Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Alcohol consumption in early adolescence: Associations with sociodemographic and psychosocial factors according to gender

  • Camille Pedroni ,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Writing – original draft

    camille.pedroni@ulb.be

    Affiliation Service d'Information, Promotion, Éducation Santé (SIPES), Research Centre in “Epidemiology, Biostatistics and Clinical Research”, School of Public Health, Université Libre de Bruxelles (ULB), Brussels, Belgium

  • Maud Dujeu,

    Roles Data curation, Resources, Writing – review & editing

    Affiliations Service d'Information, Promotion, Éducation Santé (SIPES), Research Centre in “Epidemiology, Biostatistics and Clinical Research”, School of Public Health, Université Libre de Bruxelles (ULB), Brussels, Belgium, Research Centre in “Social Approaches to Health”, School of Public Health, Université Libre de Bruxelles (ULB), Brussels, Belgium

  • Thérésa Lebacq,

    Roles Data curation, Investigation, Writing – review & editing

    Affiliation Service d'Information, Promotion, Éducation Santé (SIPES), Research Centre in “Epidemiology, Biostatistics and Clinical Research”, School of Public Health, Université Libre de Bruxelles (ULB), Brussels, Belgium

  • Véronique Desnouck,

    Roles Data curation, Investigation, Writing – review & editing

    Affiliation Service d'Information, Promotion, Éducation Santé (SIPES), Research Centre in “Epidemiology, Biostatistics and Clinical Research”, School of Public Health, Université Libre de Bruxelles (ULB), Brussels, Belgium

  • Emma Holmberg,

    Roles Data curation, Investigation, Writing – review & editing

    Affiliation Service d'Information, Promotion, Éducation Santé (SIPES), Research Centre in “Epidemiology, Biostatistics and Clinical Research”, School of Public Health, Université Libre de Bruxelles (ULB), Brussels, Belgium

  • Katia Castetbon

    Roles Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review & editing

    Affiliation Service d'Information, Promotion, Éducation Santé (SIPES), Research Centre in “Epidemiology, Biostatistics and Clinical Research”, School of Public Health, Université Libre de Bruxelles (ULB), Brussels, Belgium

Alcohol consumption in early adolescence: Associations with sociodemographic and psychosocial factors according to gender

  • Camille Pedroni, 
  • Maud Dujeu, 
  • Thérésa Lebacq, 
  • Véronique Desnouck, 
  • Emma Holmberg, 
  • Katia Castetbon
PLOS
x

Abstract

Introduction

Early alcohol consumption can irreversible damage the adolescents’ brain and may affect their quality of life. In order to better prevent such a deleterious behaviour, knowing its determinants is needed. So far, only few studies among adolescents aged <15 years exist, of which the majority failed to include gender differences. Therefore, the aim of this study was to investigate whether gender differences in the association between alcohol use and sociodemographic and psychosocial characteristics among 10-14-year olds exist.

Methods

Data came from the 2018 Health Behaviour in School-Aged Children (HBSC) study conducted in French-speaking schools of Belgium. The sample analysed here comprised 4,364 10-14-year olds from the Walloon Region. Associations of the recent alcohol consumption (at least one glass during the past month) with sociodemographic and psychosocial characteristics were estimated using gender-stratified multivariable logistic regression modelling.

Results

Prevalence of early alcohol consumption was 14% (boys: 16%; girls: 12%). Migration status and family affluence scale (FAS) were associated with early alcohol consumption only in boys. Second-generation immigrant boys (vs. natives: OR = 0.66 [0.47–0.92]) and boys from “low” FAS families (vs. “high”: OR = 0.56 [0.32–0.98]) or “medium” FAS (vs. “high”: OR = 0.63 [0.43–0.92]) were less likely to have consumed alcohol in the past month. In both genders, alcohol consumption was positively associated with age and inversely associated with school satisfaction and family support. No association was observed with family structure, peer support and life satisfaction in the multivariable models.

Conclusion

Our findings showed that gender differences may exist in the determinants of alcohol consumption among young adolescents. They will contribute to the development of public health policies and actions for the most vulnerable adolescents, which should take gender differences into account.

