Several alcohol policy measures in Sweden, such as taxation and restrictions in availability and advertisements, start out from the idea that spirits is more harmful than milder alcoholic beverages. Consequently, availability and taxation have generally been stricter for spirits, and the recent removal of the ban of alcohol advertisements in 2003 did not apply to spirits. Moreover, norms in the general population suggest that spirits is regarded as more harmful than lighter beverage types, and public opinion is significantly more in favor of increasing the availability of wine and beer than of spirits (Leifman, 2000).
The main aim of the present article is to determine whether there is any evidence for the idea that spirits consumption is more harmful than consuming milder beverage types in Sweden. As a measure of harmfulness, we apply the likelihood that a given amount of consumption is associated with both harmful drinking patterns (e.g., binge drinking and risk drinking) and actual experiences of alcohol-related harm. Differences in harmfulness by alcoholic beverage type could either result from the characteristics of those drinking the beverage or differences in the beverages (Mäkelä, Mustonen, & Österberg, 2007). In addition, if a beverage is preferred by people with a heavy drinking orientation, like younger men, it is more likely that a given amount of drinking will cause trouble than if a beverage is preferred by elderly women. However, differences in taste and the alcohol content of alcoholic beverages are examples of beverage characteristics, per se, that could influence the likelihood of drinking being harmful. For instance, strong and sweet beverages may more easily lead to intoxication and therefore to a higher risk of harm.
As to spirits, it is not theoretically unlikely that the high alcohol content attracts people with a heavy drinking orientation, e.g. by offering a quick way to intoxication. Furthermore, the notion that spirits is less suitable to consume with a meal may also imply that drinking in general becomes more oriented towards intoxication. In fact, some research suggests that alcohol poisoning is more likely to occur with spirits than with milder beverage types (Poikolainen, Leppänen, & Vuori, 2002).
There is some earlier empirical evidence that sprits consumption actually has been more closely related to alcohol-related harm in Sweden during the postwar period. Beverage-specific time-series analyses based on data for 1950-2000 showed that spirits consumption, but not consuming wine and beer, was associated with mortality outcomes such as homicide (Rossow, 2001), suicide (Norström & Rossow, 1999) and liver cirrhosis (Stokkeland, Brandt, Ekbom, Osby, & Hultcrantz, 2006). Recent analyses by Landberg (this issue) seem to support this conclusion.
According to more recent analyses based on individual-level data, however, it is not clear that spirits is more strongly linked to harmful drinking than are milder alcoholic beverages. Analyses of general population surveys from the early 1990s suggested that strong beer accounted for the highest proportion of alcohol consumed among the heaviest male drinkers, followed by spirits, whereas for women, wine and strong beer accounted for larger shares than spirits (Kühlhorn, 1998). Moreover, recent studies of heavy drinkers in treatment suggested that spirits is not more common among heavy drinkers; strong beer dominated among male patients under treatment for alcoholic cirrhosis and alcohol dependence, and wine consumption was almost at the same level as spirits consumption (Stokkeland, HiIm, Spak, Franck, & Hultcrantz, 2008). Among female patients being treated for cirrhosis, wine accounted for the major share of consumption and was of the same magnitude as strong beer in female patients under treatment for alcohol dependence. There are thus some scattered indications that, in Sweden today, spirits consumption may not be more strongly associated with harmful drinking habits than is consuming wine and strong beer. …