Background
In Australia the national guidelines state that for those aged less than 18 years of age, not drinking is the safest option [
1]. Adolescent drinking is associated with increased risk of accidents and injury, including vehicle accidents [
2], suicide and violence [
3,
4]. Regular alcohol consumption or binge drinking during adolescence also associated with a range of negative health and social outcomes including physical and mental health problems, risky sexual behaviour, poor school performance and anti-social behaviour; and long-term health effects such as alcohol dependence and poor health outcomes in early and middle adulthood [
2,
5,
6]. However, in 2011, 51 % of Australian children aged 12–17 reported consuming alcohol in the previous 12 months; 17 % in the last week [
7].
The influences on adolescent drinking behaviour are complex. There is strong evidence for the role of alcohol advertising and marketing [
8,
9], personality factors [
10,
11] and the drinking behaviours and attitudes of peers [
12‐
15], and siblings [
16]. However, a growing body of evidence indicates that parents exert significant influence over adolescents’ underage drinking behaviour; children of parents permit or accept underage drinking are more likely to consume alcohol in adolescence than those whose parents apply prohibitions or strict rules and emphasise the negative effects of alcohol [
12,
17,
18].
There are considerable variations in reports of parental supply of alcohol between countries, and between parents and adolescents within countries [
19]. However, it is clear that parents are a common source of alcohol for (particularly younger) adolescents; with this finding being replicated in a range of countries including the United States [
20,
21], Sweden [
22], Ireland [
23] and Australia [
24]. In the 2011 Australian secondary school survey, 34.9 % of 12-to-15-year-olds and 31.3 % of 16-to-17-year-olds report that their last alcoholic drink was provided to them by their parents [
7]; and in a survey of 530 secondary students in New South Wales, 40.7 % of drinkers reported receiving alcohol from their parents in the last month, with younger respondents were more likely to report that their parents were their
main source of alcohol [
25].
An analysis of data from the 2007 Australian National Drug Strategy Household Survey found that adolescent drinkers who recalled receiving their first drink from their parents had lower rates of risky drinking than those who recalled receiving their first drink from another source [
26]; and the analysis of youth data from the US National Survey on Drug Use and Health showed that those who obtained their last drink from a parent or family member drank less frequently and at lower levels than those who obtained their alcohol from an unrelated adult or purchased it themselves [
27]. Conversely, studies in Sweden [
28], Australia [
25], and concurrently in the US and Australia [
29] have shown no evidence that parental supply leads to more ‘responsible’ drinking patterns, rather that adolescents were more likely to engage in harmful drinking behaviours if their parents provided them with alcohol.
Determining the exact nature of this relationship is complicated by the combination of source of supply (parent or other), location of drinking, presence or absence of supervision, and other contextual or situational differences. For example a cross-sectional telephone survey of 6245 US adolescents found that adolescents who were with their parents last time they drank alcohol reported less frequent and more moderate drinking, whereas those whose parents or friends’ parents
1 had provided them with alcohol at a party reported more frequent and more hazardous drinking [
21]. Similarly, an Australian survey of 530 secondary students found that those who were provided alcohol by their parents for consumption without (their own) parental supervision were more likely to be risky drinkers [
25]. Finally, data from the 2011 Australian national school survey reported that current drinkers drank less alcohol per week if they obtained their alcohol from their parents and drank less if they consumed the alcohol at home; however 16–17-year-olds who drank at a party consumed significantly less if friends supplied the alcohol than if parents or someone else provided it [
7].
It is well-documented that adolescents perceive strong descriptive norms encouraging drinking and weak injunctive norms discouraging drinking [
30‐
33]. There is increasing evidence that parents perceive similar norms in relation to the provision of alcohol to adolescents and that perceived norms surrounding the drinking behaviours condoned by ‘other parents’ may influence parental attitudes [
34]. However, there are also a small number of studies that suggest that adults perceive that their own views are comparatively conservative and that the broader community is more accepting of underage drinking than they are themselves [
35,
36].
