4. The health and economic burden of alcohol consumption

Yevgeniy Goryakin
Alexandra Aldea
Marion Devaux
Yvan Guillemette
Andrea Feigl
Sabine Vuik
Alienor Lerouge

Harmful alcohol use is a leading risk factor contributing to diseases and the economic burden of diseases. Alcohol is a causal factor in more than 200 disease and injury conditions, including alcohol dependence, liver cirrhosis and cancers. Harmful use of alcohol causes approximately 3.3 million global deaths annually (or 5.9% of all global deaths), and causes 5.1% of the global burden of disease (WHO, 2018[1]).

In OECD countries, alcohol consumption per capita is about twice the world average. The 2014 World Health Organization (WHO) Global Status Report on Alcohol and Health estimated that, despite the expected decrease of alcohol consumption by 0.6 litres per capita between 2005 and 2025, the European Region will still have the highest level of per capita consumption in the world (WHO, 2014[2]).

Although the health consequences of harmful use of alcohol are well researched, evidence on its economic costs, including on health care budget spending and on labour force productivity, is scarce and context-specific. This chapter brings together the evidence from the literature on the economic costs of alcohol use, as well as main modelling outputs produced under a scenario assuming a ceiling of alcohol consumption at 1 drink1 per day for women and 1.5 drinks per day for men, compared to a business-as-usual scenario, in which alcohol consumption remains at the current levels and patterns. In addition, this chapter also estimates the burden caused by any alcohol consumption.

Previous estimates of the impact of alcohol consumption on health expenditure range from just under 1% of total health expenditure in Switzerland to just above 7% in France (see Figure 4.1). However, there is wide variation in the data sources underlying each study, in the methods used, and in the alcohol-related conditions and health care settings that are included to calculate health care costs.

For example, the study from Switzerland (Fischer et al., 2014[3]), where the lowest cost share (about 1%) was found, applied a top-down estimation approach, and only included hospital-based costs. Top-down approaches may underestimate costs as they do not account for the health care cost of minor health impairments associated with alcohol consumption, and often fail to account for the effect of comorbidities. The study from Canada (Canadian Substance Use Costs and Harms Scientific Working Group, 2020[4]) used a wider range of health care costs, including inpatient hospitalisations, day surgery, emergency department visits, specialised treatment, physician time and prescription drugs. As a result, the size of the burden estimate was higher, at about 2.3%.

In addition to methodological differences, studies also varied in other ways. For example:

  • The age groups were not consistently defined. In one study, the population of interest was people older than 15; in two studies, people aged 15-74; and in the remaining studies, “adults” with unclear age boundaries.

  • Different sets of diseases were included in the analyses. For example, 60 alcohol-related conditions were included in Switzerland, 54 in France and a slightly smaller set in Sweden, although in the case of the Swedish study the exact number of conditions was not specified. The costs for Estonia included 25 disease groups, whereas the Portuguese study included 44 alcohol-related conditions.

A number of studies go beyond health expenditure and try to estimate the impact of alcohol consumption on the wider economy. In nine reviewed studies, total non-health care costs ranged from 0.4% (Portugal and France) to 1.6% (Estonia) of GDP in the year the costs were incurred (Figure 4.2). For most countries for which data were available, the non-health care costs (excluding the social costs) ranged between 1% and 1.5% of GDP.

Among the approaches to model the non-health care costs associated with alcohol consumption, the human capital approach was the most common. This measures lost productivity, morbidity or mortality in terms of lost earnings based on wages. Again, the studies varied both in terms of the scope of the costs considered and in some other study characteristics, as discussed in Section 4.1.1 above.

For example, in some studies, only non-health care costs explicitly linked to labour market outcomes (such as labour productivity losses) were taken into account (Cesur and Kelly, 2014[12]), while in others, lost earnings resulting from premature mortality were also included (Fischer et al., 2014[3]). In several studies, other non-health care costs were included in the scope, including costs related to the use of social care (Johnston, Ludbrook and Jaffray, 2012[9]); economic losses and other costs resulting from increased crime (including policing, legal costs and costs of increased incarceration rates) (Jarl et al., 2008[7]); and other intangible costs such as pain and suffering (Johnston, Ludbrook and Jaffray, 2012[9]).

