ARTICLE IN PRESS

0277-9536/$ - se

doi:10.1016/j.so

�CorrespondE-mail addr

Social Science & Medicine 66 (2008) 663–674

www.elsevier.com/locate/socscimed

Economic strain, social relations, gender, and bingedrinking in Moscow

Tanya Jukkalaa,�, Ilkka Henrik Makinena, Olga Kislitsynab,Sara Ferlandera, Denny Vageroa,c

aStockholm Centre on Health of Societies in Transition, SCOHOST, Sodertorn University College, Huddinge, SwedenbInstitute for Socio-Economic Studies of Population, ISESP, Russian Academy of Sciences, Moscow, RussiacCentre for Health Equity Studies, CHESS, Stockholm University/Karolinska Institutet, Stockholm, Sweden

Available online 19 November 2007

Abstract

The harmful effects of alcohol consumption are not necessarily limited to the amounts consumed. Drinking in binges is a

specific feature of Russian alcohol consumption that may be of importance even for explaining the current mortality

crisis.

Based on interviews conducted with a stratified random sample of 1190 Muscovites in 2004, this paper examines binge

drinking in relation to the respondents’ economic situation and social relations. Consistent with prior research, this study

provides further evidence for a negative relationship between educational level and binge drinking. Our results also

indicate a strong but complex link between economic strain and binge drinking. The odds ratios for binge drinking of men

experiencing manifold economic problems were almost twice as high compared to those for men with few economic

problems. However, the opposite seemed to be true for women. Being married or cohabiting seemed to have a strong

protective effect on binge drinking among women compared to being single, while it seemed to have no effect at all among

men. Women having regular contact with friends also had more than twice the odds for binge drinking compared to those

with little contact with friends, while again no effect was found among men. Gender roles and the behavioural differences

embedded in these, may explain the difference. The different effects of economic hardship on binge drinking may also

constitute an important factor when explaining the large mortality difference between men and women in Russia.

r 2007 Elsevier Ltd. All rights reserved.

Keywords: Alcohol; Binge drinking; Gender; Moscow; Russia; Poverty; Social relations

Introduction and aim

In comparison with other countries, Russianshave generally been found to drink less often but toconsume larger amounts per occasion (Bobak et al.,2004; Leinsalu, 2004; Simpura, Levin, & Mustonen,

e front matter r 2007 Elsevier Ltd. All rights reserved

cscimed.2007.10.017

ing author. Tel.: +468 6084552.

ess: tanya.jukkala@sh.se (T. Jukkala).

1997). Drinking in binges, i.e. drinking until intoxica-tion within a short-period of time (InternationalCenter for Alcohol Policies, 1997), coupled with apreference for distilled spirits, are features ofRussian alcohol consumption (Chenet, McKee,Leon, Schkolnikov, & Vassin, 1998; Ryan, 1995)with deep historical roots (Makinen & Reitan,2006; White, 1996). Yet, little is known about theconcomitants of binge drinking in Russia. With this

.

ARTICLE IN PRESST. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674664

study we attempt to investigate these in relation toeconomic and social circumstances.

It has been suggested that this type of drinkingmight be one explanation behind Russia’s highlevels of alcohol-related mortality (Leon et al., 2007;Rehm et al., 2007; Room, 2005). More specifically,high levels of cardiovascular death (Britton &McKee, 2000; Shkolnikov, Chervyakov, Leon,& McKee, 2004; Shkolnikov, McKee, Chervyakov,& Kyrianov, 2002), alcohol poisoning mortality(Stickley et al., 2007), accidents (Leon et al., 2007),suicide (Makinen, 2000) and homicide (Pridemore,2002, 2004) have all been potentially associated withRussia’s binge-drinking culture. Therefore, theidentification of factors associated with this mannerof drinking might also be of potential relevance forunderstanding the Russian mortality crisis.

Economic strain and alcohol consumption

Increased alcohol consumption in Russia duringthe 1990s has been related to the political andeconomic changes of the transition period (Leon &Shkolnikov, 1998). These changes have also led to amajor increase in poverty, in particular, amongworking-age men and women (Gustafsson &Nivorozhkina, 2004). As Carlson and Vagero(1998) point out, people with a lower educationand who do manual work are those who seem tohave suffered as a result of increasing levels ofpoverty in the 1990s and from income loss in aperiod of sharply growing income differences. Thesegroups have also been found to binge drink moreoften (Carlson & Vagero, 1998; Malyutina, Bobak,Kurilovitch, Nikitin, & Marmot, 2004).

However, few studies have specifically relatedeconomic problems to heavy drinking (Bobak,McKee, Rose, & Marmot, 1999; Carlson, 2001)and the results of these have been somewhatambiguous. While the transition period has createda situation where many Russians are unable to meeteven their basic needs (Vagero & Kislitsyna, 2005),it is important to see whether and how economicstrain is related to binge drinking.

Social relations and alcohol consumption

The collapse of the Soviet state resulted in aneconomic situation where wages, pensions, andbenefit payments were often too low to cover basicneeds and in some cases they were not even paid atall (Vagero & Kislitsyna, 2005). In this situation,

social contacts that can provide material supportand various favours (blat) may be of majorimportance. The most common form of socialsupport among Russians probably comes from theinformal relations among family and friends (Rose,1995; Rose, Mishler, & Haerpfer, 1997). Those wholive alone and lack social contacts are thereforeparticularly vulnerable (Kennedy, Kawachi, &Brainerd, 1998). Strained family relations have alsobeen previously related to binge drinking in Russia(Carlson & Vagero, 1998).

