నియోనాటల్ బయాలజీ జర్నల్

నైరూప్య

Maternal Demographics and Extremely Preterm Infant Mortality in the United States

Beau J Batton, Reem Y Nubani, Christopher P Burnett, Steven J Verhulst and Daniel G Batton

Background: The mortality rate for term infants varies by location and maternal demographics, but the relationship between these factors and the extremely preterm Infant Mortality Rate (IMR) may be different. Understanding the epidemiology of the extremely preterm IMR is important because infants born before 28 weeks gestation account for more than one-third of all infant deaths.

Objectives: 1) Estimate the extremely preterm IMR for different states and locations (urban versus rural county) of the United States; 2) evaluate the impact of maternal race and education on this IMR and; 3) investigate the timing of death for extremely preterm infants who die in the first year.

Methods: Data were analyzed from the National Center for Health Statistics to investigate the influence of birth location, maternal race, and maternal education on the IMR for infants born alive in the United States at 200/7-276/7 weeks gestation from 1995-2005.

Results: The IMR was 392/1,000 live births for the 306,502 extremely preterm infants investigated. The extremely preterm IMR varied by maternal race and was highest for infants born to White mothers (397, 95% confidence interval (CI): 395, 399) and lowest for those born to Black/African American mothers (386, 95% CI: (383, 389)). Rural county birth was associated with a higher extremely preterm IMR than urban county birth (p=0.006). Maternal education had a bimodal relationship with the extremely preterm IMR such that the highest rates occurred with the least or greatest amount of education. With multi-variable analysis, birth in a rural county and maternal race negated the effects of maternal education on the extremely preterm IMR.

Conclusion: The extremely preterm IMR varies widely based on demographics known at the time of delivery. This variability is distinct from that observed for more mature infants and may partially explain differences in IMR across neonatal intensive care units.