1. Introduction

Adolescence is a key period in life characterized by many social, physical, physiological and psychological changes which may make adolescents particularly vulnerable to the adverse consequences of risk-taking behaviours, including alcohol use [1]. A report based on the 2014 Health Behaviour in School Aged Children (HBSC) survey data in 36 countries and regions in Europe and North-America [2] has shown that among 15-year-old adolescents, 28% had first started consuming alcohol at age 13 or younger (31% of boys and 25% of girls). Boys were more likely to report early alcohol initiation in most countries and regions. In the French-speaking schools of Belgium (Walloon Region and Brussels-Capital Region), one quarter of boys and one girl in five had consumed alcohol at age 13 or younger [2].

An “early alcohol consumption” is generally defined as having drunk before the age of 15 [3]. Starting to drink alcohol at such a very young age carries a variety of specific health risks, especially for the brain development [4, 5]. In addition, early alcohol use, when compared with an initiation in later adolescence, is associated with a raise in the frequency and in the quantity of alcohol consumed, and with more frequent alcohol-related troubles in later adolescence and early adulthood [6, 7]. However, while drinking habits are well documented in adolescents aged 15 years and older, studies on drinking habits in adolescents under the age of 15 are scarce, despite the specific risk of negative health-related consequences in this age group [8, 9]. As a result, the determinants of early alcohol use are barely known. These determinants are related to five domains: individual, family, peer, school and community [10]. The family- and peer-domains have been most studied [1113]. For instance, parents’ and friends’ alcohol use, peer encouragement toward consumption, or low parental monitoring have been identified as risk factors for early alcohol use [1113]. The risk factors related to the other three domains have been much less studied.

Moreover, gender disparities in various health and risk behaviours in adolescence have been identified [1416]. Another HBSC research showed, among other disparities, that girls were more likely than boys to report high levels of perceived peer support, high satisfaction with school, ate more fruits, but described fair or poor health and lower life satisfaction [14]. In turn, boys were generally reporting more positive relationships, and physical activities, but also more frequent injuries and sedentary behaviours [14]. Regarding drinking behaviors, gender differencies have been highlighted in the determinants of early alcohol use. For instance, having a secure attachment to parents and to peers was a protective factor against early alcohol use but only for girls [17]. Similarly, among girls only, a higher level of perceived parental monitoring was associated with lower odds of early alcohol consumption [18]. However, for associations related to the other three domains (individual, school and community), too few data exist to draw robust conclusions on gender-related derterminants.

In order to complete the knowledge on this topic, the aim of this study was to identify potential gender differences in the associations between alcohol use and sociodemographic and psychosocial characteristics among adolescents aged 10 to 14 years. This knowledge will be helpful to the development of actions targeted to young consumers of alcohol, to avoid serious adverse impact on their current and future quality of life.

2. Materials and methods

2.1. Sample

We used data collected for the 2018 HBSC cross-sectional survey conducted in the French-speaking schools of Belgium. The HBSC study is conducted in almost 50 countries and regions using an international standardized protocol [19]. The study is undertaken every four years in the French-speaking schools of Belgium (Walloon Region and Brussels-Capital Region). Children and adolescents were asked about their health status, wellbeing and health-related behaviours through self-administered questionnaires they filled out in the classroom according to a standardized procedure, and treated as confidential and anonymous [20]. The protocol was approved by the educational authorities of each school network (private and public) and by the ethics committee of the Faculty of Psychology of the Université libre de Bruxelles (ethics opinion n°032/2017).

A two-stage random sample was used. First, schools were randomly selected from an official list of all schools stratified per province and school network, using an allocation proportional to the school population size of each province and network. Among the 807 schools invited to participate, 266 took part in the survey, with a participation rate of 33%. Secondly, one class per grade was randomly selected in the participating schools. All selected classes were willing to participate.

In an operational point of view, a letter introducing the survey was sent to the head of each selected school. This letter contained a form to be completed to notify agreement or refusal. The schools for which no feedback was received were contacted by phone or e-mail. Then, each participating school filled out informations about the classes (number of students, education branch…) which was used to randomly select them. In addition, each school was asked to designate a coordinator to handle the distribution and retrieval of the survey materials (survey questionaires, information letters, general instructions…) and to transmit them to the research team.

