Substance Use & Misuse, 49:1619–1632, 2014 C 2014 Informa Healthcare USA, Inc. Copyright  ISSN: 1082-6084 print / 1532-2491 online DOI: 10.3109/10826084.2014.919750

ORIGINAL ARTICLE

Alcohol Consumption, Alcohol Consumption – Related Harm and Alcohol Control Policy in Austria: Do They Link Up? Irmgard Eisenbach-Stangl1 and Allaman Allamani2 1

European Centre for Social Welfare Policy and Research, Vienna, Austria; 2 Regional Health Agency of Tuscany, Florence, Italy Møller & Galea, 2012; Babor et al, 2003). This pattern is not documented for South European countries such as Italy, where a continuous decrease in alcohol consumption since 1960s–1970s took place long before any relevant policy measure was adopted. Austria represents a mixed picture; increasing consumption during the 1960s, leveling off during the early 1970s and a national alcohol control policy becoming stricter especially after the consumption peak. The interplay of alcohol consumption and alcohol control policy measures varying from one country, or even region, to the other raises the question about which factors other than, or in addition to, “planned” alcohol control policy measures are associated with changes in alcohol consumption and alcohol consumption- related harms. Previous studies focused on “unplanned” factors1 such as urbanization, women’s emancipation and economic affluence (Allamani & Beccaria, 2007; Simpura, 2001; Sulkunen, 1989). These studies neither explored factor combinations nor did they investigate their interplay with “planned” alcohol consumption control measures. The current study, part of a comparative European AMPHORA study, was designed to fill the gap and to identify the main “unplanned” and “planned” change factors for alcohol consumption and selected alcohol consumptionrelated harms in Austria between 1960 and 2010. It focuses on two questions:

The study identifies changes in selected (“unplanned”) socio-demographic and economic factors as well as in (planned) political measures that are most strongly correlated with changes in alcohol consumption and alcohol consumption-related harm between 1961 and 2006 in Austria. During the period of investigation consumption increased until the early 1970s, dropped during the next decade and have leveled off since. Increasing urbanization, female employment and average age of mothers at their child births are associated with the best time series model for the interpretation of consumption changes. The results regarding alcohol control policies and their impact on consumption were paradoxical. Study limitations were noted pointing up the necessity to improve indicators and concepts. Keywords alcohol consumption, alcohol consumption-related harms, alcohol control policies

INTRODUCTION

Changes in alcohol consumption and alcohol consumption-related harm usually are correlated to alcohol control policy measures restricting availability of alcoholic beverages or drinking in selected contexts as, for example, when driving, by the introduction of minimum alcohol blood levels. The impact of such measures on alcohol consumption and alcohol consumption-related harms is documented for North European and North American countries, although consumption in these regions has increased since the 1960s (e.g., Anderson,

1. which (“unplanned”) socio-economic and demographic changes were most strongly associated with changes in alcohol consumption and types of harms between 1961 and 2006; and

The contents of the article are solely the responsibility of the authors and do not necessarily represent the official views of the European Commission. 1 The definition of “planned” vs. “unplanned” may be regarded as being artificial and may, as such, be misleading and induce some confusion. In fact the AMPHORA study denotes “planned”(i.e., alcohol policy measures) for measures which are taken with a public health aim, and “unplanned” for all the other determinants which are supposedly able to modify alcohol consumption. However, there are some actions planned, for example, by the alcohol industry, as well as by the government, that are, or may be, able to affect the consumption of alcoholic beverages, even if they do not have a designated public health purpose. Address correspondence to Irmgard Eisenbach-Stangl, European Centre for Social Welfare Policy and Research, Vienna, Austria; E-mail: [email protected]

1619

1620

I. EISENBACH-STANGL AND A. ALLAMANI

2. considering the impact of the identified “unplanned” factors, which alcohol control policy measures were associated with the strongest effects on alcohol consumption and alcohol consumption-related harms during the period of investigation. AUSTRIA 1960–2010: MAJOR SOCIAL CHANGES AND CHANGES IN ALCOHOL CONSUMPTION AND ALCOHOL CONSUMPTION-RELATED HARM Major Social Changes

Austria is a federally organized parliamentarian republic in Middle Europe covering an area of 84,000 square kilometers, and bordering eight other European countries. It had 8.38 million inhabitants in 2010. Its population has been increasing, in part, since the 1960s, due to immigration when so called “guest workers” were recruited mainly in former Yugoslavia and Turkey. Many of them became Austrian citizens (Statistik Austria, 2012). Austria is member of the European Union since 1995 and its cross national product amounted to 301 billion Euros—35.000 Euros per capita—in 2011. It is one of Europe’s countries with the lowest unemployment rates, but it also is a country with high state debts (Statistik Austria, 2012; Eurostat, 2012). The federal republic consists of nine states, governed by state parliaments, and organized into administrative districts which do not have any legal powers. Two parties have dominated Austria’s political arena during the last hundred years: the Christian conservative party ¨ (“Osterreichische Volkspartei”) and the Social Democrats ¨ (“Sozialisitsche Partei Osterrreichs”). Life expectancy in Austria increased by 8 years since 1980 (78 years for men and 83 years for women in 2010). Main death causes are diseases of circulation and cancer. In 2010 almost 70% of Austrians evaluated their health status as being “very good” or “good” (Hofmarcher, 2012). This optimistic estimate corresponds to experts estimates on “Human Development” on behalf of UN measuring standards of living and quality of life [Human Development Index (HDI), 2013]: in 2012 and 2013 Austria occupied rank 18 among 187 countries, and it showed a tendency to improve its position (http://hdr.undp.org/en/statistics/hdi) The period of observation was not a time of dramatic major changes but a time of more or less visible continuous innovations modernizing policy, economy, society, culture, and every day life. Modernization was promoted by the access of Austria to EU as well as expressed by it. During the 1990s political traditions, organizations and leadership eroded due to decreasing loyalties of members and voters and the political system changed from a two party to a multi party system. As ties between political parties, partly leaders and supporters weakened they weakened within families between husbands and wives and parents and children. As in the public political system male supremacy was dismantled in the private family system and the law contributed considerably to this attenuation of traditions (Giddens, 1996). The “Great

