A comparative analysis of drinking patterns in six EU countries in the year 2000.
There have been surprisingly few comparative studies on differences in drinking patterns among Western European countries. To my knowledge there are only three that include samples of the general population in several countries representing different drinking cultures and with the data collection taking place at the same time period. One was conducted in 1988 in each of the 12 EC countries, with a few alcohol questions appended to the 29th Eurobarometer (see Hupkens et al., 1993), another in 1990 (Reader's Digest Eurodata, 1991; see also, e.g., Osservatorio, 1994; WHO, 1995), and the third in 1992, also as part of a Eurobarometer (37.0+37.1) (see Cassidy, 1997). Another Eurobarometer conducted in 1990 (34.1) also contained some useful data on drinking, but they have not been used in any specific analyses.
The rather few alcohol questions included in these studies, however, have made it possible to study only a few aspects of drinking patterns across the countries--that is, first of all frequency of drinking, including abstinence, and, in the 1988 survey, also the context of drinking (consumption of alcohol the previous day at breakfast, lunch, dinner, and other times). Some comparative approaches have used these survey data complemented by national data sources from countries not included in these surveys (e.g., Hanhinen, 1995; Simpura et al., 1993), and some comparative studies have used already existing national survey data not primarily collected for comparative purposes (e.g., Bloomfield et al., 1999).
In the Nordic countries, two comparative studies of drinking patterns have been made, one in 1978/79 and another in 1995/96 (Makela, 1999; Hauge & Irgens-Jensen, 1986, 1987). In addition, one study compared the drinking patterns in the Netherlands, Germany and Switzerland (Knibbe and Lemmens, 1987). Data on adolescent drinking in several countries have been collected by the so-called ESPAD 1995 and 1999 and by WHO as part of Health Behavior Among School Children (HBSC) (e.g., 1993/94 and 1997/98).
Previous analyses of survey data have shown that the proportion of abstainers and the frequency of drinking vary substantially among the countries. The number of daily drinkers in 1988 and in 1990 was highest in Southern Europe and lowest in Northern Europe (Hupkens et al., 1993; WHO, 1995). Hanhinen (1995) compared drinking habits among the Nordic countries, Italy and Germany. Daily drinking, according to the 1990 survey, was the most common in Italy (43%), followed by Germany (16%) and Denmark (14%). Separate national surveys for the other Nordic countries in the late 1980s and early 1990s showed much lower proportions of daily drinkers: 3% in Finland, 1% in both Norway and Sweden, and less than 1% in Iceland.
This was the situation ten years ago. The ECAS surveys in the six countries compare the situations in the year 2000. The frequency of drinking and abstinence are only two of several aspects of drinking habits that are studied. Results on mean consumption, average quantity consumed when drinking, heavy drinking occasions, and drinking contexts are also presented. Most analyses are age and gender specific.
One crucial question is the comparability of the survey samples. This issue was raised at the beginning of the analyses, since we found large differences among the samples in response rates and coverage rates (i.e., the self-reported survey estimate of per capita consumption in relation to the recorded per capita consumption). These issues are discussed in detail by Leifman (2001a) and are not repeated here. However, it should be noticed that the differences in especially non-response rates and underreporting limit the comparability among the study countries. As a consequence, direct comparisons among the countries in drinking are to a large extent avoided. Instead, most of the comparative analyses are based on ratios calculated for each country, such as age and gender ratios in drinking and the number of heavy drinking occasions per drinking occasion. The rationale for this is that it could perhaps be assumed that both dimensions (e.g., heavy drinking occasions and frequency of drinking) and both groups (e.g., men and women, young and old) show approximately the same degree of underestimation.
Data and methods
This study is based on data on drinking habits from general-population surveys conducted in six EU member states: Finland, France, Germany (excluding former East Germany), Italy, Sweden, and the UK. In each country, data were collected by telephone interviews. The sampling procedure (random digit dialing) differed somewhat among the countries but should in each be representative of the adult population ages 18-64 years (see Leifman, 2001a). In choosing which eligible person in a household to interview, the "birthday method" was used, which means that the person in the household who is next in line to have a birthday should be interviewed. A maximum of seven calls were made.
In all study countries, the questions were made as similar as possible. The Quantity-Frequency scale (Q-F scale) was used for estimating volume of drinking. Each country included the Q-F scale for beer, spirits and wine (and cider in all countries except Sweden and Italy). For Sweden the beer question was divided into separate questions, for strong or medium beer (> 3.5% alcohol by volume) and for beer class II (2.3%-3.5% alcohol by volume). For each of the alcoholic beverages, the respondents were asked how often they drink and, when they do drink, how much is usually consumed.
Binge drinking (heavy drinking occasions) was defined as an occasion when the respondent had consumed at least one bottle of wine, 25 centiliters of spirits, or four cans of beer. There were also questions on experiences of alcohol-related problems, context of drinking, consumption and production of home-made alcoholic beverages, private import of alcoholic beverages from abroad, informal control (if one has tried to influence other people to drink less), attitudes toward alcohol control, and alcohol and violence, in particular the excuse-value of drinking for violence.
One common measure of drinking habits is the proportion of non-drinkers (or abstainers). In the ECAS survey, nondrinkers were defined as those who said they had not been drinking any alcoholic beverages during the past 12 months. The proportions are shown in Table 1. For men the country differences are relatively small, while for women the differences are larger, with France and Italy showing the highest figures (27% and 22%, respectively) and Finland the lowest (7%).
The latest of the series of national general-population studies in Finland, conducted in year 2000, showed abstinence rates almost identical to those found in this ECAS survey: 9% for both men and women 15-69 years old (Mustonen et al., 2001). Other data from a Swedish general-population survey conducted in Spring 2000 show somewhat lower levels of 12-month abstainers ages 18-64 years: 6% men, 10% women (Leifman, work in progress).
The remaining four countries were included in the Hupkens et al. (1993) study on drinking in the European Community, based on surveys conducted in each of the 12 EC countries in 1988. For each country, the number of "never" drinkers was estimated. They were defined as those who drank neither beer, spirits nor wine, but with no defined time period. (For each alcoholic beverage, the question was how often they drink beer/spirits/wine, with five answer categories: every day or almost every day, 3 or 4 days in a week, 1 or 2 days in a week, less often, or never.) In Italy the share of never-drinkers in 1988 was 22% for women and 10% for men, quite similar to the ECAS results. In the 1990 survey (Reader's Digest Eurodata, 1991) conducted in 17 European countries, the percentage in Italy for men and women together was 19% (see WHO, 1995). In a national population survey from 1997, 15% of men and 30% of women ages 15 years or more reported being non-consumers during the past three months (Osservatorio, 1998).