Adolescents and their parents received an information letter about the study objectives, the topics covered, the confidentiality and anonymity of the data collected, and their right to refuse to participate. Parents could express their refusal for the participation of their child. On the day of the questionnaire completion, adolescents also still had the opportunity to refuse to participate in the survey. The sample included all students in the selected classes, who were present on the day of questionnaire completion and who agreed to fill it in. In all selected classes, 87.0% of the students agreed to participate and eventually completed the questionnaire.

The entire 2018 HBSC sample was composed of 14,407 adolescents of 10 to 21 years of age (Fig 1). Adolescents older than 14 years of age and those who attended schools in the Brussels-Capital Region were excluded from the study sample (n = 8,489). Since the two regions include very different population structure and lifestyle (in relation with socioeconomic and cultural characteristics), analyses were here restricted to Wallonia. Those who had missing values for one or more study variables were also excluded from the analyses (n = 1,554). The final sample therefore included 4,364 adolescents aged 10 to 14 years from Walloon schools (Fig 1).

2.2. Measures

2.2.1 Alcohol consumption.

The question on alcohol consumption was: “On how many days (if any) have you drunk alcohol in the last 30 days (at least a glass of beer, wine, cocktail, aperitif…)?” Seven-answer options ranged from “never” to “30 days” [21]. The parenthesis was added to avoid that only sipping was accounted for by the respondants, since it has previously been shown that the youngest not really drink full glasses but only take small amount of alhoholic beverages [22]. Respondents were classified into two categories: adolescents who consumed alcohol at least one day during the last 30 days versus those who did not consume alcohol during the last 30 days.

2.2.2 Family & peer support.

Family and peer support was assessed through a four-item scale from the « Multidimensional Scale of Perceived Social Support » (MPSS), whose internal coherence and factorial validity have been demonstrated, especially among adolescents [23, 24]. Family support scale included statements such as “I can talk about my problems with my family” and the peer support scale, “I can count on my friends when things go wrong”. For both scales, a five-point Likert scale ranging from “strongly agree” to “strongly disagree” was given for each item. The four-item scores were summed and the sum was divided into three categories: “low” support (1 to 2.9), “moderate” support (3 to 5) and “high” support (5.1 to 7) [25].

2.2.3 School satisfaction.

School satisfaction was measured by the question: “How do you feel about school at present?”, response options were: “I like it a lot”, “I like it a bit”, “I do not like it very much” or “I do not like it at all” [21].

2.2.4 Life satisfaction.

Adolescents’ life satisfaction was assessed through the adapted version of the Cantril ladder graduated from 0 to 10, with the value 10 equivalent to "the best life possible" and the value 0 to "the worst life possible" [26]. A binary variable was created: “low” life satisfaction (from 0 to 5) versus “average to high” life satisfaction (from 6 to 10) [21].

2.2.5 Socio-demographic variables.

Socio-demographic factors we included were: gender, age (10–12 years/13-14 years) and family structure (two-parent families/blended families/one-parent families).

Adolescents were asked to indicate their country of birth and that of their parents. A “migration status” indicator was created and divided into three categories: the “first generation immigrants” (adolescents born abroad and whose parents were both not born in Belgium); the “second generation immigrants” (those born in Belgium who had at least one parent born abroad); and the “natives” (those who had both parents born in Belgium).

In addition, the adolescents’ socio-economic status was assessed through the revised Family Affluence Scale (FAS), which is composed of six items and has been previously validated [27]. Individual FAS scores ranged from 0 to 13; they were subject to a ridit transformation and classified into three groups, i.e. “low”, “medium”, and “high”, corresponding to the 20% of adolescents with the lowest FAS scores, 60% of adolescents with intermediate scores, and 20% of adolescents with the highest FAS scores, respectively.

2.3. Statistical analyses

To improve the estimates’ representativeness, individual weighting factors were calculated according to the inclusion probability of schools in the initial sample, schools characteristics (size, province, network and socioeconomic status) and the differences observed between the adolescents who took part in the survey and those of the reference population in terms of school grade and gender. These weighting factors as well as the sample design were taken into account in all the statistical analyses.

After the description of the sample, univariate analyses were performed to identify characteristics and behaviours associated with adolescents’ alcohol consumption during the last 30 days before the survey. Statistical associations were estimated using Pearson’s chi-square tests (Rao & Scott correction for the sampling plan). Odds ratio (OR) and 95% confidence intervals [95%CI] were calculated using logistic regressions [28]. All indicators associated with alcohol consumption in the univariate analyses with a P-value<20% were included in the multivariable models. Inter-correlations between explanatory variables were tested using Cramer’s V test; all values were lower than 0.25 (S1 Appendix). All statistical analyses were performed using Stata® V.14.

