AbstractBackground:There are no published New Zealand (NZ) studies on alcohol drinking and total mortality, despite its importance to alcohol health policy.Aims:To estimate the proportion of NZ deaths caused or prevented by alcohol drinking.Methods:The proportion of current alcohol drinkers from recent NZ surveys, and pooled relative risks from a review of the international literature on alcohol and mortality, were used to calculate disease‐specific population attributable risks. The number of deaths caused (or prevented) by alcohol were calculated for 1987 New Zealand deaths. Person‐years of life lost (or saved) were calculated using recent NZ life tables.Results:The association between alcohol and total mortality was related to age. Alcohol was estimated to have caused 3.0% of all deaths among 0–14 year olds and 20.1% of deaths among 15–34 year olds, mostly from road injuries. In contrast, alcohol was estimated to have prevented 0.5% of all deaths among 35–64 year olds and 3.4% of deaths among>65 year olds due to its protective effect against coronary heart disease. For all age groups, alcohol was estimated to have prevented 1.5% of deaths. However, the number of person‐years of life lost among ages less than 35 years was greater than those saved in the older age groups, so that alcohol was estimated to have caused the loss of 9525 person‐years of life for all ages combined.Conclusions:The adverse effects of alcohol on total mortality are confined to age groups less than 35 years. Public health policy to minimise deaths from alcohol should be concentrated