Times are displayed in (UTC-04:00) Eastern Time (US & Canada) Change
11/17/2025 |
9:45 AM – 11:00 AM |
Room 4
S25: Smart Trials: Informatics at the Intersection of Ethics, Equity, and AI
Presentation Type: Oral Presentations
Recruitment, Retention, and Participant Diversity in the Context of a Heart Failure Self-Care Digital Intervention Decentralized Clinical Trial
2025 Annual Symposium On Demand
Presentation Time: 09:45 AM - 10:00 AM
Abstract Keywords: Diversity, Equity, Inclusion, and Accessibility (DEIA), Mobile Health, Chronic Care Management, Patient Engagement and Preferences
Primary Track: Applications
Programmatic Theme: Consumer Health Informatics
Background: Low participant accrual is a persistent concern in physiological disease intervention trials, inflating costs and jeopardizing the timeliness and validity of findings. Investigators are increasingly adopting decentralized methods to facilitate participation. Objective: To add to the recruitment evidence base by describing the performance of direct and remote recruitment strategies in a decentralized randomized controlled trial of a digital intervention to improve heart failure self-care behaviors. Methods: We conducted a descriptive analysis of referral, enrollment, and retention rates; cost; and sociodemographic diversity of participants across six recruitment streams. Data were collected from March 1, 2022 to January 31, 2025. Results: Decentralized recruitment channels generated 97.2% of enrollments and achieved varying success with respect to sample representativeness. Enrollment rates progressed in accordance with proposed timelines. Retention at 6 months was 88.8%. Conclusions: Decentralized recruitment strategies are feasible, cost effective, and conducive to achieving enrollment targets.
Speaker:
Emily West, MA
The University of Texas at Austin School of Nursing
Authors:
Angelica Rangel, MS - The University of Texas at Austin School of Nursing; Jasmine Zeng, MS - The University of Texas at Austin School of Nursing; Cheongin "Rachel" Im, MSN, RN - The University of Texas at Austin School of Nursing; Fatimah Shah, underway - University of Texas at Austin College of Liberals Arts; Kavita Radhakrishnan, PhD - University of Texas - Austin;
2025 Annual Symposium On Demand
Presentation Time: 09:45 AM - 10:00 AM
Abstract Keywords: Diversity, Equity, Inclusion, and Accessibility (DEIA), Mobile Health, Chronic Care Management, Patient Engagement and Preferences
Primary Track: Applications
Programmatic Theme: Consumer Health Informatics
Background: Low participant accrual is a persistent concern in physiological disease intervention trials, inflating costs and jeopardizing the timeliness and validity of findings. Investigators are increasingly adopting decentralized methods to facilitate participation. Objective: To add to the recruitment evidence base by describing the performance of direct and remote recruitment strategies in a decentralized randomized controlled trial of a digital intervention to improve heart failure self-care behaviors. Methods: We conducted a descriptive analysis of referral, enrollment, and retention rates; cost; and sociodemographic diversity of participants across six recruitment streams. Data were collected from March 1, 2022 to January 31, 2025. Results: Decentralized recruitment channels generated 97.2% of enrollments and achieved varying success with respect to sample representativeness. Enrollment rates progressed in accordance with proposed timelines. Retention at 6 months was 88.8%. Conclusions: Decentralized recruitment strategies are feasible, cost effective, and conducive to achieving enrollment targets.
Speaker:
Emily West, MA
The University of Texas at Austin School of Nursing
Authors:
Angelica Rangel, MS - The University of Texas at Austin School of Nursing; Jasmine Zeng, MS - The University of Texas at Austin School of Nursing; Cheongin "Rachel" Im, MSN, RN - The University of Texas at Austin School of Nursing; Fatimah Shah, underway - University of Texas at Austin College of Liberals Arts; Kavita Radhakrishnan, PhD - University of Texas - Austin;
Enhancing Long-Term Care Efficiency: Embedded LLMs for Clinical Report Summarization and Caregiver Support
2025 Annual Symposium On Demand
Presentation Time: 10:00 AM - 10:15 AM
Abstract Keywords: Clinical Decision Support, Large Language Models (LLMs), Natural Language Processing, Privacy and Security
Primary Track: Applications
Programmatic Theme: Clinical Informatics
Long-term care facilities face a critical shortage of nursing staff and an increasing administrative burden, reducing time for direct patient care. Generative artificial intelligence offers a promising solution to automate administrative tasks and support caregivers.
