Global alcohol consumption has fallen from almost 18 litres of pure alcohol per year and per inhabitant in 1960 to almost 11 litres in 1999
Two major sources of data allow alcohol consumption to be estimated on an individual and collective level: market studies and consumer surveys1,2.
Market studies allow global alcohol consumption to be estimated per year and per inhabitant (15 years of age and above) based on the production, importing and exporting of alcohol. Imports are added to the quantities of alcohol produced per country, and exports are deducted. After weighting according to the size of the population, a mean annual consumption rate per inhabitant is obtained, expressed in litres of pure alcohol.
This is a mean value that does not distinguish between sex, age bracket, socio-professional category or other sociodemographic criteria. Furthermore, even if this is of little relevance to France, it appears that quite a considerable proportion of alcohol production is not taken into account (not declared and, therefore, not registered). This is particularly evident in the Scandinavian countries and Canada where it can account for up to 30% of alcohol production.
In France (Table 1), following an increase between 1951 and 1957, global alcohol consumption fell by almost 40% between 1960 (17.7 litres of pure alcohol/year/inhabitant) and 2002 (10.3 litres). In twenty years, wine consumption has fallen by almost 40% and beer consumption by 15%. The consumption of spirits has also declined (with considerable fluctuations over time). This fall in global alcohol consumption is therefore due essentially to a substantial decrease in the consumption of wine.
In 1999, France ranked in third position behind Luxemburg, Ireland and in front of Portugal. Wine consumption is on the decline in Southern Europe but is increasing very markedly in Northern Europe. The differences in alcohol consumption between the Latin countries, which are traditionally wine producers and consumers, and the Anglo-Saxon and Scandinavian countries persist, but are increasingly less noticeable (Table 2, Figure 1).
Figure 1. Evolution of global alcohol consumption in Europe for 40 years.
The countries traditionally recognised for their beer and spirits consumption have also witnessed a very rapid growth in their wine consumption whereas wine-consuming countries have registered an increase in beer consumption. Beer is a newcomer to Mediterranean countries and is mainly consumed by young people. Alcohol consumption in Europe is thus becoming more evenly distributed. However, the increase in beer consumption does not offset the decrease in wine consumption, hence global alcohol consumption is declining.
Some countries (Table 3) around the world have witnessed an increase in their global alcohol consumption (Brazil, Paraguay, Turkey and Mexico in particular), whilst others have noted a significant decrease (United States, Chile, Argentina and Algeria in particular).
Over 80% of all the alcoholic beverages exported come from European countries. France remains the leading exporter of alcohol (25.9% of the world's trade) whilst the United States is by far the leading importer.
Consumption surveys highlight indicators that enable homogeneous groups of the population to be followed over time, and allow changes in attitude towards alcohol to be studied according to sex, age and other sociobiographical criteria.
From 20 years of age, more than one French person in two consumes alcohol at least once a week
In France, the Baromètre santé 20005 estimated the daily, weekly, monthly and more occasional consumption habits of the French population between 12 and 75 years of age.
Among these French people, 3.5% declared that they had never drunk any alcohol in their life. The proportion of abstainers decreases as a function of age (17% at 12-14 years and almost 2% at 45-54 years). Women abstain more frequently than men (Figures 2 and 3).
Figure 2. Alcohol consumption among males over the last twelve months according to frequency of intake and age (Baromètre santé 2000, CFES)5.
Figure 3. Alcohol consumption among females over the last twelve months according to frequency of intake and age (Baromètre santé 2000, CFES)5.
Amongst the 12-75 year-olds, almost 20% admit to drinking alcohol every day: 28% of men and 11% of women. This daily consumption pattern starts to appear in young people from the age of 20-25 years, increasing with age and reaching a peak between the ages of 65 and 75 and affecting 65% of men and 33% of women.
Wine is the most widely consumed alcoholic drink. 17.5% of the 12-75 year-olds consume it on a daily basis: 25% of men and 10% of women. This increases with age: 62% of men between 65 and 75 years of age, and 32% of women in the same age bracket.
Over the last twelve months, 90% of the 12-75 year-olds admitted to having drunk at least one alcoholic drink. The average drink units the day before the interview (Figure 4) are more likely among men than women, for lifetime.
