Assessing the Relationship between HIE and Patient Outcomes at National Scale

Presentation Type: Oral Presentation - Regular
Presentation Time: 09:45 AM - 09:57 AM

Abstract Keywords: Outcomes Improvement and Equity, Standards, Terminology, and Interoperability, TEFCA, FHIR, Health Policy, Reimbursement and Affordability, and Sustainability
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


There is a need for robust evaluations of HIE impact. We use national claims data to identify individuals that visited two unaffiliated provider organizations and compare outcomes based on whether the two organizations were connected to the same HIE network(s). HIE is associated with slightly lower probability of being admitted to the hospital after an ED visit but not other types of utilization. Efforts are needed to address factors impeding benefit realization from HIE.

Speaker(s):
Julia Adler-Milstein, PhD, FACMI
UCSF School of Medicine

Author(s):
Jordan Everson, PhD - Georgetown University School of Medicine; Ariel Linden, DrPH - UCSF; Julia Adler-Milstein, PhD, FACMI - UCSF School of Medicine;

---

Evaluating the Impact of HIE-based Intervention to Improve Care Transitions

Presentation Type: Oral Presentation - Regular
Presentation Time: 09:57 AM - 10:09 AM

Abstract Keywords: Clinical Decision Support and Care Pathways, Workforce Automation, Communication, and Workflow Efficiency, Outcomes Improvement and Equity, Innovation Partnerships, Implementation Science, and Learning Health Systems, Change Management, Human Factors and Usability, Social Determinants of Health (SDoH)
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


This study evaluates a health information exchange–based enhanced transitional care alert (ETCA) to improve care transitions for high need patients. Using the COMPLEXedex SDH algorithm and a stepped wedge design, we identified medical and social complexity and delivered real time alerts to nurse coordinators. Implementation reduced inpatient utilization compared with matched controls, demonstrating the value of interoperable, personalized cross sector transitional care management.

Speaker(s):
Sharon Hewner, PhD, RN, FAAN
SUNY Buffalo School of Nursing

Author(s):
Sharon Hewner, PhD, RN, FAAN - SUNY Buffalo School of Nursing; Suzanne Sullivan, PhD, MBA, RN - Upstate Medical University, SUNY, Syracuse; Katia Noyes, PhD, MPH - Uiversity at Buffalo, SUNY; Sabrina Casucci, PhD. MBA - University at Buffalo, SUNY, Department of Industrial & Systems Engineering; Varun Chandola, PhD - University at Buffalo, SUNY, Computer Science & Engineering;

---

A Temporal Transformer for Continuous ICU Discharge Readiness Monitoring

Presentation Type: Oral Presentation - Student
Presentation Time: 10:09 AM - 10:21 AM

Abstract Keywords: Clinical Decision Support and Care Pathways, Health Data Science, Outcomes Improvement and Equity
Working Group: Clinical Decision Support Working Group
Primary Track: Big Data for Health


Background: Timely and safe intensive care unit (ICU) discharge decisions are essential to patient safety, as premature discharge can lead to unplanned readmission or post-discharge mortality. Existing models typically provide a one-time prediction at discharge, overlooking dynamic changes in patient trajectories. We developed a temporal transformer for continuous discharge readiness monitoring and compared architectures for point-of-care prediction.
Methods: We conducted a retrospective study of adult ICU admissions (2020–2024) from hospitals in Illinois using electronic health record data. Patients were split using stratified random sampling (70% train, 15% validation, 15% test), yielding 25,744 training stays, 5,591 validation stays, and 5,483 test stays. Unsafe discharge—ICU readmission within 48 hours or post-discharge mortality—occurred in 4.3% of stays. Two paradigms were evaluated: (1) continuous monitoring, where a temporal transformer predicted at each 8-hour interval whether discharge at that moment would be unsafe, and (2) point-of-care prediction, where models used only the final 24 hours to predict actual discharge outcome.
Results: For continuous monitoring, the temporal transformer achieved AUROC 0.872 and AUPRC 0.459, outperforming temporal LSTM (AUROC 0.832, AUPRC 0.450). For point-of-care prediction, performance was similar across architectures: Transformer (AUROC 0.727), LSTM (0.721), XGBoost (0.715), Logistic Regression (0.698), and Random Forest (0.672).
Conclusion: A temporal transformer enables continuous ICU discharge readiness monitoring with strong predictive performance. For point-of-care prediction, deep learning and traditional ML approaches performed comparably. These complementary paradigms address different clinical needs: ongoing surveillance versus discharge decision support.

