Alcohol how much, how harmful?

by | Last updated May 2, 2020 | Published on Mar 8, 2020 | Odds & Ends

How much is safe?

The literature concerning how much it is safe or desirable to drink is contaminated with opinions from people who only ‘see’ the evidence they want to believe, rather than objectively assessing the evidence that is out there [1-7].

The latest evidence about the health-risks of alcohol from a comprehensive analysis published in the Lancet in 2018 [1] comes to the firm conclusion:

The total attributable burden of alcohol use was larger than

previous evidence has indicated… the level of

consumption that minimises health loss due to alcohol use is

zero. These findings strongly suggest that alcohol control

policies should aim to reduce total population-level

consumption. … there is a need for countries to revisit

their alcohol control policies and assess how they can be

modified to further lower population-level consumption.

Such is the post-truth world, and such has always been the nature of the discussion about alcohol. It is significantly driven by moral crusaders and religious contributors on the one hand, and those trying to sell alcoholic drinks on the other. Many views are influenced more by preconceptions, ignorance, and biases, than by evidence.

The alcohol lobby, represented by fronts like ‘Drinkwise’ (industry funded) has been spinning the information to the public about the health problems caused by alcohol, using tactics similar to those used by the tobacco industry.

Long ago, the WHO classified alcohol as a Group 1 carcinogen (with tobacco): the degree of risk of alcohol-related cancer is hundreds of times higher than for most ‘environmental’ carcinogens that people worry about (the current popular favourite being glycophosphate, Roundup™). The arbitrary life-time risk-limit for substances to be labelled as environmental ‘toxins’ is taken as 1 in 1,000,000.

Alcohol is about 1,000 times more significant as a carcinogen than Roundup.

Relative risks

The lifetime risk of dying in a traffic accident is around 1 in 100.

In the USA you stand a 1% chance of being shot dead by the time you reach 60 years old (40,000 gun-deaths p/a, population 327 million).

Suicide 1 in 100

Falling over 1 in 100

In relation to taking risks, if you are of puritanical inclination, then that will influence the type and degree of risk you are inclined to tolerate; if you hold to a different outlook and consider yourself closer to a hedonist, then you are likely to consider things differently.

Summarised in simple terms, one has to solve the equation which answers the question, ‘how much enjoyment can you get out of life without undue risk, including food (most people in the ‘West’ are considerably overweight) and alcoholic beverages, and how much is that going to impair your ‘wellbeing’ and health and shorten your life’ — assuming something else does not shorten it first — splat! where did that bus come from?

Medical discussions often use a benchmark of what is considered to be an ‘acceptable risk’ of premature death; that is generally taken as a level of one death for every 100 people as a guide.

I have sympathy with the opinion expressed by the famous French gastronome, Brillat-Savarin, who said ‘a meal without wine is like a day without sunshine’.

As Winston Churchill said; ‘I have taken more of alcohol than alcohol has taken out of me’.

On his way up to Scapa Flow to review the fleet during the war he asked his scientific adviser, Professor Lindeman, to calculate the depth to which all champagne he had drunk in his life would fill up the carriage they were in. For the purposes of this calculation he gave Lindeman the figure of half a bottle a day.

Basic stats on alcohol and health

Life-time risk (LTR)

Any level of drinking is associated with an increased risk of ill-health and injury and premature death.

In Australia, around 8 per cent of the population drink daily, around 41 per cent drink weekly.

All of the statistics given in this commentary have been rounded off to make them easy to conceptualise and remember, especially because the data on which they are based simply does not justify the spurious degree of accuracy to which many of these statistics are given.

Alcohol accounts for about 13 per cent of all deaths among 14-17-year-old Australians: for 15-34-year-olds, it accounts for most drug-related deaths and hospital episodes, that is more than all illicit drugs combined. That statistic alone reveals how absurd and unbalanced the illicit drug policies of most western countries presently are.

