INTRODUCTION: Early neonatal deaths are related to access to good quality prenatal, hospital delivery and newborn health care. In Minas Gerais, Brazil, hospitals are heterogeneously distributed, which may result in various levels of neonatal mortality among the State's regions. OBJECTIVE: To investigate early neonatal hospital mortality in the Hospital Information System (SIH/SUS) and assess possible associations between early neonatal hospital mortality rates (ENMR_SIH) and variables regarding care to pregnant women and the newborn in strata of homogeneous cities. METHOD: The SIH/SUS was used to obtain the number of births and deaths according to place of residence. Cities were grouped according to micro-region and population size, totaling 199 groups. The CART (Classification and Regression Tree) identified three strata of homogeneous city groups according to socioeconomic factors. For each stratus, a Spearman correlation matrix was used to evaluate associations between rates and health care indicators. RESULTS: The ENMR_SIH for Minas Gerais was 10.9/1000 live births. A greater probability of an ENMR_SIH under 8/1000 live births was observed in areas with poorer socioeconomic conditions. A positive correlation between the ENMR_SIH and cribs per woman in fertile age and the proportion of low birth weight in Strata 1, and between the ENMR_SIH and physicians per inhabitant in Strata 3 was observed. CONCLUSION: Issues concerning access to healthcare, underreporting of deaths and the misclassification of live births as stillbirths may be reflected in rates. The number of births and deaths registered in the SIH/SUS justifies studies on the data collected and investments in the quality of those records.
Neonatal mortality; Socioeconomic factors; Information systems; Death registration; Underreporting; SUS (BR)
