Background: Since the mid-1970s, a striking reduction in alcohol consumption has been observed in Italy and other developed countries. Alcohol-related mortality in Italy has been estimated for 1983 and 1996.

Methods: Alcohol-attributable and alcohol-preventable deaths were estimated by: i) data on prevalence of drinkers from two Italian surveys; ii) the parameters of meta-regression models investigating the relationship between alcohol intake and the risk of several conditions positively and negatively related to alcohol; and iii) the number of deaths from 21 alcohol-related conditions.

Results: About 68,000 and 42,000 deaths were attributed to alcohol consumption in 1983 and in 1996 respectively, mostly from hemorrhagic stroke, liver cirrhosis, cancer, and injuries. About 6,600 deaths from coronary heart disease were prevented by alcohol. Light intake (25 g/day or less) caused about 30% of deaths attributable to any consumption in women. In men, about one-half of the deaths were attributable to the highest category of intake (100 g/day or more), while a lower proportion of deaths was attributed to light intake (almost 7%). In 1996 the number of the deaths caused and those prevented by light intake was approximately the same (5,400 and 5,200 respectively) and did not significantly differ.

Conclusion: The estimated number of deaths attributable to alcohol consumption in Italy still far exceeds the number prevented for both women and men. Despite the cardiac protective effect, alcohol consumption remains a major public health problem in Italy. Both population and high risk strategies in preventing alcohol-related problems should be implemented.