Individual and Contextual Determinants of Male Mortality in Lithuania: A Multilevel Census-Linked Study

Domantas Jasilionis, Max Planck Institute for Demographic Research
Vlada Stankuniene, Institute for Social Research (ISR), Vilnius
Dalia Ambrozaitiene, Statistics Lithuania
Vladimir M. Shkolnikov, Max Planck Institute for Demographic Research and New Economic School, Russia

This study aims at assessment of the role of the selected individual and contextual determinants of mortality among Lithuanian males. The study uses the census-linked data for 2001-2005 covering 6.6 million person years and 72.5 thousand deaths for males aged 30-74. Random intercept Poisson regression was used to capture unobserved regional effects. The study found that the contextual variables were associated with individual mortality risks. In particular, unfavourable socioeconomic structures of municipalities were related to elevated mortality risk at the individual level. However, after controlling for selected major individual characteristics these area-level effects either became small or statistically insignificant. Interestingly, the share of divorced males had an opposite effect on mortality as compared to the same characteristic at the individual level. The results suggest that the regional variation in male mortality risk in Lithuania can be largely explained by the variation in the individual-level characteristics.