3. Results

Table 1 presents the characteristics of the sample, overall and by gender. Alcohol use during the month before the interview was higher in boys than in girls. The proportion of adolescents feeling supported by their family was higher in boys (83%) than in girls (76%). The opposite situation was observed for peer support (boys: 66%; girls: 76%). About a third liked school a lot (boys: 27%; girls: 35%). Almost all had a high level of life satisfaction (boys: 92%; girls: 87%) (Table 1).

thumbnail
Table 1. Characteristics of the sample overall and by gender (n = 4,364).

HBSC, French-speaking Walloon schools in Belgium, 2018.

https://doi.org/10.1371/journal.pone.0245597.t001

In univariate analyses, in boys, all characteristics and behaviours except family structure and peer support, were significantly associated with alcohol consumption (Table 2). In girls, only age, family support, school satisfaction and life satisfaction were significantly associated with alcohol consumption. In the multivariable logistic regression models (Table 2), age, family support and school satisfaction were significantly associated with alcohol consumption in both boys and girls. Adolescents aged 13–14 years (vs. 10–12 years: boys: ORa = 3.14 [2.34–4.22]; girls: ORa = 4.45 [3.15–6.28]) were more likely to have consumed alcohol in the last 30 days preceding the survey. Associations between family support and early alcohol consumption were a little different between boys and girls. Girls who felt “moderate” (ORa = 1.59 [1.04–2.43]) or “low” (ORa = 1.94 [1.17–3.21]) support by their family were more likely to have consumed alcohol during the past month compared with the “high” category. In boys, this association was only statistically significant in the “moderate” versus the “high” categories (ORa = 1.58 [1.13–2.22]). In both boys and girls, school satisfaction was inversely associated with alcohol consumption. Compared with those who liked school very much, those who liked school a bit (boys: ORa = 2.15 [1.45–3.20]; girls: ORa = 1.68 [1.08–2.60]), those who did not like school very much (boys: ORaa = 2.33 [1.25–4.33]; girls: ORa = 2.12 [1.30–3.46]), and those who did not like school at all (boys: ORa = 4.88 [2.85–8.36]; girls: ORa = 3.35 [1.85–6.08]), were more likely to have consumed alcohol in the last 30 days before the survey (Table 2).

thumbnail
Table 2. Characteristics and behaviors associated with alcohol consumption during the last 30 days before the survey among adolescents aged 10 to 14.

HBSC, French-speaking Walloon schools in Belgium, 2018.

https://doi.org/10.1371/journal.pone.0245597.t002

In boys only, migration status and FAS were significantly associated with early alcohol consumption (Table 2). Boys from “low” FAS (ORa = 0.56 [0.32–0.98]) or “medium” FAS families (ORa = 0.63 [0.43–0.92]) compared with those from “high” FAS families were less likely to have consumed alcohol during the past month. Lastly, second-generation immigrant boys (ORa = 0.66 [0.47–0.92]) were less likely to have consumed alcohol compared to natives (Table 2).

4. Discussion

The aim of this study was to identify potential gender differences in the correlates of alcohol consumption in early adolescence. We did not find any gender difference with the psychosocial indicators. The main gender differences were found for migration status and FAS, which were significantly associated with alcohol consumption in boys only. Indeed, our results showed that second-generation immigrant boys were less likely to have consumed alcohol compared to natives. Consistently with these findings in boys, a Dutch longitudinal study in adolescents age 10–14, found that being a native was a predictor of early alcohol use [9]. Some cross-sectional studies came to the same conclusion [29]. However, the results could differ depending on the country of origin of the immigrant adolescent. In fact, alcohol consumption in adolescents is closely linked to religion as well as to the culture and customs of the country of origin. For instance, prevalence of alcohol consumption is generally higher in adolescents from Eastern European countries compared with those from Northern Europe [30, 31].