Figure 4. Average drink units the day before the interview according to age and sex (Baromètre santé 2000, CFES)5.
Inebriation is most common place amongst young people (Figure 5). The difference between the sexes is most apparent in the 20-25 age bracket: 40% of men between the ages of 20 and 25 repeatedly become drunk (more than 3 times per year) compared with 24.5% of women in this particular age bracket.
Figure 5. Average number of cases of inebriation over the last twelve months per sex and per age (Baromètre santé 2000, CFES)5.
In young people, repeated bouts of inebriation are often associated with regular alcohol or cannabis consumption
Based on surveys conducted in French schools between 1993 and 19996-8 (Espad, survey conducted by means of an anonymous questionnaire on a typical population of schoolchildren), experimenting with alcohol has slightly increased in France, rising from 81% to 86% in 16 year-old boys and from 79% to 85% in girls of the same age. Repeated usage (at least ten uses in the course of one month) does not appear to have increased. As regards inebriation, the number of young people who, between 1993 and 1999, admitted to having experienced at least ten episodes of drunkenness over a 12-month period has remained stable in the 14-16 year-old bracket (5% in the boys), but has slightly fallen amongst the 17 and 18 year-olds: from 14% to 10% in boys of 18 and from 3% to 2% in girls of the same age.
In 1999 (Table 4), the proportion of French pupils who consumed alcohol during the previous twelve months was below the average figure obtained for all European countries (77% versus 83%). The same applied to the proportion of French pupils who had been inebriated in the last twelve months (36% versus 52%)(Table 5).
The French survey9, Escapad («health and consumption» survey conducted by means of an anonymous questionnaire completed during a «Journée d'Appel de Préparation à la Défense» (defence preparation seminar), shows that 17 year-old girls had their first taste of alcohol at 13.6 years, i.e. on average 6 months later than boys (13.1 years). Alcohol preceded smoking. The first experience of inebriation occurred approximately two years after initial alcohol consumption, regardless of age and sex. Girls declared having been inebriated for the first time on average 5 months after boys of their own age.
Behaviour differs between the sexes. The prevalence of smoking, alcohol consumption and inebriation appears to be linked to the early onset of experimentation. As regards concomitant usage, cannabis and alcohol are mostly linked.
According to longitudinal studies conducted in the United States and Europe focusing on adolescent alcohol consumption, the first risk factor for consumption at the end of adolescence or the onset of adulthood is the early onset of consumption. The onset of alcohol consumption at 12-14 years of age is a predictive factor for alcohol consumption at 16 or even alcohol abuse, whereas the onset of alcohol consumption at 16 is marginally predictive for adult consumption. This applies to both boys and girls.
One-third of young people between 16 and 17 years of age has experimented with alcohol, smoking and cannabis. There is a gradual shift towards experimenting with one or even two substances, followed by all three substances.
One young person in five regularly uses one of these substances. Polyconsumption increases appreciably with age, especially in the case of 19 year-old boys, 14% of whom take several substances on a regular basis.
Outings and evening functions are a fundamental aspect of a young person's life, and often lead to alcohol consumption. However, being a consumer may entice a young person to go to functions. According to Escapad, young people who frequent «techno» events account for less than 1% of the study population. Consequently, most of the alcohol consumers, smokers and people who become inebriated are recruited from amongst youngsters who have never been to a «techno» event.
The correlation (measured by the odds ratio) between the regular consumption of cannabis and regular alcohol consumption or smoking is high (3 < OR > 5) for boys and even higher (OR > 5) for girls. For both males and females, the link between repeated inebriation and regular alcohol or cannabis consumption is very marked. It is, however, more marked for girls than for boys. Overall, the risk of regular cannabis consumption is higher amongst female smokers than male smokers and amongst female drinkers than male drinkers. The risk of having been inebriated several times over the last thirty days is higher in girls who consume alcohol on a regular basis or take cannabis than amongst boys with similar habits.
