Working Paper
Short- and Long-Term Determinants of Cardiovascular Mortality

An Econometric Assessment of the Working Age Population in Russia, 1965-95

This paper examines the short- and long-term determinants of the mortality due to cardiovascular diseases (CVD) among the working-age population in Russia. Why Russia? Why CVD mortality? Why the working-age population? Russia is the Eastern European most affected by a rapid a sudden increase in mortality. A change which has hardly been recorded in peace time in the contemporary history. One of the main cause of such increase has been CVD mortality which has particularly been climbing rapidly among the working-age population in Russia ever since the onset of the transition and the dramatic economic and social shocks which accompanied it . Two possible explanations for this tremendous increase in adult mortality can be summarised as follows. (1) The demographic crisis is merely a 'catch-up effect' of a past trend which was interrupted by the anti-alcohol campaign. The long-term factor is identified with a time trend which explains almost all variance. This approach is very unconvincing, because the authors have not provided any plausible econometric evidence that this phenomenon really occurred. (2) The rise in mortality is due to short-term factors resulting from the problem-ridden initial phase of the transition, plus a long-term weakness in the offset variable, namely, health assets, exacerbated by the recent collapse in the provision of care through the health system. In this case, the long-term factor is not a trend, but a precise variable which could influence and constrain the short-term variation provoked by short-term shocks. This paper belongs to the second theoretical family. It attempts to specify and assess econometrically a relationship among these components over time in a 1970-94 sample. After a discussion of the theoretical model, the paper focuses on the evolution of both the dependent and explanatory variables, and then presents an estimate of the model.