We therefore hypothesised that social relations,by providing alternative resources in an economic-ally uncertain situation, may protect against bingedrinking. Further, social relations may also beprotective against binge drinking because theymay provide an environment for collective regula-tion against deviant behaviour (Weitzman & Chen,2005; Weitzman & Kawachi, 2000). This, of course,requires that binge drinking is regarded as beingdeviant. If not, then social relations may evenencourage this type of behaviour (Almedom, 2005;Kawachi & Berkman, 2001). Friendship ties may,in general, act as channels through which newbehavioural patterns, such as habits of alcoholdrinking spread through society. There is thus ahuman tendency to follow one’s peers and whetherthis is beneficial or harmful to one’s health dependson the norms within the network and the extent ofexternal information added to it (Ferlander, 2007).

Gender and alcohol consumption

The large gender difference in Russian lifeexpectancy (13 years in 2005) has been related togender differences in alcohol consumption (Carlson,2001). Previous studies have found that Russianmen drink alcohol in general, and binge drink inparticular, to a much larger extent than Russianwomen (Bobak et al., 1999; Malyutina et al., 2001;Nicholson, Bobak, Murphy, Rose, & Marmot,2005).

It is also possible that economic and social factorshave different effects on alcohol consumption inmales and females. In a study of eight Europeancountries Kuntsche et al. (2006) found that socialstratification and family roles were differentlyassociated with heavy drinking between men andwomen. These differences themselves varied be-tween countries and were more pronounced incountries with a low level of social welfare and highgender inequality.

ARTICLE IN PRESST. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674 665

Aim of study

Little is currently known about economic strainand social ties in relation to binge drinking inRussia. We hypothesise that they may be of majorimportance in explaining the social distribution ofthis drinking style. Thus, the aim of this study is toexamine how economic strain, the availability ofsocial support in terms of social relations, andgender are related to binge drinking in Moscow.

Data

The data used in this study come from TheMoscow Health Survey 2004, which was conductedby the Stockholm Centre on Health of Societies inTransition (SCOHOST) at Sodertorn UniversityCollege in collaboration with The Institute forSocio-Economic Studies of Population (ISESP) atthe Russian Academy of Sciences in Moscow inspring 2004. The interviewers, trained by ISESP,carried out face-to-face interviews with the respon-dents, using a structured questionnaire of somewhatmore than 100 questions. The respondents werechosen from the target population of individualsaged 18 years or over living in greater Moscow(8,540,000 individuals) using stratified randomsampling. The 125 municipal districts of greaterMoscow were divided into 10 administrative dis-tricts, and the aim was to obtain a specific numberof interviews from each district, with a specific sexand age distribution matching that of the districtpopulation. The final sample consisted of 1190respondents, and the primary response rate was47%. Non-responses from these interviews werereplaced with subjects from a ‘‘reserve list’’ whichwas randomly drawn from the same target popula-tion before the start of the interviewing.

With one exception, an analysis of non-response,based on the first 1200 attempted interviews, showedno obvious bias in the sample characteristics (Vageroet al., 2007). However, the proportions of persons atdifferent educational levels in the sample differedsignificantly from those in the Moscow City popula-tion, when compared to the figures of the 2002 Census(All-Russia Population Census, 2002). Especially, thehighly educated were overrepresented.

The dependent variable: binge drinking

The survey had several questions on alcohol con-sumption. The question concerning the frequency of

alcohol consumption had the response alternatives‘‘almost never’’, ‘‘only on special occasions’’, ‘‘onceor twice a month’’, ‘‘once a week’’, ‘‘several times aweek’’, and ‘‘every day’’. For each of the beverages,beer, wine, and spirits, there was also a questionabout the amount normally consumed per occasion,which was measured in litres of beer and wine andgrams of spirits, respectively.

‘Binge drinking’ was defined, in accordance withMalyutina et al. (2004), as drinking 80 g or more ofpure ethanol (corresponding to approximately 230 gof vodka) on one occasion for men and 60 g or morefor women. Those reporting drinking less than onceper month were excluded from the binge-drinkingcategory even if they reported drinking a highamount on any occasion.

The intake of pure ethanol in grams was calculatedfor the three different beverages using the methodpreviously used by Malyutina et al. (2001), where theamount of ethanol for 1 l of the beverages was 33 gfor beer, 88 g for wine, and 350 g for spirits. Theamount consumed per occasion was calculated usingthe highest amount of ethanol from the reportedamounts for each beverage for each respondent.

The independent variables

Economic strain was measured with four ques-tions indicating the presence of economic problems.These questions referred to the economic situationof the respondent’s family during the previous 12months. They were asked whether they ever:

(1)

Had to rely on outside help to pay regularexpenses on time (for example rent)?

(2)

Could not have meat or fish more than once ortwice a week?

(3)

Had to refrain from purchasing necessaryclothes or footwear (for adults or children inthe family)?

(4)

Involuntarily had to refrain from taking part insocial or cultural activities, like going to arestaurant, cinema, theatre, etc.?

The responses were recoded into two groups, withthe answers ‘‘sometimes’’ and ‘‘often’’ indicatingthat there were economic problems. The answers toeach question were then summed, giving a scale of0–4, which was further divided into two categories,those experiencing no or only one type of economicproblem and those experiencing two or more types.In the analyses below, this measure was used as an

ARTICLE IN PRESS

Table 1

Sample characteristics of The Moscow Health Survey 2004

Variable Men

(n ¼ 510)

Women

(n ¼ 680)

Total

(n ¼ 1190)

Age (mean) 45.2 48.6 47.1

Educational level (%)

High 56.7 50.1 52.9

Secondary 23.5 31.6 28.2

Low 19.8 18.2 18.9

Economic problems (%)

0–1 problems 73.9 56.6 64.0

2–4 problems 26.1 43.4 36.0

Marital status (%)

Non-married 33.7 50.3 43.2

Married/cohabiting 66.3 49.7 56.8

Social contacts (%)

Little contact with

friends

48.2 54.3 51.7

Regular contact with

friends

51.8 45.7 48.3

T. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674666

indicator of poverty instead of income, becausemonetary income does not reflect non-monetarysources, which are of great importance for manyRussians (Vagero & Kislitsyna, 2005).