Certain sociodemographic and educational factors are more closely linked with a high alcohol consumption. The same applies to absenteeism from school. The pupil's behaviour at school (absenteeism, school results and liking for school) is more closely linked with alcohol consumption than familial characteristics («intact», single-parent, restructured family), regardless of the type of school establishment (priority education area, public, private, town or rural location). Violent, delinquent behaviour (major violence, theft and disputes), running away and attempted suicide are associated with regular alcohol consumption. Consumption is not linked more frequently with major violence than it is with other problems such as attempted suicide or running away. The links between regular consumption and risky behaviour are always more noticeable for girls than for boys. Finally, those who do not practise any sport or who practise for more than eight hours per week are more likely to become regular consumers of alcohol than others.
Alcohol use disorders in France
The prevalence of alcohol abuse and dependency is estimated on the basis of studies conducted amongst the general population and for which the same methodological comments can be made as for the consumer studies: representative sample; concealed group; maintenance methods; nature of questions, risks of non-response that are added to the under-estimation of self-reported consumption. Comparative groups are often described but there are many biases in terms of selection, thus limiting extrapolation. According to the Baromètre santé 2000 data3, 8.6% of people in the 12 to 75 year-old bracket are or have been at risk of excessive consumption (according to the DETA questionnaire, the French version of the CAGE questionnaire): 13.3% of men and 4.1% of women. The highest prevalence is in the 45 to 54 age bracket. There is a link between the frequency of inebriation, smoking and cannabis. There was no change in dependency (according to DETA) between 1992, 1995 and 2000 for those over 20 years old.
According to a survey conducted among general practitioners10, 20% in adult outpatients (30% males and 11% females) are hazardous drinkers and 5% are dependent.
The prevalence of alcohol use disorders in hospital patients was studied in 1999 among 9 french regional military hospitals11: the prevalence of alcohol disorders (according to DETA score >= 2) is 14.9%: 18.3% among males and 6.6% among females. The prevalence of DETA scores (from 2 to 4) is higher among the ages 46-55 and in psychiatric care units.
In the working environment, men drink twice as much alcohol as women
In France, several studies have been carried out in recent years to assess the prevalence of alcohol consumption in the working environment. First and foremost, it should be noted that the collection of objective and systematic data is ethically difficult within the scope of occupational medicine. Furthermore, it should also be noted that acceptable blood alcohol levels change over time not least by current legislation for road users which is 0.5 g/l at the present time (law dated 13 November 1996).
According to an Ipsos survey conducted in September 2001, 71% of those interviewed consume alcohol during business lunches/dinners, 35% consuming more alcohol than usual. Alcohol consumption in a professional context is evident in physically demanding professions such as building, farming and maintenance works, as well as in professions dealing with the general public.
In all cases and regardless of age, the average number of glasses consumed by men in a working environment is 1.5 to 2 times higher than that of women. According to the results of the Gazel cohort study [over 20 000 EDF-GDF (gas and electricity) employees], 18.3% of men admitted to drinking 3 to 4 glasses of alcohol every day and 12.3% to 5 glasses and more per day. The equivalent percentages in women are 3.5% and 1.2%, respectively.
Surveys carried out by teams of work doctors have also facilitated assessment of the number of people experiencing alcohol-related problems. In the iron and steel sector, doctors confirmed in 1983 that 10% of workers were excessive drinkers and 8% alcoholics.
Alcohol is involved three times more often in road-traffic accidents than in accidents in the workplace
Numerous studies have investigated the presence of alcohol in various types of accident in order to establish any correlation between blood alcohol levels and the accident.
Although alcohol is associated with various types of road-traffic accidents, it also plays a major role in domestic accidents, accidents at work, in fights and in drownings, etc. In 1992 in the United States, alcohol was found to be involved in 50% of the road-traffic accidents and in fewer than 20% of accidents occurring in the work place.
In France, a multicentre study involving almost 5 000 accident victims admitted to 21 hospitals between October 1982 and March 1983 allowed the blood alcohol levels of the victims and two biological indicators, namely gamma-glutamyltransferase (GGT) and mean corpuscular volume (MCV), to be examined at the same time, confirming significant and chronic alcohol consumption. The results indicate that alcohol is very often involved in fights and road-traffic accidents (Table 6): 60% of men involved in fights have a blood alcohol level exceeding 0.50 g/l. Sporting accidents and accidents in the workplace represent the lowest incidences of elevated blood alcohol levels: 5% in men. In women, high blood alcohol levels are mainly evident in domestic accidents and fights.
