Perinatally acquired bacterial neonatal sepsis is a low-incidence, high-risk disease with a relatively benign treatment. Accurate diagnosis is difficult because there is no definitive diagnostic test; even blood cultures have an unacceptably low sensitivity. Therefore, the clinician must accept that a number of neonates will have treatment initiated for sepsis who do not have the disease. In order to treat rapidly all infants with sepsis and to minimize therapy for those without infection, historical, clinical, and laboratory data can be used together in a management approach to achieve optimal results.