Aim. To estimate the effects of changes in per capita alcohol consumption on liver cirrhosis mortality rates in various demographic groups across 14 western European countries. Method. Yearly changes in gender- and age-specific mortality rates from 1950 to 1995 were analysed in relation to corresponding yearly changes in per capita alcohol consumption, employing the Box-Jenkins technique for time series analysis. Country-specific estimates were pooled into three regions: northern, central and southern Europe. Measurements. Cirrhosis mortality data for 5-year age groups were converted into gender-specific mortality rates in the age groups 15 +, 15-44, 45-64 and 65 + and expressed as the number of deaths per 100 000 inhabitants. Alcohol sales were used to measure aggregate consumption, which were calculated into consumption (litres 100% alcohol) per year per inhabitant over 14 years of age and weighted with a 10-year distributed lag model. Findings. The country-specific analyses demonstrated a positive and statistically significant effect of changes in per capita consumption on changes in cirrhosis mortality in 13 countries for males and in nine countries for females. The strongest alcohol effect was found in northern Europe, due mainly to a large effect in Sweden. Moreover, when different age groups were analysed significant estimates were obtained in 29 of 42 cases for males and in 20 of 42 cases for females. Most of the non-significant estimates were found in older age groups. Conclusions. The results suggest clearly that a change in the overall level of drinking as a general rule affect cirrhosis mortality in different drinking cultures as well as among different demographic groups. Moreover, the findings correspond with what is expected from the collectivity theory of drinking cultures.