Adolescents’ Alcohol Use: Does the Type of Leisure Activity Matter? A Cross-National Study

The main objective of this study was to analyze the relationship between structured, unstructured, and family leisure activities on the frequency of adolescent alcohol intake across three different countries (Spain, Peru, and The Netherlands). The self-control of adolescents was also investigated as a moderator in the relationship between leisure activities and alcohol consumption. Methodology: This research involved 4608 adolescents aged between 12 and 17 from three countries (Spain, Peru, and The Netherlands). In Spain and Peru, data was collected through a self-report questionnaire which was part of the Your Life project. In The Netherlands, a self-questionnaire was used, collected by the University of Utrecht. A multiple logistic regression was performed for each country. Results: The results showed that participation in unstructured leisure activities increased the likelihood of drinking more frequently and more heavily in all three countries. Structured leisure activities, in general, did not have a significant predictive effect on alcohol consumption in any of the countries. Family leisure activities reduced the risk of engaging in yearly alcohol use and yearly binge drinking among adolescents, especially in The Netherlands and Spain. The protective effect of family leisure and unstructured leisure risk on yearly alcohol use applied especially to Dutch adolescents with a low level of self-control. Discussion: The article emphasizes the need for parents to engage in leisure activities with their child; participation in unstructured activities is not to be encouraged.


Introduction
Adolescence is a stage in the life cycle in which young people start experimenting with the use of substances, such as the initiation of alcohol consumption. Adolescence is characterized by a period in time where youngsters are not concerned about the consequences of their long-term behavior but are more concerned with the short-term situation [1]. Alcohol consumption is a common risk behavior among adolescents [2], with numerous and serious effects on the lives of young people, which can affect their physical or psychological health [3].
Alcohol consumption (AC) is common among adolescents around the world, often with high levels of consumption [2,4]. However, there are differences between countries. On average, 58% of European and Canadian adolescents start experimenting with drinking alcohol before the age of 15. For example, 29% of Spanish adolescents and 26.5% of Dutch adolescents had their first drink before age 13 [5], whereas 50% of the adolescents in South America indicated that they drank alcohol before the age of 14 [4]. Once adolescents initiate drinking, many of them are involved in so-called binge drinking, i.e., drinking 5 or more glasses within a limited period of time. Nearly three quarters (70.8%) of Dutch youth < 16 years [6] and more than half (56%) of Spanish youth < 15 years [7] who had The contribution of the type of leisure activity on drinking behavior among adolescents is not a well-studied area of research. Insight into the role of supervised and unsupervised leisure on adolescents' drinking patterns can provide tools for prevention. Moreover, to our knowledge, there have been no cross-national studies conducted to investigate the role of leisure activities on adolescents' alcohol use. The objective of this study is to evaluate the predictive effect of structured, unstructured, and family leisure activities on adolescents' frequency of alcohol drinking across three different countries (Spain, Peru, and The Netherlands). In addition, this study also investigates moderation due to adolescents' self-control in relation to leisure activities and alcohol use.

Materials and Methods
Data collected in three countries (Spain, Peru, and The Netherlands) were used in this study. Data obtained in Spain and Peru were collected in light of the same project-Your Life- [46]. Dutch data were collected in an independent project investigating alcohol use among adolescents in a community trial [47]. Therefore, in order to match data across projects, instruments were standardized based on the criteria of linguistic, semantic, and functional similarity [48].