To quantify the impact of several risk factors – including alcohol consumption – on population health and the economy, the OECD developed the Strategic Public Health Planning for non-communicable diseases (SPHeP-NCDs) model. This simulates the impact of major risk factors on disease incidence, mortality, health expenditure2 and employment and productivity (see Box 4.1 for more details on the model). The OECD SPHeP-NCDs model can be used to understand the economic burden of diseases caused by alcohol consumption, as well as the potential impact of interventions. As the model applies a standardised approach to all countries, it also allows cross-country comparison. This section presents the outputs of the OECD SPHeP-NCDs model and its estimates of the health and economic burden of diseases caused by alcohol consumption above 1 drink per day for women and 1.5 drinks per day for men, between 2020 and 2050, and estimates the burden caused by any alcohol consumption.

Alcohol consumption above 1 drink per day for women and 1.5 drinks per day for men has a considerable negative impact on population health. Specifically, drinking above these caps contributes to about 1.1 billion new cases of dependence, 37 million cases of injury, 12 million cases of diabetes, 24 million cases of CVDs, 5 million cases of cirrhosis and 10 million cases of cancer related to alcohol over the next 30 years in 52 countries. This accounts for about 88% of all cases of dependence, and 38% of all cases of cirrhosis projected for 2020-50 (Figure 4.4).

However, as alcohol consumption above the 1/1.5 drinks per day cap reduces life expectancy, it also reduces the amount of time available to develop other diseases or conditions. As a result, and because of shorter life expectancy, drinking above these caps decreases the incidence of several diseases such as MSDs, cancers not related to alcohol, chronic obstructive pulmonary disease (COPD) and dementia (Figure 4.4).

Alcohol consumption above 1 drink per day for women and 1.5 drinks per day for men can also lead to people dying prematurely – between ages 30 and 70, according to the WHO definition (WHO, 2018[28]). Specifically, the model predicts that, compared to the business-as-usual scenario, an additional 1.1 million people will die early due to diseases caused by drinking above these caps in the 52 countries each year. On average across OECD countries, 24 people per 100 000 population will die prematurely each year due to alcohol consumption above the 1/1.5 drinks per day cap (Figure 4.5). In the EU27,this average is higher, at 28 per 100 000, mostly driven by relatively high premature mortality rates in Central and Eastern European countries.

The impact of alcohol consumption above the 1/1.5 drinks per day cap on population health can also manifest itself in shorter life expectancy (Figure 4.6). On average across all OECD countries, life expectancy is 0.9 years lower over 2020-50 due to drinking above the 1/1.5 drinks per day cap. For comparison, over the last 30 years, life expectancy in OECD countries has increased by about 6.7 years (World Bank, 2020[29]), driven by changes in a large number of medical and social factors. Alcohol consumption is only one determinant of population health, but drinking within the 1/1.5 drinks per day cap could potentially contribute to about 13% of the total life expectancy gain recorded over a similar period of time in the past. The largest reductions are predicted in Central and Eastern European countries, with more than 1.5 years of life expectancy lost in Lithuania, the Russian Federation, Poland, Estonia, Latvia and Romania. Given that the current life expectancy in Lithuania is about 74 years for both sexes, and in Japan – the country with the longest life expectancy – it is about 84 years, and given that life expectancy loss due to alcohol consumption above the 1/1.5 drinks per day cap in Lithuania is greater than in Japan by about 1.6 years, alcohol consumption above these caps potentially accounts for about 16% of the life expectancy gap between these two countries.

The effect of alcohol consumption above the 1/1.5 drinks per day cap on years of healthy life expectancy (HALEs) – i.e. after taking into account the quality of life years lived through disability-adjusted weights for people with diseases – is even greater. For example, across all OECD countries, 1.13 HALEs are lost over 2020-50 due to this level of alcohol consumption, with the largest effect predicted to be in Lithuania (2.3 HALEs lost), and the smallest in Turkey (0.11 HALEs lost).

Finally, a brief comparison of the health-related burden of alcohol consumption above the 1/1.5 drinks per day cap with the health-related burden of any alcohol consumption is made in Box 4.2 and discussed further in Annex 4.B.

It should be also noted that the model produces conservative estimates as it does not take into account the impact of alcohol consumption on certain diseases, either because they represent a small part of the alcohol-attributable disease burden, or because of a lack of availability of reliable epidemiological data on these diseases as, for example, in the case of the foetal alcohol spectrum disorders (Box 4.3). In addition, alcohol consumption may lead to additional health problems – for example, by hindering effective management of medical conditions, either related or unrelated to drinking alcohol. For instance, alcohol consumption may be associated with lower adherence to medical therapies, or with the reduced likelihood of seeing a doctor (Ahmed, Karter and Liu, 2006[30]). Down the line, these are likely to increase the likelihood of disease progression or complications.

