Comprehensive Care Planning for People Living with Multiple Chronic Conditions

Presentation Type: Oral Presentation - Regular

Click to View Presentation

2026 Amplify Health Equity Presentation

Presentation Time: 10:15 AM - 10:27 AM

Abstract Keywords: Clinical Decision Support and Care Pathways, Change Management, Outcomes Improvement and Equity, Diagnostics, Human Factors and Usability
Working Group: Clinical Informatics Systems Working Group
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


People with multiple chronic conditions (MCC) face significant challenges navigating fragmented healthcare systems, with critical health data scattered across multiple providers and settings. The Agency for Healthcare Research and Quality (AHRQ) and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) collaborated to address these care coordination barriers through the development of standardized electronic care planning tools. This is the second round of eCare Plan pilot project work, incorporating more data, more vendor systems and a more comprehensive look at care coordination.

Speaker(s):
Laura Marcial, PhD
UNC

Author(s):
Jacqueline Ortiz, MS, MMCi - RTI International; Keegan Barnes - RTI International; Erin Mallonee, MS - Colorado State;

---

Relieving Pressure on the Health System by Reducing Hospital-Acquired Pressure Injuries

Presentation Type: Oral Presentation - Regular

Click to View Presentation

Presentation Time: 10:27 AM - 10:39 AM

Abstract Keywords: Clinical Decision Support and Care Pathways, Workforce Automation, Communication, and Workflow Efficiency, Quality Informatics and Lean, Change Management, Outcomes Improvement and Equity, Analytical Artificial Intelligence: ML, Digital Pathology, Imaging AI, Predictive Analytics, Governance
Primary Track: Driving Change at Scale through Effective Leadership and Governance


At Ochsner Health, we implemented a comprehensive patient-centered strategy to reduce hospital-acquired pressure injuries across a large multi-hospital integrated health system. We implemented clinical decision support tools, workflow standardization, an Ochsner-developed predictive model for emergency department boarders, and virtual nursing support. Our multi-pronged approach combined optimization of electronic health record technology tools, process redesign, and multidisciplinary engagement to achieve sustained reduction in pressure injury events while improving documentation quality.

Speaker(s):
Christopher Girardo, DO
Ochsner Health

Author(s):
Teresa Arrington, MBA LSSMBB PMP - Ochsner Health; Christopher Girardo, DO - Ochsner Health; Sean O'Connor, BS - Ochsner Health; Redessa Besse, ASN - Ochsner Health; Lisa Fort, MD, MPH - Ochsner Health;

---

Improving Infection Prevention Workforce Efficiency Through Instant Order Ring Surveillance Protocols

Presentation Type: Oral Presentation - Regular

Click to View Presentation

Presentation Time: 10:39 AM - 10:51 AM

Abstract Keywords: Workforce Automation, Communication, and Workflow Efficiency, Clinician Well-Being, Diagnostics
Primary Track: Advancing Wellness for Providers and Community with Consideration of Human Factors


A semi-automated ring surveillance protocol for Candida auris and carbapenem-resistant Enterobacterales (CRE) was developed and implemented by pairing a flowsheet with rule-based instant orders. This initiative created a flexible system that successfully reduced the workload of infection preventionists (IPs), improved adherence to institutional protocols, and demonstrated significant labor savings. The process offers scalable improvements for infection control in other healthcare settings and could be easily adapted at other institutions.

Speaker(s):
Alexander Plattner, MD, MBA
Washington University in St. Louis

Author(s):
Alexander Plattner, MD, MBA - Washington University in St. Louis; Nicholas Hampton, PharmD - BJC Healthcare; Kevin O'Bryan, MD - Washington University in St. Louis; Carole Leone, RN, MSN, CIC, FAPIC - BJC HealthCare; Carlee Hoxworth, MPH, CIC - BJC HealthCare; Ashley Lloyd, RN, BSN, CIC - BJC HealthCare; Patrick Reich, MD - Washington University in St. Louis;

---

Blood Loss Assessment and Stabilization Tool to Assess Transfusion Requirements

Presentation Type: Oral Presentation - Regular

Click to View Presentation

Presentation Time: 10:51 AM - 11:03 AM

Abstract Keywords: Analytical Artificial Intelligence: ML, Digital Pathology, Imaging AI, Predictive Analytics, Governance, Diagnostics, Health Data Science, Outcomes Improvement and Equity
Primary Track: Big Data for Health


Background:
Timely blood product transfusion is critical for survival in hemorrhagic trauma. This study evaluated physiologic and injury-based predictors of transfusion and examined model performance as a potential foundation for decision support within trauma system workflows.
Methods:
Casualties with prehospital and emergency department vital signs were identified from the Department of Defense Trauma Registry. Candidate predictors, such as lowest systolic and diastolic pressures, highest heart rate, shock index, and injury pattern, were analyzed using both multivariate logistic regression and Fast & Frugal Trees. Model discrimination, calibration, and operational performance were evaluated on held-out test sets using ROC and precision–recall curvesand confusion matrices.
Results:
Among 3,574 PH and 16,979 ED casualties, 26% and 20%, respectively, received blood. Transfused patients had significantly lower blood pressures, higher heart rates, and greater injury severity (all p < 0.001). In the PH model, shock index > 0.9, thoracic injury, abdominal injury , and extremity injury were the strongest predictors. The optimized PH model achieved ROC-AUC = 0.90, AUPRC = 0.83, sensitivity = 0.86, specificity = 0.81, and balanced accuracy = 0.84. Calibration was strong, and similar results were observed for the ED model (ROC-AUC = 0.89, AUPRC = 0.79). FFT classifiers provided transparent rule-based decisions with comparable accuracy, facilitating rapid clinical interpretation.
Discussion:
Findings demonstrate that routinely available prehospital data can reliably identify patients at high risk for transfusion, suggesting practical application in triage algorithms, transport prioritization, and resuscitation resource allocation.
Conclusion:
This model provides a foundation for automated tools to support early damage control resuscitation in both military and civilian trauma systems.

