritual behavior and Driver's (1991) more functional typology that outlines the "social gifts" of rituals. Rook (1985, p. 252) explains that ritual behavior occurs in both every day and special occasions, it is "a type of expressive, symbolic activity constructed of multiple behaviors that occur in a fixed, episodic sequence, and that tend to be repeated over time." According to Rook, rituals involve four elements: (1) a ritual artifact, such as a product like alcohol that is consumed, (2) a script, which directs the consumption practice, (3) a performance, which is enacting the expected behaviors and (4) an audience, who observe the performance. There are different types of rituals, such as "rites of passage" signifying the transition from one life stage to another, and small group rituals to establish order, cement relationships and encourage social cohesion. Driver's (1991) three "social gifts" correspond quite closely, as he sees rituals providing: (1) order, which gives a sense of direction and stability (2) shared experience, which increases a sense of community and (3) transformation, which gives the opportunity to change in the eyes of others and oneself.
Researchers who have used ritual consumption experience typically have explored binge drinking among university students. Binge drinking meets the criteria of ritual behavior (Treise, Wolburg, and Otnes 1999). Alcohol is the consumption artifact. Students follow rules, a script, on how to get alcohol (especially relevant as not all students are of legal drinking age), when to drink (e.g., time of day and week) and where to drink (e.g., dorms, house parties, bars). The performance of drinking includes what to order, how many drinks to consume and how fast, as well as how to behave while drinking. There are also roles to perform (e.g., caretaker, designated driver) and an audience (e.g., peers, bartenders, campus security). All three "social gift" are found in binge drinking (Wolburg, 2001) with transformation appearing to be the main benefit. Binging allows students to escape, feel free, experience the sensation and thrill of being out of control, gives them an excuse to break the "rules" and behave without any inhibitions. Understanding binge drinking as ritualistic behavior looks to be a fruitful approach, especially if consumer researchers can learn how rituals are initiated and how rituals can be interrupted to break the performance.

| Parental style/monitoring theory
Several researchers investigate the influence of parental style and monitoring on children's binge drinking. Baumrind's (1967) seminal work on parenting styles is based on the level of responsiveness (i.e., the degree to which parents react to their child's needs in a supportive and accepting manner), and the level of demandingness (i.e., the rules parents put in place, the expectations for children to comply, and the repercussions if rules are broken) (Arnett, 2013).
Through her studies, she identifies three parenting styles: authoritative, authoritarian, and permissive. Authoritarian parents tend to impose rules and to be extremely directive in their approach to parenting. In contrast, permissive parents tend to impose few rules and to communicate in a manner that contains fewer directives (Mallett et al., 2011 (Stanton et al., 2000). Several studies indicate that greater parental monitoring is related to less alcohol consumption and binge drinking (Veal & Ross, 2006;Larm, et al., 2018). Similar findings also support this, where specific family rules and tough parental monitoring predict lower alcohol use and abuse (Guo et al., 2001). Additionally, parental monitoring effects can have an influence from an early school age (Guo, et al., 2001;Ross et al., 1997) and research shows some gender differences. Poor parental monitoring has been related to greater alcohol use and abuse (as well as other drugs) for young males more so than females (Svensson, 2003).
The role of the family in consumer socialization (Moschis, 1985;  Italy (Fabbrizzi et al., 2017) and Russia (Neufeld et al., 2017), through to the use of simulations to understand peer influences in drinking as shown by research in The Netherlands, (Giabbanelli & Crutzen, 2013) and web and interactive voice messaging of Public Service Announcements (PSAs), in Sweden (Andersson, 2015). There also is a series of studies in Europe on the harm binge drinking can do to health in terms of liver damage (Nivukoski et al., 2019, in Finland) and breast cancer as shown in Danish research (Mørch et al., 2007).  Table 5.
What is most prescient is that around a third of studies (32%) do not consider independent variables in their research design or take a merely a descriptive analysis of binge drinking. This is reflected by the large use of descriptive research that focuses only on the prevalence of binge drinking and related issues, as is seen in 29.1% of studies.
Binge drinking itself is used as an independent variable in research, as shown by 8% of studies. This is the result of early work on binge drinking and cognitive loss among men and women (Parker et al., 1983) and later attempts to try and quantify risky outcomes associated with binge drinking (Caamaño-Isorna et al., 2017;Knott et al., 2013;Vassallo et al., 2008;Vingilis & Wilk, 2008). This can include greater risks of motor vehicle injuries (Caamaño-Isorna et al., 2017;Vassallo et al., 2008) and adverse health effects (Knott et al., 2013). Greater risk profiles for specific occupations, such as shift workers has also been studied (Dorrian et al., 2017). Another area of focus examines sexual violence associated with binge drinking (Brooks, 2014;Luke, 2009). Other negative outcomes associated with increased alcohol consumption includes, for example, poorer academic performance by teens (Renna, 2008) and risky driving behaviors (Valentine et al., 2008). These situations also involve binge drinking, but these are not included in the research design per say as an independent variable.