The proportion of never-drinkers in France, according to the Hupkens et al. 1988 data, was estimated at 22% for women and 8% for men, thus five percentage points lower than the French estimates in the ECAS survey. Ahlstrom (1999) reports the proportion of non-drinkers in France in the last 12 months, based on data from a general-population study from 1995/1996, to be 4% for men and 13% for women, and results from a national health survey made in 2000 showed a 12-month abstinence rate of 7% for men and 11% for women in the age group 12-75 years (Guilbert & Baudier, 2000), considerably lower than the ECAS estimates.
Also Germany showed lower proportions in 1988 than in 2000: in 1988, 6% of men and 9% of women (Hupkens et al., 1993), compared with 12% of men and 18% of women in 2000. Ahlstrom (1999) also reports the proportion of nondrinkers in the past 12 months obtained in a German study conducted in 1994. For men the rate was 12% and for women 18%, which corresponds well to abstinence rates obtained in the ECAS survey. However, in the latest national survey on the use of psychoactive substances conducted in 2000, the rates were much lower. In the former West Germany, with altogether 7,143 completed interviews in the age group 18-59 years, 6.5% of women and 4.7% of men reported no alcohol consumption during the past 12 months (Kraus & Augustin, 2001). This should be compared with much higher proportions for 12-month abstainers obtained in the 1997 national survey (Kraus & Bauernfeind, 1998): 15% among women and 10% among men (in former West Germany).
As concerns the UK, several general-population studies have been conducted. The 1998 General Household Survey (GHS) found that 14% of women and 7% of men ages 16 and over were non-drinkers during the past year (DoH, 2001). The latest of the ONS Omnibus Surveys, conducted in the year 2000, found that 9% of men and 11% of women ages 16-64 years had not drunk any alcohol in the last year (Lader & Meltzer, 2001). These results correspond rather well with the ECAS survey results.
Frequency of drinking
Table 2 shows the frequency of drinking days by gender, based on the most frequently consumed beverage for each respondent (including the abstainers). The results should be seen as minimum levels, since different beverages may be consumed on different occasions and days. However, it should give a rather good picture of the general trends. The frequencies differ substantially among the six countries, with the highest frequency of regular drinking (two to three times a week or more often) for both men and women in Italy, followed by the UK, France, and Germany. The lowest frequencies are found in the two Nordic countries. Also the proportion of daily drinkers is by far the highest in Italy, with France in second place. Forty-two percent of Italian men and 26% of women drink daily. In France, 21% of men and 9% of women drink daily. Germany and the UK show about the same proportions for daily drinking: approximately 10% of men and 5% of women. In Sweden and Finland, the percent age of daily drinkers is much lower: among women 1%-2% and among men 3%-4%.
A compilation of recent national surveys conducted in the six countries reveals approximately the same differences across countries in the prevalence of daily consumers. In the French study from 2000, 25% of men and 9% of women ages 12-75 years reported daily use of alcohol (Guilbert & Baudier, 2000). The 2000 ONS Omnibus Survey showed that 10% of men and 7% of women in the UK ages 16-64 years old drank almost every day (Lader & Meltzer, 2001). In the most recent German national survey conducted in year 2000, 9% of men and 2% of women ages 18-59 reported daily use of alcohol (Kraus, personal communication). The Italian report of drinking habits in 1997 (Osservatorio, 1998) presented the proportion of daily drinkers of different alcoholic beverages (but not daily drinking of any alcoholic beverage). In all age groups, for both men and women, the percentage of daily drinkers was highest for wine. Forty-six percent of all men ages 15 and over drank wine daily, and 24% of women in the same age group (Osservatorio, 1998).
According to the latest of the Finnish national series of general-population surveys, done in 2000, 10% of men and 2% of women ages 15-69 drank at least four to five times per week (Mustonen et al., 2001). In a survey in Sweden conducted in December 2000 and January 2001 (see Norstrom & Skog, 2001), 1.8% of men and 0.6% of women in the age group 20-74 years reported daily use of alcohol.
Also analyses of surveys conducted in the late 1980s and early 1990s have shown much lower frequencies in Northern Europe (see Eurodata: Reader's Digest, 1991, in WHO, 1995), especially compared with Southern Europe. The Hupkens et al. (1993) study estimated average frequencies of drinking situations in each EC country in 1988. The measure was calculated as the sum of the drinking frequency of beer, wine and spirits for each consumer. Contrary to the measure in Table 2, this measure overestimates the number of occasions, since an unknown number of consumers drink more than one alcoholic beverage during the same drinking occasion. The same frequency measure was calculated on the ECAS survey data. Abstainers were excluded in 1988 and therefore also in 2000. The results are shown in Table 3. The data for Finland and Sweden for the earlier period originate from national surveys (see Simpura et al., 1993, for Finland and Kuhlhorn & Leifman, 1993, for Sweden). Comparisons of the mean frequency between the two time periods should be made cautiously. It is perhaps more revealing to compare the countries' relative positions in 1988 and 2000.
In 1988, Italian men and French men showed the highest frequency of drinking situations, with 6.5 drinking situations per week, followed by Germans (5.0 drinking situations per week) and British (4.1). Almost the same rank order between the countries was found among women, but Italian women showed a higher frequency than French women, and Swedish women reported a lower frequency than Finnish women.
Between 1988 (1990) and 2000, the country differences have diminished. Still, Italy shows the highest frequency among both men and women, but British women show the second-highest frequency and British men almost as high a frequency as French men.
Women in Finland, Sweden and Britain show a rise in frequency between 1988 (1990) and 2000. British men show almost the same frequency in both years. The three remaining countries show a substantial decrease in the frequency of drinking situations. The result thus points at a homogenization in drinking frequencies: Northern Europe (including the UK) has approached Southern Europe (including Germany) in the frequency of drinking.
