Per capita consumption of ethanol is the most reliable measure of a population's overall alcohol-drinking behaviors

Oct 16 2004

Per capita consumption of ethanol derived from sales of beer, wine and spirits is the most comprehensive and reliable measure of a population's overall alcohol-drinking behaviors.

New findings, published in the October issue of Alcoholism: Clinical & Experimental Research, indicate that more alcohol is being consumed in the form of beer than previously thought, and the national mean alcohol content of beer is higher than previously thought.

"Alcohol consumption in the U.S. comes from three main beverage types," said William C. Kerr, associate scientist with the Alcohol Research Group at the Public Health Institute and first author of the study, "beer, wine and spirits. Beer is the largest source of alcohol consumption, though the share of each beverage type varies by state and changes over time. Market share can help better understand aspects of drinking culture related to beverage choice. For example, France and Italy are wine-preferring countries, while Russia is a spirits-preferring country."

"It may seem strange, but it has been found again and again that a small percentage change in the overall alcohol consumption of a nation predicts similar changes in the number of serious problems caused by drinking, such as deaths, injuries and illnesses," said Tim Stockwell, professor and director of the Centre for Addictions Research of B.C. at the University of Victoria. "This means a seemingly small increase in the total amount of ethanol consumed in beers can translate into a few more people dying from one of the 37 plus alcohol-related causes of death."

Researchers gathered national estimates for the years 1988 - 2001 and state-specific estimates for the years 1993 - 2001 using several different sources, including the Adams Beer Handbooks, the Washington State Liquor Control Board price lists, and the Seibel Institute of Technology Survey of Retail Market Beers and Market Beer Review.

"We found that there have been significant shifts in the types of beers consumed by Americans between 1988 and 2001," said Kerr. "Market shares of each type have varied by state, and specific brands have changed their alcohol percentage on occasion. These changes and differences have resulted in overall differences in the average alcohol percentage of beer in each state and in each year. This means that the proportion of the beer sold that is pure alcohol also differs."

Kerr and his colleagues found that the national mean alcohol content of beer was higher than the 4.5 percent figure typically used in aggregate-level research, ranging from 4.58 percent in 1993 to 4.75 percent in 1996.

"While these differences may seem small when considered in terms of a single bottle of beer," said Kerr, "they appear more substantial when applied to the six million gallons of beer sold each year in the US. For example, in 2000, using the corrected percentages results in about 10 additional standard drinks yearly for each person aged 14 or older in the U.S. State differences range from half a drink less in Iowa, where light beer is especially popular, to more than 14 drinks more in Mississippi, where malt liquor and premium beer are more popular than the average for all states."

"This study has set a new standard in estimates of per capita consumption," said Stockwell. "It is noteworthy that the method used follows guidelines first outlined by a World Health Organization expert-working group and published in 2000. The research also highlights the relatively limited range of beer strengths in the U.S.A. compared with other countries. Most of the beer consumption falls within a relatively narrow range of alcoholic strengths; there are very few low- or middle-strength varieties with nearly all brands falling in the narrow range of 4.1 to 5.1 percent alcohol content. In other countries where tax incentives have been given to producers, retailers and consumers to give preference to lower-strength beers, there is much greater variety. In Australia, for example, beers with a strength of less than four percent make up 40 percent by volume of the beer market, and there are close to 40 varieties from 0.9 percent to 3.8 percent to choose from."

Kerr concurs. "An important message is that not all beer is the same and that consumers should consider the percentage of alcohol in the brand of beer they drink when deciding how many beers to consume. For example, the light version of a particular brand may contain 4.2 percent alcohol, while the ice version of the same brand may contain as much as 5.9 percent alcohol, which is 40 percent more alcohol per can."

"Future research should study variations in alcoholic strengths in the U.S.A. and their relationship to levels of serious alcohol-related harm," said Stockwell. "Again, in Australia, strong relationships have been found between the alcohol content of beer in different regions and the local levels of night-time violence and hospital admissions for alcohol-related reasons. This study signals that more could be done to promote better public health and safety through the promotion of lower strength beers."