Procedure
The data from Spain and Peru used in this study were collected as part of the Your Life Project, which was carried out in five countries, including Spain and Peru. The aim of this project is to know what young people feel and think about relationships, love, and sexuality. An invitation with information about the project was sent by e-mail to public and private high schools. In Spain, invitations were sent to all high schools available in public databases. In Peru, local collaborators include schools with easier access. In this way, the sampling presented is non-probabilistic.
Schools that agreed to participate were further informed about the process for the study. Adolescents in the participating schools were verbally informed about the purpose of the study; thereafter, written information was provided detailing the objectives of the project. Adolescent consent to participate in the study was formally established at the beginning of the online questionnaire [49]. Participants had the option to withdraw from the study at any time. Since Spain and Peru have different requirements regarding parental notification and consent, participating schools were informed of the different alternatives they had available to meet the requirements. The framework for requesting parental permission for research with adolescents is explained in Ruiz-Canela et al. [49].
The adolescent participants filled out a self-report questionnaire in which they did not provide names or any other personally identifiable information. Only the code of the school, the type of school, and the country were recorded in the database. The surveys were designed to ensure this anonymous participation [50]. To increase the respondents' sense of privacy and promote honesty in reporting, questionnaires were administered at school, and teachers were instructed to not move around the computer room while the adolescents filled out the questionnaire.
The survey data were kept in a safe place at the project site. Access to the website is restricted to researchers directly involved in the study. The database linking the codes of the participating schools and their confidential information (name and address) was kept separate from the actual content of the survey. Data and results that allow for the identification of any school will never be published.
Study participants did not receive any incentive for their participation, but schools subsequently received overall results from their school, allowing them to use their results in specific educational programs, for the prevention of the problems identified through the survey. The design of the study was approved by the Ethics Committee of the University of Navarra, and each new participating center was asked to comply with the specific ethical requirements of the project. All relevant ethics committees in the countries participating in the project had access to the questionnaire prior to its application.
In The Netherlands, in light of a community intervention trial, the LEF program, data was collected by Utrecht University as commissioned by the municipality of Edam-Volendam [47]. For the LEF-project, two secondary schools in Edam-Volendam participated in the survey, including adolescents aged 12-17 years. Anonymous data was collected among adolescents with the passive permission of their parents. The adolescents were also given the opportunity to refuse participation and they provided active consent at the start of the online questionnaire. Participants filled out the questionnaire under an anonymous identification code, which could not be matched by the researchers involved in the study with any personal data. The online questionnaires were conducted during school-time in the school's computer rooms under the supervision of a teacher and research assistant. The students were notified that the study would be about 'factors that can stimulate the development of young people in a good and healthy way'. For this study, approval of the Faculty Ethical Review Committee from the Utrecht University (FETC18-060) was obtained.

Participants
For Spain and Peru, the data were gathered during the period between December 2016 and September 2018. In total, 18 schools for Spain and 15 for Peru participated in the study, to which the link to the website was provided, designed to provide detailed information to the participants (https://proyectoyourlife.com//, accessed on 16 August 2021). The surveys were completed in the classroom under the supervision of the teacher in charge, in all its stages. At the time of this work, the sample was made up of 3228 persons. After evaluation for inconsistency, incomplete surveys, and requirements on the variables, the sample was reduced to 2506 respondents, distributed in 1276 (50.92%) from Peru and 1230 (49.08%) from Spain. For both countries, the age of the respondents was between 12 and 17 (mean age = 14.28, SD = 1.35) and 62.4% were female and 37.6% were male.
The Dutch data were collected in May 2018. In The Netherlands, schools that offer all levels of education often have more than 1000 students. In the participating schools, 2166 adolescents filled out the questionnaire. The data was controlled for inconsistency, incomplete surveys, and the requirements on the variables; therefore, 64 respondents were excluded from the dataset. A remaining 2102 respondents between the ages of 12 and 17 participated (mean age = 14.7, SD = 1.32). More than half (52.2%) were female and 47.8% were male.

Measures
The assessment instrument used in Spain and Peru was the Your Life project self-report questionnaire [46], with three different age-related versions (13,15, and 17 years). As a reference source for the creation of our questionnaire we used the Illustrative Questionnaire for Interview-surveys with Young People [51]. The questionnaire was first written in Spanish and later adapted to English.
For the scales of yearly alcohol use (YAU) and yearly binge drinking (YBD), participants were asked to report the number of drinking occasions where, for YAU, they consumed at least one alcoholic beverage during the last year, and, for YBD, occasions where they consumed 5 or more alcoholic drinks within a few hours in the last year [52]. For Spain/Peru, responses were on a five-point scale (0 = Never, 1 = Less than 1 day per month, 2 = 1-3 days a month, 3 = 1-2 days a week, 4 = 3 days a week or more). For The Netherlands, response options were on a 14-point scale (1 = zero, 2 = 1 . . . 10 = 9, 11 = 10, 12 = 11 to 19, 13 = 20 to 39, 14 = 40 or more). Finally, both instruments were dichotomized in never and any YAU or YBD (Table 1). Playing games inside together (e.g., board games, computer games).
We do sports, outings, or excursions, play board games.
Tertiles of the average of the items.
Going somewhere together (e.g., to the city, the beach, shopping, sports matches). I usually dine with my parents.
Exercising together. They talk with you [your parents] about your interests (your hobbies, the things you like . . . ) Going for a walk together.
[Your parents:]-They have time to talk to you. Eating together. Talking about things together. Visiting family or friends together. Watching television, a movie, or a series together.

