**2. Methods**

Three TAD questionnaires were used to monitor the level of tobacco, alcohol, and drug use *via* surveys conducted in regular 4-yearly cycles from 1994 till 2018. The questionnaires were devised to map the impact of the nationwide drug prevention program "School without alcohol, nicotine and drugs" [10]. The questionnaires consisted of [number of questions]. The data from these surveys were processed by statistical package for social sciences (SPSS 20.0).

The surveys were carried out among primary school pupils from grades 5 to 9, aged 10–15 (TAD1). TAD1 questionnaires assessed 30 day, 12 months, and lifetime prevalence of drug use, together with the items on early start of drug use and family environment. The items on aggression and bullying at school were used from 2010 until 2018 [11].

A similar survey (TAD2) was carried out in secondary schools among 16–19 year olds. Apart from the traditional scheme of epidemiological information on drugs, bullying, and aggression, TAD2 contained items assessing the presence of alcohol dependence symptoms, namely screening CAGE (Cut down, Annoyed, Guilty, Eyeopener) and shortened ADS (Alcohol Dependence Scale—see [12, 13]).

Finally, TAD3 questionnaire was used asking for primary and secondary school teachers' own use of legal drugs such as tobacco and alcohol, knowledge about illegal drugs, as well as their attitudes toward and willingness to take part in school prevention programs.

All the above surveys were approved by the Ministry of Education of the Slovak republic, and the data collection was implemented by the Ministry of Health' network of the Slovak Office of Public Health in the whole country under the coordination of the first author.

From 1998, TAD2 questionnaire contained two items to capture at least some information related to the early start of the individual experience with alcohol; the questions remained unchanged in all TAD2 versions until 2018. They were as follows:

**Did it happen, that adult person offered you to drink alcohol beverage?** <sup>1</sup>□ No.

<sup>2</sup>□ Yes ! *If yes, 1st time it had happened, when I was about:........years old.* **Have you ever felt, that you were somewhat "tipsy" (or "half drunk")?** <sup>1</sup>□ No.

<sup>2</sup>□ Yes ! *If yes, 1st time it had happened, when I was about:........years old.*

The last one from the seven anonymous surveys, from which the data for the purpose of this chapter were derived, was conducted in 2018 among 15–19 year olds (n = 4042; 2194 boys and 1848 girls) from ninth grades of the primary schools and from four grades of the secondary schools of all types from grammar and vocational schools in the Slovak republic [14].

### **2.1 Design and data**

All TAD questionnaires were administered anonymously, and respondents received their paper-pencil versions, provided their answers without any personal data, which might be used to identity an individual, class, or school. Only the information about the year of birth and gender was required as obligatory to enable comparisons between boys and girls of certain age. Every respondent had right to refuse to take part in survey.

After completing, respondents returned filled questionnaires in sealed envelopes without any mark, except for the mark written by research assistants to the big

envelope with encrypted code of class and school, assigned by survey coordinator. In the seven research waves, 32,814 primary school pupils, 29,375 secondary school students, and finally, 10,180 teachers from both types of schools participated in the surveys.

All three TAD questionnaires were used across the eight main regions of Slovakia, with their samples being self-weighted for gender, school, class, region, and teaching language, taking into account also minorities as the part of the selection criteria.

The samples were created by a stratified proportional random sampling from ninth grades of primary and first to fourth grades of secondary schools, from five types of schools (primary, secondary grammar schools lasting 4 and 8 years, then secondary specialized schools with and without maturity exams, i.e., those lasting from 4 to 5 years, and the secondary specialized schools lasting 3 years only), selected proportionally from eight main regions of the whole country.

The sampling unit was school, and within each selected secondary school, four classes were randomly selected by research assistants from the first to the fourth grades; and one class from each of the existing ninth grades in selected primary schools, because part of the cohort of 15–16 year olds was still in primary schools, while approximately ¾ of them were in the secondary schools, predominantly in the first grade.

The scope of selection was defined according to the requirement for reliability (95%) and preciseness (2%), with respect to the existing numbers of primary school pupils and secondary school students still attending schools in Slovakia and with respect to the age range to be covered, that is, from 15 to 16 to 18–19 year olds.

### **2.2 Results**

At the beginning, we counted overall means of ages for offer and for effect in the same way as before, that is, for the whole sample. But during TAD2 data entries in 2018, we noticed frequent inconsistencies with previous results such as a younger age of the first experience of tipsiness than the age of alcohol offer. **Table 1** outlines the average ages for the first offer and the effect felt after any drinking of alcohol for TAD surveys from 1998 until 2018.

