Addiction (1999) 94(6), 857±866
RESEARCH REPORT
Alcohol consumption in a national sample of
the Russian population
MARTIN BOBAK,1 MARTIN MCKEE,2 RICHARD ROSE3 &
MICHAEL MARMOT1
1
International Centre for Health and Society, University College London, Place, London
European Centre on Health of Societies in Transition, London School of Hygiene and
Tropical Medicine, Keppel Street, London & 3Centre for the Study of Public Policy,
University of Strathclyde, Glasgow, UK
2
Abstract
Aims. Alcohol has been suggested as an important determinant of mortality in Russia but survey data on
individuals’ alcohol consumption in Russia are sparse. We have analysed the levels and distribution of alcohol
consumption in a national sample of the Russian population. Design. Cross-sectional survey. Participants. A multi-stage random sample of men and women of the Russian Federation (N 5 1599, response
rate 66%). Measurements. Data on frequency of drinking alcohol and the average amount consumed at
one occasion were collected in an interview. Information was also collected on smoking, self-rated health and
a broad range of socio-economic factors and political attitudes. Findings. Nine per cent of men and 35%
of women reported that they never drink alcohol; 10% of men and 2% women drink several times a week;
44% of men and 6% of women reported that they drink an equivalent of 25 cl of vodka or more at one
occasion and 31% of men and 3% of women would do so at least once a month (25 cl of vodka contains
78.5 g of absolute alcohol). There were differences in alcohol consumption between geographical areas.
Material deprivation was not related to alcohol consumption. Among men, smokers, unmarried, unemployed
and men reporting poor health consumed more alcohol; women with higher education, widows, non-smoking
and with worse health consumed less alcohol. Variables related to reaction to economic and political changes,
rating of family economic situation general satisfaction or political preferences were not related to alcohol
consumption. Conclusions. While the overall levels of alcohol consumption appeared low, possibly due to
under-reporting, the proportion of men who can be considered as ªbinge drinkersº was relatively high. The
absence of sizable socio-economic differences suggest that drinking may be spread relatively uniformly in
Russia, especially among males. Alcohol consumption seems unrelated to individuals’ perception of the recent
societal changes.
Introduction
The recent pronounced ¯uctuations in mortality
in Russia and in other countries of the former
Soviet Union have attracted considerable interest. After a decline between 1987 and 1989,
mortality in Russia increased dramatically be-
Correspondence to: Dr Martin Bobak, International Centre for Health and Society, Department of Epidemiology
and Public Health, University College London, 1±19 Torrington Place, London WC1E 6BT. Tel: 0171±391 1684;
fax: 0171 813 0242; e-mail: martinb@public-health.ucl.ac.uk
Submitted 5th June 1998; initial review completed 2nd September 1998; ®nal version accepted 6th November
1998.
0965±2140/99/060857±10 $9.50
Ó Society for the Study of Addiction to Alcohol and Other Drugs
Carfax Publishing, Taylor & Francis Ltd
858
Martin Bobak et al.
tween 1989 and 1994.1 Leon et al. have shown
that the largest relative ¯uctuations in Russian
mortality occurred, in increasing order, with cardiovascular diseases, external causes of deaths
(violence and accidents) and alcohol-related
deaths (a speci®c group of causes in the Russian
classi®cation, with alcohol poisoning as the main
cause). Mortality from cancers remained stable.2
These ®ndings suggest strongly that the mortality ¯uctuations were real, rather than due to
poor data quality.
There is indirect evidence that these mortality
¯uctuations may be related to alcohol consumption. The decline coincided with Gorbachev’s
1985 anti-alcohol campaign and the subsequent
rise occurred after the campaign had collapsed.3±5
The timings are consistent with evidence from a
wide range of sources on changes in availability
of alcohol during the late 1980s and in the early
1990s. When mortality rose markedly, the price
of alcohol fell further relative to basic foodstuffs
and alcohol supply increased markedly.6,7 The
improvement in mortality in late 1995 also coincided with a reduction in alcohol sales outlets.8
There are two problems with this theory. First,
cardiovascular diseases have made a major contribution to these ¯uctuations in all-cause mortality but alcohol is usually thought of as being
cardio-protective. 9±11 The only study of alcohol
and mortality in Russia published so far did not
®nd any excess mortality among drinkers.12 On
the other hand, that study did not speci®cally
address the effects of episodic heavy (ªbingeº)
drinking, common in Russia.3 Such drinking patterns may give rise to a range of adverse physiological changes that have the opposite effect of
regular moderate drinking, such as lipid pro®le,
coagulation or probability of arrhythmias.13
Secondly, existing data do not provide
unequivocal support for the common assumption that consumption of alcohol in Russia is
high and binge drinking is frequent, and it is not
known which factors in¯uence of alcohol consumption in Russia. Available estimates of
alcohol consumption in Russia appear too low to
produce such an enormous impact on mortality.
