10 Alcohol Consumption Among Adolescents in Estonia 1994 – 2010

Human behavior accounts for the majority of morbidity and premature mortality throughout the world. This book explores several areas of human behavior including physical activity, nutrition and food, addictive substances, gun violence, sexual transmitted diseases and more. Several cutting edge methods are also examined including empowering nurses, community based participatory research and nature therapy. Less well known public health topics including human trafficking, tuberculosis control in prisons and public health issues in the deaf community are also covered. The authors come from around the world to describe issues that are both of local and worldwide importance to protect and preserve the health of populations. This book demonstrates the scope and some of the solutions to addressing today's most pressing public health issues.


Introduction
Estonia is the smallest of the three Baltic countries on the east coast of the Baltic Sea with an area of approximately 45 215 square kilometres and a population of 1.34 million (01.01.2011) (Statistics Estonia, 2011) which was an independent state in . Estonia regained its independence in August 1991 after the collapse of the Soviet Union. This had enormous implications for health and for the political and economic transition during the succeeding years. In 2004, Estonia became a member of the European Union associated with introduction to the common European market and general pressure towards convergence in many policy areas.
The years 1991-1994 represented the period of transition. In terms of economic development, the year 1994 was characterized by significant unemployment (7.6%) and the relatively low gross domestic product (GDP) at current prices (1370 Euros per person) (Statistics Estonia, 2011). In the period of economic growth unemployment was the lowest (4.7% in 2006) and GDP the highest (12 161 Euros per person in 2008). During the following economic downturn unemployment increased to 16.9% and GDP decreased to 10 674 Euros in 2010 (Statistics Estonia, 2011).
It is estimated that alcohol consumption is responsible for about 4% of the total disease burden in the world (WHO, 2007). While regular light to moderate alcohol intake is associated with some reduction in total mortality (Di Castenuovo et al., 2006;Rehm et al., 2009), heavy drinking has been regarded as an important contributor to the high premature mortality rates in central and eastern Europe, particularly in the countries of the former Soviet Union (Chenet et al., 1998;Leinsalu, 2002;Leon et al., 1997;McKee et al., 2000; 188 280 g pure alcohol per week increased from 7.4% to 16.2%, and the proportion of women drinking more than 140 g pure alcohol per week increased from 2.7% to 4. 7% between 19967% between and 20067% between (Pärna et al., 2010. In 2006, prevalence rate of non-beverage alcohol consumption was 2.3% among men (Pärna & Leon, 2011). In Estonia alcohol-related mortality was the lowest in 1988-1991 (3.5% of all deaths) and increased to 9.1% in 2002-2005(Rahu & Pärna, 2009). At the same time, in Estonia age-standardized mortality rates of alcoholic liver cirrhosis increased from 9.7 to 37.5 per 100 000 men and from 2.2 to 16.1 per 100 000 women aged 25-64 in 1996.
Although the vast majority of alcohol-related deaths occur in middle-aged and elderly people, alcohol consumption behaviour is undeniably established in adolescence. Early adolescence is a critical time, where behavioural habits (including alcohol consumption) are developing (Lintonen et al., 2000). Moreover, individuals who begin alcohol consumption at a younger age have an increased risk of becoming regular alcohol drinkers in adulthood. According to schoolchildren's self-estimation, they consume alcohol due to boredom, stress and desire to fit in the group (Milgram, 2001). The use of alcohol may become a means of escaping from situations that youth feel powerless to change (United Nations, 2005). As among adults, alcohol consumption among adolescents is associated with road accidents, suicides, depression, memory problems, fighting, rape and unprotected sexual intercourse which could increase the risk of getting infected with HIV and other sexually transmitted diseases (Williams & Knox, 1987). Alcohol consumption may cause decrease of learning ability, which in turn causes learning difficulties due to which adolescent might fall out of school (Scheier et al., 2000). In addition, the existing evidence about the relationship between adolescent drinking behaviour and family related factors are inconsistent and even contradictory. While some studies have identified a higher risk of excessive adolescent drinking behaviour among lower socio-economic groups (Geckova et al., 2002;Lintonen et al., 2000;Lowry et al., 1996), others have found no or even inverse social gradients in schoolchildren's alcohol consumption (Shucksmith et al., 1997). Living in non-intact families and bad family relationships have been risk factors for alcohol drinking among adolescents (Challier et al., 2000;Shucksmith et al., 1997).
The aim of this chapter is to describe trends in the prevalence of alcohol consumption and episodes of drunkenness and to analyze associations between alcohol consumption, episodes of drunkenness and demographic and family related factors among schoolchildren in Estonia in 1994-2010.

