Survival and risk factors for infant mortality: a newborn non-concurrent cohort in Brazil

INTRODUCTION: The infant mortality rate (IMR) is an indicator that reflects the living conditions of the population and is associated with social determinants, as well as the conditions and maternal and child care. Brazil has the IMR higher than other Latin American countries with similar socio-economic development, and still significant differences between its regions. The major health problems for several decades were found in less developed regions of this country. Recently there has been considerable reduction in IMR, yet still remains high, in particular the preventable diseases. The present work focused on risk factors for mortality by linking live births and infant death records. METHODS: The study was conducted in the Brazil. A total of 2.879.624 live births in 2009 and 42.102 infant deaths with follow up until one year constitute a non-concurrent cohort of vital record from health Ministry of Brazil. We found 31.979 linked records (76%). Survival analysis was performed by the method of Kaplan-Meier and analysis to estimate mortality risks by model multivariate of Regression of Cox (p<0,005). RESULTS: The outcome show that highest hazard  ratios (HR) were for Apgar >7 (HR=10,70; CI 10,30-10,90), low birth weight (HR= 5,00; CI 4,84-5,16), pre-term (HR=2,79; CI 2,70 - 2,87), without prenatal (HR= 2,03; CI 1,93-2,03), illiterate (HR=1,47; CI 1,35-1,59), single marital status (HR=1,24; CI 1,20-1,27). CONCLUSIONS: The statistical model used in the present work was effective to point socioeconomic variables with risk factors for infant mortality and with influence in the antenatal care. Therefore the capitation of pregnant, quality prenatal and care during childbirth are fundamental conditions that may have an even higher impact on reducing child mortality in the Brazil.