Another question in the ECAS survey aimed at measuring differences in the context of drinking. All consumers were asked how many days during the past seven days they drank alcoholic beverages in the following contexts: at lunchtime; during an evening/late afternoon meal; at a restaurant or a bar; and at home but not in connection with meals. The answers to this question were to construct an additional measure of weekly drinking frequency. The maximum number of possible reported drinking occasions during the past seven days is thus 28 (7 x 4). This measure, together with the minimum measure of drinking frequency (number of days drinking the most frequently drunk beverage) and the maximum frequency measure (drinking situations = sum of frequency for all the beverages), is shown in Figures 1 and 2 for men and women in different age groups. The past-seven-days measure thus based on the usual number of drinking days, which may explain why the past-seven-days measure shows the highest frequencies.
As concerns men, for all three measures (Figure 1) the frequency of drinking increases with age in France, Germany and Italy. In the three other countries, the age differences are smaller. In all three countries, the oldest age group shows the highest frequency of drinking days per week and drinking situations per week. As concerns drinking occasions during the past seven days, the youngest age group reports the highest number in Sweden and the UK. In Finland the number of occasions is quite similar in all three age groups. The country differences in drinking frequencies vary substantially among the three age groups. The differences are the most marked in the oldest age group, with a clear north-south gradient, and the least marked in the youngest group. Those 18-29 years old in the UK, for example, show higher frequencies than those 18-29 years old in all other countries but Italy.
[FIGURE 1 OMITTED]
Approximately the same pattern is found among women as for men (Figure 2). The largest country differences in frequencies are among the elderly, and the smallest are among the youngest group. As a matter of fact, no clear north-south gradient can be found among the youngest and the middle-age group (30-49 years of age). In the youngest group, the British women show the highest frequency in two of the three measures, and the French women one of the lowest.
[FIGURE 2 OMITTED]
Drinking volume per drinking occasion
The analyses suggest that the volume of alcohol consumed per drinking occasion differs across the study countries. However, here the highest quantities are generally found in Northern Europe and the lowest in Southern Europe. This is shown in Table 4, which shows the average quantities per drinking occasion for each of the three main alcoholic beverages. For both men and women, at all ages, the highest beer consumption per drinking occasion is found in the UK, Sweden and Finland. As concerns spirits, Finland shows the highest quantity, particularly among men, and France and Italy the lowest. Swedish men show the second-highest spirits consumption per occasion, followed by British men, whereas the relationship is opposite for women. Also wine shows the same country pattern: highest quantities per occasion in the two Nordic countries and the UK, and lowest in France and Italy, for both men and women.
Shifting the focus to age differences, Table 4 shows that in all countries the youngest show the highest quantity of spirits consumption per drinking occasion and, in all countries but Germany, the highest beer consumption per drinking occasion. Wine consumption is more evenly distributed across age groups, but generally somewhat higher in the middle-age group. The largest percentage difference between the youngest group (18-29) and their elders is found in Finland, Sweden and the UK.
Total alcohol consumption
This study has shown that regular drinking is most common in Southern Europe, whereas the quantity per occasion is highest in Northern Europe. Table 5 shows the gender-specific total alcohol consumption in three age groups: 18-29, 30-49 and 50-64. (Total [or mean] consumption is calculated as the product between frequency of drinking occasions and volume per occasion, summed across all alcoholic beverages.) It is the combination of having one of the highest frequencies and a high average consumption per drinking occasion that produces the very high total alcohol consumption per respondent in the UK (see Leifman, 2001a). Here, however, the focus is not primarily on differences in absolute consumption levels, but on the age and gender distributions of consumption within each country, and whether these differ among the countries.
In Finland, Sweden and the UK, for both men and women, the youngest group reports the highest consumption. Earlier studies conducted in Sweden in the 1990s have shown the same pattern, with the highest consumption in the early twenties (see Leifman, 2000). The UK General Household Survey for 1996 showed the same pattern as the ECAS survey: highest in the age group 16-24 for both genders, then decreasing with each older age group (DoH, 1999). For men, the age group 45-64 consumed roughly 25% less than those 16-24 years old. For women the difference was almost 40%. A similar pattern was found in the Health Survey for England 1998 (DoH, 1999) and in the ONS Omnibus Survey for year 2000 (Lader & Meltzer, 2001). According to earlier Finnish studies in the 1990s as well as in the national survey from year 2000, the younger middle-aged (30-49) and the youngest (18/20-29) groups reported rather similar consumption levels (see Metso & Simpura, 1997; Mustonen et al., 2001).
As can be seen in Table 5, in the other three countries consumption peaks in the middle or oldest age group (except for German females). Other studies from Central and Southern Europe--from France (in SFSP, 1998), Italy (Osservatorio, 1998), Austria (Uhl & Springer, 1996) and Germany (Kraus & Bauernfeind, 1998, Kraus & Augustin, 2001)--show rather similar age differences in consumption. Simpura and Karlsson (2001) conclude, in an ECAS review of studies of European drinking patterns, that in no EU country with available data do adolescents below 20 years of age drink more than their, elders. The heaviest drinking groups are most typically males in their thirties and forties.
Table 5 also shows that, in all age groups, men consume two to three times more alcohol than do women, with the highest ratio in France (3.1) and the lowest in Italy (2.0). In the European project "Women and Alcohol" (Bloomfield et al., 1999), gender ratios were presented for four of the six study countries (Ahlstrom, 1999). Here, too, the ratio was highest for France (3.1) and lowest for Italy (1.9). However, the Italian national survey from 1997 estimated the mean consumption the day before the interview (for consumers and non-consumers) to be three times higher for men than for women (in standard glasses or units: 1.5 vs. 0.5) (Osservatorio, 1998).
The gender rate for the UK of 2.5, presented by Ahlstrom (1999), was rather similar to the ratio in the ECAS survey (2.6). Also the 1998 General Household Study in the UK showed a gender ratio of 2.6 (DoH, 2001). For Germany, the gender ratio was estimated at 2.7 in survey data from 1995/96 (see Ahlstrom, 1999), which is somewhat lower than the 3.1 obtained in the 1997 national survey of 8,020 respondents (Kraus & Bauernfeind, 1998). Both these ratios, however, are higher than the ratio of 2.2 obtained in this study. The Swedish ratio of 3.1 is higher than ratios from other studies in recent years, which showed ratios of 2.2-2.3 (e.g., Leifman, 2000; Makela et al., 1999).
