Abstract : The purpose of this thesis is to study the determinants of the health status of young children. We consider here two types of determinants: the determinants called “traditional” and those subject to a renewed interest. Of these, we focus on three in particular: social capital, income inequality and health infrastructure. After presenting then tested and discussed the role of traditional determinants, we check several groups of assumptions about the relationship between the “new” determinants and the health of young children. Our first set of assumptions asserts that social capital may affect the health of individuals through at least two mechanisms: a biological one and a behavioural one. To test this, we consider various dimensions of social capital. Our second set of assumptions focuses on income inequality and its channels of transmission on health. The main idea here is that income inequalities have an impact on health through four main channels: income, infrastructure, biology of stress and deterioration of social capital. Thirdly, we look at the differential impact of health facilities on the health of young children according to their household's income level, their mother's education and their household's social capital. We assume in this last point that the higher the parents' socioeconomic status, the greater their ability to make their children's health benefit from infrastructure. To carry out this thesis, we used two databases. The first one is a macroeconomic database of 84 countries over 21 years, while the second is a microeconomic database of 5339 Guatemalan children. Our macroeconomic indicator of health is the infant mortality rate, while the indicator chosen for microeconomic studies is stunting expressed as a z-score for children aged from 0 to 5 years. The main methods used are those of ordinary least squares with clusters, the generalized method of moments, simultaneous equations and matching propensity score. Results indicate the relevance of many of our “traditional” determinants. On the other hand, few dimensions of social capital have an impact on stunting. Concerning income inequality, we can not prove the existence of a specific channel on children's health, but of several depending on the income inequality dimension considered. Finally, we show that the highest socio-economic categories are those that derive the greatest benefit for the children from the presence of health infrastructure.