Patel: Solving science’s ‘last-mile problem’
American Heart Association Volunteer President Manesh Patel, MD, FAHA, delivered a bold challenge during Sunday’s Lewis A. Conner Lecture: Solve healthcare’s “last-mile problem” by ensuring proven medical discoveries reach patients in a timely and more equitable manner.
The Duke University School of Medicine professor highlighted a stubborn gap between scientific breakthroughs and real-world patient care. He called it “science’s last-mile problem,” and argued that too many patients continue to miss out on therapies that can prevent heart attacks, strokes and other cardiovascular events.
“Too many times in my career, I have met people for the first time on the cath lab table, while they are having a heart attack … And I see that they have not been treated with therapies we know could’ve helped prevent their disease,” said Patel, who also serves as chief of the division of cardiology and vice president of heart and vascular services for Duke Health in Durham, North Carolina.
By example, Patel pointed to direct oral anticoagulants, noting that it took more than 14 years for these evidence-based therapies to become more widely prescribed than warfarin, despite research demonstrating their benefits years earlier.
Barriers to quality cardiovascular care remain a global issue, he said, due to cost, limited healthcare access and poor adherence to treatment. Patel stressed that improving implementation science may be just as important as developing new treatments.
“In my home state of North Carolina, limb loss dramatically varies across neighboring counties and regions — evidence that a person’s Zip code can have a greater impact than their genetic code,” he said. “Even more concerning and frustrating is the fact we can show similar maps for heart attacks and many other cardiovascular conditions. And it’s not just the United States.”
Perhaps his strongest message was that of artificial intelligence’s impact on future cardiovascular care. He described AI as one of the most promising tools available to improve medical discovery and delivery of care, while emphasizing the importance of rigorous scientific guardrails. Patel noted the ADVOCATE program, an initiative supported by the Advanced Research Projects Agency for Health. It is designed to create FDA-authorized AI systems that can act as trusted digital members of clinical care teams. The envisioned AI “agents,” he said, could help patients access medications, schedule appointments, coordinate care and support personalized health management.
“AI could and should reduce the friction patients face in getting care, and we face in delivering care,” he said, but cautioned, “If we’re not careful — if we don’t get this right in this early stage — these tools can also potentially widen the gaps that we’re seeking to close.”