On average, the treatment of diseases caused by alcohol consumption above 1 drink per day for women and 1.5 drinks per day for men increases per capita medical spending by about USD PPP 61 annually in OECD countries, which accounts for about 2.4% of the overall annual health expenditure across OECD countries in 2020-50, including both public and private expenditure on health (Figure 4.7). The largest spending is predicted to happen in countries where the cost of medical treatment is the highest, such as the United States, Luxembourg and Germany, with up to USD PPP 168 spent per capita annually in the United States. The lowest amount is predicted to be spent in Turkey, where both the level of alcohol consumption and treatment costs are relatively low. In total, USD PPP 138 billion per year will be spent to treat these diseases across all the countries included in the analysis. This is equivalent to, for instance, the current health spending in Australia, or more than twice the current health spending in Belgium.

Although the health burden of alcohol use is found to be relatively high in Central and Eastern European countries (see Figure 4.6), medical expenditure attributable to diseases caused by alcohol use (expressed in USD PPP) is relatively low (Figure 4.7). This difference in findings is mostly due to the lower costs of medical care in countries in these regions. Nevertheless, this spending accounts for a very large proportion of total medical spending in some of these countries, especially in Lithuania, Latvia and Estonia. In Lithuania, diseases caused by alcohol consumption above the 1/1.5 drinks per day cap account for the largest share of total medical spending compared to all the other countries, at 4.2%.

On a disease-specific basis, alcohol consumption above 1 drink per day for women and 1.5 drinks per day for men contributes to a large increase in the costs of treating several diseases – most notably dependence, cirrhosis and certain cancers. It accounts for 87% of all dependence-related expenditure (or about USD PPP 115 billion annually in the 52 countries studied) and for 35% of all expenditure related to treating cirrhosis in 2020-50 (Figure 4.8). Alcohol consumption above the 1/1.5 drinks per day cap is also responsible for about 4% (USD PPP 17 billion) of all expenditure for treating alcohol-related cancers (including liver, breast, colorectal, oesophageal, nasopharynx, lip and oral cavity cancers).

Finally, a brief comparison of the health expenditure burden of alcohol consumption above the 1/1.5 drinks per day cap with the burden on health expenditure of any alcohol consumption is made in Box 4.4 and discussed further in Annex 4.B.

The impact of alcohol consumption on the labour market, including employment and productivity, is complex and depends on a number of different factors (Box 4.5). The approach chosen for this report involves modelling the labour market effect of alcohol consumption only through diseases and medical conditions, rather than through any other pathway, because previous OECD analyses identified the link between diseases and labour market outcomes as the strongest from a statistical point of view (Devaux and Sassi, 2015[35]).

The OECD analysis shows that diseases caused by alcohol consumption above 1 drink per day for women and 1.5 drinks per day for men reduce employment by about 0.33% annually across all OECD countries in the working-age population (ages 18-65) in 2020-50 (Figure 4.9). At the same time, there are significant regional variations in this effect; the labour markets in Central and Eastern Europe suffer the most, with up to 0.67% employment reduction attributable to diseases caused by alcohol consumption in Latvia.

In addition, diseases caused by alcohol consumption above the 1/1.5 drinks per day cap reduce productivity when employed, as measured by absenteeism and presenteeism. Specifically, across OECD countries, 0.11% of labour force productivity is lost annually because of sickness-related absences, while 0.24% is lost due to reduced productivity at work in the form of presenteeism. The effect on early retirement is generally negligible, mostly due to a weak association of alcohol-related diseases with this outcome (Figure 4.9).

Overall, chronic conditions caused by drinking more than 1 drink per day for women and 1.5 drinks per day for men affect the productivity of the labour force by reducing the workforce by about 32.7 million full-time workers per year across the 52 countries analysed, which is equivalent to 0.62% of the total workforce on average across countries.

When the impact of alcohol consumption above the 1/1.5 drinks per day cap is translated into lost employment and productivity as measured by PPP-adjusted market wages, OECD countries lose on average USD PPP 344 per capita per year (see Figure 4.10), which is about 5.5 times as high as increases in health spending attributable to diseases caused by alcohol consumption. This is equivalent to a labour-related economic loss of about USD PPP 595 billion per year in OECD countries. This roughly corresponds to the annual GDP of Belgium or Sweden. The majority of costs are due to decreases in employment, while the effect on early retirement is small.

The impact on employment and productivity varies considerably across OECD countries: the lost labour market output is highest in Ireland, at almost USD PPP 882 per capita annually, while it is lowest in Turkey, at about USD PPP 23 per capita annually. In other modelled non-OECD countries – Cyprus, Indonesia and Saudi Arabia – the effect is even lower.

