OBJECTIVE: To assess the impact of recent changes in neonatal intensive care on the mortality and morbidity of very-low-birth-weight neonates (< 1501 g). DESIGN: Prospective cohort study. SETTING: Six neonatal intensive care units in Wisconsin and Iowa. PARTICIPANTS: All very-low-birth-weight neonates who were admitted to the neonatal intensive care units the year before the availability of exogenous surfactant (n = 333), during the investigational new drug protocol for synthetic surfactant (Exosurf) (n = 347), and after the release of synthetic surfactant (n = 356) (designated as periods 1, 2, and 3, respectively). INTERVENTIONS: None. MAIN RESULTS: The percentage of neonates receiving exogenous surfactant in the three periods was 3%, 37%, and 56%, and the percentage receiving antenatal steroids was 12%, 17%, and 27% (P = .0001 for increase in both modalities). The percentage of neonates dying in the three periods was 23%, 14%, and 19% (P = .05 for downward trend). The percentage of neonates with intraventricular hemorrhage decreased in the subgroup weighing between 700 and 1350 g (35%, 28%, and 24%) (P = .04) and increased in the subgroup weighing below 700 g (8%, 41%, and 45%) (P = .03). The percentage of neonates with bronchopulmonary dysplasia increased from 21% to 36% between periods 1 and 2 (P = .003) and decreased to 27% (P = .04) in period 3. Antenatal steroid use was strongly associated with the decrease in intraventricular hemorrhage (odds ratio, 0.35) and mortality risk (odds ratio, 0.20). CONCLUSIONS: Several developments in care have contributed to changes in mortality risk, incidence of intraventricular hemorrhage, and the severity of respiratory disease in very-low-birth-weight infants.