PC2058: Disparities in Neonates, the Missing Variable: Recording Maternal Race in the Neonatal Electronic Health Record

Background: Research on disparities in follow-up and outcomes of neonates depends on certain maternal variables being consistently recorded in the electronic health record (EHR) such as maternal race and ethnicity, smoking status, and education level. This is because there is increasing literature that disparities in neonatal outcomes are linked to disparities across the life-course, including in all-cause mortality, chronic disease outcomes, and even income inequality. For instance, mothers who identify as Black Americans have higher odds of postpartum mortality from preventable causes and are more likely to have babies who are born preterm. Preterm babies in turn are at risk for life-long disparities in development, growth, and other outcomes, including mortality and income in early- to mid-adulthood. As a result, disparities research on neonatal outcomes typically uses self-reported maternal race as the preferred social risk factor, as opposed to neonatal race. However, EHRs vary in their documentation of maternal race in neonatal charts.

Methods: Our primary objective was to determine how reliably maternal race was recorded in an EPIC-program EHR at a regional specialized medical center. The goal of this was to determine how feasible collecting maternal race for neonatal disparities research through manual chart review would be at this institution. We developed a cohort list of all patients who were born between January 1st, 2014 and December 31st, 2017, who had neonatal and pediatric intensive care unit stays at a regional specialized medical center during the neonatal period, and who were referred to the high-risk neonatal developmental clinic at that center. Of this cohort, a random subset of 50 patients was selected for manual chart review of the neonatal EHR from their first perinatal admission. We hypothesized that documentation of maternal race and ethnicity in the neonate’s EHR of their initial admission period would have a higher rate of missingness than documentation of the patient’s race and ethnicity.

Results: Of the 50 patients in our sample, 70% of neonates were white and non-Hispanic, 12% were Black, 6% were Hispanic, and 6% were Asian. Maternal race and ethnicity had the highest rate of missingness, at 94%. Neonatal race and ethnicity had a much lower rate of missingness, at 2%.

Conclusion: Self-reported maternal race and ethnicity is a significant demographic risk factor for neonatal disparities in follow-up rates and developmental outcomes that is not consistently recorded in neonatal EHRs. This deficiency precludes retrospective chart review research on neonatal disparities. Neonatal race and ethnicity can be used as a surrogate, but is not ideal. Documenting self-reported maternal race in neonatal EHRs would empower us both to evaluate quality improvement initiatives and to perform generalizable disparities research, which will hopefully improve the follow-up rates and developmental outcomes of our neonatal patients.