Educational level was divided into three cate-gories: higher or incomplete higher (high), specia-lised secondary or vocational technical school(secondary), and common secondary or less (low).

As to social relations, marital status was dividedinto the categories married (or cohabiting) and non-married (divorced, widowed, or single). Further,respondents were asked whether they usually visitedfriends during their leisure time, with the answersdivided into the groups ‘‘often’’ and ‘‘sometimes’’/‘‘no’’.

As binge drinking was suspected of varying byage and sex, these were used as general controlvariables in the analyses. Age was used as acategorical variable with the following age-groupdivisions: 18–30, 31–40, 41–50, 51–60, and 61 yearsand over.

Methods

The frequency of drinking (Table 2), the typicalamounts consumed (Table 3), as well as theprevalence of binge drinking (Table 4) in differentage groups were calculated for both men andwomen. Due to the overrepresentation of the highlyeducated in our sample, standardised figures match-ing the educational-level proportions given in the2002 Census (All-Russia Population Census, 2002)for the Moscow City population (over 18 years) arealso provided for Tables 2–4.

Using logistic regression, odds ratios (ORs) forthe probability of binge drinking were then calcu-lated in relation to four independent variables,separately for men and women, and controlling forage. Table 5 (model 1) gives the resulting ORsadjusted only for age, while model 2 gives the sameORs when all other independent variables arecontrolled for. Finally (model 3), a logistic regres-sion analysis was also performed for both sexescombined, using the same dependent and indepen-dent variables, and controlling for age and gender.

Results

Sample characteristics, drinking, and binge drinking

Table 1 summarises the sample characteristics.The average age of the respondents was 47 years,

and 57% were women and 43% men. A majority(52.9%) had a high level of education while 18.9%had a low level of education. Manifold economicproblems were reported by as many as 43.4% ofwomen and 26.1% of men. Half of the women werenot presently married or cohabiting, while thecorresponding percentage among men was only athird (33.7%). Approximately half of the respon-dents reported regular contact with friends.

As regards drinking habits, Table 2 shows that15.9% of men report drinking never or only rarely.For women this proportion was more than twice ashigh (35.9%). Among men, the reported frequenciesof drinking were spread quite evenly between theresponse alternatives, with the exception of drinkingevery day, which only 5% reported doing. Menbetween 18 and 30 years seemed to drink mostfrequently. Among women there was less variation,nearly 80% reported drinking never or rarely, oronly on special occasions. Women reporting drink-ing most frequently were also found in the youngestage group.

The sex and age patterns seem stable despite thesmall number of interviews per group, with theexception of the males in the 41–50 age group,where only a small percentage reported drinkingnever or rarely. However, a high percentage in thesame age group reported drinking only on specialoccasions, and if the percentages for both of these

ARTICLE IN PRESS

Table 2

The distribution of drinking frequency (%) by age and sex in Moscow 2004

Age group Almost

never

Only on special

occasions

Once or twice a

month

Once a

week

Several times a

week

Every

day

No of subjects

interviewed

Men

18–30 yr 17.7 9.7 21.0 21.8 24.2 5.6 124 (100)

31–40 yr 15.1 19.4 19.4 23.7 18.3 4.3 93 (100)

41–50 yr 3.8 33.7 21.2 13.5 25.0 2.9 104 (100)

51–60 yr 24.4 19.8 14.0 16.3 18.6 7.0 86 (100)

61–70 yr 19.6 26.8 21.4 8.9 17.9 5.4 56 (100)

X71 yr 19.6 41.3 15.2 8.7 10.9 4.3 46 (100)

Total 15.9 22.8 19.1 16.9 20.4 4.9 509 (100)

Education-standardised

totala,b16.5 22.6 18.7 16.9 20.6 4.7 (100)

Women

18–30 yr 30.5 39.7 12.1 10.6 5.7 1.4 141 (100)

31–40 yr 27.6 36.7 16.3 16.3 3.1 0.0 98 (100)

41–50 yr 26.6 49.2 12.5 6.3 4.7 0.8 128 (100)

51–60 yr 34.2 47.0 6.8 7.7 4.3 0.0 117 (100)

61–70 yr 42.7 43.7 8.7 2.9 1.0 1.0 103 (100)

X71 yr 60.2 36.6 2.2 0.0 1.1 0.0 93 (100)

Total 35.9 42.5 10.0 7.5 3.5 0.6 680 (100)

Education-standardised

totala,b35.8 43.1 10.1 6.9 3.7 0.4 (100)

aStandardised according to the educational-level proportions of the Moscow City population given in the 2002 Census for the

population over 18 years.bw2 statistics showed that there were no significant differences between the standardised and the unstandardised proportions.

T. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674 667

categories are combined, they are then comparablewith those for the other age groups.

The average amount drunk per occasion amongrespondents who reported drinking alcohol (seeTable 3) was 85.0 g of ethanol for men and 32.6 g ofethanol for women. The youngest consumed thehighest amounts per occasion among both sexes.The average amount decreased with age, and theoldest group drank half as much as the youngest, onaverage. Due to the construction of the survey, theproportion of abstainers could not be directlycalculated. However, 15.1% of the respondentsreported not drinking any amount of any of thethree beverages, and of these 96.6% had alsoreported drinking ‘almost never’ while 2.8% (fivepersons) had reported drinking only on specialoccasions. All these could therefore be classified asbeing mainly abstainers.