Setting and sampling
This study was based on international survey of health behavior among school-aged children (HBSC survey) conducted among 11-, 13-and 15-year-old schoolchildren in 1993/1994, 1997/1998, 2001/2002, 2005/2006 and 2009/2010 academic year in Estonia. To ensure the clarity of work results, hereafter the end year of the study (1994,1998,2002,2006,2010) was used while referring to the academic year.
HBSC survey is a World Health Organization collaborative study, in 2010 43 countries participated in the survey. The main objective of HBSC survey is to collect high-quality internationally comparable data on schoolchildren's health behavior, health and wellbeing in their social context (Roberts et al., 2007). The target groups of the survey are 11-, 13-and 15-year-old adolescents, in Estonia respectively schoolchildren from grades 5, 7 and 9.
Schoolchildren fill an anonymous questionnaire in classroom during one school hour. A person outside from school is present, schoolchildren who miss the class are left out from the survey. The questionnaires are sealed in envelopes in front of schoolchildren to ensure the confidentiality of data (Aasvee et al., 2007;Maser, 2004).
Methods used in this survey are the same in all participating countries and this gives a chance to monitor the changes in schoolchildren health, health behavior and social environment over years in different countries (Aasvee et al., 2007). HBSC survey has been conducted in Estonia in the following periods: February 1994 (King et al., 1996), February-March 1998(Currie et al., 2000, November-December 2001 (Maser, 2004), February-March 2006(Currie et al., 2008, February-April 2010.

Participants
Databases from Statistics Estonia were used to compile the sample. Estonian counties were divided into 12 stratums according to language of instruction and urbanization. Sample sizes were calculated according to the number of children in grades 5, 7 and 9 in 12 stratums (Aasvee et al., 2007).
In HBSC survey it is required that the number of schoolchildren in each age group is approximately 1500 (Roberts et al., 2007). To ensure equal inclusion probabilities the selection was made in two phases. In the first phase schools were selected. Inclusion probabilities were equal to the total number of grades 5, 7 and 9 in the selected school. This means that schools having more classes of these grades had higher probability to get into the sample. In second phase one class from each parallel was randomly chosen. While compiling the age groups, 90% of schoolchildren had to be within ±6 months of the mean age for each age group and remaining 10% no more than 12 months from the mean age. In chosen schools the school board decided whether they wanted to participate in the survey or not. Schoolchildren had the right to refuse filling the questionnaire if they themselves or their parents wanted that way. Over years the non-participation rate has been below 0.5% of the whole sample.
Data files from countries that have participated in the survey have been checked and cleaned in Norwegian Social Science Data Services, where international database was created and is preserved (Aasvee et al., 2007).
This chapter focuses on 13-and 15-year-old adolescents self-assessed alcohol consumption and episodes of drunkenness in 1994-2010.

Alcohol consumption
Frequency of alcohol consumption variable was based on different alcoholic drinks (beer, wine, strong alcohol, liqueur, cider, light alcoholic beverages, alcopops) consumption question. Possible answers were 'yes, every day', 'yes, every week', 'yes, every month', 'yes, less than once a month' and 'no, never', 'For the purposes of analysis, schoolchildren were categorised as drinking at least weekly (weekly alcohol consumption) or less often.

Drunkenness
Frequency of drunkenness was assessed by asking whether schoolchildren had ever had so much alcohol that they were really drunk. Possible answers were 'no, never', 'yes, once', 'yes, 2-3 times', 'yes, 4-10 times', and 'yes, more than 10 times'. Responses to this question were grouped into two categories: drunkenness never or only once in life and at least two episodes of drunkenness.

Demographic factors
Sex, age and nationality were used as demographic factors. In accordance with age groups, schoolchildren were either 13-or 15-year-old, from grades 7 and 9 respectively. According to ethnicity two groups were formed: Estonians and non-Estonians (mainly Russians).