Speaker(s):
Saki Amagai, Student
Northwestern University

Author(s):
Glenn Fernandes, MS - Northwestern University Feinberg School of Medicine; Yikuan Li, Ph.D. - George Mason University; Catherine Gao, MD - Northwestern; Yuan Luo, PhD - Northwestern University;

---

Structuring Pediatric Total Parenteral Nutrition Workflows for Safer Transitions of Care: A Mixed Methods Study

Presentation Type: Oral Presentation - Student
Presentation Time: 10:21 AM - 10:33 AM

Abstract Keywords: Workforce Automation, Communication, and Workflow Efficiency, Human Factors and Usability, Quality Informatics and Lean, Standards, Terminology, and Interoperability, TEFCA, FHIR, Innovation Partnerships, Implementation Science, and Learning Health Systems
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


Pediatric Total Parenteral Nutrition (TPN) workflows rely on manual, fax-based processes that hinder access to current formulations and delay inpatient care. Using a mixed-methods, human-centered approach, we identified major gaps in communication, documentation, and information retrieval. Clinicians reported frequent delays, inadequate system performance, and heavy verification burden. Findings highlight the need for structured, interoperable EHR-based TPN orders to improve safety, efficiency, and continuity of care for TPN-dependent children.

Speaker(s):
Anushree Murugan, MD
University of North Carolina @ Chapel HIll

Author(s):
Anushree Murugan, MD - University of North Carolina @ Chapel HIll; Lukasz Mazur, PhD - University of North Carolina; Kathy Vu, BS - Carolina Health Informatics Program, University of North Carolina - Chapel Hill; Shuhan Lu, MS - Carolina Health Informatics Program, University of North Carolina - Chapel Hill; Harrini Ramanathan, BDS - Carolina Health Informatics Program, University of North Carolina - Chapel Hill; Sarah Kopitskie, BS - Carolina Health Informatics Program, University of North Carolina - Chapel Hill; David Bottenberg, BS - School of Information and Library Science, University of North Carolina - Chapel Hill;

---

Standardizing Beyfortus Eligibility Across Inpatient and Outpatient Care: A Cross-Disciplinary EHR Workflow Transformation at a Quaternary Children’s Hospital

Presentation Type: Oral Presentation - Regular
Presentation Time: 10:33 AM - 10:45 AM

Abstract Keywords: Clinical Decision Support and Care Pathways, Outcomes Improvement and Equity, Workforce Automation, Communication, and Workflow Efficiency, Quality Informatics and Lean
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


Eligibility for RSV immunoprophylaxis (Beyfortus) depends on nuanced maternal vaccination data that are often missing or unstructured, leading to inconsistent clinical decisions. Our multi-disciplinary team implemented a standardized EHR workflow that captures maternal vaccination status and provides real-time, logic-based eligibility guidance. Baseline administration data identified workflow gaps and informed CDS enhancements. This approach improves the accuracy of eligibility assessments and establishes a scalable model for integrating new immunoprophylaxis recommendations across both inpatient and outpatient care settings.

Speaker(s):
Osvaldo Mercado, MD
Children's Hospital of Philadelphia

Author(s):
Brooke Luo, MD - Children's Hospital of Philadelphia; Stan Krajewski, BBA, MIS - Children's Hospital of Philadelphia; Amina Khan, MS - Children's Hospital of Philadelphia;

---

Impact of an asynchronous communication tool on pre-operative efficiency

Presentation Type: Oral Presentation - Regular
Presentation Time: 10:45 AM - 10:57 AM

Abstract Keywords: Workforce Automation, Communication, and Workflow Efficiency, Quality Informatics and Lean, Human Factors and Usability
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


A common source of inefficiency in the hospital relates to teamwork and coordination, and the operating room is no exception. In this study, an intervention (SONAR)m to improve asynchronous communication about progress towards patient pre-operative readiness is assessed on increasing pre-operative efficiency. Following implementation of the tool, usage of the tool was shown to improve efficiency by 20% compared to when the tool was not utilized by the full care team. There was only a small efficiency gain (2%) in using the tool to prepare for first cases of the day, but larger efficiency gains (20%) for subsequent cases for the day. These results highlight the ability for a tool like SONAR to help aid communication by shifting some synchronous communication - phone calls - that disrupted work to asynchronous channels - the EHR - that are less disruptive for the entire team.

Speaker(s):
Silis Jiang, PhD
Weill Cornell Medicine

Author(s):
Steven Ference, MSN, RN, CNOR - NewYork-Presbyterian Hospital; Sahana Natesan, Student - Touro College of Osteopathic Medicine; Zachary Turnbull, MD, MBA, MS - Weill Cornell Medicine;

---