Medical practice

Global consensus on safe drinking limits 'required'

The Guardian reports that there is “little agreement … between countries on what is considered safe or sensible alcohol consumption”.

This story is based on a survey of international alcohol consumption guidelines, which found that there is considerable variation in a number of important alcohol-related recommendations, such as:

  • how much alcohol is in a unit
  • what is an acceptable daily or weekly consumption limit
  • the difference between what a woman can safely drink compared to a man
  • how much alcohol a woman can drink if pregnant or breastfeeding
  • the legal blood alcohol limit when driving – this latter finding is an important reminder for those who drive abroad that the laws aren’t the same everywhere – as in some countries, having any alcohol in your blood while driving is illegal

The researchers suggest that, based on what they found, standardised international guidelines may make sense, and they could include the following recommendations:

  • women should drink no more than two standard drinks per day. With a standard drink equivalent to 10g ethanol, this would mean no more than 20g of ethanol (or 2.5 UK units)
  • men should drink no more than three standard drinks (equivalent to 30g of ethanol) per day (or 3.75 UK units)
  • women should drink no more than 12 standard drinks per week (15 UK units)
  • men should drink no more than 18 standard drinks per week (22.5 UK units)
  • women and men should have at least one alcohol-free day per week
  • motor vehicle drivers should not consume any alcohol
  • pregnant and breastfeeding women should not consume any alcohol

These recommendations differ slightly from the current UK recommendations – see box for more information.

Where did the story come from?

The study was carried out by researchers from the University of Sussex. No sources of funding were reported. The study was published in the peer-reviewed journal: Drug and Alcohol Review.

The Guardian and Daily Express cover this study appropriately.

What kind of research was this?

This was a cross-sectional survey looking at alcohol consumption guidelines from countries around the world. The researchers say that many governments have developed guidelines on what constitutes “low-risk drinking”.

However, there are differences between countries in what is considered a standard drink, and this makes it difficult to compare results of alcohol-related research between countries. These differing guidelines may also be confusing for individuals.

The researchers wanted to look at alcohol consumption guidelines from different countries to see whether there was agreement on:

  • definitions of standard drinks
  • guidelines for alcohol intake
  • legal levels of alcohol consumption for drivers of motor vehicles
  • safe levels of alcohol consumption for pregnant women

This approach is a good way to get an overview of what different countries recommend.

What did the research involve?

The researchers searched government websites for 57 countries:

  • the 27 European Union member countries
  • five additional European countries
  • five countries each from Africa, the Americas, Asia, the Middle-East, and Oceania

The guidelines were only included if their recommendations could be expressed in terms of grams of ethanol (alcohol). They did not include non-governmental guidelines. Where ranges were given, or different regions within a country had different recommendations, the lowest limits were selected.

The World Health Organization’s (WHO) Global Status Report on Road Safety provided information on alcohol limits for drivers in 145 countries.

What were the basic results?

Twenty-seven countries had official low-risk drinking guidelines that could be expressed as grams of ethanol.

Many other countries had guidelines that could not be expressed in this way, for example, because they encourage moderate consumption and/or abstinence in certain circumstances, but do not define what this constitutes.

Some countries did not have readily accessible alcohol guidelines, including eight EU member states.

Recommended maximum intakes

There was variation in what was considered a ‘standard drink’ or ‘unit of alcohol’, ranging from 8g of ethanol in the UK to 14g in Slovakia (the UK unit of alcohol is equivalent to half a pint of standard strength lager).

More guidelines expressed limits in terms of daily amounts than weekly amounts. Recommended maximum limits ranged from:

  • 20g to 56g ethanol daily for men
  • 10g to 42g ethanol daily for women
  • 160g to 280g ethanol weekly for men
  • 80g to 140g ethanol weekly for women

The ratio of recommended maximum limits for men and women also varied, with women’s limits ranging from the same as men’s to half men’s limits. Where both a daily and a weekly limit were given for a country, the weekly limit was between three and seven times the daily limit. Some countries recommended having some alcohol-free days, or reducing daily consumption if drinking every day of the week.

Alcohol and driving

Of 145 countries, 14% (21 countries) reported on by the WHO allow no blood alcohol content (BAC) when driving. The countries which allow some BAC vary tenfold in what they allow.

Alcohol and pregnancy and breastfeeding

Only 14 countries had government advice on alcohol consumption in pregnant or breastfeeding women. All of these countries recommend that it is safest not to drink alcohol in pregnancy. Several countries highlight that if women do choose to drink in pregnancy, then they should limit the number of drinks per day and the number of drinking days per week. Several countries also note that women should not drink alcohol during the first few months of a baby’s life, and that they should not drink alcohol if they are planning to become pregnant.

How did the researchers interpret the results?

In summary, the researchers say that there is international variation in what was considered harmful or excessive alcohol consumption on a daily or weekly basis and when driving. There is also no agreement on what the appropriate ratio of consumption threshold for men and women should be. They conclude that international consensus on low-risk drinking guidelines is an important goal which could help people to make better-informed choices about alcohol consumption.


This research provides interesting insight into variability in alcohol consumption guidelines worldwide.

As the authors note, some of the variability they found could be due to a lack of clarity or consistency from the research evidence in terms of what levels of alcohol increase different types of risks. This may be due to different levels of alcohol having different effects on different short- and long-term outcomes (for example, cardiovascular risk, cancer risk, or injury risk), as well as differing effects of different patterns of drinking (such as binge drinking versus regular lower-level consumption).

In addition, the research is itself complicated by the use of differing international definitions of what constitutes a standard drink (or unit) and how recommended alcohol intake is defined in different countries.

It was unclear how the searches for guidelines dealt with non English-language government websites. The authors also suggest that even international standardisation of recommendations may not be enough to change people’s behaviour. However, they argue that these standards are still important for those people who do wish to moderate their intake.

Overall, there are complexities to the research evidence which mean that a certain amount of expert judgement is needed when setting recommended maximum limits. Different countries may interpret the evidence differently and therefore make differing judgements, particularly in light of different cultural contexts.

For example, they may set different safe thresholds depending on whether their priority is a short-term risk, such as death from road traffic accidents, or long-term risks such as liver or heart disease.

The authors conclude with a set of their own recommendations (as listed in the introduction), which, with some minor differences, are broadly similar to the current UK recommendations.

NHS Attribution