Of®cial sales statistics suggest an annual per
capita consumption of pure alcohol around 5
litres; this would put Russia at the bottom
among European countries but considerable
unregistered sales and production occur.3,6 Data
from surveys are limited but also show lower
than expected levels of drinking.14 Moreover, few
studies have analysed the pattern of alcohol consumption in Russian population.14,15
This paper addresses this second issue. We
have examined the pattern of self-reported consumption of alcohol in a national sample of the
Russian population, and attempted to identify
the socio-demographic factors that in¯uence it.
Methods
The data are from the 6th New Russia Barometer survey, collected by the Russian Centre
for Public Opinion Research (VCIOM) between
25 July and 2 August 1996, shortly after the
presidential elections.16 A multi-stage nationwide sample of the Russian population aged 18
years and over was selected. The Russian Federation was ®rst strati®ed into 22 regions, then into
urban and rural areas, and then, for urban areas,
into regional centres and other towns. Within
this framework, 69 urban and rural settlements
were selected randomly with a probability equal
to their share of the population. Within these
locations, primary sampling units (enumeration
districts) were randomly drawn. In each primary
sampling unit, households were listed by
address, an address was selected randomly as the
starting point, and interviewers were instructed
to seek an interview at every n-th house. At each
address the interviewer asked for a respondent
matching an age by gender by education grid.
Face-to-face interviews were conducted.
Of 3379 households with someone at home, in
965 cases no one met the requirements of the
age±sex±education sample grid. Of the 2414
remaining households, in 470 (20%) the interview was refused; in 271 (11%) the door was not
answered; in 63 (3%) the identi®ed individual
was unable to answer, and 11 interviews (0.4%)
were interrupted. The 1599 completed interviews represent a ®nal response rate of 66%, and
agrees well with the age, sex and education distribution in the Russian population (for full
details, see Rose16). For example, 13.9% of the
responders had education higher than secondary
and 37.9% had incomplete secondary education
or less; the Goskomstat (Russian statistical
of®ce) estimates of these proportions in the study
age group are 14.5% and 37.7%, respectively.
The main objective of the survey was to collect
data on political, demographic, social and economic in¯uences on electorial behaviour, but the
Alcohol consumption in Russia
859
Table 1. Percentages of the distribution of indicators of alcohol consumption in the sample. Differences
from 100 are due to rounding
Indicator of alcohol consumption
Frequency of drinking alcohol
Never
Less than once a month
About once a month
About 1±2 times a month
Few times a week
How much drink, on average, at one occasion
Never drink
A measure of vodka or less
About 2 measures of vodka
Big glass (25 cl) of vodka
A bottle (50 cl) of vodka or more
Four dichotomized indicators used in the analyses
Drinking more than once a month
Drinking at least 25 cl vodka at one occasion
Drinking at least 25 cl vodka more than once a month
Drinking at least 50 cl vodka more than once a month
interview included a short module covering
health and health behaviours.
Alcohol consumption was measured by two
questions. First, ªDo you sometimes drink
alcoholic beverages? If so, how often?º The
answers were as follows: never; less than once a
month; about once a month; 1±2 times a month;
and few times a week; secondly, ªIf you drink
alcohol, how much, on average, do you drink at
one occasion?º Possible answers were: less than
a measure (5 cl) of vodka, glass of wine or bottle
of beer; one measure of vodka (or equivalent),
two measures of vodka, big glass (25 cl) of vodka
or equivalent; bottle (50 cl) of vodka or equivalent; and more than a bottle of vodka.
From these two questions, the following four
dichotomized measures of alcohol consumption
were constructed (Table 1) and will be reported
in the paper: (1) drinking alcohol at least once a
month; (2) drinking at least 25 cl vodka (or
equivalent) in one go; (3) drinking at least 25 cl
of vodka (or equivalent) at least once a month;
and (4) drinking at least 50 cl vodka at least once
a month.
