Wednesday, 05 September 2018 13:14

Life is a risky affair wine/alcoholic beverages and disease: is a risk-free consumption possible?

A large study in the renowned journal The Lancet last week caused uncertainty among all wine drinkers.

According to the publication, even very moderate drinking is worse than not drinking at all. This conclusion seems to contradict a large body of other studies which have found that moderate drinking is associated with a healthier and longer life expectancy and lower cardiovascular events.


A recent study[1] - published in the Lancet- examining the consumption of alcoholic beverages and disease risk concluded that even light alcohol consumption has no benefits and thus meaning that no safe level of alcohol exists.

Based on the results of 694 sources of data and 592 studies, the authors of the GBD study concluded that the health risks, in particular those for cancer, outweigh any benefits of the moderate consumption of alcoholic beverages; meaning thus that in total there are no health benefits.

According to study, the consumption of alcohol should be regulated by policy makers to reduce the average alcohol consumption of the entire population.

Commentary by the Wine Information Council

  • The GDB-study is not a new, independent study but a statistical model which is based on collected data from previous studies.
  • Consequently, the GBD-study does not show any new scientific evidence, but confirms the results of previous studies: with increasing consumption, certain health risks increase (particularly the risk of certain cancers), on the other hand, however, with light to moderate consumption, health benefits (especially in cardiovascular disease and diabetes) have been observed.
  • The GDB-study claims to present a global estimate of the alcohol effects and presumes in its model that the same amounts of alcohol in all countries worldwide have the same effects, despite different environmental and living conditions as well as nutritional factors. The problem is that data from many divergent cultures were combined to determine a single association between the consumption of alcoholic beverages and health.
  • Contrary to the previous long-term studies and meta-analyses*, which show a decreased risk of the most common cause of death (heart attack) and total mortality, the GBD-study pools ALL more or less alcohol-related health risks. This is done with the same risk measure (relative risk) for rare diseases and health risks (such as tuberculosis) in industrialised countries as well as for the most common causes of death (such as coronary heart diseases, diabetes and certain cancers) – but total mortality has not been presented.
  • The results of the previous meta-analyses in industrialised countries investigating the health benefits of moderate wine/alcohol consumption are still valid for the residents of these countries and present a higher evidence level than the current model, the results of which are based on estimates of global data – including those of developing countries.
  • The policy makers should not use global average data for the risk assessment but data that are relevant for the respective population.
  • Considering all the evidence, this GBD-study gives no reason to question the established drinking guidelines. Light to moderate consumption of up to 20 g of alcohol per day for women and up to 30 g per day for men (considering the individual circumstances and contraindications) within the international average and does not need to be adjusted. Consequently, the drinking guidelines used in Wine in Moderation are compatible with the current scientific evidence and in line with the international average of drinking guidelines.**

*Meta-analysis: Summary and quantitative and qualitative analysis of several primary studies

** These are generic guidelines; individuals should always consult their physician or family doctor if they have any doubts related to their drinking patterns and their health.


Appendix with detailed information:

What was examined?

The Institute for Health Metrics and Evaluation (IHME, Seattle, funded by the Bill & Melinda Gates Foundation) built a large and complex statistical model from a vast range of data sources (using estimates of drinking prevalence, average consumption and the burden of disease attributable to alcohol use) and analysed the data.

The researchers examined:

  • 694 previous studies to estimate the global prevalence of alcohol consumption
  • 592 studies to study the health risks associated with alcohol (in 195 countries between 1990 and 2016).


  • The report states that 2.8 million deaths (4.85% of all deaths) worldwide in 2016 were alcohol-related.
  • The study examined the effects of alcohol consumption on a list of only 23 alcohol-related health outcomes, as opposed to most previous research examining the impact of alcohol consumption on overall health and all-cause mortality. Despite the limitation to certain diseases, only a small increase in relative risk was observed at low levels of drinking.
  • Researchers found that pooling the 23 outcomes indicates a 0.5% increase in relative risk for moderate drinkers (defined as people consuming up to one standard drink that is, 10g of alcohol every day), when compared to abstainers. This risk increased to 7% for people who consumed two drinks per day, and to 37% for consumption of five drinks per day.
    In absolute numbers, this means that:
    • For each set of 100,000 people who have one drink a day per year, 918 can expect to experience one of the 23 alcohol-related problems in any year.
    • Of those who drink nothing, 914 will have an issue whether they drink an alcoholic beverage or not.
    • This means that 99,082 individuals are not affected.
    • Only 4 in 100,000 individuals who consume a drink a day may have a problem caused by the drinking, according to this study.
    • At two drinks per day, the number of individuals experiencing a problem increased to 977.
  • These findings do not invalidate previous epidemiological studies that have found health benefits at low levels of consumption but confirm these results (J-shaped curve) as the example of diabetes and coronary heart disease in Figure 1 (from the appendix of the GBD-study) shows. Even at five drinks per day, which is an excessive amount, the vast majority of people are unaffected.

