More doctors must join nurses, administrators in leading efforts to improve patient safety, outcomes, experts argue

"Physicians' training and perspectives on patient care make their contributions to improvement efforts essential," says Peter J. Pronovost, M.D., Ph.D., a Johns Hopkins patient safety expert and co-author of a commentary published in the Feb. 2 Journal of the American Medical Association. "But the work of improving quality currently rests primarily with hospital administrators and nurses, with physicians taking a peripheral volunteer role, often questioning the wisdom of these efforts."

The major obstacle to recruiting physician leaders to the safety movement, he says, is the failure of medical centers to professionally and financially compensate and reward physicians for spending time on quality-improvement projects. "Such projects take away from the time physicians spend treating patients -- and generating revenue," he says. "What's needed is a system that would support a portion of a doctor's time spent managing and standardizing quality of care on a particular unit, in a role similar to what now happens with nurse managers."

Pronovost, a professor of anesthesiology and critical care medicine at the Johns Hopkins University School of Medicine, and Jill A. Marsteller, an assistant professor of health policy and management at the Johns Hopkins Bloomberg School of Public Health, note that studies show little evidence that hospital quality-improvement programs have improved patient outcomes, despite buy-in from top administrators and a push for accountability. "Hospitals will only begin to see progress if they get physicians to not just participate more but to assume leadership roles in quality improvement," Pronovost says.

Pronovost and Marsteller argue that the root of the problem is the "antiquated" relationship between hospitals and the physicians who practice inside them. "The system we have is that most physicians act almost as independent contractors, autonomously caring for individual patients and essentially renting beds, nursing care and technology from the hospital," he says. "Physicians often sit on the sidelines, as nurses and executives push for changes they hope will improve patient outcomes."

Doctors need to focus more on not just the individual patients in their care, but also on the hospital's systems for caring for the patient population as a whole, they say.

Pronovost and Marsteller's prescription is for a formalized "physician management infrastructure" that encourages doctors to design scientifically rigorous quality-improvement interventions, develop performance measures, monitor performance, implement interventions and monitor their impact.

Such an infrastructure would include physician leadership training and development, an emphasis on reliable and valid performance measures, and creation of a chain of accountability from the hospital to physician leaders.

Pronovost and Marsteller acknowledge that their prescription will cost money that hospitals rarely have. "Hospital margins are thin if they exist at all, and current health reform efforts are focused on paying hospitals less -- not rewarding them for improving patient care," Pronovost says. "This makes it difficult for hospitals to financially support physician quality-improvement leaders, and in many cases hospitals are cutting positions, not adding them."

In the long term, however, Pronovost insists, investment in physician-led quality-improvement efforts will reduce costs and at least pay for themselves. The alternative, he says, is not good for patients or hospitals. "Without monetary support for physician leaders, the quality of care for populations of patients is unlikely to improve," he says.