Among other health-related factors, information on self-rated health, smoking and history
of cardiovascular problems was also collected.
Perceived overall health in the last 12 months
was rated on ®ve-point scale: very poor, poor,
average, good and very good.
A material deprivation score was calculated by
Men
(n 5 731)
Women
(n 5 868)
9
20
21
40
10
35
36
18
10
2
9
20
27
30
14
35
44
16
5
1
50
44
31
11
12
6
3
1
adding responses to questions about how often
the participants do without food, heating or
necessary clothes/shoes; the score had values
from 0 (low) to 9 (high material deprivation).
Education was classi®ed into four groups: primary, apprenticeship, secondary (A-level equivalent) and university. Marital status was
categorized into married, single/divorced and
widowed. Because reaction to political, social
and economic changes may act as a trigger for
heavy alcohol consumption, we have also
assessed their effects; they will be speci®ed in
more detail in the relevant sections. Data were
analysed for men and women separately. Ageadjusted odds ratios were calculated by unconditional logistic regression.
Results
Overall levels of alcohol consumption
Nine per cent of men and 35% of women
reported that they never drink alcohol (Table 1);
only 10% of men and 2% of women reported
drinking several times a week. Forty-four per
cent of men and 6% of women would drink, on
average, 25 cl vodka or more (or equivalent) at
one occasion, and 14% and 1%, respectively,
would drink 50 cl. Among men, 31% and 11%
would drink 25 cl and 50 cl of vodka, respectively, at least once a month; the corresponding
proportions were 5% and 1% among women.
Martin Bobak et al.
860
(a) Men
60
Prevalence (%)
50
40
30
20
10
0
18-24 25-34 35-44 45-54 55-64 65+
Age group
(b) Women
60
Prevalence (%)
50
are large geographical differences in alcohol consumption. Figure 2 shows the levels of alcohol
consumption in broad geographical areas of the
Russian Federation. The proportion of people
drinking heavily is lowest in the Volga region and
North Caucasus, with rates about one-®fth of
those seen in the Urals. The differences between
areas were statistically signi®cant for all indicators of consumption among men but only for
drinking more than 25 cl of vodka at one session
among women. The geographical differences
were not removed by adjusting for other variables. There were no signi®cant differences by
size of settlement. The survey did not ask about
religion, but information was available on
nationality. There were no differences in alcohol
consumption by nationality among men, but the
frequency of all alcohol consumption indicators
among non-Russian women was about half that
among Russian women. Due to small numbers
of female drinkers, only differences in the prevalence of drinking alcohol more than once a
month were statistically signi®cant (6% vs. 13%,
p 5 0.028).
40
30
20
10
0
18-24 25-34 35-44 45-54 55-64 65+
Age group
Figure 1. Alcohol consumption by age group in
Russian men (a) and women (b), (j drinks . 1 per month;
6 drinks . 25 cl vodka per occasion; 7 drinks . 25 cl
vodka . 1 per month; h ªheavy drinkingº . 50 cl
vodka . 1 per month).
Both frequency and amounts of consumed
alcohol declined with age; the decline started
after age 55 in men and after age 35 in women
(Fig. 1).
Geographical differences
So far, survey data have only been available from
large cities, but it has been suggested that there
In¯uence of socio-economic circumstances
Table 2 shows the odds ratios for drinking
alcohol more than once a month by education,
marital status and material deprivation. Among
men, the frequency of alcohol consumption
among men was not related to education or
deprivation but there were differences by marital
status: unmarried men drink more often while
widowers drink less often. Among women,
alcohol consumption was unrelated to deprivation, declined with education and was lower
among widowed women. Unemployment is a
predictor of alcohol consumption in men; currently unemployed men (10% of men in the
sample) reported consuming alcohol more often
than employed subjects. Mutual adjustment for
other socio-economic factors did not affect these
estimates.
Drinking at least 25 cl of vodka (or equivalent)
more than once a month showed the same pattern (Table 3). As heavier consumption is rarer,
con®dence intervals are wider, particularly
among women (only 3% of women reported that
they drink at least 0.25 l of vodka more than
once a month). Nevertheless, the point estimates
for smoking, unemployment and self-rated
health appear stronger for the indicator of heav-
Alcohol consumption in Russia
(a) Men
North-West (n =118)
Central (n =103)
Volga-Vyatka (n =41)
C. Black Soil (n =38)
Volga (n =81)
N. Caucases (n =83)
861
tion; this association was substantially stronger
in women, and was stronger for more extreme
measures of alcohol consumption. For example,
comparing smokers and non-smokers, the ageadjusted odds ratios for drinking at least once a
month, drinking at least 25 cl of vodka at least
once a month, and drinking at least 50 cl of
vodka at least once a month were 1.9, 2.2 and
2.9 in men, and 3.6, 10.0 and 33.4 in women
(all p-values less than 0.002).
