Date of Award


Degree Type


Degree Name

Doctor of Philosophy in Pharmaceutical Sciences


Biomedical and Pharmaceutical Sciences

First Advisor

Xuerong Wen


Opioid use among pregnant women has markedly escalated in the United States in recent decades. Alarmingly, one-fifth of these women fill one or more prescriptions for opioids despite concerns about perinatal safety. Consequently, the prevalence of opioid use disorder (OUD) in pregnancy and adverse outcomes in offspring has increased in parallel. In this dissertation, we explored the potential mediating roles of adverse pregnancy and birth outcomes residing in pathways between prenatal opioid exposure and neurodevelopmental disorders (NDDs) in children. Additionally, we assessed the potential synergistic effect of opioid and antidepressant use in pregnancy on NDDs in offspring and the potential pathways linking these associations. Furthermore, we investigated the social network of pregnant women with OUD and their healthcare providers.

Manuscript 1: We explored natural direct and joint natural indirect effects (JNIE) of prenatal opioid exposure on NDDs in children mediated through pregnancy complications, major and minor congenital malformations, and adverse neonatal outcomes. We proposed a Bayesian mediation analysis with an elastic net shrinkage prior using Markov chain Monte Carlo algorithms. Simulation studies were conducted to assess model performance. We observed significant JNIE of opioid exposure across three trimesters, indicating that adverse pregnancy and birth outcomes jointly mediated the association between prenatal opioid exposure and accelerated time to NDD diagnosis. The proportion of JNIE increased as the timing of opioid exposure approached delivery.

Manuscript 2: To assess the interaction between antidepressant and opioid use in pregnancy in relation to the risk of NDDs in children, and the potential mediating role of adverse neonatal outcomes in these associations, we developed a Bayesian mediation and interaction analysis estimating total effect, natural direct effect, and natural indirect effects attributable to each exposure alone and the interaction. Confounding variables concerning exposures, mediator, and outcome were selected by incorporating Bayesian group lasso with spike and slab. Our results suggested that adverse neonatal outcomes may reside on the associational pathway between exposure to antidepressants and prescription opioids and accelerated time to NDD diagnosis in children. No evidence of interaction (additive and multiplicative) between the two exposures for direct and indirect effects was observed. However, exposure to both medications was likely attributable to elevated mediated effects compared to exposure to prescription opioids alone.

Manuscript 3: To investigate patient and provider/facility characteristics associated with the observed pattern of healthcare resource utilization and to examine the relationship between network features and the receipt of opioid agonist treatments (OATs) among pregnant women with OUD, we constructed a patient-provider/facility bipartite network study. Exponential random graph models and generalized linear models were implemented. Our results suggested that OATs that patients receive and the type of healthcare services provided by providers/facilities were associated with healthcare resource utilization among pregnant women with OUD. Patient connectedness (i.e., degree) to providers/facilities and centrality of providers/facilities were linked to access to OATs among pregnant women. Integration of multidisciplinary care and better access to buprenorphine are encouraged to improve healthcare essential to pregnant women with OUD.

Available for download on Friday, September 05, 2025