Finally, a brief comparison of the labour force productivity burden of alcohol consumption above the 1/1.5 drinks per day cap with the labour force productivity burden of any alcohol consumption is made in Box 4.6 and discussed further in Annex 4.B.

The impact of diseases caused by alcohol consumption on life expectancy, health expenditure, employment and productivity can be combined into one overall macroeconomic effect.3 To model this, the outputs of the business-as-usual scenario and the scenario in which alcohol consumption is capped at 1 drink per day for women and 1.5 drinks per day for men from the OECD SPHeP-NCDs model were fed into the OECD long-term economic model (Box 4.7). This model was used to understand the impact of diseases caused by alcohol consumption on GDP and on the overall tax rate.

On average in OECD countries, GDP will be 1.6% lower each year due to the impact of diseases caused by alcohol consumption above the 1/1.5 drinks per day cap (Figure 4.12). The impact in G20 and 24 EU countries is similar, at 1.4% for G20 and 1.9% for EU countries. The GDP impact varies by country: from 0% in Saudi Arabia and Turkey to nearly 4% in Lithuania. Across all the 48 countries included in the analysis,4 this equates to a total of USD PPP 1.6 trillion per year in the period 2020-50, which is similar to the average annual GDP of Canada or Spain. Importantly, these results do not take into account the fact that an increase in life expectancy due to drinking above the 1/1.5 drinks per day cap may mean that people will work for longer and retire later. If the retirement age is increased by two-thirds of a year for every year of additional life expectancy, the impact of diseases caused by alcohol consumption above the 1/1.5 drinks per day cap on GDP would double, with the average for OECD countries going from 1.6% to 3.4% (Annex Figure 4.B.8).

Another measure explored in the analysis of the long-term macroeconomic burden of diseases caused by drinking is fiscal pressure. Fiscal pressure is measured as government primary revenue (as a percentage of GDP) needed to stabilise the public debt ratio, and is equivalent to an overall tax rate (under the assumption that governments respond to rising fiscal pressure by raising additional revenue). Due to diseases caused by alcohol consumption above the 1/1.5 drinks per day cap, the tax rate will be 0.43 percentage points of GDP higher on average across OECD countries (Figure 4.13). The effect in G20 countries is 0.33 and in 24 EU countries is 0.48 percentage points of GDP.

The effect of the burden of alcohol-related diseases on fiscal pressure needs to be interpreted in the light of the potential loss of government tax revenue from a decrease in alcohol consumption. Findings from the model show that any alcohol consumption has a negative effect on fiscal pressure. More precisely, across OECD countries, the overall tax rate will be 0.56 percentage points of GDP higher, on average, owing to the consequences of medical conditions caused by any alcohol consumption (see Annex Figure 4.B.7). In comparison, potential losses in tax revenue from uncollected excise duties on alcohol are estimated at 0.25% of GDP on average across OECD countries (see Annex Figure 4.B.10), with variations from 0.05% of GDP or less in Austria, Switzerland and the United States, to 0.70% of GDP or more in Iceland, Estonia and Norway. In 36 countries included in this analysis, the burden on tax rate due to medical conditions caused by any alcohol consumption is greater than the potential loss in government revenue from alcohol excise duty. In addition, at lower levels of alcohol consumption, it is likely that government tax revenue from other goods and services would increase, as suggested by the analysis presented in Chapter 8. This could potentially compensate for the loss of revenue raised by value added tax on alcohol.

The impact of diseases caused by alcohol consumption on the overall tax rate can be translated into an equivalent impact on per capita taxes for the public. On average across OECD countries, every person will be subject to USD PPP 232 per year in additional taxes due to alcohol consumption above 1 drink per day for women and 1.5 drinks per day for men in 2020-50 (Figure 4.14).

Finally, it was noted in Section 4.2 that the model does not currently take into account the effect of alcohol consumption on a number of diseases. Likewise, the model does not capture such costs as greater spending on policing due to greater alcohol-related crime, the cost of property damage or the cost of the pain and suffering of victims of alcohol-related crimes. For example, in Sweden, crime-related costs of alcohol use were found to be comparable to the health care costs, and represented about 15% of total alcohol-related costs, both direct and indirect (Jarl et al., 2008[7]). In the United Kingdom (England), the crime-related costs of alcohol use represented GBP 11 billion in 2011 (more than USD PPP 15 billion) – about half of the total cost of alcohol use (House of Commons Health Committee, 2012[43]). Thus, the costs to the economy and to society shown in this chapter should be viewed as conservative.

Finally, a brief comparison of the GDP impact of alcohol consumption above the 1/1.5 drinks per day cap with the GDP impact of any alcohol consumption is made in Box 4.8 and discussed further in Annex 4.B.

