The processes and outcomes of medical care for very low birth weight (VLBW) infants (under 1500 grams at birth) vary markedly among different neonatal intensive care units (NICUs). Substantial variation in both mortality and morbidity remain even after appropriately adjusting for patient risk. Infants with very low birth weights are a high risk patient population, accounting for 46 percent of infant deaths in the US each year. Evidence-based selective referral has the potential to improve outcomes for these fragile infants. However, the current standard (set by the Leapfrog Group) has several potential problems for NICU care of these infants: (1) it does not take into consideration the multidimensionality of quality of care and (2) it is based only on provider volume. In this project, we propose to: 1) Improve the evidence base for selective referral of VLBW infants by answering the following questions: - How good is volume as an indirect quality indicator for a broad range of medical outcomes affecting VLBW infants? How much systematic variation does volume explain in a broad range of outcome measures related to mortality and morbidity for VLBW infants? - What other indirect quality indicators, in combination with volume, could better explain the variation in outcome measures for VLBW infants? 2) Evaluate alternative approaches to selective referral by answering the following questions: - What are the implications for infant mortality and morbidity of using volume versus alternative quality measures as the basis for selective referral for VLBW infants? - What are the tradeoffs of better outcomes with treatment costs and travel times of families? The study will use data from the Vermont Oxford Network (VON) and linked discharge-abstract and vital statistics data for California. VON contains 40% of the NICUs in the US and half of all VLBW births in the US each year. Of all states, California has the largest number of total births and VLBW infants born in the country each year.