| Individual psychological factors
This area of research focuses on the decisions and emotions of individuals with respect to binge drinking. How individuals define themselves in close relationships is also linked to binge drinking (Martin et al., 2013), with people favoring and valuing close-relationships found to respond more positive to health messages about moderate drinking. Attitudes to non-binge drinking alternatives, such as meeting for coffee, rather than drinking alcohol are also important in understanding the decision to drink excessively (Turrisi, 1999). Depression is associated with binge drinking and cognitive behavioral strategies which help deal with depression are found to reduce the risk of binge drinking (Martens et al., 2008).
Personality is linked to binge drinking. People who are more optimistic about the effect of heavy drinking are at risk of overindulgence (Sloan et al., 2013). Having a more agreeable personality is also more likely to be associated with binge drinking, more than personal economic circumstances (Bessey, 2018). Binge drinking is associated with more impulsive personalities (Zhang & Shrum, 2009) and those who are more hedonistic (Zhang & Shrum, 2009). Personality research suggests that some people may be more at risk of binge drinking, and that these should be specifically targeted to by policy makers.
Activating emotions of guilt and shame can assist in reducing or inhibiting binge drinking (Agrawal & Duhachek, 2010;Duhachek et al., 2012). Framing health messages with guilt appeals is more effective in combined with a gain scenario and shame appeals more in a loss situation. Resistance to public health messages has also been a focus of research (Hackley et al., 2015), with moral arguments likely to be undermined and subverted by those who heavily drink. In short, emotion research shows us that just saying no is not an appropriate health strategy. This research is also important in how messages should be framed in public health campaigns.

| Social influences
Research examining social influences examines the external interpersonal bases of binge drinking. This is important since much of binge drinking occurs within a social context. Fabbrizzi et al. (2017), Other research looks at the nature of social relationships, such as committed relationships versus casual contacts. Being in a committed relationship is found to reduce the risk of binge drinking (Duncan et al., 2006), but interestingly only for men (Pedersen & Pithey, 2018).
Romantic partners can also be "network" bridges, though more to groups with more risky behaviors such as binge drinking (Kreager & Haynie, 2011). Social proximity, such as sharing a dorm with heavy drinker, has also been found to be associated with a higher risk of binge drinking (Guang et al., 2015).
Other research on close relationships looks at the influence of parents. Research on parenting shows that drinking patterns are modelled (Zhai et al., 2017) and that the style of parenting has an effect on binge drinking behavior (Mallett et al., 2011;Veal & Ross, 2006).

