When Harlan M. Krumholz, M.D., S.M., was a young academic cardiologist in the early 1990s, colleagues advised him to pick a niche, like echocardiography or atrial fibrillation. But Krumholz, the Harold H. Hines Jr. Professor of Medicine, was interested in questions that transcended specialty: How can we measure outcomes to improve care? How can we bring hidden clinical insights into the open? “I was drawn to a bigger picture—how things are connected, how we can best learn to help patients,” he says.

Today, Krumholz’s interdisciplinary work puts him at the center of efforts to improve health care in the U.S. He founded the Yale-New Haven Hospital Center for Outcomes Research and Evaluation, which, among other activities, helps develop the Centers for Medicare & Medicaid Services’ hospital performance measures. His Yale Open Data Access Project has persuaded major companies to share proprietary clinical trial data—a huge win for scientific transparency.

Cardiology has forgiven Krumholz for not sub-specializing. He won the American Heart Association’s Distinguished Scientist Award in 2010. In 2008 he was elected to the Institute of Medicine of the National Academies. Forbes has called him “The Most Powerful Doctor You Never Heard Of.”

As a Yale undergraduate, Krumholz evaluated rural health centers, making him think about how medicine involves larger social systems. Few medical researchers held that holistic perspective at the time, but Krumholz recalls one who did.

As a Harvard medical student he attended a talk about collecting data on chest pain patients to develop admission guidelines. The speaker was Lee Goldman, M.D., M.P.H., now dean of the Faculties of Health Sciences and Medicine at Columbia University Medical Center, a pioneer in applying quantitative public-health methods to improve decisions in clinical medicine. “I asked him to be my advisor,” Krumholz says. After completing his residency in internal medicine at the University of California–San Francisco in 1989, Krumholz returned to Harvard for a cardiology fellowship, where Goldman’s support helped launch his research career.

On joining the Yale faculty in 1992, Krumholz found opportunities to pursue his interests outside the academy via a relationship with the nonprofit Connecticut Peer Review Organization, now Qualidigm. At the time, the organization was embarking on a national effort to evaluate the quality of care for Medicare patients. “I’m jumping up and down saying, ‘Count me in!’,” he recalls.

Krumholz, also professor of investigative medicine and public health, has a knack for finding insight in unexpected places. A bout of jet lag after a flight from China made him wonder if sleep disruption and malnourishment might explain why patients often get sick after a hospitalization. Last year he published a paper describing post-hospital syndrome—heightened vulnerability to accidents and illness following hospital discharge.

Today, he says, “medicine needs to learn how other industries manage data.” The goal is—as it has always been for Krumholz—better patient care. If Amazon can predict purchases so accurately that workers can pack shipments before the customer decides, Krumholz asks, why couldn’t doctors sift data buried in thousands of charts to predict a heart attack—and then preempt it?

“That’s what’s fun about what I do,” he says: “You have ideas and you think, ‘Could that work in medicine?’ ”