Risk factors for early neonatal mortality

OBJECTIVE: To assess risk factors for early neonatal mortality. METHODS: A population-based case-control study was carried out with 146 early neonatal deaths and a sample of 313 controls obtained among survivals of the neonate period in the south region of the city of S?o Paulo, in the period of 8/1/2000 to 1/31/2001. Information was obtained through home interviews and hospital charts. Hierarchical assessment was performed in five groups with the following characteristics 1) socioeconomic conditions of mothers and families, 2) maternal psychosocial conditions, 3) obstetrical history and biological characteristics of mothers, 4) delivery conditions, 5) conditions of newborns RESULTS: Risk factors for early neonate mortality were: Group 1: poor education of household head (OR=1.6; 95% CI: 1.1;2.6), household located in a slum area (OR=2.0; 95% CI: 1.2;3.5) with up to one room (OR=2.2; 95% CI: 1.1;4.2); Group 2: mothers in recent union (OR=2.0; 95% CI: 1.0;4.2), unmarried mothers (OR=1.8; 95% CI: 1.1;3.0), and presence of domestic violence (OR=2.7; 95% CI: 1;6.5); Group 3: presence of complications in pregnancy (OR=8.2; 95% CI: 5.0;13.5), previous low birth weight (OR=2.4; 95% CI: 1.2;4.5), absence of pre-natal care (OR=16.1; 95% CI: 4.7;55.4), and inadequate pre-natal care (block 3) (OR=2.1; 95% CI: 2.0;3.5); Group 4: presence of clinical problems during delivery (OR=2.9; 95% CI: 1.4;5.1), mothers who went to hospital in ambulances (OR=3.8; 95% CI: 1.4;10.7); Group 5: low birth weight (OR=17.3; 95% CI: 8.4;35.6) and preterm live births (OR=8.8; 95% CI: 4.3;17.8). CONCLUSIONS: Additionally to proximal factors (low birth weight, preterm gestations, labor complications and unfavorable clinical conditions in gestation), the variables expressing social exclusion and presence of psychosocial factors were also identified. This context may affect the development of gestation and hinder the access of women to health services. Adequate prenatal care could minimize the effect of these variables.