ORIGINAL RESEARCH INVESTIGACIÓN ORIGINAL
Alcohol consumption and burden of disease in the Americas in 2012: implications for alcohol policy
El consumo de alcohol y la carga de morbilidad en la región de las Américas en el 2012: implicaciones para las políticas relacionadas con el consumo de alcohol
Kevin D. ShieldI; Maristela MonteiroII; Michael RoereckeI; Blake SmithII; Jürgen RehmI
ISocial and Epidemiological Research Department, Centre for Addiction and Mental Health, Toronto, Ontario, Canada. Send correspondence to: Kevin D. Shield, Kevin.email@example.com
IIPan American Health Organization, Washington, DC, United States of America
OBJECTIVE: To describe the volume and patterns of alcohol consumption up to and including 2012, and to estimate the burden of disease attributable to alcohol consumption as measured in deaths and disability-adjusted life years (DALYs) lost in the Americas in 2012.
METHODS: Measures of alcohol consumption were obtained from the World Health Organization (WHO) Global Information System on Alcohol and Health (GISAH). The burden of alcohol consumption was estimated in both deaths and DALYs lost based on mortality data obtained from WHO, using alcohol-attributable fractions. Regional groupings for the Americas were based on the WHO classifications for 2004 (according to child and adult mortality).
RESULTS: Regional variations were observed in the overall volume of alcohol consumed, the proportion of the alcohol market attributable to unrecorded alcohol consumption, drinking patterns, prevalence of drinking, and prevalence of heavy episodic drinking, with inhabitants of the Americas consuming more alcohol (8.4 L of pure alcohol per adult in 2012) compared to the world average. The Americas also experienced a high burden of disease attributable to alcohol consumption (4.7% of all deaths and 6.7% of all DALYs lost), especially in terms of injuries attributable to alcohol consumption.
CONCLUSIONS: Alcohol is consumed in a harmful manner in the Americas, leading to a high burden of disease, especially in terms of injuries. New cost-effective alcohol policies, such as increasing alcohol taxation, increasing the minimum legal age to purchase alcohol, and decreasing the maximum legal blood alcohol content while driving, should be implemented to decrease the harmful consumption of alcohol and the resulting burden of disease.
Key words Ethanol; mortality; morbidity; policy; Americas.
OBJETIVO: Describir el volumen y los modelos de consumo de alcohol hasta el año 2012 incluido, y calcular la carga de morbilidad atribuible al consumo de alcohol medida según el número de defunciones y los años de vida ajustados en función de la discapacidad (AVAD) perdidos en la Región de las Américas en el 2012.
MÉTODOS: Los datos sobre el consumo de alcohol se obtuvieron a partir del Sistema Mundial de Información sobre el Alcohol y la Salud (GISAH, por sus siglas en inglés) de la Organización Mundial de la Salud (OMS). La carga del consumo de alcohol se calculó según la mortalidad y según los AVAD perdidos con base en los datos de mortalidad obtenidos de la OMS, tomando en consideración las fracciones atribuibles al alcohol. La división en subregiones se basó en las clasificaciones de la OMS del año 2004 (según la mortalidad en niños y adultos).
RESULTADOS: Se observaron variaciones regionales en el volumen total de alcohol consumido, la proporción del mercado del alcohol atribuible al consumo de alcohol no registrado, los hábitos de consumo, la prevalencia del consumo y la prevalencia de los episodios de consumo excesivo de alcohol. Los habitantes de la Región de las Américas consumieron más alcohol (8,4 litros de alcohol puro por adulto en el 2012) en comparación con el promedio mundial. La Región también experimentó una alta carga de morbilidad atribuible al consumo de alcohol (4,7% de las defunciones y 6,7% de los AVAD perdidos), especialmente en forma de lesiones atribuibles al consumo de alcohol.
