[Infant mortality due to avoidable causes in a city in Northeastern Brazil]. / Mortalidade infantil por causas evitáveis em uma cidade do Nordeste do Brasil.

This is a cross-sectional study that aimed to describe the occurrence of infant mortality in Recife (PE) between 2000 and 2009, second to avoidable causes. The population composed of cases of deaths between 2000 and 2009 among the infants of mothers living in Recife. Deaths were classified as avoidable by using the List of avoidable causes of deaths resulting from interventions within the Brazilian National Health System (SUS). Descriptive statistics were used for data analysis. A decrease in the infant mortality coefficient from 20.4 to 12.1 per 1.000 live births was observed (reduction of 40.6%). From the total of 3.743 deaths registered, 2.861 (76.4%) were classified as avoidable. It was notable that 61.2% of the deaths could have been avoided through appropriate care for the woman during the pregnancy. An approach in which avoidability is analyzed may assist in discussions relating to organization, quality and access to healthcare service, and in identifying deaths that could have been avoided through appropriate mother-child healthcare.