Differences in morbidity and mortality rates in black, white, and hispanic very preterm infants among New York city hospitals

Currently, 98%of neonatal intensive care units (NICUs) have a risk-adjusted rate of mortality equal to the best 10%of units in 2005, a statistic showcasing the significant improvements in neonatal care over the past decade. Despite improvements, research shows significant racial disparities in neonatal mortality and morbidity, with non-Hispanic black neonates and Hispanic neonates experiencing a disproportionate amount of these adverse events. Data show that non-Hispanic black infants are more likely to be born in hospitals with poor neonatal outcomes, and that risks of morbidity have been related to hospital factors. The extent that variation in hospitals contributes to racial and ethnic disparities in neonatal mortality and morbidity is unclear. This population-based retrospective cohort study aimed to measure very preterm birth (VPTB), gestational age less than 32 weeks, neonatal mortality and morbidity rates among non-Hispanic black, Hispanic, and non-Hispanic white infants, and determine to what extent these differences are related to site of delivery. Data were gathered from a linked data set including maternal procedures and diagnosis codes, all live births in New York City hospitals between 2010 and 2014, and all NICU deaths between 2010 and 2015. The primary outcome variable was neonatalmortality and severemorbidity. Risk-standardized neonatal morbidity and mortality rates were generated for each hospital, and these were ranked from lowest to highest. The distribution of black, Hispanic, and white VPTBs between these hospitals was then examined. The total cohort for analysis included 7177 VPTB infants born in 39 hospitals across New York City. Of the total 595,835 births during the study years, VPTB infants accounted for 1.5% of total births, 2.8% of black births, 1.5% of Hispanic births, and 1.0% of white births. Of the 7177 VPTB infants, morbidity and mortality occurred in 28% (n = 2011), 32.2% of black infants, 28.1%of Hispanic infants, and 22.5%of white infants (P < 0.001). Themorbidity and mortality was more than 2-fold greater for the hospitals in the highest tertile as comparedwith those in the lowest tertile.Of the proportions of VPTB neonates born in the highest morbidity and mortality tertile of hospitals, 1204 (43.4%) of 2275 black, 746 (34.4%) of 2168 Hispanic, and 325 (22.9%) of 1418white were born in this setting (P < 0.001). There was a 20%risk difference for black compared with white very preterminfants (95%confidence interval [CI], 18%-23%), and an 11%risk difference for Hispanic compared with white very preterm infants (95% CI, 9%-14%). Differences among hospital of birth explained 39.9% (95% CI, 30%-50%) and 29.5% (95% CI, 10%-49%) of the black-white and Hispanic-white disparity in outcomes, respectively. The data show that black and Hispanic VPTB infants have a higher chance to be born in hospitals in New York City with greater risk-adjusted neonatal morbidity and mortality rates compared with white VPTB infants. These disparate patterns of utilization and wide variation across hospitals contribute directly to racial VPTB morbidity and mortality in New York City.