CONCLUSIONES: El alcohol se consume de una manera perjudicial en la Región de las Américas y ello comporta una alta carga de morbilidad, especialmente en forma de lesiones. Con objeto de disminuir el consumo perjudicial de bebidas alcohólicas y la carga de morbilidad resultante, es preciso introducir nuevas políticas en materia de consumo de alcohol que sean eficaces en función de los costos, tales como el incremento de los impuestos sobre el alcohol, el aumento de la edad mínima legal para adquirir alcohol, y la disminución de la concentración máxima legal de alcohol en sangre mientras se conduce.
Palabras clave: Etanol; mortalidad; morbilidad; políticas; Américas.
Alcohol consumption causes a large health, social, and economic burden in the Americas (1-4) and results in an increasingly large burden in developing countries, especially as economic development causes the burden of disease to shift from communicable to noncommunicable diseases (NCDs) (5) and alcohol consumption to increase (6, 7). Therefore, decreasing harmful alcohol consumption and its effects has become a global priority (8).
Alcohol consumption is causally related to more than 230 three-digit ICD-101 codes.2 To address the large burden of disease attributed to this behavior, numerous suggestions have been set forth to strengthen monitoring of both alcohol consumption and its harmful effects (1113), and to increase awareness about the issue. For example, the objectives of the World Health Organization (WHO) Global Strategy to Reduce the Harmful Use of Alcohol ("Global Strategy") include 1) raising awareness of the magnitude of health problems caused by the harmful use of alcohol; 2) strengthening knowledge of the magnitude and determinants of alcohol-related harms and of effective interventions to reduce and prevent such harms; and 3) improving the dissemination of information for advocacy, policy development, and evaluations (8). The effects of alcohol consumption are also recognized in WHO's strategy for prevention and control of NCDs, which names alcohol as one of these chronic diseases' top four modifiable risk factors (4).
Several articles and reports have outlined the prevalence of alcohol consumption and its harmful effects in the Americas (1). In 2011, the Pan American Health Organization (PAHO) adopted a Regional Plan of Action consistent with the objectives of the WHO Global Strategy, and in 2014, PAHO published the 2014 Global Status Report on Alcohol and Health, which provides data for estimating the burden of disease attributable to alcohol consumption in the Americas for 2012. The objective of this study was to provide an update of the previous analyses of alcohol consumption and its harmful effects in the Americas. Specifically, this study aimed to 1) describe the volume and patterns of alcohol consumption in the Americas up to and including 2012, and 2) estimate the burden of disease attributable to alcohol consumption as measured in deaths and disability-adjusted life years (DALYs) lost (a measure of both premature mortality and years lived with disability) in the Americas for 2012.
MATERIALS AND METHODS
Alcohol exposure estimates
Alcohol exposure estimates were obtained from the Global Information System on Alcohol and Health (GISAH).3 The estimates include the adult per capita consumption of alcohol (total, recorded, unrecorded, and tourist); prevalence of current abstainers (lifetime abstainers and former drinkers) and current drinkers; prevalence of heavy episodic drinking (HED) (defined as consumption of five international standard drinks (> 60 g of pure alcohol) on one or more occasion in the past 30 days); and pattern of drinking (POD) scores (scaled from 1 to 5, with "1" representing the least risky drinking pattern and "5" representing the most risky pattern).
