Abstract: Congenital human cytomegalovirus (cCMV) infection is the most common viral infection acquired in-utero and a significant cause of neurodevelopmental abnormalities in infants and children. Although the development of prophylactic vaccines to prevent or modify this intrauterine infection is a high priority, the unique epidemiology of cCMV infections, particularly in highly seroimmune maternal populations presents a potential hurdle for vaccines that will induce immunity similar to that following natural infection. Paradoxically, the rates of cCMV infections are highest in maternal populations with the highest rates of seroimmunity. In addition, the incidence of clinically apparent congenital infections and long term neurodevelopmental sequelae in these populations are similar to those in populations with lower rates of maternal seroimmunity. In this application we will explore the hypothesis that reinfections with new strains of CMV contributes significantly to the incidence of non- primary maternal infections that result in cCMV infection and that the source of new strains of CMV are household exposures, specifically young children. Our approach will be to identify sources of virus exposures, introduce a cognitive-behavioral intervention to reduce the risks of these exposures, and to determine the impact of this intervention on the prevalence of cCMV infections in a population of pregnant women with near universal seroimmunity to CMV prior to conception.