Urals (n =102)
West Siberia (n =75)
East Siberia (n =47)
Far East (n =42)
0
10 20 30 40 50 60 70
Prevalence (%)
(b) Women
North-West (n =147)
Association with self-rated health
There was a clear gender difference in the relation of drinking to self-rated health. Among
men, the highest levels of drinking were reported
by those with the poorest perceived health.
Among women, there was a clear trend of declining alcohol consumption with declining health
(p 5 0.050, Table 2). This pattern was more
pronounced for the more extreme indicators of
alcohol consumption.
Central (n =129)
Volga-Vyatka (n =51)
Alcohol and perception of the changed situation
The survey measured a wide range of indicators
related to perception of the new situation in
Russia, such as the reaction to economic and
political changes; impact of market reform; rating of family economic situation now and compared to 5 years ago; economic expectations in
the future; role and responsibility of state; general satisfaction; and political preferences. None
of these factors were related to alcohol consumption (data not shown).
C. Black Soil (n =45)
Volga (n =98)
N. Caucases (n =96)
Urals (n =119)
West Siberia (n =88)
East Siberia (n =52)
Far East (n =42)
0
10 20 30 40 50 60 70
Prevalence (%)
Figure 2. Alcohol consumption by geographical area in
Russian men (a) and women (b), (j drinks . 1 per month
6 drinks . 25 cl vodka per occasion; 7 drinks . 25 cl
vodka . 1 per month; h ªheavy drinkingº . 50 cl
vodka . 1 per month).
ier drinking. Among those who drink at all at
least once a month (71% of all men and 30% of
women), socio-economic factors were not related
to drinking at least 25 cl of vodka (or equivalent)
more than once a month (results not shown).
Association with smoking
In both genders, smokers reported a higher
prevalence of all indicators of alcohol consump-
Discussion
In this study in a national sample of Russian
population, alcohol consumption was related to
gender, geographical area, and several socioeconomic and demographic variables, although
the associations differed between men and
women. Unexpectedly, alcohol consumption was
not related to material deprivation and reaction
to economic and political changes or rating of
the economic situation.
As with most surveys of alcohol consumption,
there is a risk that the amount drunk is underreported.17,18 As the survey was not designed
speci®cally to study alcohol, the questions have
serious limitations, although the Russian translators were asked to make sure that the questions
862
Martin Bobak et al.
Table 2. Numbers of subjects by socio-economic factors, smoking and self-rated health and age-adjusted
odds ratios (OR) and 95% con®dence intervals (95% CI) for drinking alcohol more than once a month
Men
N
Education
Primary
Vocational
Secondary
University
Material deprivation
(best vs. the worst decile)
Marital status
Married
Single/divorced
Widowed
Employment status
Currently employed
Unemployed
Smoking status
Non-smoker
Smoker
Self-rated health
Very good
Good
Average
Poor
Very poor
63
198
364
106
Women
OR (95% CI)
1.0
0.88 (0.47±1.63)
0.99 (0.53±1.84)
0.87 (0.64±2.55)
p for trend 0.96
N
92
253
406
117
1.16 (0.67±2.08)
OR (95% CI)
1.0
0.97 (0.36±2.65)
0.81 (0.29±2.29)
0.35 (0.10±1.21)
p for trend 0.036
0.97 (0.43±2.18)
564
138
29
1.0
1.38 (0.87±2.17)
0.35 (0.14±0.91)*
514
198
156
1.0
1.00 (0.61±1.64)
0.12 (0.03±0.52)**
657
74
1.0
1.93 (1.14±3.26)*
814
54
1.0
0.68 (0.26±1.80)
273
458
1.0
1.88 (1.37±2.57)***
743
124
1.0
3.57 (2.18±5.84)***
48
238
326
112
12
1.0
1.62 (0.83±3.18)
1.64 (0.83±3.26)
1.21 (0.61±2.41)
2.96 (0.76±11.6)
p for trend 0.80
23
165
426
214
40
1.0
0.38 (0.14±1.06)
0.40 (0.15±1.07)
0.26 (0.08±0.80)*
0.12 (0.01±1.13)
p for trend 0.050
Statistical signi®cance of odds ratios being different from 1: * p , 0.05; ** p , 0.01; *** p , 0.001.