Alcohol-attributable burden calculations for 2012
The number of alcohol-attributable deaths was calculated by age, sex, and cause, using alcohol population-attributable fractions (PAFs) combined with data on number of deaths, years of life lost (YLL) due to premature mortality, and years lived with disability (YLD) in the Americas obtained from WHO (3, 4). For diseases where alcohol is a necessary cause (i.e., the diseases would not occur without alcohol consumption), such as alcohol use disorders, the alcohol PAF is equal to 1 (i.e., 100% would not occur without alcohol consumption).4
Regions and population data
Statistics on alcohol consumption and the burden of disease attributable to alcohol consumption are reported to WHO by regional subgroups ("subregions") defined by WHO (14) based on patterns of child and adult mortality. Table 1 and Table 2, respectively, list these statistics by country (the 35 PAHO Member States included in this report) and Americas subregion ("A": Canada, Cuba, United States of America (very low childhood and very low adult mortality rates); "B": Antigua and Barbuda, Argentina, Bahamas, Barbados, Belize, Brazil, Chile, Colombia, Costa Rica, Dominica, Dominican Republic, El Salvador, Grenada, Guyana, Honduras, Jamaica, Mexico, Panama, Paraguay, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, Suriname, Trinidad and Tobago, Uruguay, Venezuela (Bolivarian Republic of) (low childhood and low adult mortality rates); and "D": Bolivia (Plurinational State of), Ecuador, Guatemala, Haiti, Nicaragua, Peru (high childhood and high adult mortality)). Population data by age, sex, and country were obtained from WHO (4).
Data on gross domestic product (GDP) per capita for 2012, adjusted and unadjusted for purchasing power parity (PPP), were obtained from the World Bank (15). If those data were not available for a particular country for 2012, the numbers were estimated based on the linear trend over previous years; if no GDP data for a particular country were available from the World Bank, the data were obtained from the International Monetary Fund (16).
Correlations were performed using Pearson's product-moment correlation coefficient. All statistics were performed using R version 3.1.2 (17).
Per capita consumption of alcohol
In 2012, in the Americas, total adult per capita consumption of alcohol (8.4 L of pure alcohol) was 24% higher than the global average (6.8 L of pure alcohol) and ranged from 3.2 L in El Salvador to 10.3 L in Chile in 2012 (Table 1). While unrecorded alcohol consumption5 constituted a large proportion of all alcohol consumed in the Americas (14.7%) (Table 2), this variable value was less than the global average (22.8%). Unrecorded alcohol consumption and the proportion of the market constituted by unrecorded alcohol showed considerable variation in the Americas, ranging from 0.2 L per adult in Saint Lucia to 3.1 L per adult in Ecuador (Table 1), and from 1.92% in Saint Lucia to 42.11% in Guatemala (not shown) respectively. Tourist consumption only affected countries in the Caribbean, specifically Antigua and Barbuda, Bahamas, Barbados, Dominica, Grenada, Saint Kitts and Nevis, and Saint Lucia (Table 1). In these countries, the number of tourists and their overall alcohol consumption far exceeded the number of tourists who visited other countries and their overall alcohol consumption in the host countries.
From 1961 to 2010, the recorded adult per capita consumption of alcohol in the Americas increased 11.43% (from 6.41 L to 7.14 L), mainly due to a 26.90% increase in recorded consumption in Americas subregion B (from 5.11 L to 6.48 L) (Figure A1 of the Supplementary Material). In Americas subregion D, recorded alcohol consumption decreased by 24.24% (from 4.59 L to 3.48 L), but this change did not necessarily indicate a decrease in overall consumption in the region because unrecorded consumption was not taken into account (e.g., if unrecorded consumption increases more than a recorded decrease in recorded alcohol consumption, overall consumption, will increase).
Data on beverage preferences for 2012 (by grams of pure alcohol) indicate that beer was the most popular beverage in the Americas (55.2% of all recorded consumption), followed by spirits (31.0%), and wine (13.4%) (with the remaining 0.4% corresponding to "other" types of alcoholic beverages) (not shown). The popularity of both wine and spirits versus beer has thus decreased since 1961, when the respective percentages recorded for each type of beverage for all recorded consumption of alcohol were 39.4%, 36.6%, and 24.0% (not shown).
