We’re drinking more than we thought we were

A new and more accurate way of calculating alcohol consumption shows that drinkers, especially women, are drinking substantially more than was previously believed. The new method of calculation recognizes that, nowadays, single measures of alcoholic drinks often contain more than single units of alcohol, in some cases, a lot more. The new calculations increase substantially the numbers of people exceeding the ‘sensible limits’of drinking and engaging in ‘binge’ drinking.

The government surveys that provide most of the official statistics on alcohol consumption have always assumed that there is one unit of alcohol in half a pint of beer, a glass of wine or a single measure of spirits.Whilst this is still true for standard strength beer (3.5-4%) and smaller measures (25 ml) of spirits, larger glasses and stronger wines mean that this is now a dramatic underestimate for a glass of wine. In fact one standard glass of wine (175 ml) typically contains two units – twice the old estimate. Stronger
beer, and beer sold in bottles or cans rather than pints, has also been affected.

Eileen Goddard of the Office of National Statistics,who calculated the new conversion factors, also recalculated alcohol consumption for the most recent data from the General Household Survey. Since wine was the drink most underestimated by the old system, the groups of people affected most by the revised system were wine drinkers, principally women,more affluent people and those over the age of 45. Whilst future surveys will ask for information about the size of wine glasses, the recalculation assumed a typical glass size of 175 ml.

Women:
The estimate of average weekly consumption has increased by 45%, from 6.5 units per week to 9.4. This is a much larger increase than the estimate for men, which increased by 26%, from 15.8 units per week to 19.9. The percentage of men exceeding weekly limits (21 units) was previously thought to be 24% and is now believed to be 31%, an increase of 32%. The equivalent figures for women drinking more than the weekly limit of 14 units was previously 13% and is now 21%, an increase of 65%.

The measure that has been used for binge drinking, namely the percentage of people drinking more than twice the daily limit, increased from 19% to 23% for men, an increase of 24%, and from 8% to 15% for women, an increase of 89%. These increases were smaller for the youngest age group, 16-24 year olds, who are less likely to drink wine. In this age group, the estimate of the percentage of men ‘binge drinking’ increased from 30% to 32%, an increase of 9%, and the estimate for women increased from 22% to 27%, an increase of 25%.

Affluence:
Two measures from the survey are related to ‘affluence. ’These are socio-economic classification (SEC) and income, which are both defined at a household level. The estimate of average weekly alcohol consumption of those in higher managerial or higher professional jobs increased by 45% and 43% respectively, relative to an increase of 24% for those in routine and manual jobs. The estimate of consumption for those in the highest income band (over £1000 per week) increased from 12.9 units per week to 18.2 units per week, an increase of 41%. For those in the lowest income band (up to £200 per week), the estimate increased from 9.0 units per week to 11.1 units per week, an increase of 24%.

Age group:
The survey divided people into four age groups, 16-24, 25-44, 45-64 and 65 and over. The association between getting older and drinking wine is stronger for men than for women, so the older two age groups showed the biggest increase in estimated consumption for men (29% increase in each case). For women, the biggest increase in the estimated weekly consumption was for the 45-64 year olds,with a 56% increase, closely followed by the adjacent two age groups. Only the youngest women showed a substantially smaller increase, of 21%. The older age groups also showed bigger increases in the estimates of percentage of people drinking more than the daily limit, and more than twice the daily limit.

In her report, Eileen Goddard emphasises that this new calculation does not mean that people are suddenly drinking more than they were, just that they are consuming more than was believed on the basis of previous government surveys.

National Statistics Methodological Series No. 37 Estimating alcohol consumption from survey data: updated method of converting volumes to units. Eileen Goddard December 2007