Speaker(s):
Darshan Thota, MD
United States Navy

Author(s):
Jonathan Stallings, PhD, MS - Joint Trauma System; Jan-Michael Van Gent, DO - Joint Trauma System; Jennifer Gurney, MD - Joint Trauma System;

---

Impact of Adolescent Deactivation Policy on Patient Portal Access

Presentation Type: Oral Presentation - Regular

Click to View Presentation

Presentation Time: 11:03 AM - 11:15 AM

Abstract Keywords: Population Health, Digital Therapeutics, Remote Patient Monitoring (RPM), and Digital Engagement, Health Policy, Reimbursement and Affordability, and Sustainability, Leadership and Strategy
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


This study examines how automatic deactivation of parental proxy accounts at age 13 affects adolescent and proxy patient portal access and engagement. Using data from more than 21,000 adolescents, we found that over one-third did not reactivate their accounts, with substantial variation by chronic conditions and demographics. Findings highlight potential unintended consequences of privacy-focused policies and underscore the need for approaches that balance confidentiality with sustained digital engagement.

Speaker(s):
Bryan Steitz, PhD
Vanderbilt University Medical Center

Author(s):
Shelagh Mulvaney, PhD, FAMIA - Vanderbilt University; Abigail Doyle, MSHI - Vanderbilt University; Robert Turer, MD - UT Southwestern Medical Center; Averi Wilson, MD - University of Texas Southwestern; Thomas Reese, PharmD, PhD - Vanderbilt; Adam Wright, PhD - Vanderbilt University Medical Center; S. Trent Rosenbloom, MD, MPH, FACMI, FAMIA, FAAP - Vanderbilt University Medical Center Department of Biomedical Informatics;

---

Orchestrator Multi-Agent Clinical Decision Support System for Secondary Headache Diagnosis in Primary Care

Presentation Type: Oral Presentation - Student

Click to View Presentation

Presentation Time: 11:15 AM - 11:27 AM

Abstract Keywords: Clinical Decision Support and Care Pathways, Analytical Artificial Intelligence: ML, Digital Pathology, Imaging AI, Predictive Analytics, Governance, Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM
Primary Track: Big Data for Health


Unlike primary headaches, secondary headaches need specialized care and can have devastating consequences if not treated promptly. Clinical guidelines highlight several red flags features such as thunderclap onset, meningismus, papilledema, focal neurologic deficits, signs of temporal arteritis, systemic illness, and the “worst headache of their life” presentation. Despite these guidelines, determining which patients require urgent evaluation remains challenging in primary care settings. Clinicians often work with limited time, incomplete information, and diverse symptom presentations, which can lead to under-recognition and inappropriate care. We present a large language model (LLM)-based multi-agent clinical decision support system built on an orchestrator–specialist architecture, designed to perform explicit and interpretable secondary headache diagnosis from free-text clinical vignettes. The multi-agent system decomposes diagnosis into seven domain-specialized agents, each producing a structured and evidence-grounded rationale, while a central orchestrator performs task decomposition and coordinates agent routing. We evaluated the multi-agent system using 90 expert-validated secondary headache cases and compared its performance with a single-LLM baseline across two prompting strategies: question-based prompting (QPrompt) and clinical practice guideline-based prompting (CPGPrompt). We tested five open-source LLMs (Qwen-30B, GPT-OSS-20B, Qwen-14B, Qwen-8B, and Llama-3.1-8B), and found that the orchestrated multi-agent system with CPGPrompt consistently achieved the highest F1 scores, with larger gains in smaller models. These findings demonstrate that structured multi-agent reasoning improves accuracy beyond prompt engineering alone and offers a transparent, clinically aligned approach for explainable decision support in secondary headache diagnosis.

Speaker(s):
Xizhi Wu, Master of Science
University of Pittsburgh

Author(s):
Xizhi Wu, Master of Science - University of Pittsburgh; Nelly-Estefanie Garduno-Rapp, MD, MSHI - UTSW; Justin Rousseau, MD, MMSc - University of Texas Southwestern Medical Center; Mounika Thakkallapally, Data Scientist II - University of Texas Southwestern Medical Center; Hang Zhang, MS - University of Pittsburgh; Yuelyu Ji, PhD - University of Pittsburgh; Shyam Visweswaran, MD PhD - University of Pittsburgh; Yifan Peng, PhD - Weill Cornell Medicine; Dept of Population Health Sciences; Div of Health Informatics; Yanshan Wang, PhD - University of Pittsburgh;

---