Objectives:
to analyze the contribution of the linkage between the Mortality Information System (SIM) and the Live Births Information System (SINASC) to improving the quality of data on infant mortality (IM) in five cities, one in each Brazilian macroregion.
Methods:
an analysis was conducted of the deaths of children aged under one year registered in the SIM in 2005 and live births in 2004 and 2005 included in the SINASC. The data bases were related using linkage. A statistical association was found between the linkage of data and cities and IM components, as well as the degree of completeness of the variablesbefore and after linkage.
Results:
Ninety percent of deaths were related to the live birth declaration, with the cities of Belém and Goiânia having the lowest proportions of paired records. Porto Alegre had the highest percentage of paired records, followed by Guarulhos and Recife. Both city of residence of the child and death certificate component were found to have a significant influence on the quality of information. Of the 2,464 incomplete fields for variables common to both systems, 92% were recovered after linkage of the databases.
Conclusions:
employing this technique made it possible to improve vital statistics and this demonstrates the potential of this strategy for surveillance of infant mortality at health services.
Infant mortality; Information systems; Epidemiological surveillance
