Educational differentials in activity limitations across the European Union: methodological issues and first results

Emmanuelle Cambois, Institut National d'Études Démographiques (INED)
Aïda Solé-Auró, Institut National d'Études Démographiques (INED)

Comparing the magnitude of the socioeconomic (SES) differentials in health across countries is challenging due to issues in data comparability and due to country specific association between health and SES. Social, health and educational systems have been developing differently across European countries and generations, inducing country-specific and generation-specific associations between health and SES categories. In this paper, we use EU-SILC 2009 dataset to discuss results on educational differentials in the global indicator of activity limitation (GALI) across Europe. We describe the variation in the differentials and use a logistic regression to highlight (1) the overall effect of level of education; (2) country differences in health and social context; (3) the interaction between the country and the levels of education and whether health differences are due to country specific effect towards high and/or low educational groups. The model is run for three age groups/generations to show the variation in the parameters. We found a large variation in the magnitude of the educational differentials. The distribution of the EU populations between educational groups impacts the observed level of health. The logistic model demonstrates a combination of effects; on the top of the overall effect of education, a country specific effect can be observed. For instance, higher educated have a relatively smaller health advantage in Denmark and Bulgaria; lower educated group have a relatively reduced disadvantage in Finland; southern European countries tend to have a relatively greater disadvantage for the lower educated groups. These results should be confirmed by testing other models, such as multilevel analysis to further account for contextual variables. But these first results indicate that the interaction of education in with country-specific contexts should be accounted for when interpreting educational differentials and to understand the mechanisms behind social determinants of health within the EU.