Cancer mortality in Russia and Ukraine: validity, competing risks and cohort effects.

<bold>Background: </bold>The dramatic increase in mortality in Russia and Ukraine in the late 1980s and 1990s has been due to increases in certain causes of death, particularly cardiovascular disease and accidents and violence. In contrast, there has been a slight fall in mortality from cancer.<bold>Methods: </bold>This paper presents an analysis of trends and patterns in cancer mortality and examines four possible explanations for its recent fall: changes in data collection; cohort effects; competing mortality from other causes of death; and improvements in health care.<bold>Results: </bold>All contribute to some extent to the observed changes, with each affecting predominantly different age groups. There is evidence of a significant underrecording of cancer deaths among the elderly especially in rural areas and of significant changes in coding practices in the early 1990s. Competing mortality from cardiovascular diseases and accidents can explain some reduction in male deaths from cancer in middle age. Birth cohort effects can explain some reduction among males after early middle age and among females at all ages. The impact of changes in health care are more difficult to identify with certainty but there is evidence of reduced deaths from childhood leukaemia.<bold>Implications: </bold>Recent changes in mortality in Russia are complex and their understanding will require a multidisciplinary approach embracing demography, epidemiology and health services research. [ABSTRACT FROM AUTHOR]