Men drink more often than women, and on eachoccasion they typically drink 2–3 times as muchas women. As a result, the overall prevalence ofbinge drinking (as defined above) among themale respondents was 29.6%, while among femalerespondents it was considerably less, 5.9%(Table 4). The largest percentage of binge drinkers

was found in the youngest age group, and theirproportion declined with age among both men andwomen. The differences between the sexes and agegroups seemed larger in relation to binge drinkingthan in relation to drinking in general.

The concomitants of binge drinking

Model 1 in Table 5 shows the effects of theindependent variables—controlling for age—on theprobability of binge drinking, while model 2 showsthe mutually adjusted effects. The probability ofbinge drinking was significantly greater for menwith several kinds of economic problems, even whenthe other independent variables were controlled for(OR 1.67; CI 1.06–2.63). As regards education, theodds for binge drinking were almost twice as highamong men with a secondary education and morethan twice as high among women with this level ofeducation, compared to those with a high level ofeducation. Further, being married or cohabitinghad a great and significant protective effect againstbinge drinking among women, with an OR of 0.31(CI 0.15–0.66). Contacts with friends increasedthe odds of binge drinking by a factor of 2.16

ARTICLE IN PRESS

Table 3

Average amount of alcohol consumed per occasion, by sex and

age, among those respondents who reported drinking alcohol in

Moscow 2004 (in grams of pure ethanol)

Age group Men

(n ¼ 463)

Women

(n ¼ 546)

Total

(n ¼ 1009)

18–30 yr 100.3 39.8 68.6

31–40 yr 90.0 38.6 64.6

41–50 yr 94.6 33.3 62.6

51–60 yr 79.0 32.7 52.0

61–70 yr 60.2 21.7 36.9

X71 yr 51.4 20.8 34.6

Total 85.0 32.6 56.6

Education-

standardised totala89.0 32.7 58.4

aStandardised according to the educational-level proportions

of the Moscow City population given in the 2002 Census for the

population over 18 years.

Table 4

Distribution of binge drinking by age and sex in Moscow 2004

(%)

Age group Men (n ¼ 510) Women (n ¼ 680)

X80 g of pure

ethanol at least

once per month

X60 g of pure

ethanol at least

once per month

18–30 yr 43.5 11.3

31–40 yr 34.4 9.2

41–50 yr 33.7 6.3

51–60 yr 20.9 4.3

61–70 yr 12.3 1.9

X71 yr 10.9 0.0

Total 29.6 5.9

Education-

standardised

totala,b

31.6 6.0

aStandardised according to the educational-level proportions

of the Moscow City population given in the 2002 Census for the

population over 18 years.bw2 statistics showed that there were no significant differences

between the standardised and the unstandardised proportions.

T. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674668

(CI 1.02–4.60) among women, while no significanteffect was found among men. Nagelkerke Pseudo-R2

values for the mutually adjusted models for menand women were 0.127 and 0.164, respectively.

In model 3, both sexes were analysed together.Since the effects of economic problems and mar-riage on binge drinking seemed to be different forthe two sexes, the model included two interactionterms. Both of these terms were statisticallysignificant, indicating important gender differences

in the impact of economy and marriage on bingedrinking. Even after taking these interactions intoaccount, the effect of gender was highly significantin its own right, with an OR of 0.31 (CI 0.18–0.56)for binge drinking among women compared to men.Regular contact with friends was, however, nolonger statistically significant. The Pseudo-R2 forthis model was 0.273.

Discussion

The patterns of alcohol consumption

For men, the average amount drunk per occasionwas 85.0 g of pure ethanol, which falls in the rangeof the amounts observed in earlier studies (Bobaket al., 2004; Malyutina et al., 2004). The averageamount for women was 32.6 g, which is almost thesame as observed by Malyutina et al. (2004).

The proportion of binge-drinking men was 29.6%in our study. In international terms this figureseems high—survey studies in Cracow (Poland) andKarvina-Havirov (Czech Republic) show 12% and17%, respectively (Bobak et al., 2004). However, it fitswell with the figure recorded in an earlier study byBobak et al. (1999) across the whole of Russia. Asimilar figure was also obtained in the city ofNovosibirsk by Bobak et al. (2004). Malyutina et al.(2004), using similar measures in Novosibirsk, found aslightly higher proportion of binge-drinking women,9.4%, compared to 5.9% in the present study.

In these respects, our results are in line withearlier studies in Russia. Survey responses cannot,however, serve as estimates of total consumption,since the heaviest drinkers are also those least likelyto either participate in surveys or answer surveyquestions on alcohol truthfully if they do. It istherefore likely that we have underestimated boththe prevalence of binge drinking and the typicalamount consumed per occasion. However, theresponses we obtained can nevertheless assist us inseparating heavy drinkers from more moderatedrinkers and those who do not drink, and thereforebe helpful in establishing the social determinants ofbinge drinking.

Gender differences: drinking as a male response to

hardship

Previous studies have shown that Russian mendrink relatively infrequently, but consume highamounts of alcohol per occasion (Bobak et al.,

ARTICLE IN PRESS

Table 5

The relationship between socio-economic factors, social contacts and binge drinking: an age-standardised logistic regression analysis

Variable Men (80 g or more of pure ethanol) Women (60 g or more of pure ethanol) Men and women

combineda

Model 1b Model 2c Model 1b Model 2c Model 3c

OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI

Economic problems

0–1 problems 1.00 1.00 1.00 1.00 1.00

2–4 problems 1.78 1.14–2.78 * 1.67 1.06–2.63* 0.71 0.36–1.43 0.66 0.32–1.35 1.70 1.08–2.67*

Educational level

High education 1.00 1.00 1.00 1.00 1.00

Secondary education 1.96 1.22–3.16** 1.86 1.15–3.02* 2.03 0.99–4.16 2.15 1.03–4.50* 1.94 1.29–2.90***