Structured leisure
Structured leisure Structured leisure (SL) Voluntary work (e.g., at a sports club or an institution).
Volunteering (collaborating with an NGO, charity, etc.) Tertiles of the average of the items. Artistic or cultural activities (e.g., museum/theatre visits, making music, acting, or painting).
Participating in sports, going to the mountains, etc.

Unstructured leisure
Unstructured leisure Unstructured leisure (UL) Hanging out with friends at home, without supervision of parents.
Hang out on the street, in a park, on the beach, or in other public places.
Tertiles of the average of the items. Other leisure activities (e.g., shopping, visiting sports matches, playing pool).
Meeting in a place where you are only the group of friends, without adults present. Hanging out in shopping centers, games rooms, billiards, soccer stadium, etc.
Structured leisure (SL) indicates activities that adolescents engage in under the supervision of adults, those with a certain structure, regular schedule, clearly defined goals and rules, and focused on building skills [18]. The instrument developed by Carlos et al. [46] was used to ask about engagement in a variety of activities, both structured and unstructured. Adolescents were asked how often they engaged in three different types of activities: voluntary work, artistic or cultural activities, and exercising. A five-point Likert scale was used: 1 = Every day, 2 = Multiple days a week, 3 = Approximately one day a week, 4 = Less than one day a week, 5 = Never. The responses were averaged, and these values were assigned to ordered tertiles (three groups of approximately 33% of individuals), indicating a low, middle, and high frequency of participation in SL activities, split by countries.
Family activities indicate activities that adolescents reported they engaged in with their parent(s). Adolescents were asked how often they engaged in eight different activities (The Netherlands, [53]) or four different activities (Spain and Peru) together with their parent(s), e.g., eating together or going somewhere together. Each item could be answered on a five-point Likert scale: 1 = Never, 2 = Less than 1 day a month, 3 = 1-3 days a month, 4 = 1-2 days a week, 5 = 3 days a week or more. The responses were averaged, and these values were assigned to ordered tertiles (three groups of approximately 33% of individuals), indicating low, middle, and high frequency of participation in family activities, split by countries.
Unstructured leisure (UL) was measured by asking about adolescents' involvement in two (The Netherlands) or three (Spain and Peru) different types of activities, e.g., meeting in a place with friends, without adults present (0 = Never, 1 = Less than 1 day a month, 2 = 1-3 days a month, 4 = 1-2 days a week, 5 = 3 days a week or more). The responses were averaged, and these values were assigned to ordered tertiles (three groups of approximately 33% of individuals), indicating low, middle, and high frequency of participation in UL activities, split by countries.
Self-control [54] (SC) was measured through the evaluation of three behaviors (see Table 1), using five-point Likert-type scales. For The Netherlands, the response options were 1 = Never, 2 = Rarely, 3 = Occasionally, 4 = Often, and 5 = Very often. For Spain and Peru, the response options were 1 = Never, 2 = Almost never, 3 = Sometimes, 4 = Almost always, 5 = Always. The responses were averaged, and these values were assigned to ordered tertiles (three groups of approximately 33% of individuals), indicating low, middle, and high average frequency of SC behaviors, split by countries.
The possible confounders were analyzed as follows: sex was dichotomized in females (0) and males (1); age group (AG) was categorized into early and late adolescents, following the median value of the population (14.3 years old); family status was assigned according to whether the child lived with one (0) or both parents (1). Socioeconomic status was selfreported in The Netherlands, and for the remaining countries it was reported by the school (schools indicated whether the majoritarian status of their families was high, average, or low income). However, the lower socioeconomic level was highly underrepresented in the sample, especially for Spain, which was 2.4%, or in The Netherlands, which reached only 5.4%. Therefore, it was decided to merge this level with the medium level and to only consider two values for this variable: low and high socioeconomic level.

Analysis
Initially, the data were evaluated through descriptive statistics, and then an association test for YAU and YBD was performed with all the variables by an χ 2 test.
The proposed model considers reports of YAU and YBD predicted by the three types of leisure (structured, unstructured, and familiar) and self-control, corrected for sex, age group, family situation, and the socioeconomic level. Consequently, the proposed statistical model of analysis was a multiple logistic regression for each country. The evaluation of multicollinearity between the predictors was evaluated using the statistics of variance inflation factor, tolerance, and condition index. In all cases, complete or quasi-complete separation was evaluated together with diagnostic statistics for each individual observation: predicted values, residuals, and influence. Odds ratios were estimated with and without (raw) confounders and evaluating interactions between the three types of leisure. Alpha was settled to α = 0.05. The analyses were performed with SAS v. 9.4.

