Leveraging Data Science to Understand Outcomes for Mothers and Children Affected by Opioids

PROJECT SUMMARY / ABSTRACT- PROJECT 2 Over the past two decades, the number of women with opioid use disorder (OUD) during pregnancy and the number of infants born affected by opioids has increased substantially. However, there remain key knowledge gaps in our understanding of how neonatal and postpartum treatment for opioid use impacts the maternal- infant dyad. Much of the current research around infants exposed to opioids focuses on those diagnosed with Neonatal Opioid Withdrawal Syndrome (NOWS), potentially under-identifying opioid-exposed infants that don’t develop the syndrome but are at risk for adverse outcomes. Postpartum women with OUD are at risk for outcomes that are detrimental to both mother and infant including overdose, severe maternal morbidity, and death. While we know medications for OUD, including buprenorphine and methadone, can improve short-term outcomes for women, evidence of their impact on maternal outcomes and dyadic stability after delivery is limited. Furthermore, maternal and infant wellbeing is intertwined, yet our understanding of dyadic outcomes (e.g., retaining custody) is limited and research has primarily focused on separate mother and infant wellbeing. Improved understanding of dyadic outcomes has the potential to inform tailored therapeutics and precision medicine for opioid-affected dyads. To address these key knowledge gaps we will utilize data linkages and novel data science methodologies to 1) develop and validate electronic health record-based algorithms to identify opioid-exposed infants and their mothers, identify key covariates for dyad wellbeing, and link to state data systems; 2) test the hypothesis that dyads with prenatal maternal treatment for OUD compared to dyads with untreated maternal OUD have improved birth outcomes and post-natal service utilization of recommended services in the first six months postpartum; and 3) test the hypothesis that dyads with prenatal maternal treatment for OUD compared to dyads with untreated maternal OUD have improved dyadic stability at one- year postpartum and whether infant treatment modifies those outcomes. This proposal will fill critical knowledge gaps and create enduring, scalable data science methods that will foster innovation, data sharing, and research focused on maternal-infant dyads affected by the opioid crisis.