Thus, it is clear that strategies to address the problem of underage drinking need to reach beyond targeting young people themselves and begin to address the attitudes and values held by parents and community members [
37]. Reductions in underage drinking will only come about from changes in the social and cultural environment in which our young people are learning about the role of alcohol in their lives. In recognition of the important role of parents in discouraging (or facilitating) alcohol consumption, several interventions in the US [
38,
39], and Europe [
40] have included components targeting teenagers and their parents, individually and concurrently.
This paper reports a qualitative study of adolescents and parents in a regional town New South Wales (Australia), a jurisdiction where it is not unlawful for children or adolescents to consume alcohol. Rather, it is unlawful for a person who is less than 18 years of age to purchase alcohol or to consume alcohol on licensed premises or in public places; or to sell alcohol to a person under the age of 18, or to supply them with alcohol in the absence of parental consent to do so. It is not unlawful for parents to provide alcohol to their own children for consumption in a private home or for others to provide alcohol to children for consumption on private premises (not including licensed premises), if parental consent is obtained.
The study explored knowledge, attitudes and experiences of the supply of alcohol to minor (not yet 18) children by their parents. The aim of the study was to investigate adolescents’ and parents’ perceptions of these behaviours, and their perceptions (and perceived personal relevance) of previous social marketing campaigns.
The study was informed by the Theory of Planned Behaviour (TPB), which posits that attitudes, subjective norms and perceived behavioural control predict behavioural intention [
41,
42]. The TPB has been used extensively to explain drinking and drinking intentions among young adults, such as college students [
43‐
47], more recently in studies among adolescents [
48,
49] and has potential for understanding parental supply of alcohol [
19]. For example, a study of 247 secondary school pupils (mean age 16.6 years) across a range of health-related behaviours (including drinking alcohol) found TPB, together with past behaviour, explained 62 % of the variance in health-risk intentions, and 51 % of the variance in health-protective intentions [
50].
Methods
The study was conducted in a local government area (LGA) in New South Wales, Australia, that is 120 km from the state’s capital city. The LGA includes a town with a population of approximately 20,000 and has a Socio Economic Index for Area (SEIFA; general level of socio-economic (dis)advantage of people living in the area) value of 1055, which is above the mean.
2 The town has one publically funded high school and its own weekly newspaper, and the median weekly household income is AUD1,099 AUD.
Focus group participants were recruited by advertisements placed in the local newspaper, community newsletters and on social media platforms; and posters displayed in community spaces including libraries, doctors’ surgeries and shopping centres. Twelve focus groups were conducted in venues such as libraries, the local high school and neighbourhood centre: four with parents of teenagers (n = 27) and eight with adolescents (three groups of 12–14-year-olds, three groups of 15–17-year olds, and two mixed groups; total n = 47). One parent group and one group of 12–14-year-olds were conducted concurrently and consisted predominantly of members of the same families; it is possible that some of the other adolescent participants may have been related to adult participants but did not disclose this.
Focus groups generally ran for one hour and consisted of 6–10 participants who were offered retail vouchers in return for their time ($50 for adults and $30 for adolescents). All discussions were audio-recorded and later transcribed in full. Recruitment methods, focus group discussion guides and all data collection methods were approved by the University of Wollongong’s Human Research Ethics Committee. Adult participants provided written consent, and adolescent participants provided both written assent and written parental consent; this included consent to publish the de-identified findings. The Consolidated Criteria for Reporting Qualitative Findings (COREQ) [
51] was used to provide a systematic framework for the design, analysis and reporting of this study. A comprehensive checklist against the 32-item COREQ as well as all other transcripts and data files can be obtained from the first author upon request.
Basic demographics of participants were recorded at the commencement of each focus group using a self-completion survey (see Table
1). The mean age of adolescent participants was 14.8 years (range 12–17 years) and of parents was 46.7 years (range 37–62 years); 55.3 % of adolescents and 70.4 % of parents were female. The majority of adolescents and parents were born in Australia and spoke English at home.
