Did household health behavior and private action play a role in lowering child mortality during the epidemiological transition? To measure child mortality that spans this era, we link households between censuses and identify whether young children are present in the subsequent census. Using this (near) century-long measure of child mortality, we test whether households with access to knowledge of Germ Theory experienced changes in child mortality after its dissemination. Our results suggest that two groups experienced the clearest reduction in mortality: the children of physicians and the children of parents from high socioeconomic status backgrounds. We show that neither household head economic status nor educational attainment alone was sufficient to effectively apply the insights from Germ Theory at home.