Concerning FAS, boys from “low” FAS or “medium” FAS families compared with those from “high” FAS families were less likely to have consumed alcohol during the past month. In a recent review, no clear pattern of associations between socio-economic status (SES) and alcohol consumption was identified [32]. A meta-analysis focusing on adolescents aged 10–15, indicated that the lower the SES, the higher the prevalence of alcohol use. However, results were inconclusive when regarding European studies only [33]. Some authors have suggested that these conflicting results between studies may be partly due to the measures used to assess the SES (e.g. parents’ education level, occupational status, family affluence or income) and the drinking behavior assessed (sipping, lifetime use, binge drinking, etc.) [22, 3436]. For example, several studies have shown that alcohol misuse was more prevalent among adolescents with low SES, whereas experimental alcohol use occurred more often among those with high SES [34, 35]. No study focusing on gender differences in the association between SES and early alcohol consumption was found.

The influence of the family structure on adolescents’ substance use has been widely studied [37, 38]. It has been argued that living in two-parent families could play a protective role on adolescent substance use. This hypothesis was confirmed by many cross-sectional and longitudinal studies [29, 39]. Very few studies have focused on potential gender differences however. One study, in 11-15-year-old Flemish adolescents, showed that boys from blended families and girls from one-parent families were more likely to consume alcohol at least once a week in comparison with adolescents from two-parent families [40]. In our study, even in the univariate analysis, no association was found at all. Further research therefore is needed to confirm whether any relationship between early alcohol use and family structure exists.

The influence of parenting factors (parents drinking behaviour, parenting style, parent’s rules toward alcohol, etc.) on adolescent alcohol use has also been widely investigated in previous research [4042]. Furthermore, a recent systematic review and meta-analysis of longitudinal studies focusing on modifiable parenting factors in adolescents aged 12–18 concluded that parental support was a protective factor related to both alcohol initiation and levels of later use [13]. Other studies among younger adolescents came to the same conclusion [42]. We also found an association between parental support and alcohol consumption in our study, with little difference between boys and girls.

In our study, in both genders, we did not find an association between peer support and early alcohol use. In the literature, several studies have examined the role on early drinking of certain determinants related to peer influence and some, such as peer approval, have been identified as predictors for starting to drink [11]. However, among the few studies that have examined this association, findings are divergent. For example, the Welsh study [43] found that peer support was a protective factor against binge drinking among 11-15-year-old adolescents, whereas another study among 15 year olds in the U.S. [12], showed that peer support neither contributed to nor protected against the alcohol use (based on alcohol consumption frequency during the last 30 days). These divergent findings might be due to a lack of comparability in the tools used to assess peer support or due to interactions with other indicators not included in this study such as peer involvement in the alcohol consumption. Further research, using a qualitative assessment, is highly needed to elucidate the role of peer support on early alcohol use.

We found a negative association in both genders between school satisfaction and alcohol consumption during the past month prior the survey. Two cross-sectional studies showed that 11-15-year olds who liked school were less likely to drink alcohol [43, 44]. In contrast, two other cross-sectional studies found no association [29, 41]. However, the adolescents included in these two latter were older (15 to 20 years of age) than those included in ours. It therefore may be assumed that school satisfaction may have an influence on alcohol consumption only among young adolescents.

Several studies have found a negative association between life satisfaction and early alcohol use [10, 43]. For example, an American study showed that 10-17-year olds with “low” life satisfaction were also more likely to consume alcohol in conjunction with tobacco and/or marijuana [10]. In our study, in both gender, we found a significant association between life satisfaction and early alcohol use in univariate analysis, but this association disappeared in multivariate models, when school satisfaction was added into the models.

One strength of our study lies in the focus on adolescents from the age of 10. Although a number of studies among young adolescents identified some predictors of alcohol consumption [3, 6, 11, 17], they remain too few and inconsistent between each other to be able to draw clear conclusions about such determinants. Our study also extends previous research by investigating gender differences in the association between early alcohol use and sociodemographic and psychosocial factors. Furthermore we also examined the role in this association, of some indicators which were less studied before (peer support for example). Our findings therefore provide new information on the sociodemographic and psychosocial factors possibly involved in alcohol use in early adolescence in both genders. Finally, we used a large randomized sample, which is another strength to our study.