Greater parental monitoring and communication with children is
found to reduce the risk of binge drinking of youths (Mallett et al., 2011;Veal & Ross, 2006).
Out-group behavior is one important explanation for binge drinking (Gallage et al., 2018;Trockel et al., 2008). Trockel et al. (2008) (Crawford & Novak, 2013), social norms are also powerful at the individual level. Crawford and Novak (2013) find that the social embarrassment from over imbibing leads to more moderate drinking behavior.
Being in contact with out-groups who drink moderately or abstain has also been found to change attitudes of a heavy drinking in-group (Gallage et al., 2018), partially because the out-group are admired for their sense of control and healthier lifestyle. It is not surprising to find that in many countries policy makers highlight the social influences as a means to change behavior away from excessive drinking (Walton & Wilson, 2018). Overall research on social influences shows that binge drinking is very much situational and social, and that campaigns that just target individual emotions and decisions may not be effective when social conditions have been shown to have such a powerful effect. Public policy in this area therefore needs to focus more on the environment in which consumption takes place and address long-term social norms. This is considerably more difficult than just considering the individual.

| Socio-economic and demographic factors
Socio-economic factors are concerned with external effects on binge drinking. These have included economic effects (Cotti et al., 2015;Dee, 2001; Ólafsd ottir &Ásgeirsd ottir, 2015), for example binge drinking increases with men in periods of economic stress, but not for women (Dee, 2001). Worsening economic conditions, such as a downturn in the stock market, are associated with higher levels of self-reported mental health conditions and binge drinking (Cotti et al., 2015). Economic evidence suggests that increases in income and labor market successes are associated with higher levels of binge drinking (Hoke & Cotti, 2016;Shao-Hsun & Huffman, 2007. This is especially so for young drinkers, where increases in the minimum wage are associated with higher levels of binge drinking (Hoke & Cotti, 2016). Further detailed evidence suggests that binge drinking is indeed countercyclical, as long as one stays employed during an economic crisis (Dee, 2001). The primary reason for this is that binge drinking is a reaction to stress in uncertain times.
The extent of prestige of a college or university has also been found to affect binge drinking, with higher levels of binge drinking occurring in more selective colleges (Fletcher & Frisvold, 2011).
The type of leisure activity has also been shown to be associated with binge drinking, with young men more likely to engage in excessive drinking when at concerts or sporting events (Andersson et al., 2014).
Differences and similarities between groups such as gender (Millea, 2019) as well as adults and youth have been studied (Nelson, 2008). Differences in regions such as states have been studied (De Boni et al., 2014;Nelson et al., 2005;Smith et al., 1990) along with characteristics of youths with heightened risks towards binge drinking (Yarmuth et al., 2012). Research has also examined differences in binge drinking across socio-economic groups (Fone et al., 2013), as well as assessing the extent of the problem in the defence forces in both active duty and post-deployment phases (Larson et al., 2014).
Socio-economic and demographic research is useful to policy makers as it focuses on macro forces driving excessive alcohol consumption. It does not provide a complete set of remedies without considering individual and social factors involved in consumption. Nor does it focus on the actions of industry and market factors, which may encourage binge drinking which is discussed next.

| Market and industry factors
Though only accounting for 6.5% of published research, this is an Another part of market and industry research is on media influences. This includes for example, the media's portrayal of binge drinking (Fogarty & Chapman, 2012;Valentine et al., 2008) and media's representation on public health strategies to reduce binge drinking (Campo et al., 2009). An assessment of media campaigns in themselves to reduce binge drinking has also been studied (Wolburg, 2005). Media influences are a salient area to study given the importance of public support for campaigns to reduce binge drinking.

| Policy interventions
This area of research has focused on examining the effectiveness of policy interventions to reduce or moderate binge drinking. Such research uses a field experimental design to examine the effectiveness of an intervention versus a control (LaBrie et al., 2016;Montag et al., 2015;Tinghög & Tinghög, 2016). Intervention studies have included education programs at work (Deitz et al., 2005;Tinghög & Tinghög, 2016) and in communities (Komro et al., 2017), the use of smartphone apps to monitor unhealthy alcohol use (Bertholet et al., 2017) or provide culturally tailored SMS messaging (Bush et al., 2019) or industry case studies of greater supervision of heavy drinkers (Pratten & Greig, 2005). The effect of health and taxation policies, especially for Indigenous people to reduce harmful drinking, have also been documented (Gray et al., 1999), as well as better policing of underage IDs (Zheng, 2018 (Austin & Ressler, 2012), which encourage excessive drinking of others. Pre-drinking, or "pre-loading", has also been found to be associated with binge drinking (Riordan et al., 2018). Location and type of alcohol can encourage binge drinking. Drinking in nightclubs compared to bars has been found to be associated with higher level of binge drinking, mainly because spirit consumption is involved (Callinan et al., 2014). There is also a higher risk on binge drinking as expected at holidays and at the end of the week (Reich et al., 2015). Situational factors, though under researched, are thus an important avenue for future research in binge drinking.

