抄録
The Russian population decreased between the two censuses conducted in 1989 and 2002 as a result mainly of an increase in male mortality and a decline in birth rates. My previous research focused on premature death or the high mortality rate of the male working-age population. In this paper, multi-regression analyses are applied to regional data from Russia, in order to clarify the relationship between male mortality by age and socio-economic factors in each region. The use of regional mortality rate by age as an independent variable in this paper is a new approach in this field of studies.
The result demonstrates that among socio-economic factors analyzed in this paper, a drink-related crime rate (the rate of crimes committed under the influence of alcohol per 10, 000 people) most significantly correlates to high male mortality, and especially to high mortality in the younger cohort of working-age males. As was shown by earlier research, the increase in male mortality in Russia was mainly brought about by such causes of death as trauma and poisoning, including suicide and homicide, and diseases of the circulatory system, all of which often originate in heavy consumption of alcohol. The conclusion of this paper also confirms this observation.
By comparison with this factor, the economic factors analyzed in this paper, i.e. per capita money income and the unemployment rate, have weaker correlations to high male mortality. But both factors have statistical significance in regression equations concerning working-age males and the middle age cohort (35-44 years old) among them. While per capita income has a negative correlation with mortality, with respect to the unemployment rate a negative correlation (i.e., the lower the unemployment, the higher the mortality) is observed. This might be a result of heavy stress in industrially active regions, which could enhance mortality caused by trauma and poisoning. The fact that the effect of economic factors is not so significant compared with the drink-related crime rate might be explained by the use of 2002 data in this paper, rather than data from the early 1990s, i.e., the period of economic turmoil caused by system change. These problems and validity of other socio-economic factors remain to be answered.