In the UK, it appears that gender differences are the most pronounced among the oldest, whereas the opposite seems to be the case for Sweden. In Finland and France the ratios are very similar across all three age groups. In Germany, the youngest stand out with a lower gender ratio.
Gender-specific age distributions for each of the three main alcoholic beverages are shown in Figures 3 and 4. Not surprisingly, beer is the dominating male beverage in the Central and Northern European countries, whereas wine dominates among men in France and Italy. (For cross-country comparisons of recorded alcohol consumption, see Leifman, 2001a.) The age pattern of alcohol consumption within each country, however, differs for the different beverages. Beer consumption is highest among the youngest men--except for France, with a similar level among the youngest and the middle-aged--and lowest in the oldest age group in all six countries.
[FIGURES 3-4 OMITTED]
In France and Italy, spirits consumption decreases with age, while in Finland--formerly a spirits-drinking country--the age pattern is the opposite: Spirits consumption increases with age. In both Sweden and the UK, the youngest and the oldest groups show a higher spirits consumption than the middle-age group, whereas the opposite is true for Germany.
Wine consumption among men shows almost the opposite age pattern from that of beer. In all countries, wine consumption increases with age, especially in France and Italy, the two traditionally wine-drinking countries. In France, for example, men 50-64 years old report wine consumption more than three times that of the youngest, whereas the youngest report a spirits consumption three times that of the oldest.
Also for women, beer drinking is a beverage for young people. Only Finland shows rather similar levels for all three age groups. In all other countries, beer consumption is the lowest in the oldest group. Spirits show a similar age distribution: highest among the youngest in all six countries, lowest among the oldest in all but Finland and Germany.
As with men, women's wine consumption shows the opposite age distribution from beer and spirits for most countries; this is most clearly seen in France and Italy. In both countries, wine consumption is twice as high among those 50-64 years old as among those 18-29 years old. The UK, however, does not follow this pattern, but displays a decreasing consumption with age even for wine.
Heavy drinking occasions
Table 6 shows the country- and gender-specific distribution of the frequency of heavy drinking occasions (defined as drinking at least a bottle of wine or equivalent on the occasion). It should be stressed that there certainly are cross-cultural differences in the social acceptability of heavy drinking and intoxication. This may have contributed to the differences found in the table. As concerns men, the British show the highest frequency of heavy drinking occasions. More than a third of all the British men reported that they had consumed these amounts during one drinking occasion at least once a week during the past 12 months, followed by the Finns with 16% and the Italians with 11%. Also the proportion of daily or almost daily (four to five days a week) heavy drinkers is highest among the British, but here with the Italians before the Finns. The Swedes and the French show the lowest rates of daily/almost daily as well as weekly heavy drinkers. Looking at the other end of the scale--the percentage of respondents with no heavy drinking occasion during the past 12 months--the Finns show the lowest percentage (25%), followed by the British (34%) and the Swedes (35%). In the three other countries, the proportion of respondents with no such drinking occasion during the past 12 months is much higher: in Germany almost every second, and in France and Italy slighly more than every second respondent.
Also among women, the UK shows the highest rates of relatively frequent number of heavy drinking occasions, followed by Italy. The proportion of daily or almost daily heavy drinkers is highest among the Italians (almost 3%), but the proportion drinking these amounts once a week or more often is highest among the British women (12%). The lowest proportion of weekly heavy drinkers is reported by the Swedish women (1%), followed by the German women (2%) and the French women (2%). The two Mediterranean countries and Germany show by far the highest proportion of female respondents, with no such drinking occasions during the past 12 months (about 75%), and Finland and the UK by far the lowest (about 52%).
The mean number of heavy drinking occasions per 12 months differs substantially between men and women and different age groups (Table 7). In all countries but Italy, men report approximately three to four times more heavy drinking occasions than women, and in Italy 1.6 times more. As concerns age differences, Italy also differs from the others. In all other countries, the youngest show the highest number of heavy drinking occasions, while in Italy it is the oldest, and in all countries except Italy, these age differences in the frequency of heavy drinking occasions are larger than for mean consumption (shown in Table 6).
For men, the age gradient is steepest in Sweden, followed by Germany and the UK. Even though the youngest are the most frequent heavy drinkers also in France, the difference between the youngest and the middle age groups (30-49) is small. In Sweden, the youngest (18-29) report twice as many heavy drinking occasions as those 30-49 years old, and in France 1.2 times as many.
Among women, the age gradient is almost equally steep in all countries but France and Italy. The youngest women in Finland, Germany and Sweden report approximately twice as many heavy drinking occasions compared with the middle-age group (30-49 years), and in the UK about 1.7 times more occasions. In France and Italy, the difference between the youngest and the middle-age group is much less. Compared with the oldest, however, the youngest show a much higher frequency of heavy drinking occasions in France, but significantly less frequency in Italy.
Finally, it should be mentioned that the age differences in the frequency of heavy drinking occasions are larger than for mean consumption (shown in Table 6) in all countries but Italy.
Heavy drinking per drinking occasion
The ratio of heavy drinking occasions to the total number of drinking occasions indicates the frequency with which a drinking occasion leads to intoxication. Both measures of drinking frequencies were used (frequency in the Q-F scale and number of drinking occasions during the past seven days). Table 8 shows quite a distinct north-south gradient in intoxication-oriented drinking patterns. In Southern Europe, approximately one out of ten drinking occasions (men and women together) results in heavy drinking. This is so also in Germany. In Finland and Sweden, however, roughly one-fourth of all occasions are heavy drinking occasions. In the UK the proportion is as high as 30%. In all countries, the ratio is higher among men than women. In Italy, however, the gender differences seem to be rather small. This north-south gradient has also been found in analyses on data of drinking habits among school children ages 15-16 (ESPAD data from 1995, Hibell et al., 1997; Rossow & Lemmens, 2001).
Context of drinking
Drinking is more concentrated on weekends in Central Europe and especially Northern Europe than in the Mediterranean countries. However, all countries have probably been influenced to some extent by the introduction of a five-day weekly working schedule (Simpura & Karlsson, 2001). Figure 5 shows how common it is to drink in different contexts (the number of occasions during the past seven days) at lunchtime, at dinner (afternoon/evening meal), at a bar/restaurant, and at home but not in connection with any meal, in the six EU countries in different age groups. Drinking alcoholic beverages at lunch is most common in France and Italy, and least common in Finland and Sweden. Drinking at restaurants/bars is most common in the UK, followed by Italy and France, and least common in Sweden. Also drinking at home but not in connection with meals shows the highest frequency in the UK, with Finland in second place and the lowest frequency among Italians.
