Original Investigation
Pediatrics
February 14, 2024
Hospital-Level NICU Capacity, Utilization, and 30-Day Outcomes in Texas
David C. Goodman, Patrick Stuchlik, Cecilia Ganduglia-Cazaban, et al
JAMA Netw Open. 2024;7(2):e2355982. doi:10.1001/jamanetworkopen.2023.55982
Question Is hospital-level neonatal intensive care unit (NICU) bed supply associated with higher risk-adjusted newborn utilization and better outcomes?
Findings In this cohort study with 874 280 newborns, NICU bed supply was associated with statistically higher likelihood of NICU admission and special care days among late preterm and nonpreterm newborns, but not among very low birth weight newborns. Higher bed supply was not associated with lower inpatient mortality and 30-day postdischarge adverse events.
Meaning These findings suggest that there may be overcapacity of NICUs in some health care regions and overuse of NICU care in some newborn populations; further investigation into the benefits of additional NICU capacity expansion is warranted.
Abstract
Importance Risk-adjusted neonatal intensive care unit (NICU) utilization and outcomes vary markedly across regions and hospitals. The causes of this variation are poorly understood.
Objective To assess the association of hospital-level NICU bed capacity with utilization and outcomes in newborn cohorts with differing levels of health risk.
Design, Setting, and Participants This population-based retrospective cohort study included all Medicaid-insured live births in Texas from 2010 to 2014 using linked vital records and maternal and newborn claims data. Participants were Medicaid-insured singleton live births (LBs) with birth weights of at least 400 g and gestational ages between 22 and 44 weeks. Newborns were grouped into 3 cohorts: very low birth weight (VLBW; <1500 g), late preterm (LPT; 34-36 weeks’ gestation), and nonpreterm newborns (NPT; ≥37 weeks’ gestation). Data analysis was conducted from January 2022 to October 2023.
Exposure Hospital NICU capacity measured as reported NICU beds/100 LBs, adjusted (ie, allocated) for transfers.
Main Outcomes and Measures NICU admissions and special care days; inpatient mortality and 30-day postdischarge adverse events (ie, mortality, emergency department visit, admission, observation stay).
Results The overall cohort of 874 280 single LBs included 9938 VLBW (5054 [50.9%] female; mean [SD] birth weight, 1028.9 [289.6] g; mean [SD] gestational age, 27.6 [2.6] wk), 63 160 LPT (33 684 [53.3%] female; mean [SD] birth weight, 2664.0 [409.4] g; mean [SD] gestational age, 35.4 [0.8] wk), and 801 182 NPT (407 977 [50.9%] female; mean [SD] birth weight, 3318.7 [383.4] g; mean [SD] gestational age, 38.9 [1.0] wk) LBs. Median (IQR) NICU capacity was 0.84 (0.57-1.30) allocated beds/100 LB/year. For VLBW newborns, NICU capacity was not associated with the risk of NICU admission or number of special care days. For LPT newborns, birth in hospitals with the highest compared with the lowest category of capacity was associated with a 17% higher risk of NICU admission (adjusted risk ratio [aRR], 1.17; 95% CI, 1.01-1.33). For NPT newborns, risk of NICU admission was 55% higher (aRR, 1.55; 95% CI, 1.22-1.97) in the highest- vs the lowest-capacity hospitals. The number of special care days for LPT and NPT newborns was 21% (aRR, 1.21; 95% CI,1.08-1.36) and 37% (aRR, 1.37; 95% CI, 1.08-1.74) higher in the highest vs lowest capacity hospitals, respectively. Among LPT and NPT newborns, NICU capacity was associated with higher inpatient mortality and 30-day postdischarge adverse events.
Conclusions and Relevance In this cohort study of Medicaid-insured newborns in Texas, greater hospital NICU bed supply was associated with increased NICU utilization in newborns born LPT and NPT. Higher capacity was not associated with lower risk of adverse events. These findings raise important questions about how the NICU is used for newborns with lower risk.