This paper evaluates the relevance of using a fine-tuned large language model (LLM) to address these challenges. Interviews with healthcare professionals identified key needs, leading to the selection of two use cases: caregiver-patient communication assistance and medical record summarization. To comply with privacy and security constraints, the model was deployed in an embedded scenario. Performance evaluations showed significant improvements in BLEU and ROUGE metrics for both use cases, demonstrating enhanced accuracy.
This study demonstrates the feasibility of leveraging LLMs to streamline workflows, reduce administrative strain, and improve operational efficiency. This work highlights the potential for broader AI applications in long-term care, paving the way for better working conditions for caregivers and improved patient care quality.
Speaker:
Arnaud Michelet, Bachelor of Science (BSc) in Business Information Technology
IT SLD Solutions SA
Authors:
Arnaud Michelet, Bachelor of Science (BSc) in Business Information Technology - IT SLD Solutions SA; Gaetano Manzo, PhD - NLM; Abraham Ritz, Bachelor of Science HES-SO (BSc) in Business Information Technology - IT SLD Solutions; Leo Anthony Celi, MD; Pamela Delgado, PhD - University of Applied Sciences and Arts Western Switzerland HES-SO, Switzerland; Michael I. Schumacher, PhD - University of Applied Sciences and Arts Western Switzerland HES-SO, Switzerland;
2025 Annual Symposium On Demand
Presentation Time: 10:00 AM - 10:15 AM
Abstract Keywords: Clinical Decision Support, Large Language Models (LLMs), Natural Language Processing, Privacy and Security
Primary Track: Applications
Programmatic Theme: Clinical Informatics
Long-term care facilities face a critical shortage of nursing staff and an increasing administrative burden, reducing time for direct patient care. Generative artificial intelligence offers a promising solution to automate administrative tasks and support caregivers.
This paper evaluates the relevance of using a fine-tuned large language model (LLM) to address these challenges. Interviews with healthcare professionals identified key needs, leading to the selection of two use cases: caregiver-patient communication assistance and medical record summarization. To comply with privacy and security constraints, the model was deployed in an embedded scenario. Performance evaluations showed significant improvements in BLEU and ROUGE metrics for both use cases, demonstrating enhanced accuracy.
This study demonstrates the feasibility of leveraging LLMs to streamline workflows, reduce administrative strain, and improve operational efficiency. This work highlights the potential for broader AI applications in long-term care, paving the way for better working conditions for caregivers and improved patient care quality.
Speaker:
Arnaud Michelet, Bachelor of Science (BSc) in Business Information Technology
IT SLD Solutions SA
Authors:
Arnaud Michelet, Bachelor of Science (BSc) in Business Information Technology - IT SLD Solutions SA; Gaetano Manzo, PhD - NLM; Abraham Ritz, Bachelor of Science HES-SO (BSc) in Business Information Technology - IT SLD Solutions; Leo Anthony Celi, MD; Pamela Delgado, PhD - University of Applied Sciences and Arts Western Switzerland HES-SO, Switzerland; Michael I. Schumacher, PhD - University of Applied Sciences and Arts Western Switzerland HES-SO, Switzerland;
Towards Safe AI Clinicians: A Comprehensive Study on Large Language Model Jailbreaking in Healthcare
2025 Annual Symposium On Demand
Presentation Time: 10:15 AM - 10:30 AM
Abstract Keywords: Large Language Models (LLMs), Natural Language Processing, Evaluation
Primary Track: Applications
Programmatic Theme: Clinical Informatics
Large language models (LLMs) are increasingly utilized in healthcare applications. However, their deployment in clinical practice raises significant safety concerns, including the potential spread of harmful information. This study systematically assesses the vulnerabilities of seven LLMs to three advanced black-box jailbreaking techniques within medical contexts. To quantify the effectiveness of these techniques, we propose an automated and domain-adapted agentic evaluation pipeline. Experiment results indicate that leading commercial and open-source LLMs are highly vulnerable to medical jailbreaking attacks. To bolster model safety and reliability, we further investigate the effectiveness of Continual Fine-Tuning (CFT) in defending against medical adversarial attacks. Our findings underscore the necessity for evolving attack methods evaluation, domain-specific safety alignment, and LLM safety-utility balancing. This research offers actionable insights for advancing the safety and reliability of AI clinicians, contributing to ethical and effective AI deployment in healthcare.