[FIGURE 5 OMITTED]
Table 9 shows the relative distribution of these four types of drinking contexts within each country, specified by gender. The results show a marked difference between the two Southern European countries and the others. In France and Italy, drinking occasions in connection with lunch and dinner comprise about 80% of all drinking occasions; the corresponding figure for Germany, Sweden and the UK is 50%, and for Finland 35%. Finland shows by far the highest proportion of drinking occasions at home but not in relation to a meal. These country differences hold true for both men and women.
As can be seen in Figure 5, drinking alcoholic beverages at lunch and at dinner is more common at higher ages in Germany, but especially in France and Italy. In the three remaining countries, the age differences are less and show no consistent pattern. Drinking in restaurants/bars is most common among the youngest in all six countries, whereas drinking' at home but not in connection with a meal shows different age profiles in different countries: highest among young people in Sweden, the UK, France and Italy, highest among the middle-age group in Finland, and highest among the oldest in Germany.
By and large, men report higher frequencies than do women in all six countries in all four contexts. The distributions of the four drinking contexts, however, showed some differences between men and women (Table 9). It appears that lunch and especially dinner drinking occasions contribute to a larger proportion of all drinking occasions for women than for men in most countries, whereas the opposite may hold true for drinking occasions at restaurant/bars and/or at home but not in connection with a meal. These differences are, however, not very dramatic.
This paper has repeatedly stressed caution in using comparative survey findings at face value without considering the validity problems that limit the degree of comparability. This was most evident in the comparisons of self-reported levels of alcohol consumption. Usually comparisons between survey estimates and consumption statistics indicate a large underreporting. Typically, some 40%-60% of actual alcohol consumption can be reached by surveys. What makes the problem even more difficult in an international comparison is that underreporting seems to vary among countries (Leifman, 2001a). The coverage rate in the UK sample was estimated at 96% and in the German sample at 31%, despite the use of a similar data-collection method and standardized questions.
On the other hand, there is no golden standard in terms of true country-specific per capita consumption. Nor are comparative studies based on the official (recorded) consumption statistics unproblematic. In a previous study, it was shown that unrecorded alcohol consumption has increased in the Nordic countries (Iceland not included) and in the UK during the past 10-15 years to reach an estimated 1.5-2 liters pure alcohol per person age 15+ in the late 1990s (Leifman, 2001b). In all likelihood, the proportion and level of unrecorded alcohol are higher in the Nordic countries and the UK than in Central and Southern Europe.
This study has tried to find a delicate balance between, on the one hand, not pushing the data too far and overinterpreting findings and on the other hand, not being so cautious as to neglect interesting findings. In any case, the approach taken in this study has been to be cautious in comparisons of absolute figures, especially mean consumption levels. Instead, more attention has been paid to cross-country comparisons of different ratios and of age and gender distributions.
Bearing the limitations in mind, the study has focused on several dimensions of drinking. A few results could be compared with those of earlier studies: frequency of drinking and abstinence. Generally speaking, the country differences obtained in studies in 1988 (Hupkens et al., 1993) and in 1990 (Reader's Digest Eurodata, 1991) also appeared in 2000. Regular drinking is most common in Southern Europe and least common in Northern Europe. However, a comparison of frequencies of situations of drinking (Table 3) indicated that the differences have been reduced, thanks to a substantial reduction in the frequency of drinking in France, Italy and Germany and to increases in frequency in Finland and Sweden and a stable level in the UK. The result concords with the observed pattern of homogenization in aggregate recorded alcohol consumption (see, e.g., Leifman, 2001c).
Also the quantity consumed when drinking showed an expected north-south gradient: the highest amounts in Finland, Sweden, and the UK, and the lowest in France, Italy, and Germany. Whether there has been a homogenization over time also in quantities per drinking occasion cannot be answered because of lack of earlier data. What can be said, however, is that there are systematic country differences in the age profile in drinking. In Finland, Sweden, and the UK, the youngest drink the most, not only per occasion but also on an annual basis. This age difference appears to be the most distinct in Sweden and the UK. In none of the other three countries (except women in Germany) do the young adults report the highest alcohol consumption. In France and Italy, consumption is lowest in the youngest age group, and for men it increases with each age group, at least when comparing the three age groups 18-29, 30-49, and 50-64. Other national survey data presented in this study support this contention. In other words, the differences in consumption levels among the six countries are the lowest among young people. Whether or not this is an expression of a generation effect, of age effect, or of both is difficult to say. If it is a generation effect, it implies that the homogenization process that is evident in analyses of recorded alcohol consumption in the postwar period is stronger among young people. Likewise the importance of national drinking cultures and alcohol policy in creating the differences in age distributions in the different countries is not known, but it would be of considerable importance to find out.
When it comes to intoxication, or more specifically to the frequency of heavy drinking occasions, as well as to the likelihood of a drinking occasion leading to intoxication, the youngest are the most prone in all countries but Italy. Thus, young French and Germans, as well as the youngest in Britain, Sweden and Finland, show a higher frequency of intoxication than their elders. However, since data over time are lacking, it is not possible to ascertain whether this pattern is a continuation of traditional cultural patterns or the result of new drinking patterns among the youngest.