In previous surveys, the mean age at the two points mapped by TAD2 was calculated, as it is indicated in the table above. Averages for the whole dataset showed that the first offer of alcohol came earlier, while "tipsiness" felt subjectively for the first time was observed about 1 year later. However, closer inspection of the data in part of the cases had shown that the "tipsiness" occurred *later* or during the same year, but in


### **Table 1.**

*Mean ages of the first offer of alcohol by an adult person and the age when the effect of alcohol was felt ("tipsiness").*

some smaller, but still in substantial proportion of the cases, this effect of alcohol was reported as felt *before* any alcohol beverage was offered by an adult person.

In other words, approximately one-fifth to one-quarter of the adolescents did not go through the usual ritual of alcohol drinking initiation by adults; they appeared to try alcohol in their own way, not adhering to any symbolic adult permissions.

Thus, three groups were identified within this whole sample in respect to the age of alcohol beverage offer and the age of the pharmacological effect of alcohol upon mood and/or psychological state, felt for the first time in life:

1.offer and effect occurred *during* the same year;

2.effect occurred one or more years *after* the offer;

3.effect was experienced *before* the offer from an adult person.

The first two groups appeared to have acted in line with unwritten social rules of handling situations where alcohol is offered, that is, group 1 (no difference in the age, i.e., offer and effect in the same year), group 2 (effect delayed), and finally, the third group, where offer of alcohol from an adult was not related to the effect felt (which thus is unlikely to be associated with an adult offer) (**Figure 4**) [14].

These groups were then compared with respect to some of the variables from TAD2 surveys' results to find the possible differences in lifetime prevalence of cannabis. There were no significant differences in numbers of boys and girls in any of these three groups.

As we might see, the third group differed from first and second groups in higher proportion of cannabis lifetime use (**Table 2**), in this case characterized by an almost reverse percentages of cannabis lifetime prevalence. And roughly similar third groups were identified also in the series of previous TAD2 surveys, as well as across the teenage period (**Table 3**).

### **Figure 4.**

*Average ages of the first experience with alcohol effect ("tipsiness"):* **After** *offer of alcohol; and* **before** *offer of alcohol by adult person.*


### **Table 2.**

*Cannabis use differences between three groups of respondents in TAD2 2018 survey (Chi2 significant at 0,000).*


### **Table 3.**

*Cannabis lifetime prevalence in three groups differing in the age of alcohol offer and tipsiness effects sorted by age from 16 to 19.*

The relationship between cannabis use and the tipsiness experienced before an adult offer of alcohol did not appear to be incidental as it was observable not only in the year of the last implemented TAD2 survey, but also across all surveys from 1998 till 2018. This was an interesting finding, and we wondered whether these differences of the third group from two others might be related to other variables, such as parental control, family status, school attendance, or others.

Therefore, we decided to use the two questions about the offer and the tipsiness in a forthcoming broader survey implemented across Europe, namely the ESPAD project during the Spring 2019 [9]. (See **Table 4**).

The results outlined in **Table 4** appear to confirm the trend revealed by TAD2 survey implemented in 2018: the third group as documented in ESPAD survey carried out 1 year later (n = 9338) was different from the two other groups in lifetime prevalence of cannabis use, even though these results were not fully identical.

As there are likely to be additional variables relevant to this problem, we tried to map some of them within our existing datasets. Even though our surveys, both TAD and ESPAD, were not specifically designed to examine this problem (early experience with alcohol and its subjectively felt pharmacological effect), we decided to map at least some of the variables such as the important persons from the family


### **Table 4.**

*Cannabis use in three groups of respondents in ESPAD 2019 survey differing in the age of alcohol offer and tipsiness (Chi2 significant at 0,000).*

environment, peers, older friends, classmates, or other people, as well as relevant events from close social environments of adolescents.

We tried to map the following variables:


The third group, which appeared to have a tendency to circumvent accepted social norms and conventions about the initiation and/or the entrance into the adult community, was different from the two other groups also in other variables, not only in cannabis use prevalence (and it seems at the same time, that those variables were relatively less socially desirable).

In our preliminary analysis, we examined the variables such as school attendance with three main reasons of missing at school (see **Table 5**).

As for drugs use, the second and the third group were compared by nonparametric tests, and only results indicating significant differences are included in **Table 6**:

As for alcohol and related problems, except of usual core questions on prevalence (lifetime, 12 months, and 30 days), we also used a separate module with 30 items, containing two screening scales—CAGE and ADS [12–14].

Finally, cannabis-related problems were mapped by the items of CAST, used regularly in the ESPAD survey [15, 16]—see **Figure 5**:

Cutoff point, that is, point indicating case finding for CAST (which is 7 points) [17], we have found significantly higher number of the third group members, reporting more cannabis-use-related problems compared to group 1 and 2% (Chi<sup>2</sup> significant at 0.000).


### **Table 5.**

*School attendance in three groups differing in the age of the first offer of alcohol and in the age of the first tipsiness felt in the ESPAD2019 survey.*


### **Table 6.**

*Legal and illegal drugs and related problems.*
