Inside Medicare’s Stalled Shift to Value-Based Care
Former CMMI Leader Liz Fowler Cites Rigid Federal Scoring Rules and Bureaucratic Impatience for Pilot Failures
2026 LDI-Penn Medicine Research Laboratory Project Grants Announced
Print:
The LDI-Penn Medicine Research Laboratory is a strategic granting partnership designed to advance LDI’s mission of developing generalizable health care knowledge for the nation and to generate actionable knowledge which guides Penn Medicine’s operational decisions towards ensuring that patients receive the best care. Its 2026 grant award proposals seek to improve the effectiveness, efficiency, safety, and experience of care delivery at Penn Medicine, with a special focus on how generative artificial intelligence (AI) can optimize health care delivery. Below are the three winners.
Hospital AI Evaluation Toolkit: A Framework for Rigorous, Rapid-Cycle Assessment of Generative AI Tools in Clinical Practice
Principal Investigator: Eric Bressman, MD, MSHP | Perelman School of Medicine
Generative artificial intelligence (AI) tools, including large language models (LLMs), are increasingly integrated into the electronic health record (EHR) to automate documentation and summarize clinical data. These tools promise to improve efficiency and reduce clinician burden, yet real-world evidence on their effectiveness, safety, and impact on care delivery remains limited.
Without systematic evaluation, health systems risk adopting tools that are ineffective or unsafe. This project aims to (1) develop the Hospital AI Evaluation Toolkit – a reusable framework of methods, templates, and analytic resources for rapid, rigorous evaluation of EHR-integrated AI tools in clinical practice; and (2) conduct the inaugural randomized trial using this framework to evaluate an Epic-based AI tool that drafts hospital-course summaries for discharge documentation at Chester County Hospital.
Large Language Models to Screen for Cognitive Impairment
Principal Investigator: Ari Friedman, MD, PhD | Perelman School of Medicine
Cognitive impairment (CI), including mild cognitive impairment (MCI) and Alzheimer’s disease and related dementias (ADRD), remains underdiagnosed in emergency department (ED) and primary care settings, particularly among minoritized and low-income populations. Early detection can transform care by prompting communication adjustments, safer discharge planning, and timely referrals for definitive diagnosis and treatment. Despite national guidelines calling for universal screening, routine CI assessment is rarely implemented because of time and staffing constraints.
The best-performing brief screen, the Short Blessed Test (SBT), requires five minutes to administer, creating a critical need for efficient, scalable alternatives. This project aims to refine and validate a large language model–based system capable of administering the SBT for use with patients in clinical environments.
Conversational AI Integration with EHR Phenotypes to Predict and Prevent Early Pregnancy Complications
Principal Investigator: Alice Abernathy, MD, MSHP | Perelman School of Medicine
Access, diagnosis, and timely treatment of early pregnancy complications remain constrained by a lack of integrated data to refine diagnostic tools. This gap drives avoidable emergency department visits and costs, missed opportunities for critical interventions to reduce adverse perinatal outcomes, and widens maternal health disparities—particularly for those facing systemic gaps in early pregnancy care.
This project will integrate conversational data collected from a novel conversational artificial intelligence platform (CIRCA) with electronic health record (EHR) data to elucidate care trajectories and facilitate identification of patients at high risk for pregnancy complications such as miscarriage and ectopic pregnancy. Integrating these data streams will improve clinical prediction and surveillance and enable longitudinal analyses of symptoms, care quality, and health equity.
This work is also supported by the Dean’s Innovation Fund, established by members of the Council for Discovery Science to support emerging discoveries with promise to improve human health.
More LDI News
How Hospitals Can Ease Staff Burnout—and Improve Patient Safety
A Major European–U.S. Hospital Study Finds That Changing How Hospitals Are Organized Reduces Burnout and Turnover While Improving Care Quality
ED Psychiatric Boarding of Children Worsens Nationwide
Penn LDI Senior Fellow Dominic Sisti Cites “Alarming Levels”
Methadone Use Rises—But Too Few People Get Opioid Medication
Chart of the Day: Methadone Use for Opioid Use Disorder Tripled From 2010–2020, Yet Only One in Four People With Addiction Receive Medication
Can AI Hear When Patients Are Ready for Palliative Care?
Researchers Use AI to Analyze Patient Phone Calls for Vocal Cues Predicting Palliative Care Acceptance
Medicare’s Hidden Fix for High Drug Costs
Medicare’s Payment Plan Can Ease Seniors’ Crushing Drug Costs but Medicare Buries it in the Fine Print