| Mediating variables
Surprisingly given the complexity of studies using IVs in binge drinking research, there are very few studies, which include mediating variables, nine studies or 4.7% of research. The most common are individual psychological factors (7 studies) and include factors such as self-efficacy and motivation to change as a mediating variable for an intervention of SMS messaging to prevent binge drinking (Bush et al., 2019), coping strategies (emotional and practical) in examining message appeals to prevent binge drinking (Duhachek et al., 2012) and coping with depression and conformity mediating the use of selfprotective strategies to prevent binge drinking (Martens et al., 2008).
Other studies have examined psychological mediators such as impulsivity in relation to the effectiveness self-construal on binge drinking (Zhang & Shrum, 2009). As can be seen research using mediational analysis greatly assists policy makers, as it shows the conditions when behaviors are likely to be triggered or program interventions will work.
Thus, the need for more specified research.

| Moderating variables
Likewise, there are few studies that include moderating variables in their design, only around 9.7% or 17 studies. Individual psychological factors included as moderators are most popular (6 studies) followed by demographics (4 studies). Individual and peer drinking behavior, corporate social responsibility (which moderates the link between binge drinking and online gambling, see Yani-de-Soriano et al., 2012) and the presence of a television in the bedroom (which moderates parental monitoring, see Veal & Ross, 2006) make up the remainder of moderators that have been included.

Individual psychological factors used as moderators in research
designs have included risk taking in whether family history and binge drinking predicts gambling problems (Zhai et al., 2017), impulsivity as to whether binge drinking can be influenced by hedonistic goal directed behavior (Ramanathan & Menon, 2006) and motivation to get better as a moderator for program effectiveness to manage intoxication arising from binge drinking (Arnaud et al., 2017). Other individual psychological factors include self-referencing in moderating the effectiveness of messages to reduce binge drinking within social situations from a first or third person perspective (Martin et al., 2013), and the use of protective behaviors as moderators with motives to drink and depressive symptoms as predictors of binge drinking (Martens et al., 2008).
A range of demographic moderators have been included. College size has been used as a moderator for the relationship between race and binge drinking (Wechsler & Kuo, 2003), and age for the relationship between binge drinking and automobile accidents (Vingilis & Wilk, 2008). The city /rural dichotomy has been included in research examining the link between advertising exposure and binge drinking (White et al., 2017). Gender, subsidies, and education have been studied as moderators of the relationship of price of alcohol and binge drinking with Chilean youth (Paraje et al., 2021). As can be seen, the use of moderators greatly assists researchers in understanding the complexities and nature of effects of both causes of binge drinking and the effectiveness of policy interventions to prevent it. It is a pity that not more research has considered moderating effects, especially with demographic information, which is usually collected in any standard research project.

| Dependent variables
Dependent variables in binge drinking research not only include binge drinking behavior and attitudes (28.6%), but also wider alcohol consumption behaviors (22.9%) and harm from binge drinking (19.4%).
6 | METHODOLOGY Table 5 shows the types of methods used by various disciplines to study the problem of binge drinking. The method types are based on a similar framework in the systematic review by Paul and Rosado-Serrano (2019). We coded research as descriptive in method, when only a descriptive survey approach was taken, for example the amount of binge drinking by various groups and used the term regression when survey data was analyzed to provide an explanation of factors influencing the dependent variable, which was usually binge drinking.