See Center for Disease Control data

Death (life-time risk LTR)

The commonest causes of disease related death (per 100,000 persons) in wealthier societies are:

Ischaemic heart disease 135

Stroke 120

Chronic obstructive pulmonary disease 50

Respiratory system cancer 33

Breast cancer 30, (all cancers 250)

Respiratory infections 26

Diabetes 20,

Alzheimer’s 20,

Liver cancer 20

Stomach cancer 20

Suicide 10

That suggests that if you run to the pub and back, the improvement in your health from the exercise thus taken will be greater than the impairment of your health from the alcohol that you drink whilst there. There is probably enough data now to calculate exactly health how far you would have to live away from the pub for this to be true!

Reference comparison

Guns and cars

Traffic accident deaths are around 1000 per 100,000 persons/year, i.e. about a 1% Life-time risk (LTR). And (at least in the USA) one has about a 1% LTR of being shot dead (40,000 gun-deaths p.a., population 327 million).

Unintentional poisonings and unintentional falls account for well over 2% of deaths, and Drug overdoses for another 1%.

The proportion of all deaths, attributable to alcohol is estimated at ~5% [8]. That is to say, out of every 20 people who die, one of them dies because of alcohol (but, of course, they might have died the next day of something else). The number of years of life likely to remain, and wellbeing and health during those years, is also an important dimension to consider. Some causes of death have a relatively small impact on that figure, compared to a road traffic death of a teenager.

In Europe overall ‘alcohol attributable mortality’ is: 20-40/100,000 (fig 72, [9].

Five categories account for 90% of the total alcohol attributable mortality

cardiovascular ~40/100,000

Injury (e.g. traffic injury and violence) ~10/100,000

cirrhosis ~6/100,000

cancers ~4/100,000

Mortality attributable to alcohol, by disease type

causal relationship is thought to exist with;

  1. Cancer; nasopharyngeal, oesophageal, laryngeal, liver, pancreatic, breast, colorectal
  2. Infectious: tuberculosis, pneumonia
  3. Liver cirrhosis, pancreatitis
  4. Cardiovascular: overall neutral, moderate intake decreases morbidity & mortality in ischaemic disorders, but for other causes the impact is negative and dose-related (hypertension, haemorrhagic stroke, atrial fibrillation).

Alcohol attributable mortality

Total Life-time risk (LTR) by daily consumption is nearly the same for both genders up to 60 ml/d:

35 ml/day or less, risk is less than 1% (0.01%/year)

50 ml/d 2%

60 ml/d 3%

Above 60 ml/d the risk for females is higher:

90 ml/d M = 5%, F = 8%,

120 ml/d M = 6%, F = 10%.

Overall alcohol-related death risk at 100 ml/d is 3%, of which 0.7% is disease and 2.3% injury.

Alcohol, dose-related effects

Cirrhosis life-time risk [10]: non-drinker base rate 6/100,000 (0.006%)

25 ml/d = 10/100,000 (i.e. 0.01% LTR)

50 ml/d = 15/100,000 (0.015%)

75 ml/d = 20/100,000 (0.02%)

100 ml/d = 23/100,000

125 ml/d = 24/100,000 (i.e. 0.024% LTR)

The disability-adjusted life year (DALY)

The disability-adjusted life year (DALY) is a measure of overall disease burden, expressed as the number of years lost due to ill-health, disability, or death [1]. Globally, alcohol use was the seventh leading risk factor for both deaths and DALYs accounting for ~ 5% of age-standardised deaths.

Silliness heaped upon silliness

For some decades we have had a rather silly system for estimating the amount of alcohol one consumes; that is the scheme of ‘standard drinks’. This means containers, or servings, either from various sized bottles and cans, or from the measures used in retail outlets, bars etc. have a notation like ‘contains 7.6 standard drinks’. This system was produced by a committee and is of poor workability, imprecise, and opaque to most people.

What has been defined as a ‘standard drink’ (by different committees) has varied greatly, from time to time and country to country.

So, anyone who is reading information about alcohol from a scientific paper or from health guidelines, or any source, cannot always be sure what volume of alcohol is being referred to.