Prevalence of alcohol consumption and patterns of drinking
In 2012, 38.5% of all adults (47.3% of adult women and 29.3% of adult men) in the Americas were abstainers at the time of the study ("current abstainers") versus 61.9% of the global adult population (70.9% of adult women and 53.0% of adult men). The prevalence of current abstainers and current drinkers varied greatly in the Americas. Subregionally, Americas subregion A had the lowest prevalence of current abstainers (30.5% of the adult population), whereas Americas subregion D had the highest prevalence (49.0% of the adult population) (Table 2). There was a significant association between the ratio of current drinkers among all men and all women ("prevalence ratios") and the overall prevalence ratio (among the general population region-wide) (r = -0.764; 95% confidence interval (CI): -0.875 to -0.578; P < 0.001), indicating that as the prevalence of all current drinkers increased, the gap between the prevalence ratio for men versus women decreased. A similar statistically significant (P < 0.001) correlation was found when examining the association between the absolute difference in prevalence ratio of current drinkers among all men versus all women and the overall prevalence of current drinkers.
The consumption of alcohol among current drinkers in the Americas (13.7 L per adult drinker) was less than the global average (17.8 L per adult drinker). Subregionally, Americas subregion D had the highest consumption per drinker (14.4 L) and Americas subregion A had the lowest (13.0 L) (Table 2). A weak and non-significant association was observed between the ratio of alcohol consumption among all female current drinkers to alcohol consumption among all male current drinkers and overall consumption among all current drinkers.
For 2012, HED among current drinkers was more prevalent in the Americas compared to the global average. Within the region, HED was most prevalent in subregion A (Table 2) (23.9% of all adult current drinkers engaged in HED- 16.6% of all adult female current drinkers and 30.4% of all adult male current drinkers) and least prevalent in Americas subregion D (17.5% of all adult current drinkers-6.3% of all adult female current drinkers and 25.0% of all adult male current drinkers). A weak and non-significant association was observed between the prevalence ratio of HED among adult female current drinkers to HED among adult male current drinkers and the overall prevalence of HED among all current drinkers.
The POD scores for the countries in the three Americas subregions included in this study ranged from 2 (Argentina, Bahamas, Barbados, Canada, Chile, Cuba, Jamaica, Saint Kitts and Nevis, Saint Vincent and the Grenadines, Trinidad and Tobago, and the United States of America) to 4 (Belize, Grenada, and Guatemala), with an average score for the region of 2.5. The most harmful drinking pattern was observed in subregion D, which had a POD score of 3.2, reflecting the frequency of festive drinking and drinking outside meals in the countries it comprises (Table 2).
Alcohol consumption and development
The level of development of a country, as measured by GDP-PPP per capita, was significantly associated with the adult per capita consumption of alcohol (r = 0.377; CI: 0.050 to 0.631; P = 0.026); prevalence of current abstainers (r = -0.561; CI: -0.753 to -0.279; P < 0.001); and the POD (r = -0.562; CI: -0.754 to -0.281; P < 0.001).
Alcohol-attributable burden of disease
In 2012, in the Americas, alcohol consumption was responsible for 302 860 deaths (31.7 per 100 000 people) (Table 3) and 18 410 200 DALYs lost (19.2 per 1 000 people) (Table 4), representing 4.7% of all deaths and 6.7% of all DALYs lost. Men in the Americas experienced much more harm from alcohol consumption than women, with alcohol causing 7.3% of all deaths and 10.1% of all DALYs lost among men compared to 1.8% of all deaths and 2.6% of all DALYs lost among women (not shown). There was wide variation by subregion in the number of deaths (per 100 000 people) and DALYs lost (per 1 000 people) attributable to alcohol consumption for 2012 in the Americas, with Venezuela, Guyana, and Brazil experiencing the greatest burden for both variables (51.5, 47.0, and 42.3 deaths per 100 000 people attributable to alcohol consumption and 33.2, 25.7, and 24.8 DALYs lost per 1 000 people attributable to alcohol consumption respectively) (not shown).
