The Division of Biostatistics of the Department of Preventive Medicine, UTHSC, in coordination with the Center for Biomedical Informatics, invites you to attend TODAY’s seminar.
Time: Monday, December 02, 2:00 pm-3:00 pm CT
ZOOM Virtual Room Connection: Register in advance for this meeting to get the Zoom Link
Seminar Website: https://www.eventcreate.com/e/biostatisticsseminar
Application of AI Methods to Detect Social and Behavioral Factors Associated with Frequent Emergency Department Utilization
Ramin Homayouni, Ph.D.
Professor, Foundational Medical Studies
Director, Population Health Informatics, OUWB
Director, Population Health Research, Corewell Health East
The Emergency Department (ED) offers a unique opportunity to evaluate not only the health of their communities, but also social and economic disparities. Social and Behavioral Determinants of Health (SBDH) are critical contributors to health outcomes, often leading to both emergent and non-emergent conditions that drive ED visits.
While health systems have increasingly adopted screening tools to identify SBDH, these tools often fall short in capturing the full scope of SBDH needs across entire patient populations. The ED, despite being a critical point of care where documentation of SBDH could most effectively inform population health interventions, faces significant challenges in implementing these tools due to its high patient volume, intensity of services, and staff shortages. However, robust SBDH documentation in the ED could enable strategies to address barriers such as limited access to primary care and socioeconomic factors that exacerbate chronic health conditions. To address these challenges, developing automated methods to extract and aggregate SBDH data from diverse sources—both structured fields and unstructured clinical notes within the Electronic Medical Record (EMR)—can provide a more comprehensive understanding of community SBDH needs.
In collaboration with Corewell Health East (CHE), our team has developed an approach to aggregate SBDH data from multiple EMR sources, including structured screening responses, ICD-10 Z-codes, and unstructured clinical notes. Our current efforts are focused on utilizing this approach to identify SBDH needs within our patient populations and to support outreach initiatives and wraparound services aimed at addressing these needs effectively.