All scientific papers and other ‘proper’ sources should talk about millilitres and percentages by volume.

Upsizing: spirits used to be commonly served in pubs/bars in 25/50 ml (single/double) measures, but this is now often 35/70 ml.

That is why the metric system was invented, because it is more logical and easier. To adopt any other system is perverse, inaccurate and well, silly. It is similar to the nonsense in the USA of giving a drug dose of 1 mg/kg to someone who weighs 230 lbs!

The variation in the volume of alcohol that has been defined as a ‘standard drink’ is almost threefold. It is ridiculous.

The UK has moved to a somewhat less silly system, talking about ‘units of alcohol’ where one unit is 10 ml. But why not call it what it is, 10 ml of alcohol: why invent a new term for it? This is a ‘scientific’ version of the absurd practice of saying something like ‘the asteroid approaching Earth weighs as much as 370 London buses, or 68 African elephants’.

That brings us to the next bit of silliness (this is tedious, but wilful stupidity is tedious), which is the number of sources which talk about grams of alcohol rather than millilitres. All alcohol sold is labelled with the volume in millilitres, and the % of alcohol, by volume, not by weight.

Even if, like many, you have not been adroit with mental arithmetic for some years, it is still quite simple to work out how many millilitres of alcohol are in anything that you are drinking. A bottle of wine that is 14% alcohol would have 140 mL of alcohol in a one litre (1,000 ml) sized bottle. It easy to work out that each 100 ml has 14 ml of alcohol, or that a 750 ml bottle has approximately 105 ml.

Most ‘proper’ wineglasses are designed so that if they are filled to the waist, they contain 125 ml (i.e. 6 glasses per bottle). I will not belabour the point, but it is simple to make sure you know what size the glasses in use are. It makes me cringe when I see the uninformed slopping wine into a glass without having a clue how much alcohol they are consuming (nor, I am sure, how good or bad the wine is).

Here is another little exercise in alcohol awareness: when you next watch a film, try to estimate the volume of spirits poured into the glasses. This exercise regularly makes me feel ill, because volume poured is enormous, often two ‘fingers’. A ‘whisky tumbler’, filled with minimum strength 40% whisky, to two finger-widths is ~200 ml, i.e. 80 ml of pure alcohol. That is about a whole bottle of wine (750 ml @13.5% = 100 ml), a very substantial amount in one drink.

That may make you sympathise with my appreciation of the quotation from Paul Claudel:

‘A cocktail is to a glass of wine as rape is to love’.

Think when you drink

If you do not think when you drink, and you cannot describe the wine you have been served, then, in my book, your host would be excused for declining to refill your glass. That is the big mistake: drinking out of habit, not because you appreciate what you are drinking.

Incidentally, it is useful to understand that the same principle applies to food. People who just shovel it down, and do not really care what they eat, are likely to eat greater quantities more quickly and be fatter. The biological explanation for this is partly that rapid consumption of high glycaemic index foods provides excess calories before the feedback mechanisms in the brain that register satiety have time to kick in and inhibit appetite.

In essence, if you eat and drink in a leisurely and appreciative manner you will tend to eat and drink less.

If we use an alcohol intake of 50 ml per day as an example of a significant, but not unusual intake, then it is very simple to see how much, and of what, you would have to drink to reach that total.

One beer at lunchtime 330 ml @5% = 16 ml

On getting home, one beer 330 ml @5% = 16 ml

One large glass of wine with meal, ~200 ml (~4 glasses to a bottle) @ 13% =25 ml.

Total: near enough 60 ml. That is enough to raise your lifetime chance of suffering cirrhosis of the liver three times higher than the base rate in non-drinkers (from 0.006% up to 0.015%). But don’t get too excited or sanctimonious: remember the relative risks discussed above.


Summary of perspectives

The logical course of action is to think carefully about how you balance the equation; that essentially boils down to quantity versus quality, for those who wish to drink.

