This case-control study with 132 cases and 264 controls aimed to determine predictors of neonatal mortality using hierarchical modeling. Cases were defined as newborns that died within 28 days of birth, and controls as the survivors, among infants of mothers living in Fortaleza, Ceará State, Brazil. Hierarchical logistic regression identified factors associated with neonatal death: maternal race, with brown/black race showing a protective effect (OR = 0.23; IC95%: 0.09-0.56), time spent from home to the hospital > 30 minutes (OR = 3.12; 95%CI: 1.34-7.25), time < 1h or > 10 hours between hospital admission and delivery (OR = 2.43; 95%CI: 1.24-4.76), inadequate prenatal care (OR = 2.03; 95%CI: 1.03-3.99), low birth weight (OR = 14.75; 95%CI: 5.26-41.35), prematurity (OR = 3.41; 95%CI: 1.29-8.98), and male gender (OR = 2.09; 95%CI: 1.09-4.03). In this case series, neonatal deaths were associated with the quality of prenatal care and direct care during labor.