What if the greatest risk in modern medicine isn't AI—but the fact that nobody is tracking whether our treatments actually work? Dr. Monica Bertagnolli, former Director of the National Institutes of Health and the National Cancer Institute, and newly elected member of the National Academy of Medicine, joins Dr. Rohan Ramakrishna to make the case for a true learning health system—one where every patient interaction generates knowledge that makes the next patient's care better.

Dr. Bertagnolli reframes AI not as a threat to physician judgment, but as a diagnostic tool that finally makes personalized, evidence-based medicine possible at scale. She draws a compelling analogy between health data and credit cards: we accept the risk because the utility is real. And she issues a quiet challenge to the entire field—millions of Americans are being treated according to guidelines today, and almost no one is systematically following their outcomes to see if those guidelines actually worked.

This is a conversation about trust, infrastructure, and what it means to truly learn from the patients we serve.

New episodes are released every other week, wherever you get your podcasts.

For more from Roon, visit: ⁠www.roon.com/⁠

Sign up for our substack: ⁠https://rohanramakrishna.substack.com/⁠

Find us on Instagram and X: @roondoctors

If you have a question, comment, or suggestion for a future guest, please email us: [email protected].

(00:00) Intro & Guest Background

(00:42) Who Is Dr. Monica Bertagnolli?

(02:53) Defining the Learning Health System

(04:43) How AI Makes Learning Health Systems Possible

(06:15) Why We Need Data from Millions of Patients

(07:26) Data Safety, Trust, and the Sacred Patient Relationship

(09:10) How Clinicians Can Rebuild Trust One Patient at a Time

(10:35) The Courage It Takes to Trust an Institution

(12:08) Infrastructure & EHR Networks: Where We Stand Today

(13:44) Is the Technical Capability Already There?

(14:52) AI Applications in Cancer Care Dr. Bertagnolli Is Most Excited About

(16:17) Breast Cancer Guidelines and the Learning Health System in Action

(18:03) Tracking Outcomes to Make AI and Guidelines Better

(19:15) Risks and Unintended Consequences of AI in Medicine

(20:25) The Cardinal Rule: Always Put the Patient First

(21:29) Data Stewardship: Who Really Owns Health Data?

(22:51) The Credit Card Analogy for Health Data Risk

(24:52) Will AI Weaken Physician Creativity and Individualized Care?

(26:17) The Scary Truth: Nobody Is Watching the Outcomes Today

(27:26) Lightning Round

(27:45) Most Impactful Mentor: Dr. Eugene Braunwald

(29:20) Most Memorable Experience with a Politician

(30:33) Book Recommendation: Ron Chernow's Mark Twain

(31:01) Who Should Dr. Ramakrishna Interview Next?

(31:35) Closing Remarks