Low education 1.39 0.83–2.34 1.43 0.85–2.41 1.70 0.63–4.59 1.87 0.68–5.15 1.49 0.93–2.37

(P for categorical

variable)

(P ¼ 0.019*) (P ¼ 0.034*) (P ¼ 0.140) (P ¼ 0.110) (P ¼ 0.004**)

Marital status

Non-married 1.00 1.00 1.00 1.00 1.00

Married/cohabiting 1.00 0.63–1.57 1.01 0.63–1.60 0.32 0.16–0.66** 0.31 0.15–0.66** 1.01 0.64–1.58

Contact with friends

Little contact 1.00 1.00 1.00 1.00 1.00

Regular contact 1.17 0.77–1.79 1.19 0.77–1.84 2.28 1.09–4.75* 2.16 1.02–4.60* 1.39 0.95–2.01

Gender

Man 1.00

Woman 0.31 0.18–0.56***

Interaction terms

Woman*2–4

problems

0.40 0.17–0.90*

Woman*married 0.34 0.15–0.79*

Sample size (n ¼ 507) (n ¼ 678) (n ¼ 1185)

�2 log likelihood 566.84 262.58 834.23

Pseudo R2 (Nagelkerke) 0.127 0.164 0.273

*pp0.05, **pp0.005, and ***pp0.001.a80 and 60 g of pure ethanol, respectively.bAge-adjusted effects.cMutually adjusted effects.

T. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674 669

1999; Malyutina et al., 2001; Nicholson et al., 2005).These findings were confirmed by the present study,where the mean amount consumed by men peroccasion was already higher than what is normallyconsidered as constituting binge drinking. On theother hand, the present study also showed that thefrequency of drinking among Russian men varied,with the respondents distributed fairly evenlybetween the response categories with the exceptionof daily drinking. A large majority of womenreported drinking almost never or only on specialoccasions, and only a very small proportionreported drinking several times a week or everyday. The mean amount of alcohol consumed peroccasion was also only half of what is regarded as a

level which constitutes binge drinking for women.In fact, Moscow women, on average, probablydrink less than women in London or Prague (Rehmet al., 2007). This finding accords with the largegender differences in drinking and drinking patternsthat were observed in previous studies in Russia(Bobak et al., 1999; Malyutina et al., 2001;Nicholson et al., 2005), and which are a prominentfeature of its drinking culture.

Age patterns: increasing consumption among the

young

The prevalence of binge drinking decreased withincreasing age, as did the average amount consumed

ARTICLE IN PRESST. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674670

per occasion (Tables 3 and 4). The highest propor-tions of binge drinkers, as well as the highestreported amounts consumed per occasion, wereobserved among the youngest (18–30 years). NeitherBobak et al. (1999) nor Carlson and Vagero (1998)found such age patterns in their previous studies onbinge drinking. This might therefore indicate a newdevelopment with increasing consumption amongyoung people in Russia, or at least in its capital city.Although the results may be artefactual due to thefact that young people are more likely to reportdrinking and even exaggerate it, other evidencesuggests that this is not the case. For example, arecent study on alcohol consumption in Russiashowed that the youngest age group had the highestintake per occasion (Pomerleau et al., 2005).Sidorov and Mitychlaev (in: Koposov, Ruchkin,Eisemann, & Sidorov, 2002) also claim that theaverage age for the beginning of hard drinking hasdecreased.

Further, the substantially heavier consumptionamong young women (18–30 years) compared to theolder female age groups might be indicative of anincreasing level of alcohol consumption amongwomen which has been suggested in recent studies(Stickley et al., 2007; Zaigraev, 2004).

Economic strain, gender and binge drinking

Experiencing several kinds of economic problemwas positively related to the risk of binge drinkingamong men. In contrast, women seemed less likelyto binge drink when experiencing manifold econom-ic problems. This difference is indicative of theexistence of a fundamental gender difference inRussian drinking behaviour.

Gender differences in alcohol consumption pat-terns may be strongly affected by socio-culturalfactors (Wilsnack, Vogeltanz, Wilsnack, & Harris,2000). Ashwin and Lytkina (2004) and Kiblitskaya(2000) have described how the effects of thetransition, with increased levels of unemployment,as well as wage reductions, payments in-kindor even the non-payment of wages have had ademoralising effect on Russian men. The idea ofmasculinity in Russia has been very much relatedto men’s status as primary breadwinners. Heavyconsumption of alcohol, a form of traditionalmasculine behaviour that both strengthens mascu-line identity and which provides a temporal releasefrom a stressful life situation, may be one result ofthis status loss (see Ashwin & Lytkina, 2004).

Moller-Leimkuhler (2003) has suggested that menin Eastern Europe, as a result of traditional mascu-line norms, are more prone to respond to stressfulsituations with maladaptive behaviour such asincreased alcohol consumption, while women havea more adaptive stress response. Cockerham,Hinote, and Abbott (2006) found that in Russiapsychological distress in the population did lead toalcohol consumption among men but not amongwomen, even though women reported significantlymore distress. They argued that the social role ofwomen in post-Soviet societies contains expecta-tions of being able to continue with daily life despitepersonal strain, which might discourage drinkingamong women as a response to distress.

Our results, showing a higher prevalence of bingedrinking only among men experiencing economicstress, could be seen to support the above hypoth-esis. However, our study does not investigateindividuals’ behaviour over time, so a reverserelationship, i.e. binge drinking causing economicproblems, cannot be excluded. Nonetheless, theliterature referred to above suggests that ourinterpretation of this relationship is plausible.Moreover, the gender difference in the relationmakes an inverse relationship between the variablesless probable, since it should exist among bothsexes.