Alcohol-related diseases and their broader societal implications carry considerable costs to both individuals and society over the next 30 years. Alcohol consumption above 1 drink per day for women and 1.5 drinks per day for men is associated with a number of diseases, and will reduce population-wide life expectancy by up to one and a half years in 2020-50. Countries will spend around 2.4% of their health care expenditure on treating alcohol-related diseases or injuries caused by drinking above the 1/1.5 drinks per day cap, and diseases caused by alcohol consumption above the caps will also have an impact on the labour market, effectively reducing the workforce by 33 million people across the 52 countries. In OECD countries, this will cost countries on average USD PPP 344 per capita per year in lost employment and productivity. Combined, the impact of diseases caused by alcohol consumption above the 1/1.5 drinks per day cap on life expectancy, health expenditure and labour market output will result in 1.6% lower GDP on average in OECD countries. As the overall tax rate increases, individuals face an equivalent tax of USD PPP 232 per year.

In addition to these economic costs, drinking has an impact on education – as described in Chapter 5 – which may result in further long-term effects on employment and productivity. It is therefore crucial to invest in prevention and treatment of harmful drinking and to reduce its burden on individuals and society. Countries have implemented a number of policies and interventions to prevent and reduce harmful drinking, which are described in Chapter 6. Chapter 7 uses the OECD SPHeP-NCDs model to assess the cost-effectiveness of a number of these policies to understand their impact on the health and economic burden of harmful drinking.

References

[30] Ahmed, A., A. Karter and J. Liu (2006), “Alcohol consumption is inversely associated with adherence to diabetes self-care behaviours”, Diabetic Medicine, Vol. 23/7, pp. 795-802, http://dx.doi.org/10.1111/j.1464-5491.2006.01878.x.

[10] Balakrishnan, R. et al. (2009), “The burden of alcohol-related ill health in the United Kingdom”, Journal of Public Health, Vol. 31/3, pp. 366-373, http://dx.doi.org/10.1093/pubmed/fdp051.

[4] Canadian Substance Use Costs and Harms Scientific Working Group (2020), Canadian Substance Use Costs and Harms 2015-2017, Canadian Centre on Substance Use and Addiction, Ottawa, https://www.ccsa.ca/sites/default/files/2020-06/CSUCH-Canadian-Substance-Use-Costs-Harms-Report-2020-en.pdf (accessed on 5 November 2020).

[20] Cecchini, M., M. Devaux and F. Sassi (2015), Assessing the Impacts of Alcohol Policies: A microsimulation approach, OECD, Paris, https://doi.org/10.1787/18152015.

[12] Cesur, R. and I. Kelly (2014), “Who pays the bar tab? Beer consumption and economic growth in the United States”, Economic Inquiry, Vol. 52/1, pp. 477-494, http://dx.doi.org/10.1111/ecin.12048.

[11] Cortez-Pinto, H. et al. (2010), “The burden of disease and the cost of illness attributable to alcohol drinking: Results of a national study”, Alcoholism Clinical and Experimental Research, Vol. 34/8, pp. 1442-1449, http://dx.doi.org/10.1111/j.1530-0277.2010.01229.x.

[35] Devaux, M. and F. Sassi (2015), “The labour market impacts of obesity, smoking, alcohol use and related chronic diseases”, OECD Health Working Papers, No. 86, OECD, Paris, https://dx.doi.org/10.1787/5jrqcn5fpv0v-en.

[37] Feng, W. et al. (2001), “The impact of problem drinking on employment”, Health Economics, Vol. 10/6, pp. 509-521, http://dx.doi.org/10.1002/hec.611.

[3] Fischer, B. et al. (2014), Alkoholbedingte Kosten in der Schweiz (Alcohol-related Costs in Switzerland), Polynomics, https://www.suchtmonitoring.ch/docs/library/fischer_mhrf7rn7ju.pdf.

[25] GBD 2016 Alcohol Collaborators (2018), “Alcohol use and burden for 195 countries and territories, 1990-2016: A systematic analysis for the Global Burden of Disease Study 2016”, Lancet, Vol. 392/10152, pp. 1015-1035, http://dx.doi.org/10.1016/S0140-6736(18)31310-2.

[19] GBD 2016 Risk Factors Collaborators (2017), “Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016”, Lancet Global Health Metrics, Vol. 390/10100, pp. 1345-422, http://dx.doi.org/10.1016/S0140-6736(17)32366-8.