Family Reform” enforced during the 1970s provided wife and husband with the same rights and obligations and assimilated their position in legal and economic aspects. It also emphasized the concertedly shaping of life and abolished the fatherly power to decide on all aspects of the life of minors and to represent them in the public. Thus, individual freedom of choice was enhanced at the cost of traditional dependencies, a social change in social theory referred to as “individualization” (Beck, 1986). Increased freedom of choice—which can switch into compulsion to choose—had an impact on consumption: “demand” became diversified and burdened with the expression of identity (Sulkunen, 2002). But diversified demand was simultaneously counteracted by a (globalized) “supply” providing with new products as well as with new meanings and morals of consumption (Beck, 1996). This ongoing multidimensional, dynamic, nonlinear, and levelnuanced process is the context in which the changes of alcohol production, consumption and controls described in the following chapters are to be understood. Changes in Alcohol Consumption, Drinking Pattern and Alcohol Consumption-Related Harms

Figure 1 shows that alcoholic beverage consumption in Austria increased from 1961 (10.95 l pro capita) to 1973 (16.04), when a decrease started, reaching 12.27 l in 1982. Since then, Figures were rather stable (12.20 in 2009). However, between 1961 and 2009 there was an increase in total alcoholic beverage consumption of 22.5%. The three types of alcoholic beverages consumed in Austria have maintained their relative proportions for most of the time since the 1980s. Austria is primarily a beer-drinking country. In 2009, beer accounted for 50%, wine contributed more than 30% and spirits with 15% of consumed alcoholic beverages. Attitudes toward drinking and intoxication have been traditionally tolerant as articles of prominent stakeholders of alcohol policy (mainly party politicians, lawyers, and medical doctors) and few representative studies show2 : The individual right to get drunk in private as well as in public settings was respected for decades, if not centuries, and the impact of the temperance movement with the exception of the first three decades of the 20th century was weak (Eisenbach-Stangl, 1991, 2014). Tolerance was accompanied by social pressures to drink, which authorities tried to reduce for centuries—in vain (Lehner, 1969). Tolerance was openly discussed, but not specified that only male drinker were addressed. With women tolerance was far less developed as few surveys show (Eisenbach-Stangl & Hager, 2004; Mader, 1981). During the period of observation tolerance toward drinking and intoxication seems to seem have somewhat declined: Attitudes seem to have become stricter in 2

For influental lawyers Hans Hoegel (1887), for influental medical doctors Hans Hoff (1954), for influental party politicians Otto Bauer (Bundesvorstand 1970). The most important representative studies dealing, among other things, with attitudes are Mader et al. 1981, and Uhl et al. 1996.

1621

RELATED HARM AND ALCOHOL CONTROL POLICY IN AUSTRIA 18,00

16,00

14,00

liters per capita per year

12,00

10,00

8,00

6,00

4,00

2,00

wine

beer

spirits

other

2009

2008

2007

2005

2006

2004

2002

2003

2001

2000

1998

1999

1997

1996

1994

1995

1993

1992

1991

1989

1990

1988

1987

1986

1984

1985

1983

1981

1982

1980

1978

1979

1976

1977

1975

1973

1974

1972

1970

1971

1969

1968

1966

1967

1965

1964

1963

1962

1961

0,00

Tot

FIGURE 1. Alcohol recorded consumption in Austria, in liters of pure alcohol per inhabitant 15 years and older, 1961–2009—source: WHO GISAH, 2011 (∗ ). Other = all wines that are not grape wine, all beers not made from malt, cocktails, mixed drinks, fortified wines, cider, and the like.

regard to “heavier intoxication” and to be more in favor of levels of “lighter” consumption (Eisenbach-Stangl, 2005). During the period of observation alcohol consumptionrelated behavior also slowly and cautiously became somewhat more “problematized” by the state/ by state interventions, which got even more numerous after consumption stagnation. Normal planned intervention always aimed at selected behavior and never targeted alcohol beverage consumption as such. For example, (1) in 1961 alcoholic beverage consumption and driving was regulated by the introduction of a Blood Alcohol Content (BAC) limit of eight promille, (2) respective regulations and sanctions were made stricter and increased during the following decades; (3) treatment for alcoholism3 has been contin3

Treatment can be briefly and usefully defined as a unique, planned, goal directed, temporally structured, multidimensional change process, of necessary quality, appropriateness, and conditions (endogenous and exogenous), which is bounded (culture, place, time, etc.) and can be categorized into professional-based, tradition-based, mutual-help based (AA,NA, etc.) and self-help (“natural recovery”) models. There are no unique models or techniques used with substance users- of whatever types and heterogeneities, which aren’t also used with nonsubstance users. Whether or not a treatment technique is indicated or contraindicated, and its selection underpinnings (theory-based, empiricallybased, “principle of faith-based, tradition-based, etc. continues to be a generic and key treatment issue. In the West, with the relatively new ideology of “harm reduction” and the even newer Quality of Life (QOL) and wellbeing treatment-driven models there are now new sets of goals in addition to those derived from/associated with the older tradition of abstinence driven models. Conflict-resolution models may stimulate an additional option for intervention. Each ideological model has its own criteria for success as well as failure as well as iatrogenic-related harms. Treatment is implemented in a range of environments; ambulatory as well as within institutions which can include controlled environments Treatment includes a spectrum of clinician-caregiver-patient relationships representing various forms of decision-making traditions/models; (1). the hierarchical model in which the clinician-treatment agent makes the decision(s) and the recipient is compliant and relatively passive,

uously expanded and diversified since the beginnings of the 1970s; (4) the regulations about minimum drinking age were continuously increased and were also intervened with in the private spheres. Most state interventions taken during the period of investigation followed the levels of decrease and leveling off of alcohol consumption. And as in earlier decades alcoholic beverage production and retail sales were exclusively targeted for fiscal and economic reasons with one exception; advertising was restricted in the early 1970s. Thus, though state interventions regarding alcoholic beverage consumption have enfolded during the study’s period of investigation, their selectivity indicates the extent to which alcoholic beverage consumption is still culturally well-integrated (Eisenbach-Stangl, 1991, 2011). The decrease of alcohol consumption after the 1970s and the slight change of individual attitudes and of “problem concepts” of the state during the period of investigation can plausibly be associated with the continuous flow of immigrants from countries with lower alcoholic beverage consumption since the 1960s (Bloomfield et al., 2005). An additional relevant consideration is the dynamic globalization expanding (substance) markets to drugs other than, and in addition to, alcohol which has diversified the availability and accessibility of the range of the mind altering and mood manipulating psychoactive substances to consumers. And in Austria also urbanization and the growth of the (international) service sector at the expense of the agrarian and industrial sector should be associated with the decline of alcohol beverage consumption and an increase of stricter attitudes—contrary, for example, to Nordic countries (Sulkunen, 1980). Austria (2) shared decision-making which facilitates the collaboration between clinician and patient(s) in which both are active, and (3) the ‘informed model’ in which the patient makes the decision(s). Editor’s note

1622

I. EISENBACH-STANGL AND A. ALLAMANI

45,0

40,0

Rate per 100,000 pop.