Education and binge drinking

Those with lower education tend to consumealcohol in a binge-drinking manner more often thanthose with high education, independently of theireconomic situation. The explanation for this mightlie in the difference in choice of activities and waysof socializing between educational groups. Knowl-edge about the risks related to alcohol consumptionmight also be unevenly distributed.

In the present study however, having a secondary

education was associated with a two-fold increase inthe odds for binge drinking among both men andwomen, compared to high education, while theeffect of having a low education was statisticallyinsignificant. Similar results, where the greatest riskfor binge drinking has been observed among thosewith a secondary or vocational education have beenobtained in previous studies in Russia in the cities ofArkhangelsk (Nilssen et al., 2005) and Taganrog(Carlson & Vagero, 1998).

The specific effect of secondary education isdifficult to interpret, especially as age is controlled

ARTICLE IN PRESST. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674 671

for thus making a cohort effect unlikely. It mightbe that the effects of education are in partindirect, deriving from the composition of thedifferent educational groups in terms of occupa-tional status, and perhaps from other situationalfactors among the respondents. However, inan additional analysis comparing only bingedrinkers and other drinkers, that omitted non-drinkers, those with a low education had asignificantly higher OR for binge drinking (OR1.69; CI 1.05–2.72) than those with a high educationin the model with both men and women (data notshown).

Social relations and binge drinking

It was hypothesised that social relations woulddecrease the probability of binge drinking byproviding alternative resources in situations ofeconomic hardship in terms of instrumental supportand favours. However, and in contrast to Weitzmanand Kawachi (2000) who found that membership instudent associations protected against binge drink-ing in the US, a positive relationship betweenhaving regular contact with friends and bingedrinking was not unexpected. Drinking together isa common form of socializing, especially in so-called ‘‘wet cultures’’ (Skog, 1991). According toSkog (1991), socially integrated individuals in ‘‘wetcultures’’ drink more. Surprisingly, this relationshipwas only found among women. This might beexplained by the fact that heavy drinking is acommon way of male socializing in all kinds ofsocial contexts, while for women it might only beacceptable within groups of friends, where tradi-tional social norms discouraging women’s drinkingmight not exist.

Being married or cohabiting had a strongprotective effect on binge drinking among women.This might be related to the married female’sresponsibilities. Domestic gender roles in Russiahave a highly traditional character where thewomen assume responsibility for the household’sday-to-day functioning (Ashwin & Lytkina, 2004).With this meaningful responsibility, women maychoose not to spend their time or money onbecoming intoxicated. There was no significantassociation between being married or cohabitingand binge drinking among men.

Another possibility is that the relationship stemsfrom the fact that women who are binge drinkershave more difficulties in getting married. Consider-

ing the social stigma attached to women’s drinkingand the fact that there is a major gender imbalancein the Russian population this possibility must beconsidered. This might also explain why there wasno effect of marriage on binge drinking among men.

Limitations

The highly educated (52.9%) were overrepre-sented in the study sample, compared with thegeneral population (39.3% according to the All-Russia Population Census, 2002). Education-stan-dardised figures are given in the descriptive tables inorder to correct for this. However, the differencesbetween the standardised and the unstandardisedfigures are not significant. Moreover, this shouldnot have affected the relationships observed be-tween binge drinking and the independent variables,as education was controlled for in the analysis.

In general, questions on ‘the usual amountconsumed per occasion’ are likely to result in anunderestimation of consumption, and heavy drin-kers are less likely to participate in surveys(Nemtsov, 2003, 2004). Thus, particularly as theresponse rate was rather low (47%), it is plausiblethat the levels of alcohol consumption found in thepresent study are also underestimated. The relation-

ship between binge drinking and the independentvariables might have been further affected if thenon-response was biased against binge drinkers (ornon-binge drinkers) with certain characteristics,however, we have no obvious reasons to believethat this is the case.

We were also unable to consider the use ofsurrogate alcohols, which may have influenced theresults. These are probably consumed by a sig-nificant number of Russian men, most likely thepoorest (Leon et al., 2007), as surrogate alcohols aremuch cheaper than commercial spirits (McKeeet al., 2005). If this is the case, then the associationbetween economic problems and binge drinkingobserved in the present study could be evenstronger.

Moreover, using a cut-off point of 80 g and 60 gof ethanol for men and women, respectively, as wellas putting both abstainers and moderate drinkers inthe same group is disputable. In order to detect anyeffects of these methodological choices some addi-tional analyses were also undertaken. Analyseswhere non-drinkers were omitted from the logisticregression model showed very similar results to themodel with non-drinkers included, except for low

ARTICLE IN PRESST. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674672

education which here became significant in themodel with both men and women (cf. model 3,Table 5). As binge-drinking women were relativelyfew, higher cut-off points were only tried for men.A model with men only where the cut-off point forbinge drinking was set to 100 g showed almostidentical results.

Conclusions

In sum, gender is of primary importance forunderstanding the relations between economicstrain, social relations, and binge drinking incontemporary Russia. Men with economic pro-blems drink heavily; women with economic pro-blems drink less. Married men drink as much astheir non-married friends, married women substan-tially less. Women’s drinking is concentrated in thecontext of friends, while men’s occurs everywhere.

Gender specifies the relationships, implying thatthe determinants of heavy drinking in Russia mustbe interpreted gender-wise. Thus, the oppositeassociations seen between economic strain andbinge drinking among men and women might partlyexplain the extremely large gender gap in Russianmortality statistics.

Finally, the development of drinking patterns asindicated by this study is very worrying as bothwomen and young people of both sexes now seem tobe drinking more than they did previously—whichbodes ill for the future in terms of alcohol-relatedproblems in Russia.

