The sociodemographic patterning of drinking and binge drinking in Estonia, Latvia, Lithuania and Finland, 1994-2002.(Research article)

Authors: Ville Helasoja (corresponding author) [1]; Eero Lahelma [2]; Ritva Pr�tt�l� [1]; Janina Petkeviciene [3]; Iveta Pudule [4]; Mare Tekkel [5]

Background

Despite the relatively low recorded alcohol consumption level [1, 2], the Baltic countries (Estonia, Latvia and Lithuania) and Finland suffer from harmful consequences related to the heavy use of alcoholic beverages. This may be partly due to the undercoverage of consumption statistics, but people's drinking patterns [3] are likely to contribute as well. These countries are usually regarded as sharing a common European drinking pattern, which is particularly evident in the Northern European countries [4] and is characterised by high consumption at weekends. This pattern is known to contribute to harmful consequences of drinking, including socio-economic inequalities in premature mortality in Finland [5].

In the Baltic countries heavy alcohol use is recognised as a social problem [6], and there is aggregate level evidence on the adverse consequences of drinking and alcoholism [7, 8]. In a comparative cross-sectional study in 1997, weekly alcohol consumption was associated with young age among both sexes, and with higher income among women [9]. In many Eastern and Western European countries, heavy drinking is patterned by key sociodemographic factors, such as age, sex, ethnicity, family status, socioeconomic status and living environment [10].

Little evidence is, however, available on changes in the sociodemographic patterning of heavy drinking in the 1990s and early 2000s in the Baltic countries, although mortality studies suggest that this will likely become an important issue in the near future [11, 12, 13]. Such evidence is needed to better understand these harmful drinking patterns and to efficiently implement preventive alcohol policies [1] as mere regulation of supply does not necessarily lead to diminished alcohol-related harm [14].

The Finbalt Health Monitor data [15] allow harmonised comparisons of alcohol drinking habits of the Baltic countries. Moreover, the data provide comparison to Finland, a neighbouring country that shares many long-term cultural and historical characteristics with the Baltic countries, but has particularly during the post second world war period been more close the Scandinavian and Western European countries. There is also evidence on rigorously-surveyed alcohol consumption habits and reliable consumption statistics from Finland, which thus provides both a methodological and substantial yardstick for interpretation of the comparative results from the Baltic countries.

This study aims to compare the patterning of heavy drinking and binge drinking by key sociodemographic determinants, i.e. sex, age, education, urbanisation, marital status and ethnicity in Estonia, Latvia, Lithuania and Finland in 1994-2002. More specifically, the aims are to examine:

1) What is the overall sociodemographic patterning of heavy drinking in these four countries from 1994-2002?

2) What kind of changes occurred from 1994-2002 in the sociodemographic patterning of heavy drinking in these four countries?

3) What is the overall sociodemographic patterning of binge drinking in these four countries from 2000-2002?

Methods

Data were collected from nationally representative cross-sectional postal surveys from the Finbalt Health Monitor project in 1994, 1996, 1998, 2000 and 2002 in Estonia, Lithuania and Finland. Lithuanian data on weekly alcohol consumption has been available since 1996. In Latvia, surveys were carried out since 1998. The national surveys have been accepted by respective ethics committees. In Estonia by the Tallinn Medical Research Ethics Committee, in Lithuania by the Lithuanian Bioethics Committtee, in Latvia the Ethical commitee of Riga Stradins university and in Finland by the ethical Committee of the Public Health Institute. In all the countries the covering letter of the questionnaire is formulated in a way that a respondent provides an informed consent by returning the questionnaire.

The methodology and questionnaires of the surveys have been standardised [15]. The data have also been compared to census information [16]. Generally, the youngest age group and men are slightly underrepresented but the proportion of urban and non-native population corresponds well with censuses. Despite lack of comparable information, we can assume that those with better education and married are somewhat overrepresented. Late response, as well as unit- and item- non response have also been analysed. The direction of bias is likely to be similar in all four countries.

In Estonia, Lithuania and Finland, the questionnaire measured education as the total number of years of education. In Latvia, the questionnaire measured education according to four standard educational levels. In all four countries, the level of urbanisation was based on the administrative classification of the respondent's address. Ethnic origin was dichotomized as 'natives' and 'non-natives'. In the Baltic countries, 'non-natives' comprised mainly Russians. (Additional File 1)

Additional file 1: Basic characteristics of the study material.

Weekly alcohol consumption was measured with the following question: 'How many glasses (regular restaurant portions) or bottles of the following drinks have you consumed …