35,0

30,0

25,0

20,0

15,0

10,0

5,0

m

2009

2008

2007

2006

2005

2004

2003

2002

2001

1999

2000

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

1987

1986

1985

1984

1983

1982

1981

1980

0,0

f

FIGURE 2. Liver disease mortality rate per 100,000 population, AUSTRIA 1980–2009. Source: WHO mortality data (2011).

traditionally is a producer and self supporter country in regard to all alcoholic beverages and the change of economy and living conditions addressed by both factors diminishes availability of rural products in every respect and reduces their ordinary character. A final factor associated with a consumption reduction are the complex social and cultural processes of individualization and the modernity-associated changes of gender relationships as one of its main features. Since heavy alcoholic beverage consumption and intoxication always have been a privilege of men and a proof of masculinity the weakening of paternalistic structures must have contributed to consumption decrease—a development that can also be observed with tobacco consumption (Eisenbach-Stangl, 2005). The decrease of alcoholic beverage consumption related to this factor is probably softened by the increase of life expectancy of men. Though men older than 50 years do not drink as much as those who are younger, they drink more than male youth and young adults (Eisenbach-Stangl, Bernardis, Fell¨ocker, Haberhauer-Stiedl, & Schmied, 2008). Alcoholic beverage consumption decreased and then leveled off after the early 1970s as did alcoholic beverage consumption—related harms such as liver disease mortality and transport accidents mortality during the period of investigation (Figures 2 and 3). Liver disease mortality (liver disease and liver cirrhosis) mortality indicates the extent of high and continuous alcohol consumption, whereas the number of persons killed or harmed in transport accidents indicates single intoxication events (Rehm & Scafato, 2011). Figure 2 shows a remarkable drop in mortality due to liver disease and cirrhosis in Austria, from about 40 deaths per 100,000 male inhabitants during the first half of the 1980s to less than 15 deaths during 2008–2009. A comparable decrease has occurred for females.

From 2002 to 2003 figures are dramatically decreasing, suggesting that this would be an artifact due to change in classification criteria. Such a remarkable drop is not present in Health For All (HFA) data base for liver deaths data for both genders and all ages, which gradually decreased in 1999–2000 and in the 2000s (World Health Organization, 2010). Figure 3 shows a drop in mortality due to alcohol consumption-related transport accidents in Austria, from about 50 deaths per 100,000 male inhabitants during the first half of 1980s to less than 15 deaths in 2008–2009; females manifesting a comparable decrease. DATA AND ANALYSIS Alcoholic Beverage Consumption and Alcoholic Beverage Consumption -Related Harms

Data on per capita recorded total alcoholic beverage consumption in those aged 15 years and older were provided through 2009 GISAH (Global Information System on Alcohol and Health) dataset (World Health Organization, 2009). Data on liver mortality and transport mortality were provided by WHO Health For All (HFA) Database: Austrian mortality standardized death rates (SDR) for both genders and all ages per 100,000 inhabitants were used (see World Health Organization, 2010). For Austria, there were complete liver mortality series for 1970–2010, and complete transport mortality series for 1980–2010 (see AMPHORA Work package 3, 2013: First Part, The Study, 2.4.2). Since the public health approach considers that restrictive alcohol control policies, which reduce alcohol consumption, also are able to produce a reduction in harm, the indirect effect of policy measures on harm, that is, the effects on harm mediated by consumption, was considered. In addition, the direct effect of policy measures on

1623

RELATED HARM AND ALCOHOL CONTROL POLICY IN AUSTRIA 60,0

50,0

Rate per 100,000 pop.

40,0

30,0

20,0

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

1987

1986

1985

1984

1983

1982

1981

0,0

1980

10,0

m f

FIGURE 3. Transport accidents mortality rate (100,000 population) per gender & age, AUSTRIA (1980–2009). Source: WHO mortality data, 2009 (∗ ). Data were provided through World Health Organization Mortality Database, ICD-9 B47 (Transport accidents).

harm was studied, that is, the effect of policies independently from the consumption changes (see AMPHORA Work package 3, 2013: First Part, The Study, 2.4.2). Alcohol Control Policy Measures Six main alcohol control policy measures taken by Austria, between 1961 and 2006, were selected to analyze the effect of alcohol control policies as part of the EU AM PHORA comparative study. Two of them are restrictive traffic regulations. Austria was among the first European countries to decide on a blood alcohol concentration limit in 1961 to raise traffic security which had gotten increasingly dense in the 1950s. The BAC-related sanctions were severe and they became stricter during the next decades. Austria lowered the limit for blood alcohol concentration to 0.5 in 1998; earlier than other European countries. Two additional measures target youth; both are to be classified as targeted prevention, but are based on different methods. The (repressive) minimum drinking age regulations were formulated in 1922 on a federal level; they were complemented by state laws after WW II, which were continuously made stricter during the study’s period of observation (Eisenbach-Stangl 1991; Uhl et al., 2009) Primary prevention was established by law on a national level in 1980 and was subsequently implemented by nine Austrian state institutes, whose programs were developed in accordance with known and posited protective factors4 associated with health promotion. The treatment of alcoholism, classified as a tertiary preventive measure, is also to be considered to have addi4

The reader is reminded that the concepts of “risk factors”, as well as “protective factors”, are often noted in the literature, without adequately noting their dimensions (linear, nonlinear; rates of development and decay; anchoring or integration, cessation, etc.)), their “demands”, the crit-

tional effects because it “spreads the message” about the risks of types of hazardous or problem drinking. Voluntary treatment for alcoholics was at first established in Austria’s mental asylums under welfare administration between the world wars, but since 1961 was offered in special psychiatric clinics and has been expanded since then. Alcoholic beverage advertising regulations, a relatively recent measure, were established in 1974. They are unique because the Austrian state never intervened in any part of the alcohol economy—production and sales—until this legislation. They are modestly repressive and complemented by voluntary self-regulations (Eisenbach-Stangl, 1991; Uhl et al, 2009). The time-line of the main alcoholic beverage control policies are: 1. 1960 Blood Alcohol Concentration 0.8 for drivers 2. 1961 establishment of voluntary treatment of alcoholics 3. advertising law of 1974 4. 1966 minimum age regulations on federal level of 1922 taken over (additionally) and extended and strengthened by state law after WW2 ical necessary conditions (endogenously as well as exogenously; from a micro to a meso to a macro level) which are necessary for either of them to operate (begin, continue, become anchored and integrate, change as de facto realities change, cease, etc.) or not to and whether their underpinnings are theory-driven, empirically-based, individual and/or systemic stake holder- bound, based upon “principles of faith doctrinaire positions,” personal truths,” historical observation, precedents, and traditions that accumulate over time, conventional wisdom, perceptual and judgmental constraints, “transient public opinion.” or what. This is necessary to consider and to clarify if these term are not to remain as yet additional shibboleth in a field of many stereotypes, tradition-driven activities, “principles of faith” and stakeholder objectives. Editor’s note.