Tables 3 and 4 also show the number of deaths and DALYs lost attributable to alcohol consumption by cause. Alcohol-attributable NCDs were the largest contributor to the burden of disease caused by alcohol, accounting for 55.1% of all alcohol-attributable mortality and 61.0% of all alcohol-attributable DALYs lost. For the region as a whole, digestive diseases were the largest NCD contributor to alcohol-attributable mortality, accounting for 25.7% of all alcohol-attributable deaths. In contrast to alcohol-attributable mortality, the largest NCD contributor to alcohol-attributable DALYs lost was mental and behavioral disorders, which accounted for 37.1% of all DALYs lost in the Americas. Alcohol-attributable communicable, maternal, perinatal, and nutritional conditions accounted for 7.4% and 4.1% of alcohol-attributable mortality and DALYs lost in the region respectively (with injuries accounting for the remaining 37.5% and 32.6% respectively) (not shown).
Harmful alcohol consumption is prevalent in the Americas, and the health harms attributable to alcohol consumption in the region are severe. Therefore, the prevention of alcohol-related harms should be a priority in the Americas (18).
Alcohol consumption and alcohol-attributable burden in the Americas
In the Americas, adult per capita consumption of alcohol is high, and per capita consumption, prevalence of current abstainers, and POD scores are significantly associated with economic development (as measured by GDP-PPP)-findings that corroborate the results of previous analyses (6). In addition, unrecorded consumption varied widely by subregion (subregions B and D had a higher percentage of total consumption comprised of unrecorded alcohol than subregion C). According to a systematic review, the main sources of unrecorded alcohol consumption in the Americas are "surrogate alcohol" (ethanol-containing substances that are not meant for consumption) and artisanal spirits (including home production); however, cross-border shopping and smuggling also affect unrecorded alcohol in Mexico and Canada (19).
The largest contributor to the burden of alcohol-attributable DALYs lost was mental and behavioral disorders, and the majority of this burden came from alcohol use disorders (AUDs). The prevalence of AUDs among adults in the Americas in 2010 was 46.3% higher than the world average (6.0% versus 4.1% globally) (4). For that same year, the prevalence of alcohol dependence among adults and the harmful use of alcohol were much greater in the Americas versus the worldwide averages (3.4% and 2.6% versus 2.3% and
1.8% respectively). The prevalence of AUDs among women in the Americas was particularly high, with 3.2% of all adult women reported to have an AUD versus 2.9% in Europe and 1.3% globally. The prevalence of AUDs among men was also high, with 9.0% of all adult men reported to have an AUD versus 12.6% in Europe and 7.2% globally. Drinking patterns in the Americas also contributed to the large burden of alcohol-attributable disease, especially in the countries of Latin America and the Caribbean (LAC), where they caused the largest burden of alcohol-attributable injuries (3, 4, 20). Therefore, effective interventions are needed to target AUDs in the Americas.
Comparison with previous studies
Although not directly comparable due to difference in methodology, the results of this research are similar to those of Rehm & Monteiro (1), who assessed the burden of alcohol consumption in the Americas for 2000. Like that study, the research reported here found that alcohol consumption is a major health risk in the Americas, and that alcohol-attributable injuries and AUDs accounted for most of the alcohol-attributable burden of disease. However, the methods used by Rehm & Monteiro (1) differed from the ones used in this study (e.g., this study modeled alcohol consumption as a continuous distribution, and used continuous and more up-to-date alcohol relative risk (RR) functions, and different disease categories), and some of the increase in the alcohol-attributable burden in the current study versus the prior study is due to those differences (1, 3, 4). Other differences in this study versus prior studies include the lack of controls for population size and structure (previous analyses that did control for those factors reported an increase in the burden of disease caused by alcohol for LAC and a decrease for North America (high-income countries)) (21). In addition, some previous studies included all subregions of the Americas (versus only three, in the current study); the prior studies found that alcohol was one of the top risk factors for the burden of disease in all subregions (20, 22, 23).