were understandable and acceptable, and that
the wording includes all main types of alcohol
beverages. In particular, the scales of the questions on both the frequency and amount consumed are not ideal. The frequency measure did
not include an important category between ª1±2
times a monthº and ªa few times a weekº. In the
measure of amount consumed, there is a gap
between ªtwo measures of vodkaº (10 cl) and ªa
big glass of vodkaº (25 cl). Since the people
whose consumption falls in these gaps would be
more likely to report the lower category, the
overall consumption level would be underestimated. In addition, it has been shown that questions on the ªaverage amountº consumed usually
results in underestimation.19 Consequently, the
main importance of our data is in the new perspective they offer on the pattern of drinking
within the Russian population, rather than in the
estimates of the absolute consumption levels.
Nevertheless, the lower than expected total
volume of alcohol consumed estimated in this
study in not entirely inconsistent with other data.
A survey in Moscow in 1994 also did not ®nd
exceptionally high levels of drinking. In that
study, the average alcohol consumption in
Moscow in 1994 was similar to that in Finland,
and the proportion of daily drinkers among adult
men (around 4%) was lower than in most of
Europe.14 The proportion of men who became at
least slightly intoxicated at least once a week was
also similar in Moscow and in Finland: 32% and
29%, respectively.14
There may be an alternative, or additional,
possibility to under-reporting. While, as in the
Moscow,14 the overall volume consumed per
annum may be less than in other countries, the
associated harm may be much greater because of
the episode was in which it is drunk. Such a
pattern has been described previously, for example, in Native Americans.20 Binge drinking may
be potentially important for explaining the suspected role of alcohol as a cause of cardiovascular death rates in Russia, as episodic drinking
may have different effects than regular consumption. Earlier research has shown that irregular
Alcohol consumption in Russia
863
Table 3. Age-adjusted odds ratios (95% con®dence intervals) for drinking at least 0.25 l
of vodka (or equivalent) more than once a month
Men
Education
Primary
Vocational
Secondary
University
Material deprivation
(best vs. the worst decile)
Marital status
Married
Single/divorced
Widowed
Employment status
Currently employed
Unemployed
Smoking status
Non-smoker
Smoker
Self-rated health
Very good
Good
Average
Poor
Very poor
Women
1.0
0.74 (0.35±1.58)
0.91 (0.46±1.82)
0.72 (0.33±1.57)
p for trend 0.82
1.0
0.90 (0.17±4.84)
0.70 (0.12±3.97)
0.13 (0.01±1.72)
p for trend 0.072
1.35 (0.74±2.50)
1.34 (0.32±5.66)
1.0
0.91 (0.56±1.48)
0.64 (0.23±1.78)
1.0
0.59 (0.21±1.63)
0.50 (0.09±2.64)
1.0
2.19 (1.24±3.87)**
1.0
1.57 (0.43±5.68)
1.0
2.16 (1.51±3.09)***
1.0
10.0 (4.29±23.6)***
1.0
0.99 (0.48±2.04)
1.05 (0.50±2.19)
0.99 (0.43±2.29)
4.84 (1.17±20.0)*
p for trend 0.35
1.0
0.57 (0.11±2.89)
0.35 (0.07±1.75)
0.17 (0.02±1.21)
0.00 (0.0± ` )
p for trend 0.024
Statistical signi®cance of odds ratios being different from 1: * p , 0.05; ** p , 0.01;