1624

I. EISENBACH-STANGL AND A. ALLAMANI

5. 1980 establishment of primary prevention in amendment of the drug law (Suchtgiftgesetz 1980), BGbl Nr. 31) 6. 1998 Blood Alcohol Concentration 0.5 for drivers Selected Socio-Demographic and Economic (Unplanned) Variables To increase comparability between the European countries participating in the AMPHORA study, the unplanned factors that are included concern indicators of: (a) economic changes: income; prices of the most consumed alcoholic beverages (in Austria beer and wine), (b) socio-demographic changes (proportion of older people, proportion of people living in larger cities) and (c) cultural changes, more specifically indicators of emancipation (e.g., proportion of working women; average age of mothers at their child’s birth). For Austria, it can be noted that over the whole period of this study the income (transformed into purchasing power parity), as well as the level of female employment and of female education, have generally increased, as has the proportion of older people in the population. The urban level has also increased, but with a standstill during the decade 1990–2000. The price of beer (transformed into EUROSTAT price index numbers through the study period) decreased (for more information see AMPHORA, WP3, Part 2: Austria). Other non-planned socio economic factors might also reasonably explain changes in alcoholic beverage consumption, but the limited data availability of these variables over the period under investigation did not permit them to be taken into account for the quantitative analysis. For example, a tendency toward a more nuclear society with an increase in both single household and in single parent household has been observed. An eight time increase of people declaring no traditional religion identification—from 444 p. 100,000 (1961) to 3,200 p. 100,000 (2001) have been recorded. The number of on premises selling alcoholic beverages—rate per 100,000 population—also increased from 448 in 1995 to 559 in 2007. On the other hand, an indicator of Austrian health behavior such as smoking, showed a notable increase of smokers from 27,700 in 1972 to 29,300 in 1997 (per 100,000). This trend was opposite to the trends occurring in other European countries. Analysis The time series model that was considered to be appropriate for this study’s data analysis was obtained from a linear regression with a time trend among the explanatory variables (see AMPHORA WP3, 2013; Baccini & Carreras, 2014). A core model was specified for the pro capita total alcoholic beverage consumption (that was transformed in its logarithm), for the unplanned variables. This accounted for (a) Income, and (b) Price of beer (the most consumed type of beverage in Austria over the study period). This

approach rests on the relative importance of economic factors in determining changes in alcohol consumption. The core model also took into account (c) Males over 65 in the population (percentage of elderly), so that the demographic age structure of the population was considered. All of these variables were included after a logarithmic transformation. A time trend was also added, to capture the long-term consumption behavior that could be related to unobservable factors. Three additional socially relevant variables selected for the analysis (i.e., Female employment, Mothers’ age at all childbirths, Urban level), were then inserted in the core model one at time, and were also logarithmically transformed (Baccinin & Carreras, 2014). In order to avoid the effects of a possible high correlation between the variables, resulting in unstable results if they were included in the model at the same time, the quality of three models, each including one of the three socially relevant variables, was measured by means of the AIC (Akaike Information Criterion5 , Baccini & Carreras, 2014). The Confidence Interval was set at 0.90 instead of 0.95, which is the usual value for CI, in order to capture even minor information from the data. Moreover, since the effects of the unplanned variables on alcoholic beverage consumption and harm were expected to be delayed in time, the mean value of each variable’s current year value and those in the two preceding years, was taken into account. In summary, this analysis allows conclusions about which of the unplanned factors is most strongly associated with changes in consumption, after controlling for economic factors (income and price) and age structure of the population. For each main control policy measure a model was specified (by adding a dummy variable which was equal to 0 before the policy introduction and 1 after) in order to estimate the net policy measure effect over the period between the introduction of the policy measure and after, up to the end of the study period (see Baccini & Carreras, 2014). In this study, each control policy was arbitrarily considered to be equal in effect with each other, and the impact of each policy was considered as being immediate and constant over the whole study period. This analysis allows conclusions about the extent to which alcohol control policy measures, after controlling for selected sociodemographic and economic factors, are related to changes in per capita consumption. The study’s data were also analyzed using a new methodology based on an Artificial Neural Network (ANN) architecture, that is, the Auto Contractive Maps (AutoCMs). It allows for basic improvements in both robustness of use in poorly specified and/or computationally demanding problems, and output usability and intelligibility (AMPHORA, 2013, Part 1; Buscema, Maurelli, & Massini, 2014). AutoCMs spatialize the correlation among variables by constructing a suitable embedding space where a visually transparent concept and 5

Akaike Information Criterion (AIC) is a measure of the relative quality of a statistical model for a given set of data. The model that generates the minimum AIC is the best model for the interpretation of results.

1625

RELATED HARM AND ALCOHOL CONTROL POLICY IN AUSTRIA

dimension such as closeness among variables accurately reflects their associations. Such representation is constructed by building a complex global picture of the whole pattern of variation. The connections between any couple of variables, that also contemporarily take into account all of the other connections between all the variables, have a connection weight, which can be transformed into values between 0 (minimum) and 1 (maximum), thus indicating the strength of connection. To maximize the information contained in the matrix of relationships between variables obtained from AutoCM, the graphs named Minimum Spanning Tree (MST), and Maximally Regular Graph (MRG) were calculated, where the highest-value relationships, or connections (lines in the graph) between variables ( = nodes in the graph) are expressed. In addition to graphs, a connecting value table is also used describing all the interactions among the different types of variables. Increase and decrease trends in alcoholic beverage consumption and in alcoholic beverage consumption-related deaths were treated in terms of Min/Max, independently from the actual direction of the time trend of consumption or harm in the country. For example, the variable “beerMin” is a complement of “beerMax”. While variable “beerMax” has much higher values—the higher value is the amount of beer consumed per year over time—variable “beerMin” is exactly the opposite; its values are higher if the beer consumption is less over time. The two variables are inversely related. The analyses which were provided using the ANN approach for Austria, were the following: (a) Selected socio-demographic and economic variables (Female employment, Mothers’ age at all childbirths, Urban level; Income, Price of beer; Males over 65), and main policy measures, (the five mail policy measures between 1961 and 1998) as well as consumption of beer, wine, and spirits (GISAH data, 2009, in World Health Organization, 2011) in order to check the interaction among the two sets of independent variables; (b) Selected socio demographic and economic variables, and policy measures, as well as deaths from liver disease & transport accidents (HFA database, World Health Organization, 2010). RESULTS Unplanned Variables and Alcoholic Beverage Consumption

