Pay for performance: Coming your way soon
Increasingly, physicians will be compensated based on outcomes
By James H. Herndon, MD
Pay for performance (PFP) is the latest in a series of efforts to improve patient outcomes, increase patient safety, reduce costs and reform physician payment systems. It is a major shift from the fee-for-service model and has the potential to revolutionize how doctors are paid.
Pay-for-performance is an umbrella term for the many physician-payment methodologies that provide incentive reimbursement to physicians who meet clinical performance criteria and document that care by tracking services provided, patient education and outcomes.1
The emphasis on PFP has increased rapidly in the past two years, leading to a plethora of PFP or pay-for-quality initiatives by hospitals, insurers and government agencies. Currently, more than 100 different PFP efforts are underway nationwide, involving more than 2 million people. Even the Centers for Medicare and Medicaid Services (CMS) has 10 PFP demonstration sites and is “100 percent committed to promoting pay-for-performance,” according to the CMS Administrator Mark McClellan, MD, PhD.
PFP and medical errors
One of the reasons that payers, industry watchdogs and consumer advocates have flocked to PFP programs is the continuing incidence of medical errors. Until the health care industry (including hospital executives and physicians) embrace patient safety measures, the marketplace will continue to question the wisdom of paying everyone the same amount, regardless of quality of care or outcome.
In addition, recent studies have shown that patients often do not get the best of care, even when evidence-based guidelines are available. A 2003 Rand Health study highlights the widespread underuse of recommended services: Overall, patients received only 55 percent of recommended care. (While the study measured which treatment the doctors prescribed and took into account explicit patient refusals of care, it did not reflect patients who simply ignored their doctor’s advice.)
Whether PFP is the best way to control health care costs, reduce errors and foster quality standards of care is still being debated. Lucien Leape, MD, whose work with the Leapfrog Group on patient safety has been widely hailed, is a strong supporter of such programs. Dr. Leape believes that payers, through PFP, have finally joined the patient safety team and will drive it forward.
On the other hand, Donald M. Berwick, MD, MPH, president of the Institute for Healthcare Improvement, is skeptical of PFP systems. “The problem with PFP is not that it doesn’t mold behavior,” he said. “The problem is that it does mold behavior.” He thinks that PFP could lead doctors to “game the system,” making access to care harder for noncompliant patients and turning the physician’s goal from healing to scoring.2
Early program results
In 2002, one of the first PFP programs was launched by the Integrated Healthcare Association with six California health plans, including Aetna, Blue Cross and Blue Shield of California, Cigna HealthCare of California, Health Net and PacifiCare. The performance measures covered patient satisfaction, chronic care management, preventive care and information technology (IT). Under this program, Blue Cross and Blue Shield of California paid quality-related bonuses of $28 million in 2002, $56.9 million in 2003 and $72 million in 2004.3
The Premier Hospital Quality Incentive Demonstration Project, sponsored by CMS, is a three-year program that was initiated in October 2003. It involves more than 275 hospitals nationwide. Participating hospitals receive bonuses based on their peformance on evidence-based quality measures for inpatients with heart attack, heart failure, pneumonia, coronary artery bypass graft and hip and knee replacements. (See Table 1 for a list of the hip and knee replacement performance indicators.)
Hospitals in the top 20 percent of quality for these clinical areas will receive a financial payment as a reward for the quality of their care (either a 1 percent or a 2 percent bonus of their Medicare payments for the measured condition). CMS anticipates that the cost of the bonuses will be about $7 million a year, or $21 million over the three years of the program. However, CMS is also considering reducing payments to hospitals in the lowest 20 percent of quality after the three years of the program. The requirement that CMS retain budget neutrality means that any incentive payments to top performers will ultimately affect other payments.
Bridges to Excellence (BTE), a PFP program developed in 2003, is a coalition of physicians, health plans, large employers and others. It is now the nation’s largest employer-sponsored effort to reward physicians for delivering high-quality care. Since its inception, BTE has paid out more than $1.5 million in physician awards.
BTE now hopes to involve patients by identifying doctors with proven outcomes or exemplary patient care and support systems and by providing patients with coupon rewards for seeing those doctors. In effect, this would attract patients to better performing physicians and away from those with poorer records of care. The number of programs that publicly report a physician’s results is small, but could increase rapidly if patients start to drive the effort.
Perhaps the most striking attempt to link pay to performace is the effort by HealthPartners of Minnesota. On Jan. 1, 2005, HealthPartners implemented a policy that witholds payment to hospitals for “never events,” which are defined as extremely rare medical errors that the National Quality Forum identifies as things that should never happen to a patient, including wrong-site and wrong-patient surgery.
So what’s the problem?
PFP programs are offering physicians meaningful incentives for achieving standard, recognized and attainable measures. They’re engaging consumers to support physicians who follow evidence-based care guidelines. They’re also reducing costs. According to BTE, doctors recognized as quality providers in their diabetes program delivered services at 15 percent to 20 percent below the cost of nonparticipating doctors.