[18] Global Burden of Disease Collaborative Network (2016), Global Burden of Disease Study 2015 (GBD 2015) Risk Factor Results 1990-2015, Institute for Health Metrics and Evaluation (IHME), Seattle, WA, http://ghdx.healthdata.org/record/ihme-data/gbd-2015-risk-factor-results-1990-2015 (accessed on 9 December 2020).

[41] Guillemette, Y. and D. Turner (2017), “The fiscal projection framework in long-term scenarios”, OECD Economics Department Working Papers, No. 1440, OECD, Paris, https://dx.doi.org/10.1787/8eddfa18-en.

[43] House of Commons Health Committee (2012), Government’s Alcohol Strategy, Volume I: Report, together with formal minutes and oral and written evidence, The Stationery Office, London, https://publications.parliament.uk/pa/cm201213/cmselect/cmhealth/132/132.pdf (accessed on 6 November 2020).

[17] IARC (2020), Global Cancer Observatory, International Agency for Research on Cancer, Lyon, https://gco.iarc.fr/ (accessed on 9 November 2020).

[16] IHME (2020), Epi Visualization | IHME Viz Hub, Institute for Health Metrics and Evaluation, Seattle, WA, https://vizhub.healthdata.org/epi/ (accessed on 21 December 2017).

[39] Jarl, J. and U. Gerdtham (2012), “Does drinking affect long-term sickness absence? A sample selection approach correcting for employment and accounting for drinking history”, Applied Economics, Vol. 44/22, pp. 2811-2825, http://dx.doi.org/10.1080/00036846.2011.566206.

[7] Jarl, J. et al. (2008), “The societal cost of alcohol consumption: An estimation of the economic and human cost including health effects in Sweden, 2002”, European Journal of Health Economics, Vol. 9/4, pp. 351-360, http://dx.doi.org/10.1007/s10198-007-0082-1.

[9] Johnston, M., A. Ludbrook and M. Jaffray (2012), “Inequalities in the distribution of the costs of alcohol misuse in Scotland: A cost of illness study”, Alcohol and Alcoholism, Vol. 47/6, pp. 725-731, http://dx.doi.org/10.1093/alcalc/ags092.

[8] Konnopka, A., M. Bodemann and H. Konig (2011), “Health burden and costs of obesity and overweight in Germany”, European Journal of Health Economics, Vol. 12/4, pp. 345-352, http://dx.doi.org/10.1007/s10198-010-0242-6.

[14] Konnopka, A. and H. Konig (2007), “Direct and indirect costs attributable to alcohol consumption in Germany”, Pharmacoeconomics, Vol. 25/7, pp. 605-618, http://dx.doi.org/10.2165/00019053-200725070-00006.

[5] Kopp, P. (2015), Le Coût Social des Drogues en France, Observatoire Français des Drogues et des Toxicomanies, Paris, https://www.ofdt.fr/BDD/publications/docs/eisxpkv9.pdf.

[31] Lange, S. et al. (2017), “Global prevalence of fetal alcohol spectrum disorder among children and youth: A systematic review and meta-analysis”, JAMA Pediatrics, Vol. 171/10, pp. 948-956, http://dx.doi.org/10.1001/jamapediatrics.2017.1919.

[15] Lima, E. and T. Esquerdo (2003), “The economic costs of alcohol misuse in Portugal”, Working Paper Series No. 24, Núcleo de Investigação em Microeconomia Aplicada Universidade do Minho.

[33] Lupton, C., L. Burd and R. Harwood (2004), “Cost of fetal alcohol spectrum disorders”, American Journal of Medical Genetics, Vol. 127C/1, pp. 42-50, http://dx.doi.org/10.1002/ajmg.c.30015.

[38] MacDonald, Z. and M. Shields (2004), “Does problem drinking affect employment? Evidence from England”, Health Economics, Vol. 13/2, pp. 139-155, http://dx.doi.org/10.1002/hec.816.

[36] Marchand, A., A. Parent-Lamarche and M. Blanc (2011), “Work and high-risk alcohol consumption in the Canadian workforce”, International Journal of Environmental Research and Public Health, Vol. 8/7, pp. 2692-2705, http://dx.doi.org/10.3390/ijerph8072692.

[26] Naimi, T. et al. (2017), “Selection biases in observational studies affect associations between ‘moderate’ alcohol consumption and mortality”, Addiction, Vol. 112/2, pp. 207-214, http://dx.doi.org/10.1111/add.13451.

[22] OECD (2019), SPHeP-NCDs Technical Documentation, OECD, Paris, http://oecdpublichealthexplorer.org/ncd-doc.