At the same time, some limitations should also be considered. Firstly, in our cross-sectional study, no causal link can be inferred from the observations made. Indeed, bilateral associations could exist between early alcohol consumption and some characteristics such as family support. We can hypothesize that adolescents with “low” family support are more likely to drink alcohol, but alcohol consumption may also make adolescents less close to their family and therefore, feeling less supported by them. Secondly, all data collected were self-reported which can be a source of declaration bias [45]. One of them is the social desirability bias which may lead, among others, to an underestimation of alcohol consumption [45, 46]. Another possible bias is linked to memorization due to the fact that we have asked adolescents to report their alcohol consumption during the past month. Nevertheless, some evidence is available that short recall periods like we used here lead to more accurate reporting [45].

5. Conclusion

We investigated the determinants of alcohol consumption among adolescents under 15 years of age taking gender into account. This topic has been insufficiently treated so far, with the exception of family support and peer influence. Our findings therefore provide crucial new knowledge about this public health issue, since alcohol consumption in young adolescents can have serious consequences for their current and future health and quality of life.

Whereas both genders shared common risk factors (age, family support and school satisfaction), our findings have highlighted gender differences in the following sociodemographic risk factors associated with early alcohol consumption: migration status and FAS. Nevertheless, since a number of the potential determinants we have analysed here have received little attention in the previous studies, we need research on gender-specific determinants of early alcohol consumption to draw firm conclusions. This knowledge will help the developers of public health policies and actions to better target and adapt messages to the most vulnerable adolescents.

Supporting information

S1 Appendix. Inter-correlations between variables included in the study.

https://doi.org/10.1371/journal.pone.0245597.s001

(DOCX)

Acknowledgments

The authors thank the schools and adolescents for their participation in the French-speaking Belgian HBSC survey. They also thank the HBSC International Coordination Center (University of St. Andrews, United Kingdom) and the HBSC Data Management Center (University of Bergen, Norway) for their scientific support. They are grateful to Estelle Méroc, Nathalie Moreau, Amélie Bellanger, Morgane Eggen and Omer Cimpaye for their contribution in the data collection and analysis, and to Louise Mennen for scientific editing.

