Regional Growth in US Neonatal Intensive Care Capacity and Mortality, 1991-2020

Abstract

ImportanceThe effectiveness of neonatal intensive care in very ill newborns has led to rapid growth in US neonatal intensive care unit (NICU) capacity that is uncorrelated with regional perinatal risk. It is not known if there is an association between growth of regional capacity and newborn mortality.ObjectiveTo estimate the association between change in NICU capacity and neonatal mortality across 246 neonatal intensive care regions.Design, Setting, and ParticipantsIn this repeated cross-sectional study of US infants, the association between change in regional capacity and mortality was estimated in the years 1991, 2003, 2007, 2012, 2017, 2018, 2019, and 2020 using Poisson generalized estimating equations models adjusted for maternal and newborn characteristics, with newborns as the units of analysis. Data were analyzed June 30, 2024. This study used a 25% sample of all US infants born live with a birth weight of 400 g or more and gestational age of between 22 and less than 45 weeks (N = 30 902 221 newborns).ExposureChange in regional NICU capacity, measured as both counts of neonatologists and staffed NICU beds per 1000 live births (LBs) from 1991 to the birth year.Main Outcomes and MeasuresThe primary outcome was neonatal (<28 days) mortality and the secondary outcome was 180-day mortality.ResultsFrom 1991-2020, total adjusted neonatologists and NICU beds per 1000 LBs increased from 0.44 to 1.44 (227%) and 5.43 to 8.02 (48%), respectively, while neonatal mortality decreased from 3.87 to 2.21 (−43%) and 180-day mortality decreased from 6.27 to 3.19 (−49%) per 1000 LBs. There was no meaningful correlation between change in regional capacity (neonatologists: r, −0.12; 95% CI, −0.25 to 0.00; NICU beds: r, −0.07; 95% CI, −0.19 to 0.06) and change in regional neonatal mortality. No meaningful associations with capacity were observed in multilevel models (neonatologists: adjusted relative rate [aRR], 1.01; 95% CI, 0.93-1.01; NICU beds: aRR, 1.00; 95% CI, 0.99-1.00) nor was 180-day mortality associated with capacity. No associations were observed in birth cohorts stratified by relative need based on gestational age, maternal education, or maternal race or ethnicity.Conclusions and RelevanceIn this cross-sectional study, growth in regional NICU capacity was not associated with observable mortality benefit. Additional studies are needed to investigate the costs and benefits associated with NICU care expansion.