Table 1
Participant demographics
Gender |
Male | 44.7 % | 21 | 29.6 % | 8 |
Female | 55.3 % | 26 | 70.4 % | 19 |
Country of birth |
Australia | 89.4 % | 42 | 81.5 % | 22 |
Other | 10.6 % | 5 | 18.5 % | 5 |
Language spoken at home |
English | 93.6 % | 44 | 92.6 % | 25 |
Other | 6.3 % | 3 | 7.4 % | 2 |
Religion |
Catholic | 31.9 % | 15 | 18.5 % | 5 |
Anglican | 19.1 % | 9 | 14.8 % | 4 |
Other religion | 4.2 % | 2 | 22.6 % | 6 |
No religion | 44.7 % | 21 | 44.4 % | 12 |
Focus group discussions then followed a thematic discussion guide which explored knowledge and attitudes and around alcohol consumption by, and parental supply of alcohol to, underage teenagers. Following this, the groups were shown examples of print materials from social marketing campaigns to reduce underage drinking and participants engaged in a discussion regarding their opinions on the various campaigns.
For adolescent groups, in addition to discussion guides, facilitators used activities to stimulate discussion. Groups comprising 12–14 year olds were given picture and photo sorting activities; and groups comprising 15–17 year olds adjectival word sorting activities. These activities focused on sorting and then discussing impressions of ‘drinkers’ and ‘non-drinkers’ and of parents who ‘do’ and ‘do not’ provide alcohol to their teenage children.
The first and second authors, both experienced qualitative researchers, independently examined and manually coded the data. Both coders commenced by grouping quotes under the key headings of the Discussion Guide, itself designed to elicit responses corresponding to the constructs of the TPB (attitudes towards underage drinking and parental supply, subjective norms and perceived behavioural control), and then coded for specific issues and themes. The two coders met to consider these key groupings and to resolve any differences in interpretation. The third author reviewed the draft findings, then read the verbatim transcripts to confirm the clarity and accuracy of interpretation.
Discussion
The Theory of Planned Behaviour (TPB) posits that the predictors of behavioural intention are attitudes, subjective norms, and perceived behavioural control [
41,
42]. In the context of parental supply of alcohol, this suggests that if parents believe that supplying alcohol to children and teenagers is wrong (desired attitudes), that the majority of their peers do not provide alcohol to children and teenagers (desired subjective norms) and that the provision of alcohol to children and teenagers is within their control (desired perception of behavioural control), they will not provide alcohol to their children before they reach the age of 18 (the legal alcohol purchase age in Australia).
However, it is clear from the results of this study and from existing national data [
7], that many parents
do provide alcohol to their underage children. Consistent with previous research [
20,
53,
54], adolescents in our focus groups were more likely to report that their parents provided them with alcohol than parents (some of whom came from the same families, and all of whom came from the same small community) were to report doing so.
Our participants wholeheartedly agreed with the messages (they thought) we and ‘the government’ were communicating – that supplying alcohol to teenagers is inappropriate. They interpreted ‘supply’ to refer to the provision of quantities of alcohol for unsupervised drinking, such as for consumption at parties, and clearly distanced themselves from people who would engage in this behaviour. The homogeneity of the responses was noteworthy, with all of the parents communicating that they would not ‘supply’ alcohol and almost all providing small amounts or tastes of alcohol in specific contexts. It is possible – given that this behaviour is associated with moral censure and ‘bad’ parenting – that our recruitment strategy did not attract parents who provide larger amounts of alcohol to their children and/or that some of the participants do indeed do so but were unwilling to state this in the group context.
