Interhospital Transfers of Maternal Patients: Cohort Analysis of Nationwide Inpatient Sample, 2011.

Objective The objective of this study was to estimate the annual rate of interhospital transfers of pregnant and postpartum women in the United States and analyze associated patient and health system characteristics as measures of regionalized perinatal care performance. Methods Separate weighted univariate analyses of the 2011 Nationwide Inpatient Sample (NIS) were performed for all maternal discharges, in-hospital deaths, and transfers. Multivariable logistic regression analyses for transfer dispositions adjusted for health system characteristics, maternal demographics, and diagnoses were performed. Additional perinatal service characteristics were analyzed using NIS merged with the 2011 American Hospital Association Annual Survey database. Results An estimated 18,082 patients, 0.43% of maternal hospitalizations, were transferred to an acute care hospital; 81% occurred without childbirth delivery before transfer. Transfers were toward larger, urban teaching hospitals and hospitals with higher levels of obstetrical and neonatal care and were more likely in states with 4.0 maternal-fetal medicine specialists/10,000 live births. Blacks and Native Americans were more likely and Hispanics and Asians were less likely than white patients to be transferred. Privately insured women were less likely to be transferred than were others. Transfers were associated with life-threatening maternal diagnoses and fetal indications. Conclusion Transfers reflected a risk-based regionalized system of perinatal care, with racial and payer differences.