With all these advantages, why are doctors so concerned? Are there real downsides to such programs or are physicians simply feeling threatened by the transparency PFP provides?
Both the number of programs and the varying guidelines within programs are areas of concern. With dozens of programs—each with its own measures of success—a physician’s office may be hard-pressed to maintain the records needed to verify compliance and qualify for the incentive payments.
Communicating the goals and measures of PFP programs is also problematic, leading to a mixed reaction among physicians. When the BTE program was introduced in Louisville, Ky., some doctors felt it was “shoved down” their throats.4
In fields like orthopaedic surgery, the evidence-based guidelines that govern many PFP initiatives are still being developed. Many physicians are concerned that they might be left out of the process. Although CMS, BTE and other programs do use evidence-based guidelines developed in concert with groups such as the AAOS, physicians are concerned that “efficiency” measures, which result in reductions in cost of care, may be adopted as “quality” measures.
In addition, innovative surgical methods and technological developments make it difficult to keep surgical quality measures sufficiently up-to-date to be perceived as relevant.
There is also the difficulty in measuring outcomes after surgery. A surgical patient’s outcomes may depend as much upon the patient’s comorbidities and compliance with postoperative treatment regimens as they do on the surgeon’s skill. Measures need to be under the control of the physician, and the surgeon should be held harmless for outcomes beyond his or her control when it comes to payment for services.
The link between PFP and IT is another area of concern. Most PFP programs include some IT requirements, ranging from electronic health records to computerized prescription order entry and online lab orders and results. Small physician groups and solo practitioners may find it difficult to make a massive investment in IT. Recognizing this, CMS recently announced that it would make its Vista software available for free.
Finally, physicians are concerned about the effect of PFP on reimbursements in general. If PFP is imposed on a system like Medicare that is required to be “budget-neutral,” would reimbursements be reduced across the board to create the PFP incentive fund? A major issue is retaining part of expected payments to physicians (withholds), with the expectation that they will be reimbursed only if the quality measures are met. This is similar to managed care programs of the past, which many physicians and patients found unacceptable.
Facing the unknown
Whether pay-for-performance programs will raise standards of care for everyone or widen the gap between “winners” and “losers” among health care professionals and patients is still unknown. Either way, PFP programs have the potential to significantly alter the health care payment system.
For the moment, PFP programs are still looking to physicians to set the standards. The American Medical Association has adopted a series of PFP principles and guidelines that specifically call for practicing physicians “with expertise in the area of care in question” to help design, implement and evaluate any program.5 Specialty societies such as the AAOS need to define the attributes that PFP programs should measure and the markers that demonstrate high performance care.
Earlier this year, the Alliance of Specialty Medicine—a coalition of 13 medical societies, including the AAOS—wrote to CMS on the impact that PFP might have on physician participation in Medicare, particularly if PFP is imposed without fixing the current Medicare physician payment formula. The Alliance made the following recommendations:
• Any type of system that rewards providers by improving patient care and outcomes should not be subject to budget neutrality or used as a physician volume control.
• PFP programs must not be punitive.
• Measures must be specialty-specific and developed by the physician community.
• Data collection must be reliable, easy for physicians to record and report, based on a clinicial data set and in a manner acceptable to the physician community, without creating a burden on practicies and in compliance with requirements under the Health Insurance Portability and Accountability Act.
• Incentives should be placed on optimizing quality of care and physician participation, not only on performance of specific quality measurements.
• Implementing PFP should be phased in and pilot-tested on a voluntary basis first.
The time to be proactive is now. Rather than argue the value of PFP, physicians need to accept that such programs are here and growing, and become involved in shaping them so that neither they nor their patients are negatively affected.
James H. Herndon, MD, is chairman emeritus, department of orthopaedic surgery, at Partners Healthcare System Inc., and second past president of the AAOS. He can be reached at firstname.lastname@example.org
1. Champlin, L: Congress likely to link physician reimbursement to pay-for-performance. AAFP News Now, April 4, 2005. Accessed online at http://www.aafp.org/x33684.xml on June 20, 2005.
2. Galvin, R: A deficiency of will and ambition: A conversation with Donald Berwick. Health Affairs, Web exclusive, Jan. 12, 2005. Accessed online at Health Affairs archives (subscription required) on June 20, 2005.
3. Rauber C: “Blue Cross hands out record $72M in quality bonuses to doctors.” San Francisco Business Times, July 25, 2005. Accessed online at http://www.bizjournals.com/industries/health_care/physician_practices/2005/07/25/sanfrancisco_newscolumn2.html, on July 25, 2005.
4. Pallarito, K: “A Business Prescription.” U.S. News and World Report, July 25, 2005, accessed online at: http://www.usnews.com/usnews/biztech/articles/050725/25health.htm on July 24, 2005.
5. American Medical Association: “Board of Trustees Report 5: Pay-for-Performance Principles and Guidelines.” Adopted June 21, 2005.