[42] OECD (2018), OECD Economic Outlook, Volume 2018 Issue 2, OECD Publishing, Paris, https://dx.doi.org/10.1787/eco_outlook-v2018-2-en.

[34] Popova, S. et al. (2016), “The economic burden of fetal alcohol spectrum disorder in Canada in 2013”, Alcohol and Alcoholism, Vol. 51/3, pp. 367-375, http://dx.doi.org/10.1093/alcalc/agv117 (accessed on 6 November 2020).

[32] Popova, S. et al. (2017), “Estimation of national, regional, and global prevalence of alcohol use during pregnancy and fetal alcohol syndrome: A systematic review and meta-analysis”, Lancet Global Health, Vol. 5/3, pp. E290-E299, http://dx.doi.org/10.1016/S2214-109X(17)30021-9.

[13] Rehm, J. et al. (2007), “The costs of alcohol, illegal drugs, and tobacco in Canada, 2002”, Journal of Studies on Alcohol and Drugs, Vol. 68/6, pp. 886-895, http://dx.doi.org/10.15288/jsad.2007.68.886.

[21] Rehm, J. et al. (2003), “The relationship of average volume of alcohol consumption and patterns of drinking to burden of disease: An overview”, Addiction, Vol. 98/9, pp. 1209-1228, http://dx.doi.org/10.1046/j.1360-0443.2003.00467.x.

[24] Roerecke, M. and J. Rehm (2010), “Irregular heavy drinking occasions and risk of ischemic heart disease: A systematic review and meta-analysis”, American Journal of Epidemiology, Vol. 171/6, pp. 633-644, http://dx.doi.org/10.1093/aje/kwp451.

[6] Saar, I. (2008), “The social costs of alcohol misuse in Estonia”, European Addition Research, Vol. 15/1, pp. 56-62, http://dx.doi.org/10.1159/000173010.

[23] Sassi, F. (ed.) (2015), Tackling Harmful Alcohol Use: Economics and Public Health Policy, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264181069-en.

[27] Stockwell, T. et al. (2016), “Do ’moderate’ drinkers have reduced mortality risk? A systematic review and meta-analysis of alcohol consumption and all-cause mortality”, Journal of Studies on Alcohol and Drugs, Vol. 77/2, pp. 185-198, http://dx.doi.org/10.15288/jsad.2016.77.185.

[40] Thørrisen, M. et al. (2019), “Association between alcohol consumption and impaired work performance (presenteeism): A systematic review”, BMJ Open, Vol. 9/7, p. e029184, http://dx.doi.org/10.1136/bmjopen-2019-029184.

[28] WHO (2018), European Health Information Gateway: Premature mortality, World Health Organization, Geneva, https://gateway.euro.who.int/en/indicators/h2020_1-premature-mortality/ (accessed on 16 July 2020).

[1] WHO (2018), Global Status Report on Alcohol and Health 2018, World Health Organization, Geneva, https://apps.who.int/iris/bitstream/handle/10665/274603/9789241565639-eng.pdf?ua=1 (accessed on 10 September 2019).

[2] WHO (2014), Global Status Report on Alcohol and Health 2014, World Health Organization, Geneva, https://apps.who.int/iris/bitstream/handle/10665/112736/9789240692763_eng.pdf?sequence=1.

[29] World Bank (2020), DataBank: The World Bank, World Bank Open Data, https://databank.worldbank.org/.

Analyses carried out with the OECD model use data from the Global Burden of Disease Study (GBD 2016 Alcohol Collaborators, 2018[25]) and account for a protective effect of alcohol consumption on ischaemic CVDs and diabetes for some age groups. However, these effects are debated: some studies conclude that there is no protective cardiovascular effect once lifetime abstainers are separated from those who quit and do not drink for health reasons (Naimi et al., 2017[26]; Stockwell et al., 2016[27]). To account for this uncertainty around relative risks, a sensitivity analysis was carried out to take out any protective effect. Results from the modified version of the model conclude that under the assumption of no protective effect of alcohol consumption:

  • Any alcohol consumption continues causing greater population health harms than drinking above the 1/1.5 drinks per day cap. For instance, any alcohol consumption contributes to lowering life expectancy by 1.1 years, compared to 0.8 years for the burden of drinking above the 1/1.5 drinks per day cap (Annex Figure 4.A.1).

  • Medical conditions caused by drinking any alcohol lead to higher medical spending (USD PPP 58 per capita per year, in OECD countries) than only drinking above the 1/1.5 drinks per day cap (USD PPP 52 per capita per year, in OECD countries) (Annex Figure 4.A.2).