Descriptive Statistics
The main characteristics of the participants are presented in Table 2. More than half of the participants were female (59%). In The Netherlands, 57.9%, of the participants were between 15 and 17 years old, while in the other two countries the majority were between 12 and 14 years old (58.5% in Peru and 55.2% in Spain). In all three countries, more than 80% of the young people involved had both parents and more than 60% had lower socio-economic status. Peruvian and Spanish adolescents showed a higher level of self-control (34.8% and 35.7%, respectively) compared to Dutch adolescents (28.6%).  Table 3 presents the results of the tests for the association between predictors and the use of Yearly Alcohol Use (YAU) and Yearly Binge Drinking (YBD) without correction by confounders. It is observed that, for both YAU and YBD, their prevalence was highest in The Netherlands (45.7% and 31.6%, respectively) and lowest in Peru (21.4% and 8.6%). As for the sex of the respondents, no associations with the use of both alcohol outcomes were observed (p = 0.071 and p = 0.057, in both data sets). In terms of the age group of respondents, YAU was about four times higher in the older age group compared to the younger age group. For YBD, the ratio rose by nearly 7 times. By relating two-parent families with one-parent families, a higher prevalence was observed, for both YAU and YBD, in single-parent families. Finally, for sociodemographic variables, the socioeconomic level was only associated with the use of YBD, where this practice was more prevalent for high socioeconomic levels. Regarding the behavioral variables, the protective effect of self-control and family leisure was observed for both YAU and YBD. This is contrary to the UL, which consistently shows its effect as a risk factor for these types of alcohol consumption. A different result was obtained for SL, which only showed an effect on the use of YAU, with the intermediate levels for those with the highest prevalence of use.

Association Tests
The models evaluated for alcohol consumption, in general, showed the same behavior in the three countries ( Table 4). The protective effect of having two parents was highlighted in all countries (with and without adjustment for confounders). The highest frequency of alcohol consumption was in late adolescence, with an OR between 5.4 and 9 in comparison to early adolescence. The effect of alcohol consumption as a function of the level of UL was also consistent, with the OR between 2.1 and 3.3 when comparing low levels of UL with medium levels, and between 3.2 and 6.2 when comparing low levels with high levels. As expected, family leisure was confirmed in the three countries as a protection factor, especially the high levels of family leisure with obtained ORs between 0.40 and 0.43 compared with low levels of this variable. In Peru, the average levels of family leisure were not enough to be protective when compared to the low levels, contrary to the other two countries. On the other hand, SL, in general, did not protect against alcohol consumption in the three countries; there was an exception in Spain, where intermediate levels of this type of leisure were a risk factor. Although no significant OR values were detected, the estimates for Peru and Spain showed values greater than one. Income was identified as a risk factor only in Peru with an OR of 2.14 in relation to lower income levels. Finally, the sex of the respondent did not show a significant relationship with alcohol consumption in Peru and Spain.  Regarding YBD (Table 5), there was a response similar to YAU, with a protective effect of family leisure or having both parents, and unstructured leisure and belonging to the late adolescent age group as risk factors. In Peru, high income was a risk factor, and the average levels of family leisure did not protect against this type of consumption. There was no relationship between structured leisure and YBD in all countries. Regarding differences, in relation to the YAU, it was found that for Spain having two parents did not have a significant protective effect. It is also noteworthy, for this type of consumption, that the ORs by sex already showed females with risk values in relation to males in The Netherlands (1.47). ORs associated with structured leisure were not significant (p > 0.05). Table 5. Odds ratio and 95% confidence limits for yearly binge drinking (YBD) in each country, raw and adjusted by confounders.

Independent Variables
The  When evaluating the model including the interactions of self-control with the different types of leisure, the presence of interaction for The Netherlands was observed with unstructured and family leisure ( Table 6). In the case of unstructured leisure (Figure 1), people with a low level of unstructured leisure benefited from medium and high levels of self-control (w = 10.27; p = 0.04). In other words, if the level of self-control is low, the probability of alcohol consumption increases despite not being exposed to medium or high levels of unstructured leisure. Regarding the interaction between family leisure and self-control (Figure 2), it is noted how children with low levels of self-control especially benefited from increases in the levels of family leisure-even at the highest levels of family leisure-since this condition manages to neutralize the negative effects of the lack of SC (w = 13.75; p = 0.01) ( Table 6).

