Pediatrics, July, 2002, by Jeffrey D. Horbar, Gary J. Badger, Joseph H. Carpenter, Avroy A. Fanaroff, Sarah Kilpatrick, Meena LaCorte, Roderic Phibbs, Roger F. Soll
In the 1990s, new approaches emerged for both the obstetric management of preterm birth and the neonatal care of the prematurely born infant. These included the widespread use of antenatal corticosteroids for women at risk for preterm delivery, (1) surfactant for the prevention and treatment of neonatal respiratory distress syndrome, (2) postnatal steroids for chronic lung disease (CLD), (3) and new modes of respiratory support for neonates with respiratory distress. (4-6) Furthermore, structural changes in the health care system have resulted in the deregionalization of perinatal and neonatal care. (7) It is uncertain how these developments have affected routine care and what impact they have had on patient outcomes. In particular, it is unclear whether the historical trend of steadily improving mortality for very low birth weight (VLBW) infants that has been observed over the past decades has continued throughout the 1990s. (8-10)
To address this question, we used the Vermont Oxford Network Database to identify trends in medical practices and patient outcomes for infants with birth weights of 501 to 1500 g born from 1991 to 19997 Results for infants born in 1990 have been previously reported. (12)
Vermont Oxford Network
The Vermont Oxford Network is a voluntary collaboration of health professionals whose mission is to improve the quality and safety of medical care for newborn infants and their families through a coordinated program of research, education, and quality improvement. (11) In support of this mission, the Network maintains a database for infants with birth weights of 401 to 1500 g. Members adhere to uniform definitions included in the Network's Database Manual of Operations. (13)
Infants with birth weights of 401 to 1500 g (501 to 1500 g before 1996) are eligible for inclusion in the database if they were born at a Network center or were transferred to it within 28 days of birth. This report is based on an analysis of the data for all eligible infants born from 1991 and 1999 with birth weights between 501 and 1500 g.
Statistical analyses were performed using SAS statistical software version 8.1 (SAS Institute, Cary, NC). The significance associated with changes over time in the dichotomous intervention and outcome measures were first analyzed based on logistic models containing only time as a predictor variable. Next, multivariate models were fit to evaluate changes over time while adjusting for changes in infant characteristics (gestational age, race, gender, location of birth, multiple birth, and size <10th percentile for gestationa