Acknowledgements

The Royal Swedish Academy of Sciences andThe Swedish Foundation for Baltic and EasternEuropean Studies have both contributed financiallyto this research. Thanks to Natalia Rimachevskayaand Ludmila Migranova at ISESP for helpingto set up the study, and to the 30 interviewerswho carried out the interviews. Thanks also toDr. Andrew Stickley and Dr. Per Carlson fortheir valuable comments on an earlier version ofthis paper.

References

All-Russia population Census. (2002). Available at: /http://

www.perepis2002.ru/index.html?id=11S.

Almedom, A. M. (2005). Social capital and mental health: An

interdisciplinary review of primary evidence. Social Science &

Medicine, 61(5), 943–964.

Ashwin, S., & Lytkina, T. (2004). Men in crisis in Russia. The

role of domestic marginalisation. Gender & Society, 18(2),

189–206.

Bobak, M., McKee, M., Rose, R., & Marmot, M. (1999).

Alcohol consumption in a national sample of the Russian

population. Addiction, 94(6), 857–866.

Bobak, M., Room, R., Pikhart, H., Kubinova, R., Malyutina, S.,

Pajak, A., et al. (2004). Contribution of drinking patterns in

rates of alcohol related problems between three urban

populations. Journal of Epidemiology and Community Health,

58(3), 238–242.

Britton, A., & McKee, M. (2000). The relation between alcohol

and cardiovascular disease in Eastern Europe: Explaining the

paradox. Journal of Epidemiology and Community Health,

54(5), 328–332.

Carlson, P. (2001). Risk behaviours and self rated health in

Russia 1998. Journal of Epidemiology and Community Health,

55(11), 806–817.

Carlson, P., & Vagero, D. (1998). The social pattern of

heavy drinking in Russia during transition—Evidence from

Taganrog 1993. European Journal of Public Health, 8(4),

280–285.

Chenet, L., McKee, M., Leon, D., Schkolnikov, V., & Vassin, S.

(1998). Alcohol and cardiovascular mortality in Moscow; new

evidence of a causal association. Journal of Epidemiology and

Community Health, 52(12), 772–774.

Cockerham, W. C., Hinote, B. P., & Abbott, P. (2006).

Psychological distress, gender, and health lifestyles in Belarus,

Kazakhstan, Russia, and Ukraine. Social Science & Medicine,

63(9), 2381–2394.

Ferlander, S. (2007). The importance of different forms of social

capital for health. Acta Sociologica, 50(2), 115–128.

Gustafsson, B., & Nivorozhkina, L. (2004). Changes in Russian

poverty during transition as assessed from microdata from the

city of Taganrog. Economics of Transition, 12(4), 747–776.

International Center for Alcohol Policies. (1997). The limits of

binge drinking. ICAP reports 2. Available at /http://www.

icap.org/portals/0/download/all_pdfs/ICAP_Reports_English/

report2.pdfS.

Kawachi, I., & Berkman, L. (2001). Social ties and mental health.

Journal of Urban Health—Bulletin of the New York Academy

of Medicine, 78(3), 458–467.

Kennedy, B. P., Kawachi, I., & Brainerd, E. (1998). The role of

social capital in the Russian mortality crisis. World Develop-

ment, 26(11), 2029–2043.

Kiblitskaya, M. (2000). ‘‘Once we were kings’’: Male experiences

of loss of status at work in post-communist Russia. In

S. Ashwin (Ed.), Gender, state and society in Soviet and post-

Soviet Russia (pp. 90–104). New York: Routledge.

Koposov, R. A., Ruchkin, V. V., Eisemann, M., & Sidorov, P. I.

(2002). Alcohol use in adolescents from northern Russia: The

role of the social context. Alcohol and Alcoholism, 37(3),

297–303.

Kuntsche, S., Gmel, G., Knibbe, R. A., Kuendig, H., Bloomfield,

K., Kramer, S., et al. (2006). Gender and cultural

differences in the association between family roles,

social stratification, and alcohol use: A European cross-

cultural analysis. Alcohol and Alcoholism, 41(Suppl. 1),

i37–i46.

ARTICLE IN PRESST. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674 673

Leinsalu, M. (2004). Troubled transitions: Social variation and

long-term trends in health and mortality in Estonia. Stockholm:

Almqvist & Wiksell.

Leon, D. A., Saburova, L., Tomkins, S., Andreev, E., Kiryanov,

N., McKee, M., et al. (2007). Hazardous alcohol drinking and

premature mortality in Russia: a population based case-

control study. Lancet, 369(9578), 2001–2009.

Leon, D., & Shkolnikov, V. M. (1998). Social stress and the

Russian mortality crisis. Journal of the American Medical

Association, 279(10), 790–791.

Malyutina, S., Bobak, M., Kurilovitch, S., Nikitin, Y., &

Marmot, M. (2004). Trends in alcohol intake by education

and marital status in an urban population in Russia between

mid 1980s and the mid 1990s. Alcohol and Alcoholism, 39(1),

64–69.

Malyutina, S., Bobak, M., Kurilovitch, S., Ryizova, E., Nikitin,

Y., & Marmot, M. (2001). Alcohol consumption and binge

drinking in Novosibirisk, Russia, 1985–95. Addiction, 96(7),

987–995.

Makinen, I. H. (2000). Eastern European transition and

suicide mortality. Social Science & Medicine, 51(9),

1405–1420.

Makinen, I. H., & Reitan, T. C. (2006). Continuity and change in

Russian alcohol consumption from the tsars to transition.

Social History, 31(2), 160–179.

McKee, M., S +uzcs, S., Sarvary, A., Adany, R., Kiryanov, N.,

Saburova, L., et al. (2005). The composition of surrogate

alcohols consumed in Russia. Alcoholism: Clinical and

Experimental Research, 29(10), 1884–1888.

Moller-Leimkuhler, A. M. (2003). The gender gap in suicide

and premature death or why are men so vulnerable?

