338 - The Impact of Pattern of Payment for Opiate in Pregnancy on Rate of Adverse Neonatal Outcomes in Arkansas

Poster Number: 338
Publication Number: 338.1660

Background: Prior reports on opiate utilization among pregnant women are largely based on administrative databases and can underestimate opiates purchased by out-of-pocket payment.

Objective: Evaluate the patterns of opiate purchases by payment source among pregnant women in Arkansas (AR) and assess the impact of payment variation on the rate of adverse neonatal outcomes.

Design/Methods: All liveborn singleton infants, born at 22 to 41 weeks' gestation in AR in 2016, were identified from the Birth Certificate Database (BCD). The BCD was linked to the AR All State Payer Database (APCD), a state database that includes claims from private and public payers, and the Prescription Drug Monitoring Program (PDMP), a database that tracks the dispensing of controlled substances irrespective of payment source. Infants were excluded if they could not be uniquely linked to their maternal records, if the beginning of the pregnancy was earlier than 1/1/2016 or if their mothers did not have continuous insurance during pregnancy. Neonatal and maternal characteristics were extracted from BCD and APCD. The characteristics of purchased opiates were assessed from APCD and PDMP. The study cohort was divided into pregnant women who never purchased opiates vs. those who purchased opiates at least once during their pregnancy. The buyers’ group was subdivided into those who only used insurance vs. those who paid at least once out-of-pocket to purchase their opiate prescriptions. We compared maternal and neonatal characteristics and evaluated the rate of adverse neonatal outcomes. Logistic regression models were fit to assess the difference in rates of adverse neonatal outcomes, controlling for maternal characteristics.

Results: The final study cohort included 4986 unique infants and mothers where almost 20% of pregnant women bought opiates during pregnancy and 16% of those purchased at least one opiate out-of-pocket. Opiate buyers were more likely to smoke and have a substance use disorder (Table 1). Out-of-pocket opiate buyers purchased more opioids than those using insurance exclusively (Table 2). Logistic regression analyses revealed that infants born to mothers who used insurance were more likely to have a NICU admission and both insurance and out-of-pocket opiate buyers were less likely to breast feed (Table 3).

Conclusion(s): Opiate buyers vs. non-buyers were at higher risk of adverse neonatal outcomes, though out-of-pocket users were not at higher risk of these outcomes in comparison to insurance users. Current work is still in progress to evaluate the association of the different subgroups with risk of neonatal withdrawal symptoms.