Family related factors
Family structure, family wealth, family affluence scale and family relationships were used as family related factors, that influence health and health behaviour of schoolchildren.
Based on family structure, schoolchildren were divided into four groups: 1) two biological parents; 2) one parent; 3) one parent and one step-parent; 4) other combinations (i.e. parent and grandparent, foster home etc.).
Based on family wealth, schoolchildren responses were distributed into three groups: 1) bad (very bad and bad); 2) average; 3) good (very good and good).
Family affluence scale (FAS) was calculated on the basis of four items 1) does your family own a car (0, 1, 2 or more), 2) how many times did you travel away on holiday with your family during the past 12 months (0, 1, 2, 3 or more), 3) do you have your own bedroom for yourself (0, 1) and 4) how many computers does your family own (0, 1, 2, 3 or more). A composite FAS score was calculated by summing the responses to these four items ranging from 0 to 9 (Richter et al., 2006). The scores were subsequently recoded into tertiles and respondents were divided into three groups respectively. In 2010 the FAS tertiles were as follows: low (0-4 points); middle (5-6 points); high (7-9 points) (Currie et al., 2008).

Statistical analysis
Prevalence of alcohol consumption and episodes of drunkenness were calculated separately for 13-and 15-year-old boys and girls. Logistic regression analysis was applied to assess the associations between weekly alcohol consumption, at least two episodes of drunkenness and demographic and family related factors among 13-and 15-year-old schoolchildren. Weekly alcohol consumption (I model) and at least two episodes of drunkenness (II model) were used as binary variables and demographic and family related factors were used as independent variables in logistic regression analysis models. Odds ratios (OR) and corresponding 95% confidence intervals (CI) were computed for both models. OR that equals 1 refers to the base comparison group. OR's of weekly alcohol consumption and at least two episodes of drunkenness were adjusted to all demographic and family related factors in logistic regression analysis.
The present analysis is based on 13-and 15-year-olds (N=12244), 5861 boys and 6383 girls through the study waves (Table 1). Questionnaires, where the correspondents hadn't answered about their age (10 questionnaires) and alcohol consumption (99 questionnaires) were left out from the analysis. Questionnaires that lacked information about episodes of drunkenness were left out from the drunkenness analysis (62 questionnaires). Additional questionnaires that lacked information about questions related to demographic and family related factors were excluded from logistic regression analysis.

Demographic and family related characteristics of respondents
In 2010, the number of 13-and 15-year-old adolescents studied was almost equal (Table 2). There was about 77% of Estonian and nearly one fourth non-Estonian schoolchildren. About 63% of adolescents were living with both biological parents, approximately 20% with one parent, 15% with parent and a step-parent, and 2% had some other family structure. Almost half of the schoolchildren evaluated their family wealth to be good, 46% found it was average and about 4% said it was bad. Approximately 38% of schoolchildren had high, one third middle and 27% low FAS. About 77% of adolescents evaluated their family relationships good, 19% average and 3% bad.

Drunkenness
The prevalence rate of at least one episode of drunkenness mostly increased from 1994 to 2006 and then decreased. Compared to boys, prevalence rate among girls was only slightly lower in 2010 (

Alcohol consumption and drunkenness by demographic and family related factors
In 2010, weekly alcohol consumption and repeated drunkenness were higher among 15-year-olds (versus 13-year-olds) and among Estonian (versus non-Estonian) boys and girls (Table 7). Family structure was not associated with drinking alcohol weekly. Compared to schoolchildren living with both parents, odds for repeated drunkenness was higher among girls living with one parent, and among boys and girls living with parent and step-parent. Family wealth and FAS were associated with alcohol consumption and drunkenness only among girls. Odds to drink alcohol weekly and to have at least two episodes of drunkenness was higher among girls living in families with lower perceived wealth, but having higher FAS. Compared to adolescents with the good family relationships, odds to consume alcohol weekly and to have at least two episodes of drunkenness was higher among boys and girls with average and bad family relationships.

Discussion
The present chapter focused on alcohol consumption among adolescents in Estonia in 1994-2010.

Limitations and strength of the survey
Before discussing the results, one has to consider the limitations of the survey. Limitations are mainly related to questionnaire survey and its validity. Some studies have demonstrated that there is tendency to under-report when asking questions on such a sensitive risk taking behaviour like alcohol consumption (Newell et al., 1999). To deal with this potential bias of self-reporting all possible efforts to guarantee anonymity of responses were made. Despite of some limitations, this study provides basic data and several inferences can be drawn.
Strength of this survey is related to the possibility to make cross-national comparisons as standard approach was employed according to the study protocol to use the same questions in each participating country.