TABLE 1 The proportion (%) of abstainers (non-drinkers during the past 12 months) in the study countries in age group 18-64 years, by gender Finland Sweden Germany UK France Italy Men 7 7 12 11 13 11 Women 8 13 18 15 27 22 Total 7 10 15 13 20 16 TABLE 2 Frequency (%) of drinking; based on the beverage that has the highest frequency for the respondent Daily 4-5 2-3 Once a 2-3 Once days a days a week days a a week week month month Men: Finland 4 4 20 32 19 7 Sweden 3 4 16 24 23 12 Germany 12 6 24 18 11 11 UK 9 16 31 18 8 4 France 21 5 19 23 7 5 Italy 42 3 17 14 4 4 Women: Finland 2 2 7 22 22 14 Sweden 1 1 5 17 24 17 Germany 5 2 13 20 15 10 UK 5 6 18 22 12 10 France 9 3 10 16 9 12 Italy 26 4 10 12 8 4 Once or Never a few days a year Men: Finland 8 6 Sweden 12 7 Germany 7 12 UK 4 11 France 8 13 Italy 6 11 Women: Finland 24 8 Sweden 23 13 Germany 17 18 UK 11 14 France 14 27 Italy 14 22 Sweden: Including separate questions for strong beer (>3.5% alcohol by volume) and beer class II (2.8%-3.5% alcohol by volume). TABLE 3 Frequency of drinking situations (1) according to surveys 1988 and 2000 (abstainers excluded) Men 1988 (2) 2000 (3) Frequency Index Frequency Index (situations/ 100= (situations/ 100= week) Total week) Total Finland (4) 1.4 33 2.3 68 Sweden (5) 1.8 43 2.1 (6) 62 Germany 5.0 119 2.9 85 UK 4.1 98 3.9 115 France 6.5 155 4.1 121 Italy 6.5 155 5.1 150 Women 1988 (2) 2000 (3) Frequency Index Frequency Index (situations/ 100= (situations/ 100= week) Total week) Total Finland (4) 0.9 38 1.3 65 Sweden (5) 0.8 33 1.0 (6) 50 Germany 2.8 117 1.7 85 UK 2.5 104 2.4 120 France 3.6 150 2.2 110 Italy 4.1 171 3.4 170 (1) Sum of frequencies of beer, wine and spirits for each respondent. (2) 1988 survey; see Hupkens et al., 1993. (3) The ECAS survey conducted Spring 2000, according to the frequencies in the Q-F scale. (4) Finland: mean of 1984 and 1992 (calculated from Simpura et al., 1993). (5) Survey conducted Spring 1990 by SIFO with face-to-face interviews with a random sample of 1,111 ages 16-75 (ages 18-64; N=1,035). (6) Including so-called ordinary strength beer (2.3%-3.5% alcohol by volume). TABLE 4 Volume (cl. 100% alcohol) per drinking occasion for each alcoholic beverage Finland Sweden Germany UK France Italy Men: Beer: 18-29 9.0 9.7 3.0 8.9 3.1 2.0 30-49 4.8 6.4 3.3 6.1 2.4 1.9 50-64 3.7 2.5 2.8 5.4 1.5 1.2 Total 5.9 6.3 3.1 6.7 2.4 1.7 Wine: 18-29 2.9 3.3 3.0 3.7 2.4 2.1 30-49 3.6 3.5 3.2 3.9 3.1 2.4 50-64 3.6 2.9 2.5 3.6 2.9 2.5 Total 3.4 3.3 3.0 3.8 2.8 2.3 Spirits: 18-29 7.4 6.9 3.3 4.0 3.4 1.9 30-49 7.0 4.1 2.5 2.4 2.2 1.1 50-64 6.7 3.1 1.7 2.6 1.3 1.0 Total 7.1 4.7 2.5 2.9 2.4 1.3 Women: Beer: 18-29 4.5 4.5 1.9 3.4 1.2 1.2 30-49 2.2 2.2 1.6 1.9 0.6 1.2 50-64 1.5 1.1 1.1 1.1 0.5 0.8 Total 2.0 2.5 1.5 2.0 0.8 1.1 Wine: 18-29 3.4 3.9 2.8 3.8 1.5 1.5 30-49 3.3 3.2 2.5 3.8 1.8 1.9 50-64 2.8 3.0 2.2 3.0 1.5 1.6 Total 3.2 3.3 2.5 3.6 1.6 1.7 Spirits: 18-29 2.7 2.4 2.0 3.8 1.1 1.1 30-49 2.4 1.8 1.2 2.2 0.9 1.3 50-64 2.5 1.4 1.1 1.4 0.4 0.9 Total 2.5 1.9 1.4 2.3 0.8 1.1 TABLE 5 Total (mean) alcohol consumption (liters 100% alcohol/year) by gender and age Men Study Age Mean Index country group consumption Total=100 Finland 18-29 8.2 117 30-49 6.2 89 50-64 6.8 97 Total 7.0 100 Sweden 18-29 7.5 142 30-49 4.7 89 50-64 3.8 72 Total 5.3 100 Germany 18-29 4.3 81 30-49 6.1 115 50-64 5.2 98 Total 5.3 100 UK 18-29 16.0 122 30-49 11.1 85 50-64 13.2 101 Total 13.1 100 France 18-29 6.1 81 30-49 8.0 107 50-64 8.6 115 Total 7.5 100 Italy 18-29 6.4 90 30-49 7.0 99 50-64 7.7 108 Total 7.1 100 Women Study Age Mean Index Ratio country group consumption Total=100 men/women Finland 18-29 2.9 121 2.8 30-49 2.2 92 2.8 50-64 2.4 100 2.8 Total 2.4 100 2.9 Sweden 18-29 1.9 112 3.9 30-49 1.6 94 2.9 50-64 1.6 94 2.4 Total 1.7 100 3.1 Germany 18-29 3.2 133 1.3 30-49 2.2 92 2.8 50-64 2.1 88 2.5 Total 2.4 100 2.2 UK 18-29 7.9 155 2.0 30-49 5.0 98 2.2 50-64 3.1 61 4.3 Total 5.1 100 2.6 France 18-29 1.8 82 3.4 30-49 2.4 109 3.3 50-64 2.4 109 3.6 Total 2.2 100 3.4 Italy 18-29 2.5 71 2.6 30-49 3.8 109 1.8 50-64 3.8 109 2.0 Total 3.5 100 2.0 TABLE 6 Distribution of heavy drinking occasions during the past 12 months, by gender Daily 4-5 2-3 Once a 2-3 Once days a days a week days a a week week month month Men: Finland 0.4 0.4 3.3 11.7 18.8 15.8 Sweden 0.0 0.2 1.5 5.9 13.0 15.9 Germany 1.4 0.2 1.3 6.3 4.7 8.5 UK 2.7 5.2 11.1 15.0 7.6 10.2 France 0.2 0.2 2.1 5.5 5.8 10.2 Italy 2.4 2.5 3.2 3.1 4.81 5.1 Women: Finland 0.0 0.0 1.0 1.7 5.2 8.9 Sweden 0.2 0.0 0.0 1.2 4.1 6.5 Germany 0.1 0.2 0.5 0.9 1.1 4.6 UK 0.2 1.3 3.9 