Discussion
This study investigated the predictive effect of structured (SL), unstructured (UL), and family leisure activities on the frequency of annual alcohol consumption (YAU) and excessive alcohol consumption (YBD) of adolescents in three countries (Spain, Peru, and The Netherlands). In addition, this study also investigated moderation due to adolescents' self-control in relation to leisure activities and alcohol use. Findings demonstrated that, in line with our hypothesis, involvement in UL activities increased the likelihood of drinking alcohol more frequently and intensively (binge drinking) across all three countries. However, SL was not negatively associated with alcohol use in any of the countries. Moreover, family leisure activities only lowered the risk of involvement in yearly and binge drinking among adolescents in The Netherlands and Spain, not in Peru. The protective association between family leisure and the risk of UL on YAU (not binge drinking) particularly applied to Dutch adolescents with a low level of self-control.
For both alcohol outcomes (YAU and YBD), family leisure could be considered to be a protective effect, whereas unstructured leisure generally appeared to be a risk factor. These results are in line with those obtained in another study [35], in which no protective effect of extracurricular activities on the initiation of alcohol use was found, while playing sports was associated with a decrease in the likelihood of initiating marijuana and tobacco use. In contrast, they differ from those found by other studies, in which SL activities had a significant protective effect against alcohol use in young people [22,55,56]. The protective effect of engaging in leisure activities with parents has also been found by other studies [23]. According to Barnes et al. [24], family time is a protective factor against problem behaviors, whereas time spent with peers is a risk factor for problem behaviors. In addition, access to alcohol may also be related to family leisure activities as a protective effect and to unstructured leisure as a risk factor. Thus, in three countries within and outside Europe, the importance of young people engaging in family leisure activities was shown, while unstructured leisure activities could be considered a risk factor for adolescent drinking. [22]. Though a higher level of self-control is associated with a lower level of drinking across all three countries (except for YBD in Peru), only in The Netherlands did adolescents' self-control influence the relation between family leisure and UL on YAU (not binge drinking). That is, the positive character of family leisure and the risk of unstructured leisure on yearly alcohol use was strongest among Dutch adolescents with a low level of self-control. This implies that, in The Netherlands, parents spending time with their kids should be encouraged, and involvement in unstructured activities should be discouragedparticularly among adolescents with a low level of self-control; this subgroup is particularly vulnerable to involvement in a variety of risk behaviors [57]. According to Hofstede's rating [58], The Netherlands is considered an individualist society which is characterized by people who are supposed to look after themselves and their direct family only. Spain and Peru are more collectivist societies where people belong to 'in-groups' that take care of them in exchange for loyalty [59]. This may indicate that in more individualistic countries, such as The Netherlands, self-control plays a more important role in adolescents' behavior than in countries where a more collectivistic, familial structure is present. It seems that in individualistic societies, it is even more relevant for parents to do activities together with their child, and to discourage the involvement of their kids in activities that are unsupervised and unstructured. Parents should be made aware of the importance of doing activities together with their child. In addition, involvement in unstructured activities should not be promoted-particularly in more individualistic countries where there are at-risk adolescents, i.e., those with a lower level of self-control, who are more vulnerable to drinking alcohol.
Some comments are worthy concerning the response rates in our analysis. The Your Life project is not a representative survey, being more akin to a convenience sample, where response rates are not relevant. In such studies, representativeness is not a goal and thus the response rate is not an issue, but follow up rates are. Since the analysis we are presenting does not entail follow up data, the presence of confounding factors is addressed by multivariate adjustment, so large sample sizes are more important, and this goal has been achieved in this study. The Netherlands data has an estimate of the response rate because the initial target sample consisted of two schools. This was thus a closed sampling frame. However, also in their case it was more important to adjust for confounding factors, as representativeness was not an objective either.
This study has some limitations. The cross-sectional design does not necessarily allow the direction of the association to be described. However, these results link unstructured activities as a risk factor for YAU and YBD drinking, and family leisure activities as a protective factor. The responses of the questionnaires are self-reported, but social desirability should be minimized using anonymous questionnaires.
Despite these limitations, this study has several strengths. First, the large sample size allowed for adjustment for possible important confounding factors and allowed interactions to be assessed. Second, the association of YAU and YBD with SL and UL was analyzed, along with each specific leisure activity and with self-control.

Conclusions
Adolescents' participation in UL activities increases the likelihood of drinking alcohol more frequently in all three countries (YAU and YBD). Family leisure had a protective effect, whereas unstructured leisure was presented as a risk factor for drinking. A higher level of self-control predicted a lower level of alcohol consumption in all three countries. In The Netherlands, a difference was found, where the protective effect of family leisure and the risk of unstructured leisure on annual alcohol consumption was stronger among adolescents with low self-control.