The methods used in this report are the best methods for calculating the burden of disease attributable to alcohol consumption and comparing the estimated burden across countries (24). However, as with all burden of disease studies, these methods incur certain limitations in terms of the quality of the exposure and outcome data and the estimation procedures used, including various types of bias and the use of unadjusted data.
To minimize potential survey bias, the survey data on alcohol consumption were triangulated with per capita consumption data. However, survey biases can also affect prevalence results for current drinkers, former drinkers, lifetime abstainers, and current drinkers that engage in HED (6). Currently, there is no way to quantify or correct for these biases. Mortality and disability data can also be affected by accuracy biases, especially in low- and middle-income countries, where the infrastructure for tracking and recording mortality and morbidity is limited or does not exist (25).
The use of cross-sectional estimates of alcohol consumption was another source of potential bias due to the biological latency period between the alcohol consumption and chronic disease development and mortality, as in the case of liver cirrhosis and cancer (10). Despite that inherent limitation, the burden of chronic diseases attributable to alcohol was modeled using cross-sectional estimates for three primary reasons: 1) most surveys only measure alcohol consumption in a cross-sectional manner and do not measure lifetime alcohol consumption history (6); 2) most meta-analyses assessing the relationship between alcohol consumption and chronic diseases use cross-sectional estimates for alcohol consumption (9); and 3) although there is a biological latency period, some of the effects of decreases in alcohol consumption on the burden of chronic diseases can be observed at the population level immediately, such as the effects of the Gorbachev alcohol reforms on liver cirrhosis deaths (26).
Finally, the PAFs used in this study were designed using unadjusted alcohol RR functions (27). The impact of this limitation may have been minimal, however, as alcohol RR functions adjusted for confounding factors, such as smoking status, have been observed to be very similar to unadjusted alcohol RR functions (9). Furthermore, alcohol RR functions are based on the reference category of lifetime abstainers. Although the accuracy of this reference group is questionable, with misclassification of former drinkers as lifetime abstainers potentially creating bias, the absolute risk of a disease related to alcohol consumption among misclassified lifetime abstainers who are in fact former drinkers is likely to be not much higher than the absolute risk of a disease related to alcohol consumption for correctly classified lifetime abstainers (28).
Alcohol policy recommendations
Cost-effective measures to reduce the health, social, and economic burden of alcohol consumption in the Americas (29, 30), taking into consideration statistics on alcohol consumption and alcohol-attributable harms, should focus on the following four areas: 1) reducing per capita consumption of alcohol through measures such as increased taxation, minimum pricing, reduced availability, and alcohol marketing control; 2) decreasing the burden of alcohol-attributable injuries through measures such as lowering the maximum legal blood alcohol content (BAC) while driving, increasing the minimum legal age for the purchase and/or consumption of alcohol, and instituting and/or increasing the frequency of BAC roadside checks; 3) given the high prevalence of AUDs, increasing the identification of people in the general population with these disorders and providing them with evidence-based interventions, treatment, and rehabilitation (taking into consideration the fact that the interventions will require reliable health care systems and are not as cost-effective as other methods, such as increases in alcohol taxation); and 4) introducing new policies, such as restricting the times when alcohol can be sold, restricting sales to intoxicated persons and underage individuals, legislation to improve enforcement and monitoring, and methods of regulating and reducing the unrecorded alcohol market.
Evidence used in determining the optimal measures may require some adjustment for developing countries, as most research on alcohol policies is derived from data collected in high-income countries (18). For example, in developing countries with a large unrecorded alcohol market, increases in taxation and decreases in availability will be less effective without regulations to control the informal and illicit market.
Furthermore, although little has been done in terms of alcohol policy reform in the Americas (3, 4), more countries are now adopting and reforming national alcohol strategies (4), following WHO's Global Strategy and PAHO's Regional Plan of Action for reducing the harmful effects of alcohol (31). Nevertheless, additional action is needed to reduce the burdens created by alcohol, including adopting policy options and implementing national monitoring systems in lowlowand middle-income countries where these systems do not yet exist.
