Per capita alcohol consumption and total mortality: an analysis of historical data

Since total mortality is a classical proxy for the overall health status of the population, its degree of association with per capita alcohol consumption is of great interest. Existing evidence, based on historical data from the turn of the century, is mostly in graphical form. These data are analysed using modern statistical tools. The results suggest a significant alcohol effect; a 1-l increase in consumption is expected to increase mortality (middle-aged men) by about 1%. This effect might seem fairly modest but it is noted that it may well be locally substantial because of its concentration to specific categories of the population. The alcohol effect is also compared with the impact of a factor that is a surrogate for a large number of etiological agents, namely real wages. Although the latter factor seems to be the more important one the difference is not overwhelming. The shift in the cause of the death panorama during this century, with an increasing share of CHD-mortality, may well have attenuated the aggregate relationship between alcohol and mortality.