Speaker:
Hang Zhang, MS
University of Pittsburgh
Authors:
Hang Zhang, MS - University of Pittsburgh; Qian Lou, PhD - University of Central Florida; Yanshan Wang, PhD - University of Pittsburgh;
2025 Annual Symposium On Demand
Presentation Time: 10:15 AM - 10:30 AM
Abstract Keywords: Large Language Models (LLMs), Natural Language Processing, Evaluation
Primary Track: Applications
Programmatic Theme: Clinical Informatics
Large language models (LLMs) are increasingly utilized in healthcare applications. However, their deployment in clinical practice raises significant safety concerns, including the potential spread of harmful information. This study systematically assesses the vulnerabilities of seven LLMs to three advanced black-box jailbreaking techniques within medical contexts. To quantify the effectiveness of these techniques, we propose an automated and domain-adapted agentic evaluation pipeline. Experiment results indicate that leading commercial and open-source LLMs are highly vulnerable to medical jailbreaking attacks. To bolster model safety and reliability, we further investigate the effectiveness of Continual Fine-Tuning (CFT) in defending against medical adversarial attacks. Our findings underscore the necessity for evolving attack methods evaluation, domain-specific safety alignment, and LLM safety-utility balancing. This research offers actionable insights for advancing the safety and reliability of AI clinicians, contributing to ethical and effective AI deployment in healthcare.
Speaker:
Hang Zhang, MS
University of Pittsburgh
Authors:
Hang Zhang, MS - University of Pittsburgh; Qian Lou, PhD - University of Central Florida; Yanshan Wang, PhD - University of Pittsburgh;
Bridging Minds: A Novel Collaborative Approach to Enhancing Multi-Site Applied Clinical Informatics Research Trials
2025 Annual Symposium On Demand
Presentation Time: 10:30 AM - 10:45 AM
Abstract Keywords: Clinical Decision Support, Administrative Systems, Informatics Implementation, Population Health, Governance
Primary Track: Applications
Programmatic Theme: Clinical Informatics
Multi-site trials offer key advantages but can introduce complexity. We developed a structured governance model for a cluster-randomized, stepped-wedge trial of an EHR-agnostic clinical decision support tool in the emergency department across three large health systems. Our approach included a multidisciplinary leadership and study team structure, structured meeting cadences, and digital collaboration tools. This model streamlined decision-making, improved coordination, and ensured data standardization, offering a scalable framework for enhancing cross-site research efficiency and implementation fidelity.
Speaker:
Lynn Xu, MPH
NYU Grossman School of Medicine
Authors:
Amelia Shunk, MMCi - NYU Grossman School of Medicine; Yuhan Cui, MS - NYU Langone; Angela Mastrianni, PhD - NYU Langone Health; Nicholas Genes, MD, PhD - NYU Grossman School of Medicine; Gregory Simon, MD - NYU Langone Health; Charles Cleland, PhD - NYU Grossman School of Medicine; Katherine Dauber-Decker, PhD - Northwell Health; Jeffrey Solomon - Northwell Health; Nidhi Garg, MD - Northwell Health; Usman Mir, MBBS, MPH - Baylor College of Medicine; Sundas Khan, MD - Baylor College of Medicine & Michael E. DeBakey VA Medical Center; Ynhi Thomas, MD, MPH, MSc - Baylor College of Medicine; Michael Diefenbach, PhD - Northwell Health; Thomas McGinn, MD MPH - Dignity Health / CommonSpirit Health Corporate; Safiya Richardson, MD, MPH - New York University;
2025 Annual Symposium On Demand
Presentation Time: 10:30 AM - 10:45 AM
Abstract Keywords: Clinical Decision Support, Administrative Systems, Informatics Implementation, Population Health, Governance
Primary Track: Applications
Programmatic Theme: Clinical Informatics
Multi-site trials offer key advantages but can introduce complexity. We developed a structured governance model for a cluster-randomized, stepped-wedge trial of an EHR-agnostic clinical decision support tool in the emergency department across three large health systems. Our approach included a multidisciplinary leadership and study team structure, structured meeting cadences, and digital collaboration tools. This model streamlined decision-making, improved coordination, and ensured data standardization, offering a scalable framework for enhancing cross-site research efficiency and implementation fidelity.
