Variation in Neonatal Intensive Care and Post-Discharge Outcomes in Military-Insured Newborns

OBJECTIVE: Neonatal care, including intensive care, varies across health care regions. This study measured regional risk-adjusted use of military-insured newborn care and associations with 7-day post-discharge hospital care. PATIENTS AND METHODS: A total of 420 175 live births to military-insured TRICARE beneficiaries, from October 1, 2015 through December 31, 2020, were assigned to regional catchment areas (n = 103) based on birth location. Administrative records provided sponsor/parental/newborn factors associated with neonatal mortality that were used to adjust regional neonatal intensive care unit (NICU) use rates. The associations of these rates and 7-day post-discharge hospital care were estimated with weighted Pearson and Spearman coefficients. RESULTS: Adjusted regional NICU admission rates and numbers of special care days (SCDs) generally varied little for very preterm (VPT) and moderate preterm (MPT) newborns (coefficients of variation [CVs; SD/mean]<10). Regional variation was high for all types of use in late preterm (LPT) and nonpreterm (NPT) newborns (CVs generally >20). Associations between regional NICU admission and 7-day discharge events were weakly negative for LPT (weighted Pearson correlation coefficient, -0.42; 95% CI, -0.56 to -0.25; R2 0.17) and NPT (-0.35; 95% CI, -0.49 to -0.20; R2 0.12) newborns. Correlation coefficients between NICU admission rates and 7-day post-discharge events for VPT and MPT newborns and for the number of SCDs across all subcohorts were not statistically significant. Similar correlations were observed for 30- and 90-day post-discharge outcomes. CONCLUSIONS: Regional variation in NICU use was high for LPT and NPT newborns in both military and civilian hospitals and had weak or unassociated risk of post-hospital care.