Not long ago, people who advocated rewards for physician performance were voices crying in the wilderness. Now many are singing the praises of pay-for-performance (P4P) programs, and the momentum shift is noticeable.
Essentially, P4P uses financial bonuses to encourage physicians to meet standardized quality measures and boost clinical outcomes and patient satisfaction. Its potential to spur an increased use of patient registries, e-prescribing, electronic medical records and health information exchanges has widespread implications for health IT executives.
The number of P4P programs nationwide grew 25 percent from 2004 to 2005, according to Med-Vantage Inc. Through surveys, the San Francisco consulting firm counted 112 programs in 2005, including 81 sponsored by commercial health plans. The Centers for Medicare and Medicaid Services (CMS) has six programs, including a three-year P4P pilot program for physicians who treat Medicare patients.
While encouraged by CMS' involvement, employer groups pushing for change must continue to innovate, stresses Francois de Brantes, program leader for healthcare initiatives at General Electric Co., Fairfield, Conn.
As an officer and director of the employer-sponsored Bridges to Excellence (BTE) P4P program, de Brantes approached health plans a few years ago, he recalls: "No one wanted to do it, so we decided to do it ourselves. Now we've demonstrated the effectiveness, and we're starting to hand it off to the health plans."
BTE, now in its third year, has developed P4P programs involving diabetes, cardiac care and physician IT use and is planning new ones around back care and cancer. Several insurers have licensed the BTE model for use in their markets, including UnitedHealthcare, Minneapolis; CIGNA HealthCare, Bloomfield, Conn; and CareFirst Blue Cross Blue Shield, Owings Mills, Md.
The dozens of P4P projects under way have become a great laboratory for experimentation with performance measurements and payment plans. As medical studies and return-on-investment analyses begin to appear, P4P sponsors are reaching consensus on the key challenges: encouraging physician involvement, developing more meaningful incentives and creating a robust IT infrastructure. Meanwhile, vendors are working on more sophisticated products to help with data reporting and integration issues.
P4P is moving beyond primary care into specialists' offices. Currently, 95 percent of incentive programs target primary care physicians, according to Med-Vantage's surveys. Other, smaller P4P programs also offer incentives to specialists and/or hospitals. Also still in the formative stages are performance report cards that consumers are expected to use to comparison-shop for physicians and hospitals.
Engaging physicians
Although insurers and employer groups are moving ahead enthusiastically, progress is being hampered by doctors who remain skeptical about P4P, either because they disapprove of the concept or don't trust the data.
"Some see the value of financial incentives, while others find it insulting that they're being offered more money to do their job," says Anne-Marie Audet, M.D., quality improvement vice president at the Commonwealth Fund, NYC, a foundation that supports research on health and social issues.
Other issues include which measures should be used and how data should be gathered. Most P4P programs pull data from insurance claims. Yet the BTE program asks doctors to gather data from their charts, which de Brantes claims is key to physician acceptance. "When they do the self-analysis, they have their 'Aha moment' and realize perhaps they weren't doing as well as they thought in certain areas," he says.
Doctors must approve of the underlying data and methodologies if they are going to accept being rated, notes Thomas Mahoney, M.D., CEO of the Rochester Independent Practice Association (RIPA) in Rochester, N.Y., which has approximately 900 primary care physicians.
RIPA has partnered with Excellus Blue Cross Blue Shield, Rochester, N.Y., to provide doctors with reports comparing their performance and efficiency to that of their peers. "Folks don't like getting report cards. My docs don't like it any more than my kids do," Mahoney says. "But the challenge is to make it clinically sound and nonjudgmental. If we can say to them, 'Here's the best that's been achieved by your peers,' then we can have a conversation that induces change."
In 2005, the Excellus/RIPA program became one of the first projects to identify a return on investment. After spending $1 million on technology for physician profiling and patient registries, the project has since attributed almost $3 million in diabetes care cost savings to the investment.
The jury is still out on whether P4P will work well outside managed care settings, said Suzanne Delbanco, CEO of the Leapfrog Group, a Washington, D.C.-based group administering an $8.8 million initiative to help develop and measure P4P efforts. "It's much easier to engage physicians when they're in a tighter relationship, such as an HMO," she says. "In the PPO model, doctors are engaged with many health plans. They hear a lot of noise from all over the place. That makes it hard to get their attention."
Defining incentives
P4P advocates are still experimenting with payment levels. Some suggest incentives need to be 10 percent of a physician's annual income, but many are currently paying only 1 to 2 percent.
BTE's De Brantes says it's crucial to make sure the reward amount is clear. "If you have no idea how much you'll make, you won't expend the effort," he said. "We tell doctors how much they'll make per patient if they meet the performance goals and how many patients with a condition such as diabetes we have attributed to them, so they can see the total nut."
Another unresolved issue is payment discrepancies between absolute performances and relative improvement levels. In October, the Journal of the American Medical Association published a study that examined a P4P program at PacifiCare of California, Cypress, Calif. One physician group improved on cervical cancer screening from 23 percent to 34 percent and only got $26,000 in bonuses, while another group that improved from 53.6 percent to 56 percent got $400,000.
"This study showed that rewards went to groups that were already doing well, while the groups that improved the most got little money," Audet notes. The next generation of P4P is expected to feature hybrid plans that reward doctors for both hitting fixed targets and showing improvement toward those goals.
The IT factor
Some studies are finding a correlation between IT adoption and improved clinical outcomes.
Integrated Health Association, Walnut Creek, Calif., is conducting the largest pay-for performance experiment in the country: its P4P program involves a consortium of seven California health plans and 225 medical groups. IHA found that groups that received full credit on IT measures had clinical scores that were about 9 percentage points higher than medical groups that showed no evidence of IT adoption. The correlation drove IHA to increase the weight that IT measures are given in the P4P program.
"It's quite clear that IT is a significant driver of a group's ability to do well on the clinical measures," says Tom Williams, IHA's executive director.
IHA plans to add an efficiency component that measures resource use as well as quality. It will also expand its measures from processes to outcomes.
Whether or not future research finds evidence of clear financial return on investment from P4P efforts, its proponents are convinced using financial incentives is increasing the use of evidence-based medicine. "Collecting data that's meaningful will yield tremendous change in outcomes," Williams says. "And P4P is the wind behind the sails."
David Raths is a freelance writer in Portland, Maine.