Speaker:
Lynn Xu, MPH
NYU Grossman School of Medicine
Authors:
Amelia Shunk, MMCi - NYU Grossman School of Medicine; Yuhan Cui, MS - NYU Langone; Angela Mastrianni, PhD - NYU Langone Health; Nicholas Genes, MD, PhD - NYU Grossman School of Medicine; Gregory Simon, MD - NYU Langone Health; Charles Cleland, PhD - NYU Grossman School of Medicine; Katherine Dauber-Decker, PhD - Northwell Health; Jeffrey Solomon - Northwell Health; Nidhi Garg, MD - Northwell Health; Usman Mir, MBBS, MPH - Baylor College of Medicine; Sundas Khan, MD - Baylor College of Medicine & Michael E. DeBakey VA Medical Center; Ynhi Thomas, MD, MPH, MSc - Baylor College of Medicine; Michael Diefenbach, PhD - Northwell Health; Thomas McGinn, MD MPH - Dignity Health / CommonSpirit Health Corporate; Safiya Richardson, MD, MPH - New York University;
Exploring Heterogeneous Treatment Effects of First-Line Antihypertensive Medications via Target Trials
2025 Annual Symposium On Demand
Presentation Time: 10:45 AM - 11:00 AM
Abstract Keywords: Causal Inference, Precision Medicine, Real-World Evidence Generation, Clinical Decision Support
Primary Track: Applications
Programmatic Theme: Clinical Research Informatics
Personalized treatment for hypertension can potentially improve blood pressure control rates. To overcome limitations of traditional comparative effectiveness trials, we used causal machine learning to identify clinical factors that contribute to heterogeneous treatment effects of antihypertensive medications classes in terms of blood pressure reduction. Additionally, we used trial emulation to quantify heterogeneous treatment effects in different patient populations based on those factors. We found significant variations in certain patient groups for diuretics.
Speaker:
Jingzhi Yu, BA
Northwestern University Feinberg School of Medicine
Authors:
Abel Kho, MD, FACMI - Northwestern University; Norrina Allen, PhD - Northwestern University Feinberg School of Medicine; Luke Rasmussen, MS, FAMIA - Northwestern University; Lucia Petito, PhD - Northwestern University Feinberg School of Medicine;
2025 Annual Symposium On Demand
Presentation Time: 10:45 AM - 11:00 AM
Abstract Keywords: Causal Inference, Precision Medicine, Real-World Evidence Generation, Clinical Decision Support
Primary Track: Applications
Programmatic Theme: Clinical Research Informatics
Personalized treatment for hypertension can potentially improve blood pressure control rates. To overcome limitations of traditional comparative effectiveness trials, we used causal machine learning to identify clinical factors that contribute to heterogeneous treatment effects of antihypertensive medications classes in terms of blood pressure reduction. Additionally, we used trial emulation to quantify heterogeneous treatment effects in different patient populations based on those factors. We found significant variations in certain patient groups for diuretics.
Speaker:
Jingzhi Yu, BA
Northwestern University Feinberg School of Medicine
Authors:
Abel Kho, MD, FACMI - Northwestern University; Norrina Allen, PhD - Northwestern University Feinberg School of Medicine; Luke Rasmussen, MS, FAMIA - Northwestern University; Lucia Petito, PhD - Northwestern University Feinberg School of Medicine;