European Archives of Psychiatry and Clinical Neuroscience,

253(1), 1–8.

Nemtsov, A. V. (2003). Alcohol consumption level in Russia:

A viewpoint on monitoring health conditions in the Russian

Federation (RLMS). Addiction, 98(3), 369–370.

Nemtsov, A. V. (2004). Alcohol consumption in Russia: Is

monitoring health conditions in the Russian Federation

(RMLS) trustworthy? Addiction, 99(3), 386–387.

Nicholson, A., Bobak, M., Murphy, M., Rose, R., & Marmot,

M. (2005). Alcohol consumption and increased mortality in

Russian men and women: A cohort study based on the

mortality of relatives. Bulletin of the World Health Organiza-

tion, 83(11), 812–819.

Nilssen, O., Averina, M., Brenn, T., Brox, J., Kalinin, A., &

Archipovski, V. (2005). Alcohol consumption and its relation

to risk factors for cardiovascular disease in the north-west

of Russia: The Arkhangelsk study. International Journal of

Epidemiology, 34(4), 781–788.

Pomerleau, J., McKee, M., Rose, R., Haerpfer, C. W., Rotman,

D., & Tumanov, S. (2005). Drinking in the commonwealth of

independent states—Evidence from eight countries. Addiction,

100(11), 1647–1668.

Pridemore, W. A. (2002). Vodka and violence: Alcohol con-

sumption and homicide rates in Russia. American Journal of

Public Health, 92(12), 1921–1930.

Pridemore, W. A. (2004). Weekend effects on binge drinking and

homicide: The social connection between alcohol and violence

in Russia. Addiction, 99(8), 1034–1041.

Rehm, J., Sulkowska, U., Manczuk, M., Boffetta, P., Powles, J.,

Popova, S., et al. (2007). Alcohol accounts for a high

proportion of premature mortality in central and eastern

Europe. International Journal of Epidemiology, 36(2),

458–467.

Room, R. (2005). Commentary: Pattern of drinking and the

Russian heart. International Journal of Epidemiology, 34(4),

788–790.

Rose, R. (1995). Russia as an hour-glass society: A constitution

without citizens. East European Constitutional Review, 4(3),

34–42.

Rose, R., Mishler, W., & Haerpfer, C. (1997). Social capital in

civic and stressful societies. Studies in Comparative Interna-

tional Development, 32(3), 85–111.

Ryan, M. (1995). Russian report: Alcoholism and rising mortality

in the Russian federation. British Medical Journal, 310(6980),

648–650.

Shkolnikov, V. M., Chervyakov, V., Leon, D. A., & McKee,

M. (2004). Russian mortality beyond vital statistics:

Effects of social status and behaviours on deaths

from circulatory disease and external causes—A case–control

study of men aged 20–55 years in Udmurta, 1998–99.

Demographic Research, Special Collection, 2(Article 4),

71–104.

Shkolnikov, V. M., McKee, M., Chervyakov, V. V., & Kyrianov,

N. A. (2002). Is the link between alcohol and cardiovascular

death among young Russian men attributable to misclassifi-

cation of acute alcohol intoxication? Evidence from the city of

Izhevsk. Journal of Epidemiology and Community Health,

56(3), 171–174.

Simpura, J., Levin, B. M., & Mustonen, H. (1997). Russian

drinking in the 1990s: Patterns and trends in international

comparison. In J. Simpura, & B. M. Levin (Eds.), Demystify-

ing Russian drinking. Comparative studies from the 1990s

(pp. 79–107). Helsinki: STAKES.

Skog, O. J. (1991). Implications of the distribution theory for

drinking and alcoholism. In D. J. Pittman, & H. Raskin

White (Eds.), Society, culture and drinking patterns reexa-

mined (pp. 576–596). New Brunswick, NJ: Rutgers Centre of

Alcohol Studies.

Stickley, A., Leinsalu, M., Andreev, E., Razvodovsky, Y.,

Vagero, D., & McKee, M. (2007). Alcohol poisoning in

Russia and the countries in the European part of the former

Soviet Union, 1970–2002. European Journal of Public Health,

17(5), 444–449.

Vagero, D., & Kislitsyna, O. (2005). Self-reported heart

symptoms are strongly linked to past and present

poverty in Russia: Evidence from the 1998 Taganrog inter-

view survey. European Journal of Public Health, 15(4),

418–423.

Vagero, D., Kislitsyna, O., Ferlander, S., Migranova, L.,

Carlson, P., & Rimashevskaya, N. (2007). Moscow Health

Survey 2004—Social surveying under difficult conditions.

Manuscript submitted for publication.

Weitzman, E. R., & Chen, Y.-Y. (2005). Risk modifying effect of

social capital on measures of heavy alcohol consumption,

alcohol abuse, harms, and secondhand effects: National

survey findings. Journal of Epidemiology and Community

Health, 59(4), 303–309.

Weitzman, E. R., & Kawachi, I. (2000). Giving means receiving:

The protective effect of social capital on binge drinking on

college campuses. American Journal of Public Health, 90(12),

1936–1939.

White, S. (1996). Russia goes dry: Alcohol, state and society.

Cambridge: Cambridge University Press.

ARTICLE IN PRESST. Jukkala et al. / Social Science & Medicine 66 (2008) 663–674674

Wilsnack, R. W., Vogeltanz, N. D., Wilsnack, S. C., & Harris,

T. R. (2000). Gender differences in alcohol consumption

and adverse drinking consequences: Cross-cultural patterns.

Addiction, 95(2), 251–265.

Zaigraev, G. (2004). The Russian model of noncommercial alcohol

consumption. In A. Haworth, & R. Simpson (Eds.), Moonshine

markets: Issues in unrecorded alcohol beverage production and

consumption (pp. 32–40). New York: Brunner-Routledge.