  • Medical conditions caused by any alcohol drinking contribute to a loss of employment and productivity (USD PPP 506 per capita per year); this is higher than drinking above the 1/1.5 drinks per day cap (USD PPP 334 per capita per year).

  • Estimations of the burden of any alcohol drinking obtained in the sensitivity analysis are higher than those from the analysis assuming some protective effects.

This annex presents results comparing the burden of disease caused by any alcohol consumption to the burden caused by consumption above 1 drink per day for women and 1.5 drinks per day for men (discussed in the main part of this chapter).

Annex Figure 4.B.1 shows the impact on disease incidence caused by any alcohol consumption, compared to consumption above the 1/1.5 drinks per day cap. The burden caused by any alcohol consumption includes an additional 152 million cases of dependence, 48 million cases of injury and 10 million cases of cancers related to alcohol over the next 30 years in all 52 countries, compared to alcohol consumption above the 1/1.5 drinks per day cap. At the same time, the OECD SPHeP-NCDs model also predicts, in the same scenario, 69 million extra cases of CVDs and 17 million extra cases of diabetes. The incidence of cases of MSDs, which are currently assumed to be unrelated to alcohol consumption, is also predicted to increase, mainly owing to people living longer.

Compared to alcohol consumption above the 1/1.5 drinks per day cap, the impact of any alcohol consumption on the rate of premature death is greater, with an additional 4.2 people per 100 000 population dying prematurely each ear across all 52 countries (Annex Figure 4.B.2). Thus, despite some protective effect that alcohol consumed at lower levels might have on the incidence of certain diseases, findings show that alcohol consumed at any level will contribute to the risk of dying early.

In addition, any alcohol consumption leads to an extra two-month drop in life expectancy across the 52 countries modelled in this study compared to alcohol consumption above the 1/1.5 drinks per day cap (Annex Figure 4.B.3). It should be noted that life expectancy estimates apply to all people, and not just those who consume alcohol. For alcohol drinkers only, the effect on life expectancy is stronger.

The burden on health expenditure from any alcohol consumption is smaller than that found in the scenario capping alcohol consumption at 1 drink per day for women and 1.5 drinks per day for men. On a per capita basis, the average annual medical costs caused by drinking above the 1/1.5 drinks per day cap would be about USD PPP 49, while the costs caused by any alcohol consumption would be about USD PPP 40 (i.e. 19% lower) in 2020-50 (Annex Figure 4.B.4). The main reason for this difference is that in the scenario assessing the burden of any alcohol consumption, any potential protective effects of alcohol consumption at lower levels are eliminated. In addition, decreases in life expectancy are greater in the scenario assessing the burden of any alcohol consumption compared to the scenario assessing the burden of consumption above the 1/1.5 drinks per day cap (see Annex Figure 4.B.3). This further contributes to lower medical expenditure.

Finally, of interest is a comparison between the impact on employment and productivity of diseases caused by any alcohol consumption compared to consumption above the 1/1.5 drinks per day cap. USD PPP 125 extra annual per capita PPP-adjusted market wages can be attributed to any alcohol consumption compared to consumption above the 1/1.5 drinks per day cap, which is approximately equivalent to 45% extra damage (see Annex Figure 4.B.5). The main reason for the difference in results between the labour market and health expenditure outcomes is that labour market outputs are more highly correlated with productivity in the prime years, when people are still employed.

Notes

← 1. One drink in this report refers to the equivalent of 12 grammes of pure alcohol.

← 2. Health expenditure measures the final consumption of health care goods and services for personal health care, including curative care, rehabilitative care, preventative care, ancillary services and medical goods but not long-term care.

← 3. The calculation of the cost presented in this report does not take into account some dimensions. For example, the analysis does not include the following costs: i) the cost of justice (e.g. alcohol-related violence and injuries); ii) expenditure on lobbying and litigation to avoid the implementation of policies incurred by the industry; iii) the cost to counter industry-led actions incurred by the government and civil society organisations; iv) the social burden of alcohol use related to, for example, unwanted teenage pregnancies and the long-term consequences of foetal alcohol syndrome; and v) broader factors related to social bonding and pleasure of drinking in moderation, maintenance of the landscape and vineyards, tourism, and potential population resistance to stringent policy decisions.

← 4. The analysis of the impact on GDP includes 48 countries, while the analysis of the impact on fiscal pressure covers 46 countries. Four countries were not included in the OECD long-term economic model and could not be included in the analysis of the impact on GDP (Croatia, Cyprus, Malta and Peru). For the same reason, two additional countries (Colombia and Costa Rica) could not be included in the analysis of the impact on fiscal pressure.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2021

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.