6.5 6.5 10.7 France 0.0 0.0 0.5 1.4 1.0 4.2 Italy 1.5 1.3 2.5 2.2 1.9 2.1 Once or Never (1) a few days a year Men: Finland 24.4 25.2 Sweden 28.9 34.6 Germany 28.6 49.0 UK 14.1 34.1 France 22.7 53.3 Italy 26.7 52.3 Women: Finland 30.4 52.8 Sweden 26.4 61.7 Germany 17.9 74.6 UK 19.4 51.5 France 16.4 76.6 Italy 16.4 72.0 (1) Including 12-months abstainers. TABLE 7 Heavy drinking occasions in the six study countries; mean values among all respondents during the past 12 months, by gender and age Men Study Age Mean Index country group frequency Total= 100 (occasions/ 12 months) Finland 18-29 25 123 30-49 18 88 50-64 18 90 Total 20 100 Sweden 18-29 20 161 30-49 10 83 50-64 7 54 Total 12 100 Germany 18-29 19 140 30-49 13 98 50-64 7 53 Total 14 100 UK 18-29 61 131 30-49 43 92 50-64 37 79 Total 47 100 France 18-29 13 118 30-49 10 97 50-64 8 76 Total 11 100 Italy 18-29 22 96 30-49 18 80 50-64 29 128 Total 23 100 Women Study Age Mean Index Ratio country group frequency Total= 100 men/women (occasions/ 12 months) Finland 18-29 11 197 2.2 30-49 5 90 3.5 50-64 3 48 6.5 Total 6 100 3.5 Sweden 18-29 7 167 2.8 30-49 3 72 3.3 50-64 3 74 2.1 Total 4 100 2.9 Germany 18-29 7 183 2.9 30-49 3 75 4.9 50-64 3 78 2.5 Total 4 100 3.8 UK 18-29 28 175 2.2 30-49 16 101 2.7 50-64 6 40 5.8 Total 16 100 2.9 France 18-29 3 118 3.8 30-49 3 111 3.3 50-64 2 64 4.5 Total 3 100 3.8 Italy 18-29 11 78 2.0 30-49 9 65 2.0 50-64 22 160 1.3 Total 14 100 1.6 TABLE 8 Mean number of heavy drinking occasions and heavy drinking occasions in relation to drinking frequency, by gender (1) (2) (3) Mean binge Mean number Mean number occasions of drinking of drinking Country per year occasions past occasions past 7 days (1) 12 months (2) Men: Finland (n=496) 20.4 1.8 70 Sweden (n=493) 12.4 1.4 37 Germany (n=372) 13.5 2.3 97 UK (n=403) 46.6 3.9 118 France (n=475) 10.6 4.2 121 Italy (n=486) 22.6 6.6 179 Women: Finland (n=508) 5.8 1.1 35 Sweden (n=506) 4.3 0.7 24 Germany (n=628) 3.6 1.6 54 UK (n=581) 15.9 2.4 73 France (n=525) 2.8 1.7 62 Italy (n=514) 13.9 3.9 121 (4) (5) Binge (1) per Binge (1) per 100 drinking 100 drinking Country occasions occasions (past (occasions past 7 days (2)x52) 12 months (3)) Men: Finland (n=496) 29 22 Sweden (n=493) 33 17 Germany (n=372) 14 11 UK (n=403) 40 24 France (n=475) 9 5 Italy (n=486) 13 7 Women: Finland (n=508) 17 10 Sweden (n=506) 18 11 Germany (n=628) 7 4 UK (n=581) 22 13 France (n=525) 5 3 Italy (n=514) 11 7 (1) Sum of number of drinking occasions past seven days at lunchtime, during evening/late afternoon meal, at a restaurant or a bar, at home but not in connection with a meal. The highest possible number of occasions per day is thus 4. (2) Based on the alcoholic beverage most frequently drunk for each respondent. TABLE 9 Distribution of context of drinking (%) Country Lunch Afternoon Restaurant Home but not or evening or bar in connection meal with a meal Finland: Men 14 15 26 46 Women 15 23 20 42 Sweden: Men 9 41 20 31 Women 5 60 16 18 Germany: Men 12 40 18 29 Women 7 44 22 26 UK: Men 10 36 27 28 Women 5 45 22 28 France: Men 31 40 17 13 Women 37 46 11 5 Italy: Men 40 44 12 5 Women 46 44 7 3
AHLSTROM, S. (1999). Drinking patterns of women in nine European countries: Descriptive findings, in: BLOOMFIELD, K., AHLSTROM, S., ALLAMANI, A., CHOQUET, M., CIPRIANI, F., GMEL, G., JACQUAT, B.J., KNIBBE, R., KUBICKA, L., LECOMTE, T., MILLER, P., PLANT, M. & SPAK, F. (Ers.), Alcohol Consumption and Alcohol Problems Among Women in European Countries. Project Final Report (Berlin, Free University of Berlin, Institute for Medical Informatics, Biostatistics & Epidemiology).
BLOOMFIELD, K., AHLSTROM, S., ALLAMANI, A., CHOQUET, M., CIPRIANI, F., GMEL, G., JACQUAT, B.J., KNIBBE, R., KUBICKA, L., LECOMTE, T., MILLER, P., PLANT, M. & SPAK, F. (Eds.) (1999). Alcohol Consumption and Alcohol Problems Among Women in European Countries. Project Final Report (Berlin, Free University of Berlin, Institute for Medical Informatics, Biostatistics & Epidemiology).
CASSIDY, T.M. (1997). Alcohol in Ireland: The Irish solution (The University of Chicago, Department of Sociology, dissertation).
Doll (1999). Statistics on alcohol: 1976 onwards. Statistical Bulletin 1999/24. Department of Health.
Doll (2001). Statistics on alcohol: 1978 onwards. Statistical Bulletin 2001/13. Department of Health.
Eurobarometer 34.1 (1990). In: European Drug Prevention Program (Brussels: INRA). (Results accessible also at http://www.gesis.org/).
GUILBERT, R. & BAUDIER, F. (2000). Barometre sante (http://www.cfes. sante.fr/30000/pdf/07Alcool.pdf).
