During the last decade a number of studies have been published that investigate how the most fundamental aspect of political organization, the regime type, affects population health. The results unanimously show that citizens of democracies live longer and healthier lives than citizens of non-democracies. Many explanations for this have been suggested, among these are that democracies redistribute more and invest more in salutogenic resources, and that the tendency of dictatorships to control the media negatively affects the ability to spread information crucial to public health. When these mechanisms are controlled for, however, it turns out that democracy has a large residual correlation with for example life expectancy, which suggests that other mechanisms are also involved.In this paper two new mechanisms regarding the possible psychosocially mediated health effects of the regime type are investigated, namely political repression, and the possible negative effects this might have on the levels of chronic stress, and autonomy, which connects to a large previous literature in social epidemiology. In the paper an ecological cross-country design is used and country-level data, provided mainly by the World Bank and Freedom House, is analyzed with a simple multiple OLS-regression model.