DO NEONATAL INTENSIVE CARE UNITS SAVE LIVES? LONGITUDINAL EVIDENCE FROM UNITED STATES COUNTIES, 1976 - 2013

Abstract

Given the United States’ high healthcare spending and the currents of costcontainment, many medical care programs are being reevaluated. One such program is neonatal intensive care, which has emerged as a life-saving but expensive field of medicine. Although infant and neonatal mortality rates have declined substantially over the past century, it is unclear how much of the decline can be attributed to neonatal intensive care units (NICUs). This study examines the effect of NICUs on infant and neonatal mortality rates in the United States from 1976 to 2013. I use a fixed effects model to analyze data from the American Hospital Association’s Annual Survey of Hospitals and Compressed Mortality Data from the National Center for Health Statistics. I find that NICUs opened in 1976 to 1989 were associated with a reduction of 0.561 deaths per 1000 live births (4.6%) in infant mortality rate and 0.335 deaths per 1000 live births (3.7%) in neonatal mortality rate. Conversely, NICUs opened after this period did not impart significant reductions in mortality. The alignment of the late period with the emergence of deregionalization of perinatal care suggests that deregionalization indeed obstructed the beneficial effects of NICUs. Additionally, the leveling off of NICU benefits despite increased utilization of interventions and care highlights the need to reevaluate the use of the least cost-effective technologies. However, because a large body of work suggests that returns to neonatal intensive care are quite high despite high costs, the reorganization rather than the reduction of care may be preferred. NICUs have indeed saved lives, but their diminished effect over time indicates that these strategies ought to be pursued.