Alcohol is responsible for approximately 2700 deaths per year on the road
In France at the present time, alcohol is involved in one-third of fatal road-traffic accidents, regardless of whether the driver is male or female. The total number of people killed on the road in one year is approximately 8 000 and the number injured, 160 000. Alcohol is linked with around 2 700 deaths and 24 000 injuries on the road each year. In 1970 and 1980, alcohol played a role in 40% and 30% of fatal accidents, respectively.
In France, 60% of the accidents where drivers have an illegal level of alcohol in the blood (> 0.5 g/l) occur between midnight and 04.00. The number of alcohol-related cases in which women are responsible for fatal accidents is always 3 to 4 times less than that of men. However, the number of female drivers arrested for breaking the law and nighttime accidents involving one single vehicle is on the increase amongst women in Anglo-Saxon countries. Preventive actions should take into consideration the behavioural differences between men and women before driving.
Alcohol could be responsible for 10 to 20% of accidents in the work place
Although this is a recurring problem in the working environment, there have been no recent, precise studies of the implication of alcohol in accidents in the workplace. Some American studies reported in an international working review (Organisation internationale du travail, OIT international working organisation) that alcohol and drugs trigger between 20% and 25% of accidents in the workplace and up to 30% of work-related deaths.
In France, according to the Association nationale de prévention de l'alcoolisme (Anpa, 2000-2001) (national association for the prevention of alcoholism), alcohol is directly responsible for between 10% and 20% of accidents in the work place, taking all socioprofessional categories into account.
At the SNCF (French railway), alcohol is thought to be involved in 20% of the 13 500 work-related accidents that occur each year, although it has been noted that most of the accidents involved non-alcohol-dependent agents.
The cost of loss of income associated with illness or premature death costs four times more than health expenditure
Analysis of the economic consequences associated with excessive alcohol consumption is an area of vast, fertile research (most of these publications nevertheless originating from Anglo-Saxon countries). The diversity of the potential effects of alcoholism on the economy in fact implies studies that exceed the strict framework of assessing the repercussions of the disease on the health of individuals. In addition to the costs of the disease, expenditure attributed to alcohol-related criminal offences and road-traffic accidents is also taken account in similar research.
In France, several studies have been devoted to investigating the cost of alcoholism. One of the studies estimates the direct medical costs of alcoholism at 2.4 billion Euros in 1996. Another study exceeds the restrictive framework of analysing care and also includes, in addition to loss of income and production, expenses incurred by criminal offences and road-traffic accidents as well as loss of mandatory samples due to excessive alcohol consumption. The overall amount of losses attributable to alcohol is thus estimated at 17.6 billion Euros. Health costs represent 15% of this total, thus far behind the loss of income and production (50%) and the expenses that road-traffic accidents incur for insurance companies (20%).
These studies, conducted both in France and abroad, emphasise the heavy toll of alcoholism on society and the extent of indirect costs. Conversely, direct medical costs, which are mostly investigated, represent only 10% to 15% of the overall cost of alcoholism according to the studies. These studies also demonstrate the importance of carrying out sound, epidemiological studies in order to estimate the aetiological ratios of costs.
Alcohol-related pathologies
As in many other countries, alcohol misuse is linked with cardiovascular and tumoral pathologies.
Alcohol-related mortality
France falls into the intermediate range in the European Union for per capita deaths directly associated with alcohol (alcohol cirrhosis of liver, alcoholic psychosis and upper aero digestive tracts cancer), with 37 (per 100,000 deaths) relatively far behind Austria (57) and Italy (56). More deaths in other categories may be (partly) attributed to alcohol and must therefore be taken into account in estimating total deaths attributable to alcohol (Table 8). The mortality rate attributable to alcohol thus increases from 37 to 51.6, and is much higher among men (80.9) than women (23.7): 3.4 times higher.
From 1993 to 1993, the number of deaths associated with alcohol use is quite stable, among males and females (Table 9).
Among males and females, the highest percentages of deaths attributable to alcohol are between the ages of 45-54 and 55-64: among men, it is more than one out of three deaths attributable to alcohol (from 35.4 to 38.7, Table 10).