As shown in Table 1, in Austria the model with “urbanization” is the best model for the interpretation of results. This analysis suggests that in Austria during the study’s whole period, 1961–2006, the increase of urbanization—that occurred over the study period, but with a steep peak in the 2000s—would have affected 23.6% in the decrease of total alcoholic beverage consumption which that since the early 1970s was at first decreasing and then leveled off and remained rather stable. The other two models (female employment and older

TABLE 1. Regression coefficients describing the relationship between 15+ per capita recorded alcohol consumption (source: WHO 2009) and three selected socio-demographic indicators in Austria∗ , with 90% confidence intervals (CI 90), and AIC (Akaike information criterion) values∗∗ Socio-demographic variable 1- Urbanization 2- Female employment 3- Mother’s age, all childbirths

Coefficient 23,6 0,53 1,73

CI90

AIC

(19.23, 27.98) −149.08 (0.06, 1) −126,13 (0.37, 3.08) −125.86

Results from four separate regression models for each sociodemographic factor, adjusted for time trend, income, proportion of males >65 of age, and prices of beer and wine ∗∗ lower AIC values indicate better quality model which explains consumption changes.

mother’s age at all childbirths) positively correlate with a small decrease of alcoholic beverage consumption—by 0.5% and 1.7% respectively. Control Policy Measures In order to study the effects of planned intervention indicators, the effect of each of the six control policies were estimated, one at a time, in a model adjusted for the unplanned indicator that provided the minimum AIC in Austria, that is, urbanization. Table 2 summarizes the results of the models for single control policies. Since GISAH total alcoholic beverage consumption data in Austria are available only from 1961, the effects of the first two policies introduced in 1960 and in 1961 was not studied. The coefficient is the logarithm of the differences between consumptions after and before each policy measure. Larger numbers indicate larger effects. Each coefficient estimated for the policy effect should be interpreted together with its confidence interval. The latter measures the variability of the estimates. If the confidence interval contains the value 0, that is, it is between a lower negative value and an upper positive value, then the absence of any effect of the policy is plausible. If, however, the confidence interval is placed above, or below zero, the policy measure has either a positive or negative effect. TABLE 2. Correlation coefficients of 15+ per capita recorded alcohol consumption (source: WHO 2009) and selected policy measures in Austria∗ , with 90% confidence intervals (CI 90)∗ Austria policy 1—1960 BAC at 0.8 policy 2—1961 alcoholism treatment policy 3—1966 minimum age policy 4—1974 ads law policy 5—1980 prevention policy 6—1998 BAC at 0.5

Coefficient

IC 90

NA∗∗ NA∗∗ 0,19 0,10 −0,13 −0,17

(NA, NA) (NA, NA) (0.15, 0.23) (0.05, 0.16) (−0.18, −0.07) (−0.21, −0.14)

effects adjusted for lower AIC unplanned indicator (urbanization) NA = not analysed due to time limitations.

1626

I. EISENBACH-STANGL AND A. ALLAMANI

FIGURE 4. Upper graph: total alcohol consumption. The vertical dotted lines indicate the occurrence of policy measures. Lower graph: estimated policy effect with 90% confidence interval (Austria).

The results of the analysis on policy measures (see Figure 4) show that: the establishment of primary prevention in 1980 (Policy 5), and the establishing of a 0.5 BAC level while driving in 1998 (Policy 6) were associated with decreasing alcoholic beverage consumption over time, by about 10% and 20%, respectively. On the other hand, according to the analysis, the 1966 minimum age regulations (Policy 3), and the advertising law of 1974 (Policy 4) appear to be associated with increased consumption, by 20% and 10% respectively. The ANN analysis qualifies some of the outcomes presented above. In the graph expressed in Figure 5, as well as in Table 3, the two effects of unplanned variables and policy measures are analyzed simultaneously.

The 1998 BAC limits policy measure shows a relatively weak connection (0.90) with the decrease of spirits, while the other four measures in 1961, 1966, 1974, and 1980 have a similar pattern of very strong connection (0.98–0.99) with the consumption decrease in spirits, a strong connection (0.97–0.98) with the increase in beer consumption, and to a lesser extent (0.95–0.97) with the increase in wine consumption. Among the unplanned factors, income is well connected (0.96–0.97) with the decrease in spirits consumption, and with the increase in consumption of beer and wine; while both urbanization and women’s employment are rather well connected with the decreased drinking of spirits (0.96) and the increase of beer drinking (0.95), with a weaker connection in the increase of wine drinking (0.91–0.93). Older mothers’ age

TABLE 3. Consumption of beer, wine, spirits and policy measures and unplanned variables in Austria (1960–2006) – ANN analysis, connection values

Wine Max-increase Beer max-increase Spirits Min-decrease

Mothers Employ 1961 Price Price age at ment Nationw. M65 Urbanisation beer wine Income childbirth Female treatment

1974 Ads law

1966 Minimum age

1998 BAC (0.5)

1980 Nationw. prevention

0.91

0.93

0.83

0.93

0.95

0.85

0.91

0.97

0.96

0.97

0.73

0.95

0.92 0.93

0.95 0.96

0.8 0.75

0.92 0.9

0.97 0.97

0.91 0.93

0.95 0.96

0.98 0.98

0.98 0.99

0.98 0.98

0.83 0.9

0.97 0.98

The figures in the table are weight values, which represent the degree of non linear association between two variables. Every weight of the matrix of association is scaled between 0 and 1.

