Labelling alcoholic drinks: percentage proof, original gravity, percentage alcohol or standard drinks?

Drivers who wish to stay 'under the limit', problem drinkers wishing to control their drinking and literally anyone who drinks alcohol and is concerned about their health are all increasingly exhorted to monitor their alcohol intake by counting 'standard drinks' (each containing 8-14 g, depending on the country in question). Unfortunately, the evidence presented in this paper suggests that this system permits many errors. In particular, it requires two assumptions to be met: (1) that drinks of the same beverage type (i.e. beer, wine, fortified wine or spirits) normally contain the same percentage of alcohol by volume; and (2) that people serve, or are served, alcoholic drinks in standard serves. It is shown that in practice the strength of drinks available for sale of a given beverage type varies widely and that 'atypical' strengths form a significant proportion of alcohol sales. Furthermore, whether drinking occurs in a private residence or on licensed premises, it is usual for quantities greater than the supposed Australian standard of 10 g to be served. In practice, most people are unaware of the strengths of different beverages or the rough equivalences between them. Even if they are taught the standard drink system, they cannot make allowances for 'atypical' variations in strength. It is suggested these problems could be readily overcome if all alcohol containers were labelled in terms of standard drinks. The benefits of such a labelling system are discussed with regard to health promotion, accident prevention and the accuracy of surveys of alcohol use.