Enhancing Screening for Perinatal Mood and Anxiety Disorders in Critical and Emergency Care: A Multidimensional Approach Leveraging Machine Learning and System Dynamics

Perinatal Mood and Anxiety Disorders (PMAD) encompass a spectrum of conditions, including postpartum depression, anxiety, stress, perinatal psychosis, obsessive-compulsive disorder, and post-traumatic stress disorder. Among these, postpartum depression is one of the most common obstetric complications and a leading risk factor for maternal suicide. Caregivers of infants admitted to neonatal intensive care unit (NICU) are particularly vulnerable, with up to 45% screening positive for depressive symptoms. Early detection of PMAD is essential, as untreated symptoms can contribute to infant neurodevelopmental delays, recurrent emergency department (ED) visits, and long-term maternal–infant complications. To address persistent gaps in screening and treatment, the PMAD team at Children's National Hospital was established to strengthen psychological support for caregivers in both NICU and pediatric emergency department (PED) settings. Despite these initiatives, screening rates remain suboptimal, highlighting the need for a system-level understanding of the barriers and opportunities within the current process. This dissertation applies a multidimensional approach to enhance the PMAD screening system through three interconnected studies. The first develops and evaluates machine learning algorithms to identify caregivers at high risk for depression, enabling prioritized screening. The second employs a system dynamics group model–building approach to identify the drivers and barriers influencing PMAD screening in NICU and PED workflows. And finally, the third study develops a system dynamics simulation model to test potential interventions and assess how system-level adjustments can improve screening efficiency and caregiver support. Collectively, these studies advance the understanding of PMAD screening as a complex system and provide actionable insights for improving early detection and intervention. The findings have the potential to guide the design of more efficient, data-driven screening systems and to position the PMAD program at Children's National Hospital as a model for promoting maternal and infant well-being in similar healthcare settings.

Name thesis:degree_name

Doctor of Philosophy

Level thesis:degree_level

doctoral

Discipline thesis:degree_discipline

Industrial and Systems Engineering

Department dc:contributor.department

Industrial and Systems Engineering

Grantor dc:publisher

Virginia Tech

Year dc:date.issued

2026

Author dc:creator

• Sadjadpour, Fatemeh

Chair dc:contributor.committeechair

• Hosseinichimeh, Niyousha

Committee members dc:contributor.committeemember

• Ghaffarzadegan, Navid
• Abedi, Vida
• Lourentzou, Ismini

• Perinatal Mood and Anxiety Disorders
• Neonatal Intensive Care Unit
• Pediatric Emergency Department
• Screening System
• Machine Learning
• System Dynamics

Statement dc:rights

• In Copyright

Licence dc:rights.uri

• http://rightsstatements.org/vocab/InC/1.0/

Language dc:language.iso

en

Dc Identifier Other

vt_gsexam:45051

OAI identifier oai:identifier

oai:vtechworks.lib.vt.edu:10919/140928