RELATED HARM AND ALCOHOL CONTROL POLICY IN AUSTRIA

1627

FIGURE 5. Consumption of beer, wine, spirits and policy measures and unplanned variables in Austria (1960–2006)- Austria (1960–2006). Figures in the graph are weight values, which represent the degree of nonlinear association between two variables. Every weight of the matrix of association is scaled between 0 and 1.

at childbirth has weaker connections with spirit decreased consumption and beer increased drinking. Consumption—Related Harms In order to study the effects of the six selected policy measures, the effect of each policy was estimated individually in a model adjusted for the best unplanned indicator in Austria (urbanization). Figure 6 illustrates an association between a direct effect (not mediated by consumption) between the introduction of 1998 BAC 0.5 measure and a reduction of liver disease and cirrhosis deaths in Austria. Figure 7 indicates

that no correlation exists between the introduction of the policy measures studied and the modifications in transport accident-related mortality, but an indirect effect (mediated by the total alcoholic beverage consumption changes) between the1980 establishment of primary prevention and the decrease of transport accident-related mortality. The ANN analysis qualifies some of the outcomes presented above. See both Figure 8 and Table 4. The timespan of the analysis of the two selected alcoholic beverage consumption-related was 1980–2006. After 1980, only one main control policy measure (1998 BAC law) was introduced in Austria, which was analyzed.

1628

I. EISENBACH-STANGL AND A. ALLAMANI

FIGURE 6. Regression coefficients of direct and indirect liver deaths∗ and selected policy measures and their year of introduction in Austria, with 90% confidence intervals (continuous line = direct effect; dotted lines = indirect effects). ∗ direct = policy effect on harm; indirect = policy effect on harm mediated through consumption effect ∗∗ effects adjusted for lower AIC unplanned indicator (urbanization).

In Austria, the decrease in liver mortality is well linked with the 1998 BAC law (connecting values = 0.97), as well as with increased income, older age of mothers at their childbirths, and female employment. The decrease of transport accident-related mortality is even more strongly

FIGURE 7. Regression coefficients of direct and indirect transport deaths∗ and selected policy measures and their year of introduction in Austria, with 90% confidence intervals (continuous line = direct effect; dotted lines = indirect effects). ∗ direct = policy effect on harm; indirect = policy effect on harm mediated through consumption effect ∗∗ effects adjusted for lower AIC unplanned indicator (urbanization).

linked (connecting values = 0.97–0.98) to the same unplanned factors, as well as to the BAC law. Liver deaths, and especially transport deaths, are well linked with the decrease of consumption of spirits.

FIGURE 8. Chronic liver disease and transport accidents mortality and policy measures, unplanned variables, consumption of beer, wine, spirits in Austria (1980–2006)—ANN analysis. Figures in the graph are weight values, which represent the degree of non linear association between two variables. Every weight of the matrix of association is scaled between 0 and 1.

1629

RELATED HARM AND ALCOHOL CONTROL POLICY IN AUSTRIA

DISCUSSION

the introduction of the 1998 BAC measures is also associated with the decrease of liver death. These rather paradoxical results highlight the limits and shortcomings of an analysis focusing on aggregate comparative data: Due to a lack of comparative data the effects of the introduction of a 0.8 BAC level in 1961 during a period of continuous increasing alcoholic beverage consumption were not investigated. An analysis and comparison with the effects of the reduction of the BAC level in 1998 might have revealed less or contrary effects on alcoholic beverage consumption and on alcohol consumption-related harm. It might have also solved the enigma of a policy measure, which aims at single intoxication events but has an impact on effects of high and continuous alcohol consumption. The analysis of the effects of primary prevention and of minimum age regulations fell short of the long, complex, conflicting, nonlinear, and not (never) terminated development policy measures might undergo after their first enforcement on national level—the method chosen required precise time information. Thus, primary prevention, as a goal, was indeed established in 1980—but only legally. It took another 15 years until the (national) goal “on paper” had been translated into (regional) strategies and special prevention facilities had been founded in all Austrian provinces. The foundation of the nine provincial prevention facilities had been accompanied by controversial political debates on their targets and tasks and on their position within the field of prevention in general mainly “owned” by the police until this time (Eisenbach-Stangl et al., 2008). The facilities whose task is to prevent “addiction” and who aim at less risky (substance) consumption patterns give different weight to alcohol drinking-related problems and focus on youth (Fell¨ocker & Franke 2000; Springer & UHl, 1995). Thus, it seems to be an artifact that primary prevention contributed to the decrease of alcohol consumption, though in the long run it will without doubt have multifold effects. Minimum age regulations which in the current study appeared to operate as a consumption promoting measure developed during an even longer period of time, and since they are used to react to actual youth behavioral—related problems, are more or less under steady reconstruction. The minimum drinking age regulations were, for example, immediately raised in Lower Austria when also Austrian youth during the first years of the new millennium

Results of the analysis of the selected socio-economic indicators and alcoholic beverage consumption document that urbanization, and to a minor extent, female employment and mothers giving birth to their children at a later age, are correlated with the decrease of total alcoholic beverage consumption after the early 1970s. These results create various problems of interpretation, among them the variations of the impact of urbanization on alcohol consumption in different cultural—geographical environments. There is only one metropolitan agglomeration in Austria—the capital Vienna and its environment—where about one quarter of all of Austria’s population live. Vienna is located in the wine growing and wine drinking eastern part of the country where—as indicated by a secondary analysis of few surveys carried out some years ago (Eisenbach-Stangl, Hager 2004)—the decrease of alcoholic beverage consumption was stronger than in the beer preferring western part, if it was not restricted to it. For various reasons—among them the temperance movement in the first quarter of the 20th century and immigration above average—alcoholic beverage consumption in the capital since decades is lower than in the surrounding provincial and rural areas of Lower Austria and Burgenland where the main Austrian vineyards are. It seems that urbanization in this provincial wine growing East of Austria was most strongly associated with a consumption decrease whereas in the beer drinking West it only played a minor role—if at all. In regard to the results of the analysis of the two other socio—indicators mentioned above it should be added that it is more than plausible that the gender relationship changes indicated by the increase of female employment and mothers giving birth to their children at a later age had an impact on the total decrease of alcoholic beverage consumption. But it should be added that according to other analysis men contributed more to the decrease than women by drinking less frequently, less often up to intoxication and more often preferring beverages with lower contents of alcohol (Eisenbach-Stangl, 2005). Results of the time series analysis of policy measures show that the establishment of primary prevention control policies in 1980 and the establishing of a 0.5 BAC level in 1998 were associated with decreasing alcoholic beverage consumption whereas the minimum age regulations of 1966 and the legal regulations of advertising of 1974 were associated with an increase. According to the results

TABLE 4. Chronic liver disease and transport accidents mortality, and policy measures, unplanned variables, consumption of beer, wine, spirits in Austria (1980–2006) – ANN analysis, connection values

Liver cirrhosis Min–decrease Transport accidents Min–decrease

Wine Max -incr.

beer Max -incr.

spirits Min decr.

