In the 2014 global status report on alcohol and health, the World Health Organization estimated that alcohol was responsible for 5.9% of deaths worldwide or 3.3 million people annually. Alcohol-related cirrhosis is responsible for 493,300 deaths. [1] Thus, alcohol is the leading cause of cirrhosis and liver-related deaths. 2 , 3 Despite the huge disease burden, there are no approved treatments for alcoholic liver disease. Yet, theoretically, alcohol-related disease is 100% preventable, which makes the role of preventive measures central in decreasing the impact of excessive alcohol consumption on society. Liver fibrosis is a condition that progresses slowly, and it usually takes decades before cirrhosis develops or decompensation occurs. However, the natural history of alcoholic liver disease has not been fully elucidated. Population data are crucial to help create evidence-based health policies to reinforce efforts to reduce alcohol related injury. In this issue of the Journal of Hepatology, Hagström et al. analyze the relationship between alcohol consumption in men at the time of conscription (aged 18–20) and the probability of severe liver disease 40 years later. They found drinking patterns were associated with a higher probability of alcohol-related liver damage. However, interestingly, there was no evidence for a “safe” threshold in this initial analysis, as the risk was significant for daily alcohol consumption as low as 6 g/day (less than one unit). In a second analysis, after adjustment for body mass index, tobacco consumption, the use of narcotics, cardiovascular fitness and cognitive ability, the association between alcohol consumption and risk of severe liver disease became significant for daily consumptions of >31 g/day.