Neonatal Intensive Care Outcomes in the Military Health System: Comparison of Military and Civilian Hospital Births.

Abstract

Introduction: Military Health System (MHS)-insured newborns receive care in military and civilian hospitals. Differences in delivery location and corresponding payment schemas raise questions regarding possible health system effects on utilization and outcomes. We hypothesize that newborn utilization and clinical outcomes differ between military and civilian hospitals and that the differences may be more pronounced among lower risk newborns (i.e., late preterm and non-preterm infants).

Material and methods: The newborn cohort comprised live births captured in DoD Birth and Infant Health Research program data from October 2015 through December 2020. Population characteristics, hospital measures, and newborn clinical outcomes were examined using administrative medical data. Descriptive statistics for birth hospitalization and post-discharge events were calculated by the birth hospital (military or civilian) and gestational age cohort (very preterm, 23-31 weeks; moderate preterm, 32-33 weeks; late preterm, 34-36 weeks; and non-preterm ≥37 weeks). Risk-adjusted Poisson regression models compared select birth hospitalization events by birth hospital type, accounting for differences in the newborn population with regard to predicted mortality and diagnoses/procedures associated with the use of special care. Adjusted risk ratios (aRRs) and 95% confidence intervals (CIs) were stratified by gestational age cohort.

Results: Overall, 470,175 singleton live births were included, and the majority of births occurred at civilian vs. military hospitals (63.2% vs. 36.8%), with civilian hospitals caring for a higher percentage of preterm infants (7.2% vs. 5.4%). The use of ancillary imaging studies was higher across all gestational age cohorts at civilian hospitals, whereas hospital admission or an emergency room visit within 30 and 90 days of discharge from the birth hospitalization was more likely to occur among infants born at military hospitals. Compared with newborns born at military hospitals, late preterm and non-preterm infants born at civilian hospitals demonstrated an increased risk for longer birth hospitalizations (late preterm aRR = 1.21, 95% CI, 1.17-1.25; non-preterm aRR = 1.04, 95% CI, 1.03-1.05), more special care days (late preterm: aRR = 1.38, 95% CI, 1.31-1.45; non-preterm: aRR = 1.22, 95% CI, 1.17-1.28), and neonatal intensive care unit admission (late preterm: aRR = 1.31, 95% CI, 1.27-1.35; non-preterm: aRR = 1.42, 95% CI, 1.38-1.45); differences were not observed for very and moderate preterm infants.

Conclusions: In this study of MHS-insured newborns, we observed longer lengths of stay, more special care days, and increased neonatal intensive care unit admissions among late preterm and non-preterm infants born at civilian vs. military hospitals. Across all gestational age cohorts, we observed lower rates of ancillary imaging studies and higher rates of post-discharge hospital admission and emergency room visits among military hospital births. Differences by birth hospital type highlight both improved care opportunities and cost considerations for MHS leadership regarding direct and purchased care for this population.