Is teenage heavy drinking more hazardous than we thought?

Alcohol consumption is a common lifestyle behavior in many parts of the world (http://www.who.int/substance_abuse/activ ities/gad/en/). Excessive alcohol consumption, however, can lead to several somatic diseases, including cirrhosis of the liver [1]. Indeed, alcohol-related liver disease is the major cause of cirrhosis in most countries [2]. Nevertheless, most persons with a high consumption of alcohol do not develop cirrhosis despite heavy alcohol intake [3,4], highlighting the importance of detecting at an early time the subgroup of individuals at a higher risk of cirrhosis. High consumption of alcohol is common early in life, with 50–80% of teenagers in western countries reporting any alcohol consumption and 20% reporting heavy episodic (binge) drinking [5,6], usually defined as drinking five or more drinks during a short period in men [7]. Alcohol consumption early in life is associated with an increased risk of alcohol use disorder later in life [8–10]. However, the importance of adolescent alcohol consumption on the risk of developing cirrhosis of the liver later in life has not been adequately studied. Recently, we investigated a population-based cohort of all men undergoing conscription into military service in Sweden in 1970 (N = 43,296) [11]. Given that Sweden had a mandatory military service during that period, the sample covered 97% of the male population in the studied age cohort. As part of the conscription process, conscripts completed a questionnaire that included detailed questions on the total dose and frequency of consumed alcohol. The men were followed in national registers to ascertain cases of cirrhosis and complications during 39 years of follow-up. A dose-response effect of alcohol use on risk of cirrhosis was observed and no clear ‘safe’ threshold level was apparent. In abstainers, 0.4% developed cirrhosis. This figure can be compared with 0.9% in men who consumed as little as 6–10 g of alcohol per day, which is comparable with less than one glass of wine per day. In men consuming more than 50 g per day, more than 4% developed cirrhosis. Alcohol consumption at conscription was associated with an increased risk of a diagnosis of alcohol use disorder during follow-up, even after excluding men with a diagnosis of viral hepatitis. Hence, the risk of a high alcohol consumption in the development of cirrhosis later in life is present already from an early age and this increased risk is likely associated with a longer exposure to alcohol as compared with starting to drink later in life. Although this issue was not investigated as part of the current publication, previous follow-up investigations in the cohort found alcohol use at the time of conscription predictive of alcohol use disorder later in life [10]. Likely, individuals with a longer history of alcohol consumption have an increased risk for liver-related diseases later in life. A Danish cohort study in which lifetime alcohol consumption was assessed retrospectively found that present consumption of alcohol appeared to be more important for the development of cirrhosis than consumption occurring early in life, suggesting that increased alcohol consumption over time is a major contributor to the development of cirrhosis [12]. Additional to the total dose of ingested alcohol, specific ‘risk behavior’ questions were included in the questionnaire at conscription. Two of these questions (‘Have you ever been questioned [by a parent, teacher, law enforcement officer] for your level of intoxication of alcohol?’ and ‘If you have been hungover before, have you ever used alcohol to alleviate your symptoms?’) were associated with an increased risk of cirrhosis, independent of the total amount of alcohol consumed [13]. Thus, it seems possible to target a subset of adolescent men with a particularly high risk of cirrhosis later in life by identifying those who drink excessively to the extent that they are being approached for intoxication or use alcohol to relieve common hangover symptoms. Although our results do not imply causation, knowledge of risk factors does allow for identification of a group with a particularly high risk for future development of cirrhosis. The above studies on conscripts are restricted to men only and thus preclude conclusions about women. Women have an average lower body weight and higher percent body fat than men, as well as lower levels of alcohol dehydrogenase. Consequently, women not only get more intoxicated than men after drinking the same amount of alcohol but are also more vulnerable to liver damage compared with men [14]. Another limitation is that because changes in alcohol consumption habits in the general population over time are likely, estimations on the future burden of liver disease are precluded. For instance, consumption of alcohol in Swedish teenagers has increased after the 1970s, but a significant reduction has been seen in the past 5-year period in the general population, particularly in men. Alcohol consumption in young men now averages 8 g of alcohol