However, it's still not clear "whether the return on investment and the quality gains outweigh the financial and human effort," according to the Robert Wood Johnson Foundation, which funded the study.

Pay-for-performance programs have been gaining traction lately; the CMS (Centers for Medicare and Medicaid Services) has launched several programs in hospitals, chronic care management, and outpatient physician services.

For example, one program increase payment rates by 1 percent or 2 percent if hospitals perform well on 10 measures, like giving heart attack patients aspirin at arrival and discharge or getting antibiotics to pneumonia patients within four hours of their admission.

The projects represent the "largest experiment in the country" for assessing pay-for-performance, said Suzanne Delbanco, CEO of the Leapfrog Group, a consortium of health care purchasers that provided technical assistance to the programs.

The projects, she said, "provide some of the first tangible evidence that P4P incentives can raise the quality of patient care."

Resource Library:

The study suggests that incentives need to be at least $5,000 per physician or perhaps as high as 10 percent of a physician's income to make a difference.

However, additional staffing and supportive infrastructure can also motivate physicians to hit quality targets.

The study had other recommendations. Successful pay-for-performance programs must involve physicians early in planning stages, make sure their quality measures are clinically relevant, and provide regular feedback.

Making physicians aware of quality improvement demands is also essential. In one case, physicians unaware of the program threw their bonus checks in the trash.

The study did not address whether P4P can work in traditional Medicare settings or a preferred provider network.

Most Americans get their care from PPOs, a loose network of providers that are not directed by any one plan.

However, Blue Cross of California is implementing a P4P project in San Francisco and plans to expand it statewide.

The P4P projects were extremely varied in their goals and how physicians were rewarded.

Two projects found P4P programs reduced costs. The Excellus/Rochester Individual Practice Association found that a $1 million investment in health information technology reduced cost trends by nearly $3 million.

The BTE (Bridges to Excellence) program rewarded physicians for hitting a range of quality measures for diabetic care.

The doctors that most rigorously hit the quality measures delivered care at 15 percent to 20 percent lower guidelines.

A Michigan Blue Cross Blue Shield project decreased rates of life-threatening infections in the intensive care unit by 45 percent.

Massachusetts Health Quality Partners increased rates of breast cancer screening and allowed doctors to compare their performance with their peers.

The California-based Local Initiative Improving Results improved the rates of teen doctor visits and well-baby visits for Medicaid recipients.

The Integrated Healthcare Association put forth a host of preventive measures and said that some health plans "had seen a 40 percent increase in patient visits, with reduced hospitalizations."

Funds for the projects came from the RWJF (Robert Wood Johnson Foundation), the CHCF (California HealthCare Foundation) and the Commonwealth Fund.