HANHINEN, S. (1995). Nordic, Italian and German drinking habits. A comparison between surveys made since the 1980s. Nordic Alcohol Studies, 12 (English Supplement), 14-30.
HAUGE, R. & IRGENS-JENSEN, O. (1986). The relationship between alcohol consumption, alcohol intoxication and negative consequences of drinking in four Scandinavian countries. British Journal of Addiction, 81, 513-524.
HAUGE, R. & IRGENS-JENSEN, O. (1987). Age, alcohol consumption and the experience of negative consequences of drinking in four Scandinavian countries. British Journal of Addiction, 82, 1101-1110.
HIBELL, B., ANDERSON, B., BJARNASSON, T., KOKKEVI, A., MORGAN, M., NARUSK, A. (1997). The 1995 ESPAD Report (Stockholm: CAN).
HUPKENS, C.L.H., KNIBBE, R.A. & DROP, M.J. (1993). Alcohol consumption in the European Community: Uniformity and diversity in national drinking patterns, Addiction, 88, 1391-1404.
KNIBBE, R. & LEMMENS, P. (1987). Korrelate des Alkoholkonsums in der Schweiz, Deutschland und den Niederlanden, Drogalcohol, 11, 27-41.
KRAUS, L. & AUGUSTIN, R. (2001). Population survey on the consumption of psychoactive substances in the German adult population 2000, Sucht, 47, S3-S87 (Supplement).
KRAUS, L. & BAUERNFEIND, R. (1998). Population survey on the consumption of psychoactive substances in the German adult population 1997, Sucht, 44, S3-S82 (Supplement).
KUHLHORN, E. & LEIFMAN, H. (1993). Alcohol surveys with high and low coverage rate: A comparative analysis of survey strategies in the alcohol field. Journal of Studies on Alcohol, 54, 542-554.
LADER, D. & MELTZER, H. (2001). Drinking: Adults' behaviour and knowledge in 2000 (London: Office of National Statistics).
LEIFMAN, H. (2000). Ungdomar och alkohol: treader och individuell utveckling--en genomgang av litteraturen pa omradet. Stockholm: Socialdepartementet (OAS).
LEIFMAN, H. (2001a). Validity problems in alcohol surveys, with special emphasis on the comparative six-country ECAS survey. Stockholm: Stockholm University, Centre for Social Research on Alcohol and Drugs (SoRAD). Submitted.
LEIFMAN, H. (2001b). Estimations of unrecorded alcohol consumption and trends in 15 European countries. Nordic Studies on Alcohol and Drugs, 18 (English Supplement 1), 54-70.
LEIFMAN, H. (2001c). Homogenization in alcohol consumption in the European Union, Nordic Studies on Alcohol and Drugs, 18 (English Supplement 1), 15-30.
MAKELA, P., FONAGER, K., HIBELL, B., NORDLUND, S., SABROE, S. & SIMPURA, J. (1999). Drinking Habits in the Nordic Countries, SIFA rapport Nr. 2/99 (Oslo: SIFA [National Institute for Alcohol and Drug Research]).
METSO, L. & SIMPURA, J. (1997). De finlandska dryckesvanorna under 90-talet (Finnish drinking habits in the 1990s). Nordisk Alkohol- & Narkotikatidskrift, 14, 154-166.
MUSTONEN, H., MAKELA, P., METSO, L. & SIMPURA, J. (2001). Changing drinking habits: Hopes bad reality. Paper presented at the 27th annual Alcohol Epidemiology Symposium of the Kettil Bruun Society for Social and Epidemiological Research on Alcohol (Toronto, 28 May-1 June).
NORSTROM, T. & SKOG, O.-J. (2001). Attityder till lordagsoppet 1999 och 2000 (Institutet for Social forskning, Stockholm Universitet, Senter for hoyere studier Det Norske Videnskapsakademi).
OSSERVATORIO PERMANENTE SUI GIOVANI E L'ALCOOL (1994). Young people and alcohol in Europe, by agreement with the EC Commission (Rome: Edizioni Otet).
OSSERVATORIO PERMANENTE SUI GIOVANI E L'ALCOOL (1998). Gli Italiani e l'alcool/The Italians and alcohol, vol. 11 (Rome: Vignola Editore).
READER'S DIGEST EURODATA (1991). A consumer survey of the 17 European countries (London: Reader's Digest Association Ltd.).
Rossow, I. & LEMMENS, P. (2001). Alcohol consumption and patterns of drinking--a global perspective. Manuscript.
SFSP (1998). Overview of national alcohol policies in the 15 countries of the European Union (Nancy: Societe Francaise de Sante Publique).
SIMPURA, J. & KARLSSON, T. (2001). Trends in drinking patterns among adult populations in 15 European countries. Nordic Studies on Alcohol and Drugs, 18, 31-54, English Supplement 1.
SIMPURA, J., MUSTONEN, H. & PAAKKANEN, P. (1993). Nya drycker--nya dryckesmonster? Modernisering av de finlandska dryckesvanorna jamfort med trender i EG-lander. Nordisk Alkoholtidskrift, 10, 265-276.
UHL, A. & SPRINGER, A. (1996). Studie uber den Konsum von Alkohol und psychoaktiven Stoffen in Osterreich unter Berucksichtigung problematischer Gebrauchsmuster (Vienna: Bundesministerium fur Gesundheit und Konsumentenschutz 1/96).
WHO (1995). Evaluation and Monitoring of Action on Alcohol, WHO Regional Publications, European Series No. 59 (Copenhagen: World Health Organization Regional Office for Europe).
HAKAN LEIFMAN is an assistant professor at the Centre for Social Research on Alcohol and Drugs, Stockholm University (Sveaplan, S-106 91 Stockholm, Sweden; email@example.com). A sociologist who conducts the monthly monitoring survey of alcohol purchasing in Sweden and directs a national survey on alcohol problems, Leifman coedited Statistics on Alcohol, Drugs and Crime in the Baltic Sea Region (2000).
|Printer friendly Cite/link Email Feedback|
|Publication:||Contemporary Drug Problems|
|Date:||Sep 22, 2002|
|Previous Article:||The six-country survey of the European comparative alcohol study: comparing patterns and assessing validity.|
|Next Article:||Alcohol consumption and the experience of adverse consequences--a comparison of six European countries.|