Differences in income-related health inequalities in Brazil and the USA: a comparative cross-sectional study

IntroductionHealth inequality is a persistent, policy-relevant problem shaped by multiple socioeconomic factors, among which income and healthcare policy are especially prominent and often vary substantially between countries. The goal of the present study is to quantify the relationship between income and health in two countries with different healthcare systems, Brazil and the USA.MethodsWe compared health conditions in relation to income in Brazil and the USA through cross-sectional, nationally representative, household-based surveys using data from the Pesquisa Nacional de Saúde (PNS) for Brazil and the Panel Study of Income Dynamics (PSID) for the USA collected in 2019. We used survey-weighted logistic regression with cluster-robust standard errors, adjusted for urban/rural residence, sex, education, and age.ResultsBoth countries showed income-related health inequality, with lower-income groups generally reporting worse self-rated health. A formal test of the country-by-income interaction showed a significantly steeper income gradient in the USA than in Brazil for diabetes and obesity and a significant reversal in the direction of the association for psychiatric conditions, with higher income associated with higher prevalence in Brazil but lower prevalence in the USA. No significant cross-country difference was found for hypertension, self-rated health, or cancer diagnosis. Geographically, a north-south divide in inequality is apparent in Brazil, while income disparities are less uniform in the USA. States with a larger share of low-income people experience worse health conditions in the USA, while this effect is less visible in Brazil.DiscussionTaken together, income-related health inequality is more pronounced in the USA than in Brazil for several, though not all, of the conditions examined. This pattern may be associated with differences in healthcare systems and lifestyles between the two countries, although the cross-sectional design does not permit causal inference.