Aims: This paper explores the relationship between income and health among adults in Sweden. An analysis was made as to what extent the association differs when one studies individual earnings and equivalent disposable income, as well as gender differentials. Further, a study was undertaken to investigate how, and by what magnitude, the income - health relationship changes when one controls for other structural factors, such as education and class. Finally the functional form of the relationship was scrutinized, because of its obvious policy impact. Methods: Data came from the 1996-97 Swedish Living Condition Surveys, which include individuals aged 25 - 64 (n=7,201). Logistic regression was used, including various polynomial terms of the income variable. Results: The results show that both earnings and disposable household income are strongly related to health, a finding that holds for both women and men. The strength of the association becomes somewhat weaker when one controls for other structural factors, but in the final model the association is in fact about the same as the bivariate association, owing to the impact of age. Moreover, a curvilinear association was revealed by the authors' analyses. Conclusions: A clear association was found between income and health, also when other structural variables are controlled for. This indicates that income, as such, is of great importance for the risk of illness. The shape of the association between income and health is consistent with earlier debates concerning the relation between income distribution and population health indicators, and, as such, indicates that income-equalizing policies may have an impact on health.