Taking welfare state regime research globally: An application of the Wood and Gough typology to individual health

Abstract

Seldom do studies on welfare regimes and health take a global perspective. Usually most research on this genre concentrates on high-income countries only, few studies include low- and middle-income countries, and even fewer examinations combine a large number of countries to obtain a worldwide view. For this, we need global welfare regime typologies that would allow for such examinations. Given this, we investigated the usefulness of the Wood and Gough welfare regime typology on 49 countries by examining the association between welfare regimes and self-reported individual disability worldwide. The World Health Survey data were examined on 207 818 people from 49 countries using the welfare regime classification developed by Wood and Gough. Multilevel logistic regression was used to investigate links between disability and welfare regimes while also accounting for individual-level socioeconomic factors. Variations in individual disability prevalence were found within the different welfare regimes. For example, odds of poor health prevalence for citizens within the insecurity Sub-Saharan African regime as compared with the European-conservative regime were OR=1.83 95 per cent CI: 0.85–3.95. Living in a state-organized regime was associated with lower odds of disability prevalence, as higher odds of disability prevalence were observed in all non-state regimes (with the exception of the productivist regime). For instance, the people in the productivist regime of East Asia reported similar prevalence of poor health odds as compared with the European-conservative regime (OR=0.94 95 per cent CI: 0.45–1.99). This short report found that the Wood and Gough typology enables the study of welfare regimes and health globally, and appears to be a useful tool in welfare regime type research. The productivist regime seems to have health protective features that are on par with European welfare state regimes. We recommend that this finding is investigated further in future empirical analyses.