Australians pay a substantial price for the pleasure they derive from alcohol. According to the latest estimate, on a net basis, alcohol accounts for 2.2% of the total disease burden in Australia.1 This is marginally more than illicit drugs (2.0%),1 but alcohol’s toll would be even higher if the 3430 deaths of young adults caused by alcohol-related road crashes, accidents, assaults, suicide and other causes were not partially offset by 2345 deaths from heart disease counted (some would argue erroneously)2,3 as having been averted by moderate drinking in adults over the age of 65.1

The health costs of risky alcohol use are not the only social costs. According to a National Drug Strategy study,4 in addition to $1.98 billion in health costs in the 2004–05 financial year, alcohol misuse cost the Australian economy $3.58 billion in lost productivity in the workplace, $1.57 billion in lost productivity in the home, $1.61 billion in crime-related costs, and $2.20 billion in road accidents, with a total social cost in excess of $15 billion.

For the past decade, illicit drugs have overshadowed alcohol as a political and public health issue. Policies to reduce alcohol-related harms (such as taxation and reduced trading hours)5 have been effectively opposed by an alcohol industry that has had the ear of federal and state governments, both of which have major conflicts of interest in alcohol policy. The federal government derives substantial income from alcohol excise, while state and territory governments impose licensing fees and taxes on gambling in pubs and clubs. Governments are also aware that the alcohol industry plays an important role in the Australian economy, especially in regional areas (eg, the wine industry in South Australia).

The recent change of federal government has brought a renewed policy interest in alcohol use. The alcohol industry claims that this interest is misplaced, citing data which suggest that per capita alcohol use and frequency of consumption, as indicated by household surveys, have been relatively stable for over a decade.6,7 However, there are doubts about the quality of these alcohol trend data, and the most recent survey data certainly indicate there is cause for concern. In 2007, one in five Australians over the age of 14, including nearly half (44%) of young men and around a third of young women aged 20–29 years, reported drinking in risky ways monthly or more often.7 The alcohol industry derives substantial profits from risky drinking, with — on conservative estimates — two-thirds of all alcohol, and 80% of alcohol used by young people aged 14–24 years, consumed in ways that put the drinker’s (and others’) health at risk.6

It is essential that debates about alcohol policy are informed by good data on alcohol use and alcohol-related harm in the Australian population.8 The World Health Organization has recommended that public health monitoring of alcohol use should include credible estimates of per capita alcohol consumption derived from alcohol sales data, in addition to well conducted population surveys of drinking patterns.8 Despite this, Australia is now in danger of failing to collect alcohol sales data at a time when community concern about alcohol is increasing and the quality of survey data may be declining.

For many years, the Australian Bureau of Statistics (ABS) has published national estimates of per capita alcohol consumption9 based on import clearance, excise and domestic alcohol sales data. Until 1997, the ABS estimates were complemented by state and territory alcohol sales data collected by liquor licensing authorities. However, after a High Court ruling that state liquor licensing fees were unconstitutional, most jurisdictions stopped collecting these data; only Western Australia and the Northern Territory continue to do so.6 In its most recent report, the ABS has indicated that it may no longer report on national alcohol consumption estimates derived from sales data.9 If this were to happen, Australia would be the only OECD (Organisation for Economic Co-operation and Development) country to not collect national alcohol consumption data.

The collection of alcohol sales data should be improved rather than abandoned. Without these data, policymakers, community action groups and public health researchers would lack essential information to monitor trends in per capita alcohol use, which is strongly related to adverse health outcomes such as liver cirrhosis, motor vehicle crashes, and suicide.5 National sales data are essential for monitoring trends in per capita consumption, and they facilitate studies of the relationships between changes in the level of per capita alcohol consumption and both population health outcomes and social harms (eg, arrests for assault and public disorder).5

Sales data also provide a benchmark to gauge the accuracy of national alcohol consumption surveys.10 Household surveys of self-reported alcohol use, such as the National Drug Strategy Household Surveys and the National Health Surveys, provide important information on drinking patterns among population subgroups, but, at best, they complement rather than substitute for sales-based data. For example, the National Drug Strategy surveys and the National Health Surveys are conducted every 3 years and 5 years, respectively, and the alcohol use they measure accounts for only about 60% of the alcohol that is sold.10 The response rates of the most recent National Drug Strategy surveys (in 2004 and 2007) have been less than 50%, raising concerns about the representativeness of survey data for population alcohol use.

State and territory sales data should be collected in ways that enable the sales volumes of each beverage type to be estimated at local levels. These local-level sales data can be used to evaluate the effectiveness of community initiatives to reduce alcohol-related harm, such as those in northern and central Australia, and the effects of liquor licensing changes on alcohol consumption. State-level data could be used to evaluate the effect of new alcohol management strategies such as pub “lock outs”, restrictions on types of alcohol that can be sold in Aboriginal communities, and the tax increase on premixed spirits-based drinks.

The collection and reporting of alcohol sales data would entail minimal cost to the alcohol industry, which already provides these data to commercial market research companies. Given that alcohol harms a substantial proportion of people who use it and can also adversely affect the safety and amenity of those who do not misuse it, the federal government should require alcohol sales data to be provided by those who are licensed to sell this intoxicating and addictive commodity.