Harmful consumption of alcohol is prevalent in the Americas, and the resulting health burden is large. Therefore, there is an urgent need for the countries in the Americas to implement cost-effective policies to reduce the harmful consumption of alcohol and the resulting social, economic, and health burdens.
Acknowledgments. The authors thank the World Health Organization and the Pan American Health Organization for the data used in this study. The authors would also like to thank Michelle Tortolo for her work on the references for this article.
Funding. This work was funded by the Pan American Health Organization.
Conflicts of interest. None.
Disclaimer. Authors hold sole responsibility for the views expressed in the manuscript, which may not necessarily reflect the opinion or policy of the RPSP/ PAJPH or PAHO.
1. Rehm J, Monteiro M. Alcohol consumption and burden of disease in the Americas: implications for alcohol policy. Rev Panam Salud Publica. 2005;18(4-5):2418.
2. Rehm J, Mathers C, Popova S, Thavorncharoensap M, Teerawattananon Y, Patra J. Global burden of disease and injury and economic cost attributable to alcohol use and alcohol use disorders. Lancet. 2009;373(9682):222333.
3. World Health Organization. Global status report on alcohol and health. Geneva: WHO; 2011. Available from: http://www.who.int/substance_abuse/publications/global_alcohol_report/msbgsruprofiles.pdf
4. World Health Organization. Global status report on alcohol and health. Geneva: WHO; 2014. Available from: http://apps.who.int/iris/bitstream/10665/112736/1/9789240692763_eng.pdf
5. World Health Organization. Prevention and control of NCDs: priorities for investment. Discussion paper for the First Global Ministerial Conference on Healthy Lifestyles and Noncommunicable Disease Control, Moscow, 2829 April 2011. Geneva: WHO; 2011. Available from: http://www.who.int/nmh/publications/who_bestbuys_to_prevent_ncds.pdf
6. Shield K, Rehm J. Difficulties with telephone- based surveys on alcohol in high-income countries: the Canadian example. Int J Methods Psychiatr Res. 2012;21(1):1728.
7. Room R, Jernigan D, Carlini BH, Gmel G, Gureje O, Mäkelä K, et al. El alcohol y los países en desarrollo. Una perspectiva de salud pública. Mexico City: Fondo de Cultura Económica, Pan American Health Organization; 2013.
8. World Health Organization. Global strategy to reduce the harmful use of alcohol. Geneva: WHO; 2010. Available from: http://www.who.int/substance_abuse/msbalcstragegy.pdf
9. Rehm J, Baliunas D, Borges GL, Graham K, Irving HM, Kehoe T, et al. The relation between different dimensions of alcohol consumption and burden of disease: an overview. Addiction. 2010;105(5):81743.
10. Shield KD, Parry C, Rehm J. Chronic diseases and conditions related to alcohol use. Alcohol Res. 2013;35(2):15573.
11. Monteiro MG. Alcohol and public health in the Americas: a case for action. Washington: Pan American Health Organization; 2007. Available from: http://iris.paho.org/xmlui/handle/123456789/2826
12. Beaglehole R, Bonita R. Alcohol: a global health priority. Lancet. 2009;373(9682):21734.
13. Collins PY, Patel V, Joestl SS, March D, Insel TR, Daar AS, et al. Grand challenges in global mental health. Nature. 2011;475(7354):2730.
14. World Health Organization. The world health report 2000. Health systems: improving performance. Geneva: WHO; 2000. Available from: http://www.who.int/whr/2000/en/whr00_en.pdf
15. World Bank. World development indicators. Washington: World Bank; 2014. Available from: http://data.worldbank.org/sites/default/files/wdi-2014-book.pdf
16. International Monetary Fund. World economic and financial surveys: World Economic Outlook database [Internet]. Washington: IMF; 2014. Available from: http://www.imf.org/external/pubs/ft/weo/2014/02/weodata/index.aspx Accessed on 1 November 2014.