The provision of alcohol to children and adolescents can broadly be described as three behaviours: allowing children to sip or taste alcohol; allowing them to have a drink of alcohol at home; and supplying them with alcohol to take to a party. While each of the behaviours may be potentially harmful, the degree of harm and the consistency of research evidence for harm (or benefit) vary, as do parental and community attitudes. The findings of the present study are consistent with previous evidence that parents see a clear distinction between ‘sipping’ and ‘drinking’; for example, an earlier Australian study which found that parents reported having strict rules prohibiting their children from drinking but allowed them to ‘sip’ or ‘taste’ alcohol [
55]. Parents interviewed for this study clearly perceived their own behaviour – providing (small amounts of) alcohol to their teenage children – to be a fundamentally different (and appropriate) behaviour. They were ‘teaching their children to drink’ and their children were learning that their parents would provide them with alcohol for consumption in ‘safe’ places, in ‘safe’ quantities, with their families or with other ‘good’ children. These parents do not perceive any correlation between their behaviour and excessive, problematic or unsafe alcohol use and they thus do not identify with the characters or scenarios depicted in recent fear-based social marketing campaigns – which they see as targeted at problematic ‘other’ parents and children. As a result, they remain uninfluenced by campaigns targeting parental supply.
In turn, many of the (older) adolescents in our focus groups perceive that their parents condone or even encourage them to consume alcohol prior to reaching 18 years of age and, like their parents, do not perceive that underage drinking campaigns are targeted at people like them. It appears these children had learned from their parents that supplying alcohol does not include allowing children to taste or sip alcoholic drinks at home and that underage drinking does not include drinking that takes place in the presence, or with the permission, of parents.
The findings of this study have important implications for the development of communication materials and social marketing campaigns targeting underage drinking, and particularly parental supply. It is essential to ensure that the target audience perceive themselves to be the target audience, and are not in fact commenting on the effectiveness of the message for ‘other’ people. The frame of reference for, and interpretation of terms by, target audiences can be fundamentally different to that of message developers. High-fear graphic images may test well with adolescents and parents because they genuinely believe these are the most effective strategy for ‘those’ children and ‘those’ parents. Following their recommendations is, then, likely to result in campaigns that the intended target audience will not perceive as relevant to them, or as addressing their own behaviour. These messages will not impact on their attitudes or behaviours because they believe they are already doing the ‘right’ thing.
Thus, the first goal of such campaigns should be to ensure that the target audience perceives themselves to be the target audience; which can be achieved by utilising images and words that ‘nice’ parents with ‘good’ children identify with. For example, in our subsequent development of messages for this community, rather than utilising stock images that are reminiscent of previous high fear advertising campaigns we are taking photographs of clean-cut children in recognisable local environments. We are also using the words of our focus group participants in messages and taglines; for example, as one parent stated when she realised that her own children were also at risk of harm: “Bad things happen to good kids too”.
The findings also have implications for addressing the entrenched behaviour of parental supply of alcohol to teenagers by their parents (which is both legal and socially accepted in Australia). Strategies to reduce parental supply, and thus underage drinking, need to increase parents’ awareness of the negative short- and long-term effects of any actions that condone underage drinking in any context, including providing small amounts of alcohol to children and teenagers in private homes.
Limitations
This study was conducted in one regional town in eastern Australia, and thus the results may not be generalizable to larger cities or other regions; future research in other locations could explore similarities and differences in perspectives. The use of written materials (in English) for recruitment limited our ability to attract participants who were not fluent in English, had lower levels of literacy, or were otherwise hard-to-reach groups. While the use of Facebook as an additional recruitment strategy broadened our reach, we note that all of our participants were fluent English speakers, with 93 % speaking English at home, and all were sufficiently literate to read the participant information sheet and provided informed consent. We utilised mixed-gender focus groups to explore these behaviours as young people typically drink in mixed-gender social groups and parents make decisions about supply with their spouses. However, the use of mixed-gender focus groups meant that we did not explore gender differences in the participants’ responses. Future research could usefully explore differences between male and female adolescents (and between mothers and fathers) in both perceptions of the risks associated with underage drinking and the most salient social marketing messages.