Doubtless there are many different perspectives about how that balance is best achieved, but medical science sometimes seems to pay insufficient attention to that and is obsessed with statistics like those above. Likewise, there are doctors, perhaps too many, who seem unduly concerned with exerting authority over people and telling them how they should order their lives.

Some might wonder what net gain there is avoiding one or other of the increased risks from consuming alcohol when you might only live a year or two longer, but with less pleasure, before dying of something else (note, that is partly factored into the concept of ‘DALY’ above). Have you thought of reducing the distance you travel in a motor vehicle by 30% (which most people could easily do); that would reduce you chance of premature death and injury more than reducing your alcohol consumption. Have you thought of running to the shops, or when walking your dog?

As the above figures illustrate, if you stand a 3% chance of dying by falling over, getting run over, or shot, then what is the point, or how much will you gain, by avoiding a 0.1% chance of dying of alcohol-related cirrhosis?

I am sure there are many people besides me who are smiling to themselves and remembering the famous quip in response to the above idea which expressed the notion that, whether or not you do live longer by following all the recommended health precautions, it will certainly seem longer. There are those (I am told) who do not find such things as food and wine a major part of their life-pleasure, and for them medical science advances the prediction that a Diogenes-like diet of water and porridge will prolong their existence. However, humans living longer is bad for the planet since there are far too many humans on it already — so the sooner more humans die, the better off the planet will be.

There are those for whom there are specific adverse consequences, both emotional and physical, who are best off avoiding alcohol (described by the pithy Australian phrase ‘a three-pot screamer’), just as they would avoid any drug that upsets them.

When attempting the estimate of how much pleasure and enjoyment alcohol might add to your life, I am assuming, which is not true for everyone, that one is drinking ‘good stuff’ for sensory pleasure, and appreciating it.

Clearly that equation is irrelevant for those who drink cheap alcohol habitually to soothe the dis-ease of an unsatisfactory existence. In such cases the appropriate comparison then becomes what are the comparable potential harms from consuming benzodiazepines or other legal prescription drugs, such as SSRIs. The answer is probably that there is not a great deal of difference in the overall risk-benefit ratio.

Pay your money, make your choice.



1. Collaborators, G.B.D.A., Alcohol use and burden for 195 countries and territories, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet, 2018. 392(10152): p. 1015-1035.

2. Bagnardi, V., et al., Light alcohol drinking and cancer: a meta-analysis. Ann Oncol, 2013. 24(2): p. 301-8.

3. Bagnardi, V., et al., Alcohol consumption and site-specific cancer risk: a comprehensive dose-response meta-analysis. Br J Cancer, 2015. 112(3): p. 580-93.

4. Cao, Y., et al., Light to moderate intake of alcohol, drinking patterns, and risk of cancer: results from two prospective US cohort studies. BMJ, 2015. 351: p. h4238.

5. Chen, W.Y., et al., Moderate alcohol consumption during adult life, drinking patterns, and breast cancer risk. JAMA, 2011. 306(17): p. 1884-90.

6. Nelson, D.E., et al., Alcohol-attributable cancer deaths and years of potential life lost in the United States. Am J Public Health, 2013. 103(4): p. 641-8.

7. Zhao, J., et al., Is alcohol consumption a risk factor for prostate cancer? A systematic review and meta-analysis. BMC Cancer, 2016. 16(1): p. 845.

8. Rehm, J. and S. Imtiaz, A narrative review of alcohol consumption as a risk factor for global burden of disease. Subst Abuse Treat Prev Policy, 2016. 11(1): p. 37.

9. Shield, K.D., M. Rylett, and J. Rehm, Public health gains and missed opportunities. Trends in alcohol consumption and attributable mortality in the WHO European Region, 1990-2014. Report to the WHO European Region, 2016.

10. Lachenmeier, D.W., F. Kanteres, and J. Rehm, Epidemiology-based risk assessment using the benchmark dose/margin of exposure approach: the example of ethanol and liver cirrhosis. Int J Epidemiol, 2011. 40(1): p. 210-8.

apple podcasts
Visit Us
Follow Me
Research Gate
Google Scholar