Trends in alcohol consumption and drunkenness
Weekly alcohol consumption increased from 1994 to 2002 and thereafter decreased among boys and girls in Estonia. Only among 13-year-old girls drinking alcohol weekly increased during the whole study period. Nevertheless, compared to the year 1994, prevalence of weekly drinking was much higher in 2010. In 1994-2002, similar increasing trend of alcohol consumption was found among adolescents in neighbouring countries Latvia and Lithuania (Zaborskis et al., 2006).
Similar pattern emerged with drunkenness. Prevalence of at least two episodes of drunkenness increased from 1994 to 2006 and thereafter slightly decreased among boys and girls. Among 13-year-old girls weekly alcohol consumption increased during the whole study period. Throughout the study period prevalence of drunkenness increased more among girls. Again, compared to the year 1994, prevalence of weekly alcohol consumption was much higher in 2010.
At the same time, Simons-Morton et al. (2009) reported world-wide cross-cultural patterns in alcohol consumption. The Northern European countries showed a declining trend, but Eastern European countries (including Estonia) experienced increasing trend in alcohol consumption and drunkenness. The variability of trends by country might reflect alcohol and marketing policy differences by country. Estonia has experienced significant political and economic changes during the last two decades. A lot of effects on alcohol related issues in Estonia could be associated with economic changes of the former Soviet Union as well as with the joining the European Union in 2004.
Gender differences in alcohol consumption among adolescents were pronounced. Throughout the study period, drinking and drunkenness remained higher among boys compared to girls, but the gap between boys and girls declined and girls appeared to be catching up with boys especially among 13-year-olds. Also, gender gap was shrinking in almost all countries participating in HBSC Survey (Simons-Morton et al., 2009). The variability in trends by gender in Estonia could be due to increased effectiveness of contemporary marketing practices or relative ineffectiveness of policies and programs with girls. However, these changes in prevalence could also be due to changes in the social roles of women, allowing girls greater autonomy and wider range of social options (Rahav et al., 2006).
Prevention of alcohol consumption among adolescents is very important because it prevents problems among adults in the future (Hingson et al., 2006). It has been found that the most effective prevention measures are those that are targeted to schoolchildren and to their parents at the same time (Smit et al., 2008;Wu et al., 2003). However, it has to be taken into account that every country needs it's own implementation of a specific program due to cultural differences (Koning et al., 2010). Simons-Morton et al. (2010) compared alcohol consumption of adolescents in USA, Canada and Netherlands and found that there were higher odds to consume alcohol and get drunk in the Netherlands, where minimum age to purchase alcohol is 16 compared to Canada (minimum age 19) and USA (minimum age 21).