| Major methodological approaches by discipline
Across all disciplines, the major research methods used are quantita-

| DIRECTIONS FOR FUTURE RESEARCH
In total, this paper has provided a systematic review of 175 journal articles on binge drinking research across multiple disciples. Again, following the TCCM framework, which is common practice (Chen et al., 2021;Paul, Parthasarathy, & Gupta, 2017), we outline a research agenda to encourage further progress in the area. Next, we draw on and extend our work by discussing some of the opportunities for future research presented in Table 6.

| Theory
Much of the research in binge drinking is not conducted with an identified theory. Only 44% of papers having a theoretical basis. Binge drinking behavior is defined and risk factors identified, but explanations as to why consumers engage in heavy episodic alcohol consumption and therefore how best to tackle the issue remains unclear. A range of theories are used with the Theory of Planned Behavior (Ajzen, 1991) and Social Norms Theory (Berkowitz, 2003) identified as the most common. Self-control and the social context play a role. People binge partly because they want to belong, they mistakenly think it's expected and everyone else is doing it.
Less is known about binge drinking as an impulsive, hedonic consumption experience. Recent research by Vosgerau et al. (2020) shows self-control failure does not necessarily mean sacrificing pleasure and identifies several strategies to assist consumers in limiting harmful consumption. Related to hedonic pleasure is Belk et al. (2003) work on desire who conceptualize desire as a recurring emotion with both positive and negative components involving a search for belongingness yet also experiencing a sense of being different than oneself, an aspect of risk yet also unreachable. Viewing binge drinking in terms of desire may open new ways of addressing the issue. Further research can consider binge drinking as a "sticky" consumption journey (Siebert et al., 2020) involving easy, low-cost entry and a variety of unpredictable, exciting experiences. It also may be beneficial to take a life course perspective (Moschis, 2007) which incorporates changes in life events (or transitions from one stage to another), socialization, stress, coping, identity, and knowledge influence consumption decisions.
As the social aspect of binge drinking is a main factor, theories of inter-group behavior such as Social Identity Theory (Tajfel & Turner, 1979) may bring further insights. Substantial evidence shows that once individuals feel that they belong to a social group, their selfesteem becomes tied to the group and motivates them to support their "in-group" norms. Rather than examine "in-group" influences, Much of binge drinking is studied within university and high school contexts. Extending the research to different workplaces is called for, especially in sectors where significant health and safety risks exist, such as the police force and shift work in medical system, oil and mining industries (Austin & Ressler, 2012;Dorrian et al., 2017).
Excessive, episodic alcohol use may increase errors and serious accidents and put additional stress on workers in already high-pressured situations. Research is also needed to investigate binge drinking in public spaces, such as in pubs and clubs, as well as in domestic and other informal contexts (Valentine et al., 2008). Despite the established research indicating retail store atmospherics add hedonic value to the consumer shopping experience (Ballantine et al., 2010), few studies examine environmental factors on binge drinking. One exception by Carlini et al. (2014) shows 'all you can drink' service, multiple dance floors, and higher sound levels are all associated with binge drinking, though "pre-loading" before going out has a stronger effect suggesting the home environment is an important factor. While young adults and adolescents may see the home as providing a safe place, one where they have access to cheaper alcohol and more control to be adventurous and experiment (Ander et al., 2015), binge drinking at home for older adults has increased due to the financial and mental health stressor related to the COVID-19 pandemic (Weerakoon et al., 2021).