M65

0.86

0.85

0.96

0.95

0.95

0.83

0.93

0.97

0.97

0.97

0.97

0.9

0.92

0.98

0.94

0.95

0.91

0.95

0.98

0.98

0.98

0.97

Price Urbanisastion beer

Price wine

Mothers’ age at Employment income childbirth Female

1998 BAC (0.5)

1630

I. EISENBACH-STANGL AND A. ALLAMANI

started to meet spontaneously in public to socialize and drink (Eisenbach-Stangl et al., 2008). The selection of 1966 for introducing minimum age regulations was an arbitrary compromise and presumably a main reason for the paradoxical results. The effects of primary prevention, and of minimum age regulations on availability and accessibility of alcoholic beverages, are in need of more detailed investigations. These need to consider, among other things the relevant continuously changing character, the political and societal interests and concerns, and the associated stakeholders behind them. The same is true for the effects of the advertising regulations, introduced in 1974 which in the current study also appear as a factor promoting alcoholic beverage consumption. Study Limitations and Conclusions

In summary, the generalizability of the current comparative analysis of Austria’s alcoholic beverage drinking trends and patterns are limited. These limitations presumably have resulted in paradoxical findings especially in regard to the effects of political control policy measures on alcohol beverage consumption.6 The requirement of equivalent data in all countries studied did, for example, not allow considering the effect of a major alcohol consumption political event such as the introduction of the 0.8 BAC level introduced in 1961 that is very early compared to other European countries. The same requirement enforced the use of a date of implementation of a control measure disregarding, for example, the unpredictable time gap between legislating a law, on paper, to its becoming a law in action with its necessary continuous change processes for sustaining viable adaptations of the measures. This temporal gap is an unpredictable issue without raising the actual and potential factors associated with a person’s law-abidingness. A final study limitation is associated with the simplifications necessary for the chosen type of comparisons. The search for indicators associated with changes in alcohol consumption were limited to quantitative measurable factors on a national level and excluded qualitative factors operating at regional and local levels and which were, and are, closer to every day life, functioning and adaptation. It is therefore not surprising that the study did not bring about new insights but rather reinforced assumptions already formulated by earlier studies. This study does, however, point to the direction for further needed research. The paradoxical results of the analysis of the effects of selected policy measures on alcoholic beverage consumption point to the necessity to develop new indicators for alcohol consumption related measures and their effects, and thus to

6

The reader is referred to Hills’s criteria for causation which were developed in order to help assist researchers and clinicians determine if risk factors were causes of a particular disease or outcomes or merely associated. (Hill, A. B. (1965). The environment and disease: associations or causation? Proceedings of the Royal Society of Medicine 58: 295–300.). Editor’s note.

invest scientifically in the development of more complex alcohol control policy concepts and viable processes. Declaration of Interest

The authors declare no conflicts of interest. The authors alone are responsible for the content and writing of the article. The AMPHORA project has received funding from the European Commission’s Seventh Framework Programme (FP7/2007–2013) under grant agreement no. 223059—Alcohol Measures for Public Health Research Alliance (AMPHORA). Participant organizations in AMPHORA can be seen at http://www.amphoraproject.net. THE AUTHORS Irmgard Eisenbach-Stangl, Sociologist and group analyst, teaches at the University of Vienna and is lead researcher in the research area “alcohol, drugs, addiction” at the European Centre for Social Welfare Policy and Research, Vienna.

Allaman Allamani, M.D., Psychiatrist; Family Therapist; Researcher. He has been coordinatorof the Alcohol Centre, Florence Health Agency (1993–2009); since 2009 he has been consultant to the Region of Tuscany Health Agency for research on social epidemiology and prevention policy, First non-alcoholic trustee of Italian Alcoholics Anonymous (1997–2003). He is a member of the editorial board of “Substance Use and Misuse.” Coordinator of a few Italian projects on alcohol prevention and policies, he has co-lead work package 3 of the European Commission-funded AMPHORA project. Author and co-author of many articles, editor and co-editor of 17 books and special issues.

GLOSSARY

Alcohol Consumption-Related Harms: In AMPHORA study, only Chronic liver disease and cirrhosis deaths, and transport accident deaths, are considered. Alcohol Control Policies: Measures that are planned by governments to control consumption and drinking patterns, and alcoholic beverage-consumption-related harm.

RELATED HARM AND ALCOHOL CONTROL POLICY IN AUSTRIA

REFERENCES Allamani, A., & Beccaria, F. (Eds.). (2007). Changes in the consumption of alcoholic beverages in Italy. Studies of the decrease in consumption between, 1970 and 2000. Contemporary Drug Problems, 34, special issue, summer. Allamani, A., Voller, F., Decarli, A., Casotto, V., Pantzer, K., Anderson, P. . . . Gmel, G. (2011). Contextual Determinants of Alcoholic beverage consumption Changes and Preventive Alcohol Policies: A 12-Country European Study in Progress. Substance Use & Misuse, 46, 10:1288–1303 print/1532–2491 online DOI: 10.3109/10826084.2011.572942. AMPHORA Workpackage3 (2013). Report of an analysis of European alcohol-related cultural, social and policy interactions and their impact on alcohol consumption and alcohol-related harm http://amphoraproject.net/w2box/data/Deliverables/AMPHO RA WP3 D3.2.pdf Anderson, P., Møller, L., & Galea, G. (2012). Alcohol in the European Union. Consumption, harm and policy approaches. Copenhagen: World Health Organisation, Regional Office for Europe. Babor, T. F., Caetano, R., Casswell, S., Edwards, G., Giesbrecht, N., Graham, K., . . . Rossow I. (2003). Alcohol: No ordinary commodity. London: Oxford University Press. Baccini, M., & Carreras, G. (2014). Analyzing the effect of policy measures and socio-demographic factors on alcohol consumption in Europe within the AMPHORA project: statistical methods. Substance Use & Misuse, 49. Beck, U. (1986). Risikogesellschaft. Auf dem Weg in eine andere Moderne, Frankfurt a. Main: Suhrkamp. Beck, U. (1996). Die “Individualisierungsdebatte. In B. Sch¨afers (Ed.), Soziologie in Deutschland (pp. 185–198). Opladen: Leske + Budrich. Bloomfield, K., Ahlstr¨om, S., Allamani, A., Choquet, M., Cipriani, F., Gmel, G., . . . Spak, F. (2005). Gender, Culture and Alcohol Problems: A Multi- national Study. Berlin: An EU – Concerted Action, Project Final Report, Charite, Campus Benjamin Franklin. Bundesvorstand. (1970). 65 Jahre Kampf gegen den Alkoholismus, In Der Abstinent (vol. 11, pp. 11–12). Buscema, P. M., Massini, G., & Maurelli, G. (2014). Artificial neural networks: An overview. Substance Use and Misuse, 49. Eisenbach-Stangl, I. (1991). Eine Gesellschaftsgeschichte des Alkohols. Produktion, Konsum und soziale Kontrolle alkoholischer ¨ Rausch- und Genussmittel in Osterreich 1918–1984, Campus, Frankfurt/Main Eisenbach-Stangl, I. (2005). M¨anner, Frauen und der Konsum psychoaktiver Substanzen. In I. Eisenbach-Stangl, S. Lentner, R. Mader (Eds.), M¨anner, Frauen, Sucht (pp. 13–36), Wien: Facultas. Eisenbach-Stangl, I. (2011). Comparing European Alcohol Policies, Policy Brief. Vienna: European Centre for Social Welfare Policy and Research. Eisenbach-Stangl, I. (2014). Passion and Insanity. A twofold concept of addiction in Austria 1850 – 1937 (forthcoming), Social History of Alcohol and Drugs. Eisenbach-Stangl, I., Bernardis, A., Fell¨ocker, K., HaberhauerStiedl, J., & Schmied, G. (2008). Jugendliche Alkoholszenen. Konsumkontexte, Trinkmotive. Vienna: Pr¨avention, occasional publication series, European Centre for Social Welfare Policy and Research. Eisenbach-Stangl, I., & Hager, I. (2004). The Trend towards the Middle. Drinking Patterns of Men and Women in Austria in a Period of Stagnating Alcohol Consumption. Presentation at