Alcohol consumption and drunkenness by demographic and family related factors
Weekly alcohol consumption and repeated drunkenness were associated with demographic factors like gender and age as described above. In addition, compared to non-Estonian adolescents alcohol drinking was more pronounced among Estonians. At the same time, weekly alcohol consumption was higher, but the amount of pure alcohol consumed per week was lower among adults of ethnic majority in Estonia . However, in Lithuania, in schools with Lithuanian teaching language, Polish and Russian adolescents were more exposed to alcohol consumption (Šumskas et al., 2010).
Drunkenness, but not weekly alcohol consumption was associated with family structure among boys and girls. Living in non-intact families was a risk factor for repeated drunkenness. Also, Bjarnason et al. (2003) reported that schoolchildren living with both biological parents engaged less frequently in heavy alcohol consumption than those living in other arrangements. According to world-wide literature, higher supervision in intact families and supportive family environment might be associated with lower alcohol consumption among adolescents in these families (Cookston, 1999;Shucksmith et al., 1997).
Similar association was found between alcohol consumption, repeated drunkenness and family wealth among girls. Lower family wealth was a risk factor for heavy alcohol use among girls in Estonia. Zaborskis et al. (2006) found inverse relationship between alcohol consumption and the perceived family wealth in all three Baltic countries. In 1994-2002, girls in Estonia, but boys in Latvia and Lithuania from the families perceived by them as wealthy were more likely to drink weekly as compared to adolescents from the families perceived by them as not wealthy. This inconsistency in these findings might be explained by time difference in these studies (they were conducted in different years).
Weekly alcohol consumption as well as repeated drunkenness was associated with family affluence scale among girls, but not among boys. At the same time, there was not found any relationship between repeated drunkenness and FAS among 11-15-year-olds in Estonia in 2002 (Richter et al., 2006). Again, this inconsistency in these results might be explained with different age groups and study years used in these studies. Moreover, Richter et al. (2006) found very limited evidence for a close consistent relationship between episodes of drunkenness and parental FAS in almost all other participating countries (27) in the same study. Kuntsche et al. (2004) have pointed out that while adults problem drinking seems to be more common in less affluent groups, this direction might be reversed for adolescents, where accessibility to financial resources is more limited. Also, higher FAS as well as family wealth might be not directly associated with higher education. Inconsistency in association of alcohol drinking with family wealth and FAS could be explained by the fact, that distribution of perceived family wealth among adolescents is quite different as compared to FAS. Adolescents who estimated the family wealth 'not so well-off' or 'not at all well-off' was in 2010 survey 3.2%, by FAS tertiles low economic situation was in 27% of students. Evidently, FAS shows more objectively the financial situation of the family.
In addition, there might be other factors during adolescence, which may have a greater impact on drinking behaviour than family related factors and parental socioeconomic status. For instances, the specific character of adolescence as a stage of experimenting with behaviours associated with adult status (Richter et al., 2006) or situations that youth feel powerless to change (United Nations, 2005).
Weekly alcohol consumption and repeated drunkenness were associated with worse family relationships among boys and girls in this study. According to world-wide literature, a supportive family environment (Shucksmith et al., 1997) and parent-child communication is associated with lowered prevalence of alcohol consumption (Luk et al., 2010).
In 1996-2006 the prices of alcoholic beverages increased in Estonia (1.3 times for domestically produced beer and 1.4 times for vodka). However, the average price increase has been slower than the increase in the consumer price index, as well as slower than the income increase of inhabitants, which most likely has also contributed to the increase of alcohol consumption (Estonian Institute of Economic Research, 2008).
Compared to the 2004 level, excise tax increased 45% by the beginning of 2010. The highest tax increases (30% altogether) occurred in 2008 when the economic crisis started to affect the Estonian economy (Lai & Habicht, 2011). This was the first time when affordability of alcoholic beverages decreased after many years.
A nation-wide restriction on the time of off-premise sales of alcoholic beverages was introduced in the summer of 2008. Currently, off-premise sale of alcoholic beverages is prohibited from 10 p.m. to 10 a.m. throughout Estonia.
There is still wide availability of alcohol sales outlets: 198 alcohol retail shops per 100 000 inhabitants in Estonia in 2010, while 6.5 in Finland and 4.5 in Sweden (Estonian Institute of Economic Research, 2010).
Since 2008 advertising of alcoholic beverages on television and radio has been prohibited from 7 a.m. to 9 p.m. in Estonia. Before, prohibition of TV advertisement ended at 8 p.m. for alcoholic beverages other than spirits (Lai & Habicht, 2011).
Interventions on a personal level such as alcohol counselling have had very low focus in Estonia. A pilot study to evaluate the methods of early identification of risk drinking and counselling in the primary health care was carried out in the beginning of 2010. It was found that early identification of risk drinking and counselling are feasible. Continuous practical training, considering the specificity of primary health care, is necessary in order to promote the counselling skills of primary health care specialists (Saame et al., 2011).
In Estonia, further alcohol policy actions should include the reduction of the density of alcohol outlets, more comprehensive advertisement bans, clearer separation of alcoholic beverages from other goods in retail stores and full implementation of brief alcohol interventions in primary health care (Lai & Habicht, 2011).
In addition to previously mentioned policy actions, there should be an alcohol consumption prevention program targeted to adolescents and their parents. Higher price for light alcoholic beverages and consistently asking for ID-card while selling alcohol would lower www.intechopen.com Alcohol Consumption Among Adolescents in Estonia 1994 -2010 201 alcohol consumption among adolescents. Root beer (0.5% and 0.8%) and non-alcoholic cider (1.2%) should be considered as alcoholic beverages. Also raising the minimum age to purchase alcohol from 18 to 21 would be an effective preventive measure.

Conclusion
Alcohol consumption among adolescents in Estonia is a serious public health problem. Demographic and family related factors influence alcohol use of adolescents, especially among girls. The results of this study may guide the development of policy and interventions tackling alcohol consumption among adolescents in Estonia.
Additional research is needed on the nature of differences in drinking attitudes and patterns among boys and girls and their parents. Also, it might be interesting to look at the alcohol consumption within the context of other, possibly more relevant factors such as peer or school influence and other parental factors.

Acknowledgment
This study was carried out in the framework of the Health Behaviour in School-aged Children (HBSC) Survey, WHO/EURO collaborative international study. This chapter reports data from Estonia (Principal Investigator Katrin Aasvee