*** p , 0.001.
consumption of large quantities of alcohol produces, in contrast to that seen with regular moderate drinking, adverse cardiovascular effects
including an atherogenic lipid pro®le, increased
risk of thrombosis and increased probability of
arrhythmias. 13 A prospective study in Finland
reported that binge drinking (of beer) was
related to an approximate 6.5-fold increase of
risk of myocardial infarction and three-fold
increase risk of all cause mortality.21 The same
study in Finland also found that frequent hangovers were associated with increased cardiovascular mortality.22
Our data suggest that binge drinking is relatively frequent in Russia. Although there is no
agreed de®nition of ªbinge drinkingº, consumption over 70 or 80 g of alcohol per occasion has
usually been used.23 Kauhanen et al. also considered six bottles of beer (equivalent to 25 cl of
vodka, or to 78 g of ethanol) as binge drinking.21
The proportion of subjects who ful®l this
de®nition in our sample was relatively high, particularly among men; 44% of men and 6% of
women reported to drink an equivalent of 25 cl
of vodka or more at one occasion, and 31% of
men and 3% of women would do so at least once
a month. For comparison, in the US in 1997 the
prevalence of binge drinking (at least once in the
last month) was estimated as 24% among men
and 9% among women.24 In England, 38% of
men and 12% of women reported to drink at
least six units of alcohol once a month or more25;
as six units contain about a half the alcohol
contents of 25 cl of vodka, the prevalence of
ªbinge drinkingº (using the cut-off point of 25 cl
of vodka) in England would also be lower than in
Russia.
This study is inconclusive as to the hypothesized harmful effects of alcohol on heart disease.
Had we found clear evidence that binge drinking
is common in Russia, the hypothesis would be
strengthened. In fact, the data suggest some
pattern of binge drinking, but it was not dramatically more common than elsewhere. However,
given the possible non-response bias (resulting
from missing the heaviest drinkers), the probability remains open that the real occurrence of
episodic drinking in Russia is high. At the least,
864
Martin Bobak et al.
however, the pattern of drinking found in this
study was not that of moderate daily drinking,
which is seen as protective against heart disease,26 only 10% of men and 2% of women in
our study reported drinking several times a
week. This is consistent with the absence of any
protective effect of alcohol in the Russian Lipid
Research Clinic Study.12
In our study, the factor most closely associated with all indicators of drinking was gender.
This is consistent with previous reports. The
Moscow study found that the mean alcohol consumption was more than 10 times higher in men
than in women.14 In the Russian Lipid Research
Clinics Study, 40% of men but only 4% of
women reported they drunk more than 12 g of
ethanol during the last week.12 In our study,
35% of women were classi®ed as non-drinkers.
This is more than in the other studies. On the
other hand, in the Russian Lipid Research Clinics Study 72% of women reported that they did
not have a drink during the last week. It is
possible that in our study, women who drink on
rare occasions classi®ed themselves as nondrinkers. Again, this would underestimate absolute levels of alcohol use but should not
introduce a serious bias in studying the distribution of ªheavierº alcohol consumption in the
population.
The differences in drinking between genders
were not limited to consumption levels. Education did not predict drinking among men but
was strongly inversely related to drinking by
women, although this was largely due to the low
level among those with a university education.
This is consistent with the gender differences
observed in the pattern of alcohol related mortality in Moscow.27 Unemployment was associated with drinking by men, but not women. The
differences in drinking cultures of men and
women in Russia14 warrant more attention in
future. In contrast, men and women behaved
similarly in terms of marital status, with
signi®cantly lower rates of drinking among those
who are widowed. Single and divorced men
reported higher, but not signi®cantly so, frequencies of alcohol consumption. This provides
only weak support for the role of alcohol in the
®ndings that unmarried men are the most vulnerable group in the former communist societies.28,29 The ®nding that alcohol consumption
was lower in women with worse self-rated health
but higher in comparable men requires more
detailed examination. While this difference may
re¯ect major differences in the responses to ill
health of men and women, with implications for
policy, the cross-sectional nature of the data,
and relatively small numbers of female drinkers
preclude ®rm conclusions.
The data are consistent with the view that
differences in alcohol consumption exist
between geographical areas of Russia. The differences were not signi®cant among women,
probably due to small numbers of female
drinkers. We were unable to explain these differences by adjustment using the data available in
this survey on socio-economic factors or
nationality, and other factors may be important.
In Russia, the probability of a woman drinking
heavily is greatest if she is young, Russian, and
smokes. This is consistent with a scenario in
which heavy drinking is associated with a breakdown in traditional values. The same pattern is
seen with smoking.30 It has been suggested previously that the decline in health status in the
late 1980 could be related to increases in economic hardship,31,32 and that alcohol could have
played a role in this process.33 If this were true,
and alcohol was related to disillusion and dissatisfaction, one would expect higher levels alcohol
among those perceiving the current situation
more negatively. This study is consistent with
other evidence that unemployment has played
an important part in the increase in Russian
mortality. However, alcohol consumption was
not related to any of the broad spectrum of
perceptions and attitudes tested in the analyses.
