756 - The impact of maternal race and ethnicity on the resuscitation of infants born at the edge of viability

Publication Number: 2737.756

Abigail Whitney, Children's Hospital of Philadelphia, Philadelphia, PA, United States; Brielle Formanowski, Children's Hospital of Philadelphia, Philadelphia, PA, United States; Diana Montoya-Williams, Childre's Hospital of Philadelphia, Philadelphia, PA, United States; Scott A. Lorch, The Children's Hospital of Philadelphia, University of Pennsylvania, Philadelphia, PA, United States

Background: It has been shown that racial and ethnic differences exist in the care provided to infants born at the earliest gestational ages. As the landscape of periviable resuscitation has evolved with higher rates of resuscitation at 22 weeks' gestation, it is important to evaluate if and how rates of resuscitation have changed by maternal race and ethnicity.

Objective: To compare the rates of intensive resuscitation (as defined by intubation, central line placement, or inpatient length of stay >1 day) for neonates born in the periviable period by maternal self-reported race and ethnicity.

Design/Methods: We obtained maternal hospital records linked to infant birth certificates or fetal death records and infant discharge or death records from 2000-2020 in 5 states. Only live birth and in-hospital fetal deaths between 22w0d-24w6d gestational age (GA) with a birth weight >350g were included (n=15,758). Multivariable mixed-effects logistic regressions were performed to determine the association between maternal race/ethnicity and the odds of intensive resuscitation, adjusting for maternal age, education, insurance, prenatal care, previous births, smoking, co-morbidities, infant sex, and birth weight. Finally, birth hospital was included as a random effect to control for hospital-level practice variation.

Results: Across the entire cohort, 69.2% of infants underwent intensive resuscitation. Table 1 shows the distribution of selected infant and maternal characteristics by resuscitation status. Infants born to non-Hispanic Black (NHB) women had the highest rates of intensive resuscitation with 32.4%, 75.7%, and 92.6% of infants at 22, 23, and 24 weeks GA, respectively, undergoing resuscitation (Figure 1). When controlling for maternal/infant characteristics, the odds of resuscitation were higher among births to NHB women (aOR=1.47; 95% CI 1.32-1.64) and lower in Hispanic (aOR=0.86; 0.77-0.96) and Asian/Pacific-Islander births (aOR=0.72, 0.61-0.85) when compared to non-Hispanic White births (Figure 2). Multilevel models including birth hospital showed only births to NHB women were significantly different in odds of resuscitation (aOR=1.28, 1.13-1.45). Among infants born at 22 weeks GA, those born to NHB women had the highest odds of resuscitation, after controlling for maternal/infant characteristics (aOR=1.66, 1.32-2.08) and hospital (aOR=1.45, 1.12-1.89).

Conclusion(s): National variation exists in odds of resuscitation in the periviable period by maternal race/ethnicity. Infants born to NHB women had the highest odds of undergoing intensive resuscitation, independent of maternal/infant characteristics and birth hospital.

Table 1
Distribution of maternal and infant characteristics by intensive resuscitation status (defined by intubation, central line placement, or neonatal intensive care unit length of stay >1 day) for live births and in-hospital fetal deaths in the periviable (22w0d-24w6d) period.

Figure 1
Rates of intensive resuscitation (defined by intubation, central line placement, or neonatal intensive care unit length of stay >1 day) by gestational age and maternal race/ethnicity.

Figure 2
Odds ratios (unadjusted; adjusted* for maternal/infant characteristics including maternal age, education, insurance, prenatal care, previous live births, smoking, medical co-morbidities, infant sex, and birth weight; and adjusted** for maternal/infant characteristics and birth hospital) of intensive resuscitation (defined by intubation, central line placement, or neonatal intensive care unit length of stay >1 day) by maternal race/ethnicity when compared to non-Hispanic White births, stratified by completed weeks of gestation. Error bars represent 95% confidence intervals.