Figure 1 Relative risk of diabetes and coronary heart disease and alcohol consumption (red square: moderate consumption of alcoholic beverages) (1)



However, the authors of the present study stated that the protective effect of alcoholic beverages for some health conditions was offset by the risks for others and since overall the health risks increased in line with consumption, concluded that there is therefore no safe level of alcohol consumption (see figure 5 from the paper reproduced).



According to the scientific Board of the Wine Information Council, the study has several limitations:

  1. Estimates instead of measurements:

While it brings together a very large number of data sources, it nonetheless relies on assumptions to model figures for where data are lacking. I.e. the presented consumption levels are based on sales estimates.

These assumptions are not equally appropriate for all contexts, and the approach results in some of the estimates being imprecise.

  1. Drinking pattern and difference between alcoholic beverages not considered

The pattern of drinking (regular moderate versus binge drinking), the drinking with or without food, the type of beverage (e.g., wine versus spirits), the intention of the individual (drinking to get drunk versus drinking to enhance meals) was not taken into consideration.

  1. No measurable clinical end-points

The outcomes are no measurable clinical endpoints (such as death, heart attack, stroke, diabetes complications, etc) but a randomly defined heterogeneous group of alcohol-attributable health outcomes (i.e.***)

  1. Global instead of country-specific aspects

The mortality and DALY (“one year of lost health”) estimates included are useful for tracking trends and comparing between countries. However, they are less useful when pooled and interpreted at the global level.

  • Types of risk and disease burden vary significantly across countries, reflecting national health profiles, life expectancies, income levels and other socio-demographic factors.
  • The cultural context, eating and drinking patterns as well as lifestyle factors are lacking in this analysis. Many lifestyle and cultural factors strongly modify the relation of a given amount of alcohol to health and diseases.  Such modifying factors include especially the socio-economic status of the individual, the level of physical activity, etc. When these factors are not considered, the estimated intake of a certain amount of alcohol, provides an incomplete assessment of the effects of alcohol on health.
  • Also, countries with the highest prevalence of alcohol consumption are also those with the highest life expectancies, and those with the lowest consumption have the lowest life expectancies. This demonstrates that health outcomes are the result of many factors and are best interpreted at the national level.
  1. Global average values not relevant for policy members:

Policymakers seeking to prevent alcohol-related harm in their countries may be better informed by the results of research conducted in their specific context than by findings that rely partly on global averages.

  1. Theoretical determination of a "no safe level"

The concept of ‘no safe level’ is theoretical and arbitrarily set and could be applied to almost any activity or lifestyle factor.

  1. Putting percentages/numbers in perspective

Only 4.85% of all deaths are alcohol-related.


What significance do the results have for moderate wine drinkers? Are the results of this study an argument to abstain from alcohol?

Prof. Sir David Spiegelhalter (Winton Centre for Risk and Evidence Communication)

The risks of alcohol (again)

Life is a risky affair! Putting the results into perspective!

The risks of moderate consumption of alcoholic beverages are very low.

A "no safe level" is not an argument for abstaining from alcoholic beverages.

Also in other situations, there is "no safe level": There is no safe level of driving, but government do not recommend that people avoid driving.

Presumably people who choose to drink alcohol moderately get some pleasure from it, and any risk needs to be traded off against this enjoyment.

When it comes to policy recommendations, it is notable that the authors recommend public health measures to reduce total consumption at a population level, such as "excise taxes on alcohol, controlling the access to and availability of alcohol and the hours of sale, and controlling alcohol advertising". There is no mention of information campaigns or targeting heavy drinkers, which may be less effective ways of reducing average consumption.



***e.g. Hypertensive and coronary heart disease, diabetes, liver cirrhosis, hemorrhagic stroke tuberculosis, cancer, interpersonal violence, injuries)

[1] Source: Alcohol use and global burden of disease, 1990-2016; GBD 2016 Alcohol Collaborators Study, August 2018;



The authors have taken reasonable care in ensuring the accuracy of the information herein at the time of publication and are not responsible for any errors or omissions. Read more on our disclaimer and Privacy Policy.