Mustonen found that in Moscow 40% of men
report that alcohol makes them more optimistic
about life;34 we speculate that this would be
more important in relation to the individual’s
family situation.15
To design an effective policy to reduce
alcohol-related problems, policy makers need to
know the important determinants of drinking.
Our study could not provide unambiguous
answers to this question. Only age, geographical
area, smoking and unemployment were strongly
related to drinking. Perhaps surprisingly,
material deprivation was not an important predictor of alcohol consumption. Drinking,
especially among men, may indeed be spread
relatively uniformly across socio-economic
groups in Russia. There is a suggestion that a
breakup of traditional society is related to
adverse health behaviour among women, but
Alcohol consumption in Russia
further data are needed. Alcohol control must be
a key priority for Russian public health policy
and more research is urgently required.
Acknowledgements
The New Russian Democracy Barometer survey
was supported by grants from the European
Commission and the Paul Lazarsfeld Society.
MM is supported by the Medical Research
Council Research Professorship. MMcK’s work
in Russia is supported by the UK Department
for International Development although DfID
can take no responsibility for views expressed.
References
1.
SH KOLNIKOV, V., MESLE, F. & VALLIN, J. (1995)
Health crisis in Russia: I. recent trends in life
expectancy and causes of death from 1970 to
1993, Population, 4±5, 907±944.
2. LEON, D. A., CHENET, L., SHKO LNIKOV, V. et al.
(1997) Huge variation in Russian mortality rates
1984±94: artefact, alcohol, or what? Lancet, 350,
383±388.
3. RYAN, M. (1995) Alcoholism and rising mortality
in the Russian Federation, British Medical Journal, 310, 646±648.
4. SH KOLNIKOV, V. M., MESLE, F. & VALLIN, J.
(1997) Recent trends in life expectancy and
causes of death in Russia, 1970±1993, in:
BO BADILLA, J. L., COSTELLO, C. A. & MITCHELL,
F. (Eds) Premature Death in the New Independen t
States, pp. 34±65 (Washington DC, National
Academy Press).
5. SH KOLNIKOV, V. M. & NEMTSOV, A. (1997) The
anti-alcohol campaign and variations in Russian
mortality, in: BOBADILLA, J. L., COSTELLO, C. A.
& MITCHELL, F. (Eds) Premature Death in the
New Independent States, pp. 239±261. (Washington, DC, National Academy Press).
6. LEVIN, B. M. (1997) Main factors of alcohol
consumption under conditions of rapid social
change, in: SIMPURA, J. & LEVIN, B. M. (Eds)
Demystifying Russian drinking. Comparative studies
from the 1990s, pp. 33±58. (Helsinki, STAKES).
7. WHITE, S. (1996) Russia Goes Dry, (Cambridge,
Cambridge University Press).
8. MC KEE, M., SANDERSO N, C., CH ENET, L.,
VASSIN, S. & SHKOLNIKO V, V. (1998) Seasonal
variation in mortality in Moscow, Journal of Public
Health Medicine, 20, 268±274.
9. MARMOT, M. & BRUNNER, E. (1991) Alcohol and
cardiovascular disease: the status of the U shaped
curve, British Medical Journal, 303, 565±568.
10. DOLL, R. (1997) One for the heart, British Medical Journal, 315, 1664±1668.
11. THUN, M. J., PETO, R., LO PEZ, A. D. et al. (1997)
Alcohol consumption and mortality among middle-aged and elderly US adults, New England
Journal of Medicine, 337, 1705±1714.
865
12. DEEV, A., SH ESTOV, D., ABERNATHY, J.,
KAPUSTINA, A., MUHINA, N. & IRGING, S. (1998)
Association of alcohol consumption to mortality
of middle aged US and Russian men and women,
Annals of Epidemiology, 8, 147±153.
13. MCKEE, M. & BRITTON, A. (1998) The positive
relationship between alcohol and heart disease in
eastern Europe: potential physiological mechanisms, Journal of the Royal Society of Medicine, 91,
402±407.
14. SIMPURA, J., LEVIN, B. M. & MUSTONEN, H.
(1997) Russian drinking in the 1990s: patterns
and trends in international comparison, in: SIMPURA, J. & LEVIN, B. M. (Eds) Demystifying Russian drinking. Comparative studies from the 1990s,
pp. 79±107 (Helsinki, STAKES).
