In Latin America mortality estimates and studies are limited by data quality (i.e. Nunez and Salguero, 1982; Moser, 1985; Coale and Kisker, 1986; Dechter and Preston, 1991; Timaeus, Chackiel and Ruzicka, 1996; Popolo, 2000; Jorge and Gotlieb, 2000; Hil, 2003; Banister and Hill, 2004). The most common problems are incomplete coverage of vital registration systems and errors in age declaration of both in total population and mortality records (United Nations, 1983; Preston, Elo, Stewart, 1999), which are essential to the estimation of mortality rates. To circumvent these problems, different methodologies have been developed to evaluate data quality and produce estimates of infant, child and adult mortality. The data limitation has implications on several instances, such as the simple fact of measuring and targeting the New Millennium Development Goals (infant mortality, and maternal mortality) as to the development of substantive epidemiological and academic studies on mortality in Latin America, which could help us understand trends, regional differences, that are linked to the study of causes and consequences of mortality changes. This project intends to create series of consistent mortality data by age, sex, and whenever possible by cause and region in five Latin American countries. These series will be provided to users in two different ways. First, the data as it has been collected (raw data) and the proposed corrections by under-registration, using demographic techniques.