Association of neighborhood social determinants of health, race and ethnicity, and severe maternal morbidity on the frequency of emergency department visits and preventable emergency department visits among pregnant individuals

Abstract

BackgroundThe relationship between emergency department (ED) use and SDoH (both individual or neighborhood) factors is complex, and critical factors such as racioethnicity and obstetric comorbidities may moderate this relationship among pregnant individuals. The public health implications of this complex relationship are important for pregnant women because frequent ED visits (both non-preventable and preventable) increase the likelihood of adverse maternal and infant health outcomes and resource burden to communities.MethodsOur quantitative study analyzed clinical, billing, and census data about pregnant individuals from a Tertiary Medical Center (TMC) between 2017 and 2020. To classify visits as preventable, we used an updated New York University ED algorithm. The address of the patient during their ED visit was linked to an area-level deprivation measure to represent neighborhood SDoH. Race and ethnicity data were extracted from the electronic health record and clinical diagnosis data was extracted for obstetric comorbidity ICD-10 codes related to increased risk of severe maternal morbidity (SMM). The date of a clinical diagnosis was used to determine if a specific set of comorbidities were present during a pregnancy. Other sociodemographic and clinical variables were extracted for model adjustment. A negative binomial regression was used to fit the data (n=13,357) to examine the frequency of ED and preventable ED visits based on neighborhood SDoH, race and ethnicity, obstetric comorbidity, and the interactions of these variables.ResultsAdjusted model estimates indicated that individuals who identified as non-Hispanic Black experienced higher frequency of ED use across all levels of neighborhood deprivation, and the ED use among non-Hispanic Black individuals in least deprived neighborhoods were higher than or similar to individuals who identified with other racial and ethnic groups who lived in the most deprived neighborhoods. Non-Hispanic Black individuals had the highest frequencies of ED use compared to individuals who identified with other race and ethnic groups whether SMM was present or not, and the frequencies of ED use among non-Hispanic Black individuals with an absence of obstetric comorbidity was higher than individuals who identified with other race and ethnic groups with a comorbidity. Model estimates also indicate that the probabilities of preventable ED visit did not vary race and ethnicity intersected by SDoH. Individuals with obstetric comorbidities had higher probability of a preventable ED visit compared to those not at risk of SMM regardless of different levels of SDoH opportunity. Our study quantifies these differences in estimates between neighborhood SDoH, race and ethnicity, and SMM risk.ConclusionsExamination of interventions to address higher ED use among pregnant individuals require an intersectional lens through which policymakers can gain a nuanced perspective on how ED use is influenced by SDoH, race and ethnicity, and risk of SMM among vulnerable individuals.