| Characteristics
The review identifies a wide variety of independent variables have been examined, with binge drinking itself is an independent variable in around 7.2% of studies, mainly as a predictor of risky and harmful behavior associated with it. Other independent variables range from wider socio-economic factors, individual psychological factors and to a lesser extent social and parental influences and situational factors.
The extraordinary situation caused by the COVID-19 global pandemic calls for studies examining the interplay between stress, depression and excessive alcohol consumption (Pedersen, 2013). As research shows binge drinking to be mainly a social activity, more research needs to be done with regards to social influences. Examining the role of social media, peer networks and the relationship between online and offline behaviors drinking may uncover new approaches to tackling binge drinking (Guilbeault and Centola, 2021;Lee and Ahn, 2017;Zhang and Centola, 2019). More research focusing on family dynamics (Zhai et al., 2017), especially role of parental style, monitoring and communication (LaBrie et al., 2016;Mallett et al., 2011) is critical to furthering our understanding.
Another important area of research not covered in depth by researchers has been the influence of the market and industry in promoting or unintentionally providing conditions for binge drinking to occur. Research directed towards product features, such as flavored alcohol (Albers et al., 2015), warning labels (Coomber et al., 2015), glass shape (Angela et al., 2012), and the combination of alcohol with energy drinks (Marczinski et al., 2017;Newcombe et al., 2020) is warranted. Further studies examining environmental factors (Hackley et al., 2015), including location, alcohol outlet density (Ahern et al., 2016), and atmospherics may also assist in understanding what triggers or facilitates binge drinking will assist in the development of The review also revealed a lack of research considering mediating and moderating variables in research designs (5.9% and 11.1% of studies respectively). Where research did include these factors, the findings can be seen as more robust and generalizable. The lack though of more rigorous research designs and the limited context in which binge drinking research has occurred means generalizability of findings and advice to policy makers is at this stage limited.
• More longitudinal (Vassallo et al., 2008;Vingilis & Wilk, 2008) and cohort studies (Caamaño-Isorna et al., 2017) are needed to understand the long-term effects of binge drinking. • Conduct in situ observational studies (Carlini et al., 2014) and field experiments (Zhang and Centola, 2019) to test consumer reactions to binge drinking interventions. • Event history analysis (Mayer and Tuma, 1990;Moschis, 2012) may assist in identifying key factors influencing changes in attitudes and behaviors as consumers transition from one life stage to another.

Recommendations
• How can technology assisted interventions, such as smartphone applications, help to address unhealthy alcohol use (Bertholet et al., 2017) • Further elaboration on government strategies to set minimum prices on alcohol products and the effects on reduction of binge drinking (e.g., Hawkins & McCambridge, 2014;Gray et al., 1999) • Examine the extent to which stricter control on the minimum drinking age affects alcohol purchases, consumption, and associated consequences (e.g., Kypri et al., 2014;Nelson et al., 2005;Hackley et al., 2013) • How can policy influence young people's focus on the long-term health risks of binge drinking (e.g., Fogerty & Chapman, 2012) • Consider standardized use of terminology in the implementation of educational interventions and communications campaigns, to reduce confusion about excessive alcohol consumption and health warnings (Jones et al., 2016;Coomber et al., 2015) • Identify ways in which all levels of governments can work with the liquor industry, especially with hospitality venues and retail outlets, to reduce binge drinking and associated antisocial behavior (Pratten & Greig, 2015) • Explore use of moral claims and how they can be effectively used in social marketing campaigns and endorsed by policy makers (e.g.,  challenge is that insights around binge drinking do not come necessarily when it is studied as a dependent variable. The opportunity is that studying research concerned with alcohol consumption and harm from binge drinking may provide insights not readily considered.

| Methodology
Our review shows that binge-drinking research tends to be studied with a limited set of methods, namely, surveys, regression analysis and experiments. Studies also tend to use a single method and data source, which brings in the issues of validity, reliability, and common method bias (Campbell and Fiske, 1959;Podsakoff et al., 2003 (Carlini et al., 2014) or conducted field experiments (Zhang and Centola, 2019), both of which may be effective ways to test consumer reactions to binge drinking interventions. We also believe much can be gained regarding the long-term effects of binge drinking by conducting more longitudinal (Vassallo et al., 2008;Vingilis & Wilk, 2008) and cohort studies (Caamaño-Isorna et al., 2017). Event history analysis (Mayer and Tuma, 1990;Moschis, 2012) may prove useful in detecting key factors influencing changes in attitudes and behaviors as consumers transition from drinking no alcohol to moderate consumption to binge drinking as well as from binge drinking to moderate drinking. Combining qualitative and quantitative methods can lead to novel explanations and deeper appreciation of the behavior, attitudes, triggers and motivations related to binge drinking.