References

  1. 1. Steinberg L. Cognitive and affective development in adolescence. Trends Cogn Sci (Regul Ed). 2005;9:69–74. pmid:15668099
  2. 2. Inchley J, Currie D, Vieno A, Torsheim T, Ferreira-Borges C, Waber MM, et al. Adolescent alcohol-related behaviours: trends and inequalities in the WHO European Region, 2002–2014. 2018. Available from: https://www.euro.who.int/en/publications/abstracts/adolescent-alcohol-related-behaviours-trends-and-inequalities-in-the-who-european-region,-20022014-2018.
  3. 3. Donovan JE, Molina BSG. Childhood risk factors for early-onset drinking. J Stud Alcohol Drugs. 2011;72:741–51. pmid:21906502
  4. 4. Squeglia LM, Jacobus J, Tapert SF. The effect of alcohol use on human adolescent brain structures and systems. Handb Clin Neurol. 2014;125:501–10. pmid:25307592
  5. 5. Tapia-Rojas C, Mira RG, Torres AK, Jara C, Pérez MJ, Vergara EH, et al. Alcohol consumption during adolescence: A link between mitochondrial damage and ethanol brain intoxication. Birth Defects Res. 2017;109:1623–39. pmid:29251843
  6. 6. Guttmannova K, Hill KG, Bailey JA, Lee JO, Hartigan LA, Hawkins JD, et al. Examining explanatory mechanisms of the effects of early alcohol use on young adult alcohol dependence. J Stud Alcohol Drugs. 2012;73:379–90. pmid:22456243
  7. 7. Yuen WS, Chan G, Bruno R, Clare P, Mattick R, Aiken A, et al. Adolescent Alcohol Use Trajectories: Risk Factors and Adult Outcomes. Pediatrics;146:e20200440. pmid:32968030
  8. 8. Donovan JE. The burden of alcohol use: focus on children and preadolescents. Alcohol Res. 2013;35:186–92. pmid:24881327
  9. 9. Visser L, Winter AF de, Vollebergh WAM, Verhulst FC, Reijneveld SA. Do child's psychosocial functioning, and parent and family characteristics predict early alcohol use? The TRAILS Study. Eur J Public Health. 2015;25:38–43. pmid:24935960
  10. 10. Lew D, Xian H, Qian Z, Vaughn MG. Examining the relationships between life satisfaction and alcohol, tobacco and marijuana use among school-aged children. J Public Health (Oxf). 2018.
  11. 11. Donovan JE. Adolescent alcohol initiation: a review of psychosocial risk factors. J Adolesc Health. 2004;35:529.e7–18. pmid:15581536
  12. 12. Branstetter SA, Low S, Furman W. The influence of parents and friends on adolescent substance use: A multidimensional approach. J Subst Use. 2011;16:150–60. pmid:21747736
  13. 13. Yap MBH, Cheong TWK, Zaravinos-Tsakos F, Lubman DI, Jorm AF. Modifiable parenting factors associated with adolescent alcohol misuse: a systematic review and meta-analysis of longitudinal studies. Addiction. 2017;112:1142–62. pmid:28178373
  14. 14. Inchley J, Currie D, Young T, Samdal O, Torsheim T, Augustson L, et al. Health behaviour in school-aged children (HBSC) study: International report from the 2013/2014 survey. http://www.euro.who.int/__data/assets/pdf_file/0003/303438/HSBC-No.7-Growing-up-unequal-Full-Report.pdf?ua=1.
  15. 15. Chen P, Jacobson KC. Developmental trajectories of substance use from early adolescence to young adulthood: gender and racial/ethnic differences. J Adolesc Health. 2012;50:154–63. pmid:22265111
  16. 16. Cheng HG, Cantave MD, Anthony JC. Taking the first full drink: Epidemiological evidence on male-female differences in the United States. Alcohol Clin Exp Res. 2016;40:816–25. pmid:27038595
  17. 17. Danielsson AK, Romelsjo A, Tengstrom A. Heavy episodic drinking in early adolescence: gender-specific risk and protective factors. Subst Use Misuse. 2011;46:633–43. pmid:20964532
  18. 18. Nelson KM, Carey KB, Scott-Sheldon LAJ, Eckert TL, Park A, Vanable PA, et al. Gender differences in relations among perceived family characteristics and risky health behaviors in urban adolescents. Ann Behav Med. 2017;51:416–22. pmid:27995548
  19. 19. HBSC International Coordinating Centre Child & Adolescent Health Research Unit. About HBSC study. Available from: http://www.hbsc.org/about/index.html.
  20. 20. HBSC International Coordinating Centre Child & Adolescent Health Research Unit. Survey Methods. Available from: http://www.hbsc.org/methods/index.html.
  21. 21. Currie C, Inchley J, Molcho M, Lenzi M, Veselka Z, Wild F. Health Behaviour in School-aged Children (HBSC). Study protocol: Background, methodology and mandatory items for the 2013/14 Survey. St Andrews: CAHRU. 2014.
  22. 22. Wadolowski M, Bruno R, Aiken A, Stone C, Najman J, Kypri K, et al. Sipping, drinking, and early adolescent alcohol consumption: A cautionary note. Alcohol Clin Exp Res. 2015;39:350–4. pmid:25684054
  23. 23. Zimet GD, Dahlem NW, Zimet SG, Farley GK. The Multidimensional Scale of Perceived Social Support. J Pers Assess. 1988;52:30–41.