1631

the fifth Working Meeting of the EU Project GENACIS, June. Helsinki. Eurostat. (2012). Retrieved from website http://epp.eurostat.ec. europa.eu/portal/page/portal/statistics/search/ database, March 2012. Fell¨ocker, K., Franke, S. (Eds.). (2000). Suchtvorbeugung in ¨ Osterreich, Wien-New York: Springner. Giddens, A. (1996). Leben in einer possttraditionellen Gesellschaft. In U. Beck, A. Giddens, & Ch. Lash (eds.), Reflexive Modernisierung. Eine Kontroverse (pp. 113–194). Frankfurt/Main: Suhrkamp. Hoegel, H. (1887). Die strafrechtliche Behandlung der Trunkenheit. In Jurisitscche Bl¨atter (pp. 595–609). vol. 50 and 51. Hoff, H. (1954). Der akute und der chronische Alkoholismus. Medizinische Klinik, 76, 1425–1429 (part 1); 37, 1461–1466 (part 2). Hofmarcher, M. M. (2012). Gesundheitssysteme im Wandel: ¨ Osterreich 2012. Br¨ussel: European Observatory on Health Care Systems. Human Development Index (HDI) (2013). Retrieved from http://hdr.undp.org/en/statistics/hdi ¨ Lehner, O. (1969). Drogenpolitik in Osterreich, dargestellt am Beispiel der Alkoholgesetzgebung vom 15. bis ins fr¨uhe 19. Jahrhundert. In W. Ogris, & W. Rechberger (Eds.). Geda¨achtnisschrift f¨ur Werner Hofmeister (pp. 347–368). Wien, Manz#che Verlags- und Universit¨atsbuchhandlung. Mader, R., Mittendorfer, Ch., Pavlis, L., & Springer, A. (1981). ¨ Osterreichische Trinksitten. Wien: Konsumation- Einstellung – Gef¨ahrdung, Schriftenreihe des Ludwig Boltzmann-Institutes f¨ur Suchtforschung, 4, Br¨uder Hollinek. Nelson, J. P. (2010). Alcohol, unemployment rates and advertising bans: international panel evidence, 1975–2000. Journal of Public Affairs, 10, 74–87. Rehm, J., & Gmel, G. (2001). Aggregate time series regression in the field of alcohol. Addiction, 96, 945–954. Rehm, J., & Scafato, E. (2011). Indicators of alcohol consumption and attributable harm for monitoring and surveillance in European Union countries. Addiction, 106(Suppl. 1), 4–10. Simpura, J. (2001). Trends in alcohol consumption and drinking patterns: sociological and economic explanations and alcohol policies. NAD, 18, 3–13 (English Supplement). Springer, A., & Uhl, A. (1995). Prim¨arpr¨avention des Drogenmißbrauchs und ihre Evaluation,. Erfassung der o¨ sterreichischen Situation und Diskussion der Rahmenbedingungen in Euroopa und den USA, Wien: Ludwig BoltzmannInstitut f¨ur Suchtforschung. Statistik Austria. (2012). Retrieved from website http://www. statistik.at/, March 2012. Sulkunen, P. (1980). Die Kosmopolitiserung und Kommerzialisierung des Alkoholgebrauchs – die neuen, sozialen Bedeutungen des Alkohols. In: G. V¨olger. et al. (Eds.): Rausch und Realit¨at. Drogen im Kulturvergleich (pp. 134–137). K¨oln: Rautenstrauch-Museum. Sulkunen, P. (1989). Drinking in France 1965–1979. An analysis of Household Consumption Data. British Journal of Addiction, 84, 61–72. Sulkunen, P. (2002). From pastoral welfare state to consumer emancipation. In S. Henderson, & A. Petersen (Eds.), Consuming Health – commodification of health care (pp. 67–87). London: Routledge. Uhl, A., Bachmayer, S., Kobrna, U., Puhm, A., Springer, A., Kopf, ¨ N., . . . Musalek, M. (2009). Handbuch Alkohol – Osterreich, Wien: Bundesministerium f¨ur Gesundheit. Uhl, A., & Springer, A. (1996). Studie u¨ ber den Konsum ¨ von Alkohol und psychoaktiven Stoffen in Osterreich unter

1632

I. EISENBACH-STANGL AND A. ALLAMANI

Ber¨ucksichtigung problematischer Gebrauchsmuster. Wien: Bundesministerium f¨ur Gesundheit und Konsumentensschutz. World Health Organization. (2009). Global Information System on Alcohol and Health (GISAH) [online database]. Geneva: Author. World Health Organization. (2010). European Health for All Database (HFA-DB). Copenhagen: WHO Regional Office

for Europe. Retrieved February 2010, from http://www.euro. who.int/hfadb. World Health Organization. (2011). Global Information System on Alcohol and Health (GISAH) [online database]. Geneva: World Health Organization. Retrieved from http://who. int/globala/alcohol.

Notice of Correction: A typographical error in the title of this article has been corrected since the original online publication date of July 1, 2014.

Copyright of Substance Use & Misuse is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.