Establishing a Screening Tool for Early Detection of PTSD following Complicated Childbirth

About This Grant

PROJECT SUMMARY: Maternal mental health disorders are the leading identifiable cause of childbirth-related maternal death in the U.S. and a significant contributor to maternal morbidity and health disparities. Among these disorders, childbirth-related posttraumatic stress disorder (CB-PTSD) remains an underdiagnosed and debilitating condition. CB-PTSD affects approximately 240,000 American women annually who experience traumatic childbirth. It is marked by symptoms that emerge soon after delivery and are triggered by reminders of the traumatic birth, with the child often becoming an unfortunate symptom enhancer. This dynamic can impair mother-infant bonding during critical developmental stages. The urgency of addressing CB-PTSD is heightened by its intersection with life-threatening complications during childbirth and severe maternal morbidity (SMM), which are highest in the U.S. among Western countries and disproportionately affect minoritized women. Hence, the need to understand and mitigate the psychological aftermath of medically complicated childbirth. Despite recommendations for universal maternal mental health screening in U.S. hospitals, no tools currently exist to identify women at risk for CB-PTSD before symptoms fully develop. This proposal seeks to address this critical gap by developing an innovative, multi-modal screening tool for early detection of CB-PTSD. Leveraging the unique opportunity afforded by childbirth trauma, a clearly defined event with early symptom onset, we propose a prospective, longitudinal study to assess women with complicated, potentially traumatic deliveries. Beginning in the first postpartum days, we will collect oral unstructured short childbirth narratives at the bedside depicting the subjective birth experience. We will measure psychophysiological responses during recollection of the childbirth trauma using validated methods to assess objective emotional reactivity. Obstetrical data, a proxy of the magnitude of the traumatic childbirth event, will be obtained from electronic medical records. Participants will be followed at multiple postpartum time points (Week 2, and Months 1, 3, 6) to assess acute and chronic CB-PTSD via validated psychometric tools and diagnostics. In our prior work, the potential for artificial intelligence (AI) models to detect CB-PTSD via childbirth narratives is demonstrated. We will expand this innovative line of research in this application. Using state-of-the-art AI and large language models (LLMs), we will analyze these narratives and develop a machine learning framework that integrates subjective accounts, objective physiological markers, and obstetrical indicators to predict CB-PTSD risk. The anticipated outcomes include a cost-effective, scalable screening tool to identify CB-PTSD risk early, enabling timely interventions and setting a new standard for trauma-focused postpartum mental health care. This project addresses an unmet clinical need, aligning with national priorities to reduce adverse maternal and child outcomes and to mitigate racial disparities in maternal health.