P3.155: Evaluation of Perinatal Drug Testing Practices from a Social Determinants of Health Perspective

Background: Identification of newborns with in-utero exposure to illicit drugs is critical to providing adequate medical care in the delivery room, nursery, and post discharge. Drug testing protocols to identify newborns with in-utero exposure to illicit substances (e.g., universal drug testing, risk-based screening, provider discretion) vary significantly between institutions, likely due to the potential legal, social, and privacy implications associated with it. The goal of this study was to explore whether disparities exist in drug testing practices during the peripartum period at our institution where drug testing occurs based on clinician discretion.

Methods: This retrospective study took place at a large academic health system in central Pennsylvania. Data was collected from the electronic medical record using a combination of ICD-10 codes and order query. Pregnant individuals between 12 years old and 51 years old who delivered a live newborn at our institution between 4/2021- 4/2022 were included in the study, along with their newborns. Manual chart review was performed to verify the accuracy of the data extraction and to remove patients who did not meet inclusion criteria. Odds ratios (OR) and corresponding 95% confidence intervals (CI) were calculated to assess the likelihood of drug testing based on race and ethnicity. Bivariate analysis was performed to evaluate for an association between sociodemographic variables and chance of being drug tested.

Results: Complete data sets were available for 2008 mother-baby dyads, among whom 147 mothers were drug tested, and 133 infants were drug tested. Compared to non-Black counterparts, Black mothers were 2.2 times more likely to be drug tested (OR 2.2, 95% CI [1.4, 3.5]) and Black newborns were 2.4 times more likely to be drug tested (OR 2.4, 95% CI [1.5, 3.9]). Multiracial newborns were 2.2 times more likely to be drug tested than non-multiracial newborns (OR 2.2, 95% CI [1.4, 3.6]). Additionally, mothers were more likely to be drug tested if they had public insurance, under 35 years old, English was the preferred language, history of substance use, had a diagnosed mental health condition, earlier gestational age at the time of delivery, history of sexually transmitted infection, or their infant was was admitted to the neonatal intensive care unit (NICU). Infants were more likely to be drug tested if they were female, had public health insurance, mother with a diagnosed mental health condition, mother without diabetes, and if the mother was drug tested.

Conclusion: Disparities were identified when drug testing was performed based on clinical discretion, raising concern for underlying bias. Our study adds evidence for standardizing screening guidelines for substance use in pregnancy to minimize bias and mitigate the role of structural racism. Further exploration of the criteria used when clinical discretion is employed to determine who to drug test is warranted.