Abstract
This thesis examines the socioeconomic health disparities in the postcommunist countries of Central and Eastern Europe and the former Soviet Union, in which two decades of transition have resulted in social, political, economic, and health changes. A conceptual model was developed to guide the empirical work of this study in relation to its main objectives: (1) to identify the main individual-level and country-level predictors of health disparities in post-communist countries, and (2) to determine whether there are gender differences in the established predictors in these countries. Two types of data were used: (1) cross-sectional individual-level data derived from surveys (EUREQUAL) conducted on nationally representative samples in 13 post-communist countries (a total of 15,643 respondents aged 18+) and (2) macrolevel (country level) data collected from external sources and computed from the EUREQUAL survey data. The following statistical analyses were performed to answer the research questions: (1) descriptive data analyses, (2) analyses of correlations, (3) logistic regression analyses and (4) multi-level modelling. As a result of this study, a comprehensive picture of the main country- and individual-level predictors of health in the post-communist countries has emerged. The main country-level health predictors were economic development and social cohesion for both genders. However, the combinations of individual-level predictors are somewhat different between men and women with the main differences related to childhood socioeconomic circumstances, level of education, smoking and informal social networks. The health of both genders at the individual level is better predicted by a combination of factors that represent relative as well as absolute material disadvantage. Among all factors, relative material disadvantage was most strongly associated with health. Social and public health policies could be strengthened by understanding the factors that influence health in these countries and which have been identified in this study.