  24. 24. Canty-Mitchell J, Zimet GD. Psychometric properties of the Multidimensional Scale of Perceived Social Support in urban adolescents. Am J Community Psychol. 2000;28:391–400. pmid:10945123
  25. 25. Zimet GD. Multidimensional scale of perceived social support (MSPSS)—Scale items and scoring information. Available from: http://gzimet.wix.com/mspss.
  26. 26. Levin KA, Currie C. Reliability and validity of an adapted version of the Cantril Ladder for use with adolescent samples. Soc Indic Res. 2014;119:1047–63.
  27. 27. Torsheim T, Cavallo F, Levin KA, Schnohr C, Mazur J, Niclasen B, et al. Psychometric validation of the revised family affluence scale: a latent variable approach. Child Indic Res. 2016;9:771–84. pmid:27489572
  28. 28. Rao JNK, Scott AJ. The analysis of categorical data from complex sample surveys: chi-squared tests for goodness of fit and independence in two-way tables. J Am Stat Assoc. 1981;76:221–30.
  29. 29. Harakeh Z, Looze ME de, Schrijvers CTM, van Dorsselaer SAFM, Vollebergh WAM. Individual and environmental predictors of health risk behaviours among Dutch adolescents: the HBSC study. Public Health. 2012;126:566–73. pmid:22607981
  30. 30. Inchley J, Currie D, Vieno A, Torsheim T, Ferreira-Borges C, Waber MM, et al. Adolescent alcohol-related behaviours: trends and inequalities in the WHO European Region, 2002–2014. Available from: http://www.euro.who.int/__data/assets/pdf_file/0007/382840/WH15-alcohol-report-eng.pdf?ua=1. 2018.
  31. 31. Kraus L, Guttormsson U, Leifman H, Arpa S, Molinaro S, Monshouwer K et al. ESPAD Report 2015. Results from the European School Survey Project on alcohol and other drugs. ESPAD and EMCDDA. Luxembourg. Publications Office of the European Union. 2016. Available from: http://www.espad.org/sites/espad.org/files/ESPAD_report_2015.pdf.
  32. 32. Hanson MD, Chen E. Socioeconomic status and health behaviors in adolescence: a review of the literature. J Behav Med. 2007;30:263–85. pmid:17514418
  33. 33. Lemstra M, Bennett NR, Neudorf C, Kunst A, Nannapaneni U, Warren LM, et al. A meta-analysis of marijuana and alcohol use by socio-economic status in adolescents aged 10–15 Years. Can J Public Health. 2008;99:172–7. pmid:18615935
  34. 34. Sweeting H, Hunt K. Adolescent socioeconomic and school-based social status, smoking, and drinking. J Adolesc Health. 2015;57:37–45. pmid:26095407
  35. 35. Shackleton N, Milne BJ, Jerrim J. Socioeconomic inequalities in adolescent substance use: Evidence from twenty-four european countries. Subst Use Misuse. 2019;54:1044–9. pmid:30648460
  36. 36. Torikka A, Kaltiala-Heino R, Luukkaala T, Rimpela A. Trends in alcohol use among adolescents from 2000 to 2011: The role of socioeconomic status and depression. Alcohol Alcohol. 2017;52:95–103. pmid:27507821
  37. 37. Hoffmann JP. Family structure and adolescent substance use: An international perspective. Subst Use Misuse. 2017;52:1667–83. pmid:28605218
  38. 38. McArdle P, Wiegersma A, Gilvarry E, Kolte B, McCarthy S, Fitzgerald M, et al. European adolescent substance use: the roles of family structure, function and gender. Addiction. 2002;97:329–36. pmid:11964109
  39. 39. Brown SL, Rinelli LN. Family structure, family processes, and adolescent smoking and drinking. J Res Adolesc. 2010;20:259–73. pmid:20543893
  40. 40. Vanassche S, Sodermans AK, Matthijs K, Swicegood G. The effects of family type, family relationships and parental role models on delinquency and alcohol use among Flemish adolescents. J Child Fam Stud. 2014;23:128–43.
  41. 41. Balazs MA, Piko BF, Fitzpatrick KM. Youth problem drinking: The role of parental and familial relationships. Subst Use Misuse. 2017;52:1538–45. pmid:28406347
  42. 42. Johnson B, McBride D, Hopkins G, Pepper S. An examination of parent–child relationships and teen substance use: A brief report. J Child Adolesc Subst Abuse. 2014;23:210–6.
  43. 43. Desousa C, Murphy S, Roberts C, Anderson L. School policies and binge drinking behaviours of school-aged children in Wales—a multilevel analysis. Health Educ Res. 2008;23:259–71. pmid:17639119
  44. 44. Phillips-Howard PA, Bellis MA, Briant LB, Jones H, Downing J, Kelly IE, et al. Wellbeing, alcohol use and sexual activity in young teenagers: findings from a cross-sectional survey in school children in North West England. Subst Abuse Treat Prev Policy. 2010;5:27. pmid:21067562
  45. 45. Brener ND, Billy JOG, Grady WR. Assessment of factors affecting the validity of self-reported health-risk behavior among adolescents: Evidence from the scientific literature. J Adolesc Health. 2003;33:436–57. pmid:14642706
  46. 46. Davis CG, Thake J, Vilhena N. Social desirability biases in self-reported alcohol consumption and harms. Addict Behav. 2010;35:302–11. pmid:19932936