15. CARLSON, P. & VAGERO, D. (1999) Alcohol and
the Russian mortality crisis, European Journal of
Public Health, in press.
16. ROSE, R. (1996) New Russia Barometer VI: after
the presidential election, Centre for the Study of
Public Policy: studies in public policy, no. 272
(Glasgow, University of Strathclyde).
17. EMBREE, B. G. & WHITEHEAD, P. C. (1998) Validity and reliability of self-reported drinking
behaviour: dealing with the problem of reporting
bias, Journal of Studies on Alcohol, 54, 334±344.
18. DUFFY, J. C. (1992) The measurement of alcohol
consumption, in: DUFFY , J. C. (Ed.) Alcohol and
Illness, pp. 19±25. (Edinburgh, Edinburgh University Press).
19. REHM, J. (1998) Measururing quantity, frequency, and volume of drinking, Alcohol: Clinical
and Experimental Research 22 (suppl.), 4S±14S.
20. WELTY, T. K., LEE, E. T., YEH , J. et al. (1995)
Cardiovascular disease risk factors among American Indians: the Strong Heart Study, American
Journal of Epidemiology, 142, 269±287.
21. KAUH ANEN, J., KAPLAN, G. A., GOLDBERG, D. F.
& SALONEN, J. T. (1997) Beer binging and mortality: results from the Kuopio ischaemic heart
disease risk factors study, a prospective population based study, British Medical Journal, 315,
846±851.
22. KAUH ANEN, J., KAPLAN, G. A., GOLDBERG, D. D.,
COH EN, R. D., LAKKA, T. A. & SALONEN, J. T.
(1997) Frequent hangovers and cardiovascular
mortality in middle aged men, Epidemiology, 8,
310±314.
23. Anonymous (1997) The Limits of Binge Drinking,
ICAP reports no. 2. (Washington, International
Center
for
Alcohol
Policies)
(Internet:
http:WWW.icap.org).
24. SUBSTANCE ABUSE AND MENTAL HEALTH SERVICES ADMINISTRATION (1997) Preliminary Results
from the 1997 National Household Survey on Drug
Abuse, (Washington DC, USDHHS) (Internet:
http:www.samhsa.gov/oas/nhsda/nhsda97).
25. PRESCOT-CLARKE, P. & PRIMATESTA, P. (Eds)
(1997) Health Survey for England 1995, vol. I.
Findings (London, The Stationary Of®ce).
26. ANDERSON, P., CREMO NA, A., PATON, A.,
TURNER, C. & WALLACE, P. (1993) The risk of
alcohol, Addiction, 88, 1493±1508.
27. CHENET, L., LEON, D., MCKEE, M. & VASSIN, S.
866
28.
29.
30.
31.
Martin Bobak et al.
(1998) Death from alcohol and violence in
Moscow: socioeconomic determinants, European
Journal of Population, 14, 19±37.
HAJDU, P., MC KEE, M. & BOJAN, F. (1995)
Changes in premature mortality differentials by
marital status in Hungary and in England and
Wales, European Journal of Public Health, 5, 259±
264.
WATSO N, P. (1995) Explaining rising mortality
among men in Eastern Europe, Social Science
Medicine, 41, 923±934.
MC KEE, M., BO BAK, M., ROSE, R., SH KOLNIKOV,
V., CHENET, I. & LEON, D. (1998) Patterns of
smoking in Russia, Tobacco Control, 7, 22±26.
CORNIA, G. A. (1997) Labour Market Shocks, Psychosocial Stress and the Transition’s Mortality Crisis
, Research in Progress 4 working paper, United
Nations University World Institute for Develop-
ment Economics Research (Helsinki, UNU/
WIDER).
32. BOBAK, M., PIKHART, H., HERTZMAN, C., ROSE,
R. & MARMOT, M. (1998) Socioeconomic factors, perceived control and self-reported health in
Russia. A cross-sectional survey, Social Science
Medicine, 269±279.
33. WALBERG, P., MC KEE, M., SHKOLNIKOV, V., CHENET, L. & LEON, D. A. (1998) Economic
change, crime, and the Russian mortality crisis: a
regional analysis, British Medical Journal, 317,
312±318.
34. MUSTONEN, H. (1997) Positive and negative experiences related to drinking, Moscow, 1994, in:
SIMPURA, J. & LEVIN, B. M. (Eds) Demystifying
Russian Drinking. Comparative studies from the
1990s, pp. 125±147. (Helsinki, STAKES).