Editorials

Alcohol intake: measure for measure

It's hard to calculate how much you are drinkingbut you should know

The festive season is a testing time, and those who wish to drink sensibly might use the "unit of alcohol"a glass of wine or beer or a single measure of spiritsas a yardstick. But what is a unit, and how many is it safe to drink?

The discerning drinker could calculate the dose of ethanol in a drink knowing its volume and ethanol concentration. However, even the sober can find this difficult. Firstly, there are several ways of defining concentration. It is expressed as percentage ethanol by volume (% v/v) in Europe and as percentage proof in the United States, where 100% proof is 50% v/v (in England 100% proof was 57% v/v). The density of ethanol is 0.79 g/ml at room temperature, so, for example, 100 ml of ethanol 10% v/v contains almost 8 g of ethanol. Secondly, concentration can differ widely among apparently similar drinks. The strengths of beers range from about 3.4% to 9% v/v; white wine from 8% to 13% v/v; and spirits from 37.5% v/v for mass market vodka to 57.3% v/v for cask strength Laphroaig. Subjective impressions of alcoholic strength are fallible.¹

Establishing the volume of a drink can also be hard. In the United Kingdom a single pub measure of spirits is now 25 ml (it was 1/6th gill (1/24th pint) in England and 1/4 gill in Scotland). A half pint of beer is 284 ml. Bottles and cans of beer hold anything from 250 to 500 ml. A glass of wine in a pub contains 175 ml, but the large tulip glasses seen in fashionable restaurants contain twice that much. A small bottle of weak beer could contain 8 g of ethanol and a large can of strong beer 35 g; a pub glass of thin Rhine wine might contain 11 g, and your host's generous glass of Pouilly Fuissé nearly 40 g. In the United States a standard drink is 12 ounces of beer, 5 ounces of wine, or 1.5 ounces of 80 proof distilled spirits (an American ounce being 29.6 ml). American and British units therefore differ substantially, which makes it hard to compare epidemiological studies.

The relation between dose and the resulting concentration in blood is also very variable. It depends on the rates of absorption and elimination and the volume of distribution (the ratio between total amount in the body and blood concentration). The volume of distribution can be estimated from age, sex, height, and weight,² but the other variables are harder to define. It is correspondingly hard to predict what dose is likely to raise the blood ethanol concentration above the statutory limit for driving (80 mg/100 ml in the United Kingdom, 50 mg/100 ml in many other countries, and 20 mg/100 ml in a few).

View larger version (19K): [in this window] [in a new window]   The dose of ethanol (g) versus ethanol concentration (% by volume) for different volumes of drink

What effects might ethanol have? Acutely, it depresses the central nervous system and can also precipitate cardiac arrhythmia. Modest concentrations depress inhibitory neuronsturning the introvert into a garrulous exhibitionist. Higher concentrations impair cerebellar functioncausing slurred speech, poor hand-eye coordination, and unsteadiness. Subsequently, sensation, consciousness, and then brainstem functions are depressed. The effects on cerebellar function, seen increasingly as concentrations exceed about 35 mg/100 ml, are important. Admiral Jellicoe noted that "by careful and prolonged tests, the shooting efficiency of the men was proved to be 30% worse after the rum ration than before"³ (the rum ration was 1/8th pintabout 70 ml). The apparent effects of a given blood ethanol concentration, however, vary greatly among individuals. In some cases 500 mg/100 ml can be lethal, while in others much higher concentrations may cause few signs: a woman with a serum ethanol concentration of 1510 mg/100 ml (20 times the UK legal limit) was alert and responsive to questions.⁴

Advice to limit ethanol consumption to a specified number of units per week implies a threshold dose below which ethanol is harmless. Indeed, "the strong negative association between ischaemic heart disease deaths and . . . wine consumption" in developed countries encouraged the hope that moderate drinking might be beneficial.⁵ Several prospective studies, including one of British doctors,⁶ show a J or U shaped relation between coronary heart disease mortality and ethanol intake.⁷ Total mortality, though, increases remorselessly with intake above 12-16 g ethanol per day. ^{6 8} Since the protective effect relates to ischaemic heart disease, those at low risk of this, including premenopausal women, may not benefit even at these levels.

So what should we do? Well, those who will be driving home, operating machinery, or operating on patients should know what they are drinking (see figure): even 10 g of ethanol will be enough to exceed statutory levels in some jurisdictions and could impair performance. One more sobering thought for Christmas: binge drinking can cause arryhthmia and sudden death⁹or, as recently pointed out by England's chief medical officer,¹⁰ lead ultimately to cirrhosis of the liver.

R E Ferner, director. 

West Midlands Centre for Adverse Drug Reaction Reporting, City Hospital, Birmingham B18 7QH (fernerre{at}bham.ac.uk)

Jacky Chambers, director of public health. 

Birmingham Health Authority, Birmingham B16 9RG