An innovative payment program that financially rewarded clinical teams for their overall performance helped boost collaboration and reduced the tendency of clinicians to “dump” complicated patients—but it was also the source of “substantial frustration."
Researchers looked at a compensation model that Minneapolis-based Fairview Health Services introduced in July 2010. The pay-for-performance structure linked 40% of primary-care providers' annual pay to clinic-wide quality performance. The findings were published this month in the Annals of Family Medicine.
The researchers caution healthcare systems not to be too hasty as they design and adopt such models in the growing demand to move away from fee-for-service payment.
Aligning payment to quality is “a great concept,” said lead study author Jessica Greene, associate dean for research in the George Washington University School of Nursing. But it's a radical departure from previous payment systems, she said. “And it is much more complicated than the designers of these programs had imagined.”
Fairview aimed to encourage teams of physicians, nurse practitioners and physician assistants to work together to manage population health, strengthen accountability and reduce the likelihood of “patient dumping,” or shifting complex patients to others.
The system used quality performance metrics providers are required to report under Minnesota state law—such as care for diabetes, cardiovascular and asthma patients, and certain cancer screenings. Each member of the primary-care team at the clinic could receive percentage increases in salary depending on how well the clinic performed overall on the state benchmarks.
To gauge physician perceptions, researchers conducted in-depth interviews with 48 clinicians six months after the program started and again at the one-year mark. Another 150 primary-care clinicians completed a survey sent in the summer of 2013.
In both instances, clinicians noted greater collaboration and boosts in quality performance for the team. But they overwhelmingly noted feeling less control over compensation and expressed “significant frustration” with poor performers driving down potential pay increases.
Only 15% of the surveyed respondents felt basing incentives entirely on team performance was the right thing to do. But they didn't want the incentives based solely on individual performance either; only 7% supported that option.
“The team component made an impact on people,” Greene said. “It was too much team, and they wanted a mix.” Nearly three-quarters liked the idea of a combined approach. In October 2014, the health system switched to a “hybrid” model and added individual performance back to the mix.
Green was joined on the study by Ellen Kurtzman, also of George Washington University; Judith Hibbard, a senior health policy researcher at the University of Oregon; and Valerie Overton, vice president of quality and innovation at the Fairview Medical Group.
For now, Greene said, the hybrid model sounds like "a really nice compromise." But, she added, it will take more research to understand the design of these types of incentive programs and track how well they work.
Greene also led a separate analysis of Fairview's experience. That study, published last month in the policy journal Health Affairs, concluded Fairview physicians performed no better on quality metrics than comparable medical groups throughout the state, even though 40% of their compensation was riding on their performance.
“It will be very interesting to see ultimately what the experience is,” she said. “Unfortunately, they haven't been as effective to date as most people had hoped.”