| Recommendations
Finally, our review brings forward possible future research topics on how to address the difficult and complex issue of binge drinking.
Most of the studies focus on going upstream and suggest policy recommendations. Countries differ on the legal minimum age to purchase and consume alcohol. Globally most set the minimum age somewhere between 16 and 19 years with only a handful, the US included, mandating 21 years as the legal age. Research is needed to examine the extent to which stricter control on the minimum drinking age affects alcohol purchases, consumption, and associated consequences (Kypri et al., 2014;Nelson et al., 2005;Hackley et al., 2013). Other research suggests examining whether government strategies to set minimum prices on alcohol products are effective in reducing binge drinking (Hawkins & McCambridge, 2014;Gray et al., 1999). Policy recommendations also consider how can policy influence young people's focus on the long-term health risks of binge drinking (Fogerty & Chapman, 2012). Going midstream, some recommend the implementation of parental training, educational, and communications campaigns, to reduce confusion about excessive alcohol consumption and health warnings (Jones et al., 2016;Coomber et al., 2015). Finally, a number go downstream to the individual consumer level and ask how technology assisted interventions, such as smartphone applications, help to address unhealthy alcohol use (Bertholet et al., 2017). Of course, these are just some of the possible recommendations available and are not meant to be an exhaustive list.

| LIMITATIONS
No study or systematic review is without limitations, and this is no exception. This systematic review only considers research identified as binge drinking or heavy episodic drinking. This review does not examine other types of alcohol consumption, such as abstinence, moderate and excessive drinking. It may well be that different factors influence these other alcohol consumption behaviors. Although a line must be drawn, somewhere when conducting a systematic review in terms of search terms, further studies building on this may also wish to consider the broader aspects of alcohol consumption. The search terms used may have also limited the scope of the review. We included research as search term, and this may have excluded papers, which contributed to binge drinking research but did not explicitly contain this keyword or search term.
Another limitation is that many worthy pieces of published research may have been excluded from the review due to the academic requirement of including higher impact journals. Policy papers and reports have also not been referenced in this review, nor have more recent conference papers. Future research may wish to include these papers in a meta-analysis of findings and examine the effectiveness of policy interventions (Kubacki et al., 2015).

| CONCLUSION
It seems that many of the individual factors leading to binge drinking are reasonably well studied, but some matters still warrant further research such as effective coping strategies and the interplay between stress, depression, and excessive alcohol consumption. In addition, the more complex, and some would say more potent causes of peer pressure and the broader social environment, also require greater attention. Following this paper, a meta-analysis of findings would also add considerably to our understanding, notably, as in this paper, if a crossdisciplinary analysis is used and results from policy interventions included. It is also crucial for such research to consider the integrity of qualitative research and other methodological approaches not typically found in a meta-analysis. These provide essential insights into the behavior of this phenomenon.
Binge drinking is a wicked social problem that requires a crossdisciplinary understanding to address it adequately. Policymakers deserve a comprehensive and considered view of this issue rather than a narrow, blinkered one. Researchers also need to look beyond their areas to find a more thorough explanation of this phenomenon.
There is much to be gained from a more holistic view of the boundaries of expertise. Only then can meaningful progress be made in research and policy in this area.

ACKNOWLEDGMENTS
The authors wish to acknowledge the excellent editorial work on this paper by Mark Filmer and thank the editor and reviewers for their input into this manuscript. Open access publishing facilitated by University of Tasmania, as part of the Wiley -University of Tasmania agreement via the Council of Australian University Librarians.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. The data that support the findings of this study are openly available from the University of Tasmania Library as this is a systematic review.