17. R Development Core Team. R: a language and environment for statistical computing (version 3.1.0). Vienna: R Foundation for Statistical Computing; 2014.
18. Room R, Babor T, Rehm J. Alcohol and public health: a review. Lancet. 2005;365(9458):51930.
19. Rehm J, Kailasapillai S, Larsen E, Rehm MX, Samokhvalov AV, Shield KD, et al. A systematic review of the epidemiology of unrecorded alcohol consumption and the chemical composition of unrecorded alcohol. Addiction. 2014;109(6):88093.
20. Lim SS, Vos T, Flaxman AD, Danaei G, Shibuya K, Adair-Rohani H, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 19902010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380(9859):222460.
21. Institute for Health Metrics and Evaluation. GBD Compare: global, DALYs; both sexes, all ages, 2010. Seattle: University of Washington; 2013. Available from: http://vizhub.healthdata.org/gbd-compare Accessed on 11 February 2015.
22. World Health Organization. The world health report 2002: reducing risks, promoting healthy life. Geneva: WHO; 2002. Available from: http://www.who.int/whr/2002/en/
23. World Health Organization. Global health risks: mortality and burden of disease attributable to selected major risks. Geneva: WHO; 2009. Available from: http://www.who.int/healthinfo/global_burden_disease/GlobalHealthRisks_report_full.pdf
24. Murray CJ, Lopez AD. On the comparable quantification of health risks: lessons from the Global Burden of Disease Study. Epidemiology. 1999;10(5):594605.
25. Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJ. Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data. Lancet. 2006;367(9524):174757.
26. Bhattacharya J, Gathmann C, Miller GJ. The Gorbachev anti-alcohol campaign and Russia's mortality crisis. Cambridge, MA: National Bureau of Economic Research; 2012.
27. Ezzati M, Lopez AD, Rodgers A, Murray CJ, editors. Comparative quantification of health risks: global and regional burden of disease attributable to selected major risk factors. Volume 1. Geneva: World Health Organization; 2004. Available from: http://apps.who.int/iris/bitstream/10665/42792/1/9241580348_eng_Volume1.pdf
28. Rehm J, Irving H, Ye Y, Kerr WC, Bond J, Greenfield TK. Are lifetime abstainers the best control group in alcohol epidemiology? On the stability and validity of reported lifetime abstention. Am J Epidemiol. 2008;168(8):86671.
29. Anderson P, Chisholm D, Fuhr DC. Effectiveness and cost-effectiveness of policies and programmes to reduce the harm caused by alcohol. Lancet. 2009;373(9682):223446.
30. Babor TF, Caetano R. Evidence-based alcohol policy in the Americas: strengths, weaknesses, and future challenges. Rev Panam Salud Publica. 2005;18(45):32737.
31. Pan American Health Organization. Plan of Action to Reduce the Harmful Use of Alcohol. 51st Directing Council of PAHO, 63rd Session of the Regional Committee, Washington, D.C., USA, 2630 September 2011. Washington: PAHO; 2011. Available from: http://www.paho.org/hq/index.php?option=com_docman&task=doc_view&gid=20295&Itemid=
Manuscript received on 22 January 2015.
Revised version accepted for publication on 4 August 2015.
1 International Classification of Diseases, 10th Revision.
2 An overview of diseases, injuries and other health conditions causally related to alcohol consumption can be found elsewhere (9, 10).
3 For more details on the information sources for the alcohol exposure estimates, see the GISAH website (www.who.int/globalatlas/alcohol) and the Supplementary Material ("Web Appendix") supplied by the authors.
4 For more information on calculating the alcohol PAF, see the Supplementary Material.
5 Alcohol consumption not recorded by retail or government statistics.
The supplementary material is available in pdf: [Supplementary material]