$1.4M Awarded to Study "Last Mile" of Healthcare Data Interoperability

When patients visit the emergency room, their medical history is often scattered across different doctors’ offices, hospitals, and specialists they’ve seen over the years. While health care systems can now often share this information electronically, there's a critical gap: doctors often don't have time to dig through all that data to find what they really need to provide the best care.

Led by A Jay Holmgren, PhD, MHI, assistant professor and associate chief for research for the UCSF Division of Clinical Informatics, and Nate Apathy, PhD, assistant professor at the University of Maryland School of Public Health, a new project funded by the National Library of Medicine focuses on what researchers call the "last mile" of data sharing.

The study, titled “The Last Mile of Interoperability: Integrating Outside Data into Clinical Workflows to Improve Care,” will use innovative methods to track exactly when and how doctors view outside patient records at two care delivery institutions over the next four years. By combining quantitative methods developed in economics with machine learning and AI tools, the team will identify which clinical situations benefit most from clinicians’ viewing outside records and develop guidelines for when this information is most valuable.

Despite predictions that robust data sharing could save the healthcare system over $80 billion annually, the reality has fallen short. The research team believes this gap exists because external patient data often sits unused or is too difficult for clinicians to access and interpret during patient visits.

Holmgren and Apathy will analyze how doctors’ treatment decisions shift when they have access to comprehensive patient information from other healthcare facilities, such as urgent care visits, specialist appointments, or hospital stays at different institutions. Using advanced data analysis techniques, the team will determine which medical situations are most improved by having this complete picture of a patient's health history and identify when accessing external records leads to better patient outcomes.

"Our goal is to make sure that when a doctor treats a patient, they can quickly identify and surface the most relevant information from that patient's medical history, no matter where they received care before," explains Holmgren.

The findings will be particularly important as health care organizations increasingly adopt AI tools to help manage the growing amount of patient data available to clinicians. The research team's work will provide evidence-based guidance for developing and evaluating these next-generation clinical tools, informed by robust causal evidence rather than historical patterns of use that have not led to improved care.

“When health data travels seamlessly between institutions, there is immense promise to drive down health care costs and reduce health care use and duplicative paperwork, all of which can improve patient health and satisfaction,” Apathy says. “We hope this research and the open-source tools we will create from it will contribute to improved decision-making and health outcomes.”

This award is supported by a National Institutes of Health (NIH) grant: NIH/NLM R01LM014770. 

Read more about the project from the University of Maryland here.

View the RePORTER page project details here.