This paper examines the effect of income polarization on individual health. We argue that polarization captures much better the social tension and conflict that underlie some of the pathways linking income disparities and individual health, and which have been traditionally proxied by inequality. We test our premises with panel data for Spain. Results show that polarization has a detrimental effect on health. We also find that the way the relevant population subgroups are defined is important: polarization is only significant if measured between education-age groups for each region. Regional polarization is not significant. Our results are obtained conditional on a comprehensive set of controls, including absolute and relative income. © 2009 The Authors. Journal compilation © International Association for Research in Income and Wealth 2009.