Pay for Performance: A Decision Guide for Purchasers
Phase 1. Contemplation
Purchasers contemplating the adoption of pay for performance should initially consider the three key questions discussed in this section:
Question 1. Is our community ready?
P4P initiatives are currently underway in a number of different types of communities, involving large and small purchasers, for-profit and not-for-profit providers and all types of market structures. The Centers for Medicare & Medicaid Services (CMS) demonstration project with Premier, Inc. hospitals, for example, includes urban, rural, and even critical access hospitalsi (go to: http://www.cms.hhs.gov/HospitalQualityInits/35_HospitalPremier.asp).
Although there have been no studies of the type of community in which P4P is most likely to succeed, two factors are likely to be important: sponsor influence and the pre-existing capacity of local providers to engage in quality measurement and improvement.
Pay for performance is most likely to be effective when it is introduced by a powerful stakeholder in the market. Purchasers representing a large share of targeted providers’ patients are better positioned to introduce significant changes in behavior than those representing a small share of patients.7,8 For example, in Hawaii, other than Kaiser, the Hawaii Medical Service Association (HMSA) covers almost all patients with commercial insurance. Large market share facilitated HMSA’s introduction of a P4P program in 1998, well before most other purchasers were even considering this option. State Medicaid agencies, which cover roughly half of long-term care spending, would be similarly positioned to implement meaningful incentives for facilities such as nursing homes.9
In an equal and opposite way, highly organized providers with high market share are more able to resist changes in incentives that they do not like.10 The presence of powerful providers does not rule out P4P, however, but implies that the purchaser must consider the provider groups’ input.
Another major factor that can facilitate or inhibit a P4P initiative is the local market capacity for quality measurement and improvement. There may be little value in establishing ambitious performance targets based on process or outcome measures if providers have weak information systems and poor office systems for managing patient care. Purchasers in such communities might initially focus on rewards based on measures that do not require well-developed information system capacity, such as patient ratings of their experience with care or measures of infrastructure (so-called structural quality measures).
For example, the Bridges to Excellence program and the California Integrated Healthcare Association (IHA) both reward physicians for information technology adoption, in addition to other dimensions of quality. The Medicare Payment Advisory Commission (MedPAC) also has taken this position in its recommendation that CMS focus on encouraging information technology adoption as a first step in introducing P4P for physicians in traditional Medicare.11 In Massachusetts, under the Massachusetts Healthcare Quality Partnership, capacity building to support P4P has focused on quality measurement and data aggregation across a set of participating health plans. Similarly, purchasers in communities where quality information is scarce might initially focus on rewards for reporting of measures as was done by CMS in the Hospital Quality Alliance (HQA) program.
i. Critical access hospitals are rural, acute-care hospitals that are eligible for cost-based reimbursement by Medicare based either on State designation as a "necessary provider" or distance from the nearest acute-care facility.
Question 2. Should we partner with other purchasers or go it alone?
Purchasers face tradeoffs in the decision of whether to coordinate with other purchasers or undertake their own P4P program.ii Reasons that a purchaser might find coordination for P4P desirable include:
- The purchaser’s market share is small, such that any unilaterally imposed incentive would likely have little or no effect.
- Providers will be frustrated with multiple, uncoordinated data requests from different sources.12
- If the purchaser acts on its own, other purchasers will benefit from the P4P program without directly participating and there may be a "free-rider" problem because of overlapping networks.7,8
- For these reasons, purchasers may want to coordinate with each other and a broader group of stakeholders to establish agreements about what quality indicators to collect. While there may be substantial benefits from coordination, purchasers will need to consider carefully the antitrust implications of doing so (Box).
Aside from minimizing antitrust issues, some purchasers may be prepared to "go it alone" for the following reasons:
- For a commercial health plan, P4P could be used as a differentiating factor in attracting business from employers and enrollees.
- Purchasers can align their program with their own or broader ongoing data collection and quality improvement efforts. For example, purchasers could design their P4P programs around the data collected by hospitals on cardiac and lung patients for the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), for the Centers for Disease Control and Prevention (CDC) on hospital-acquired infections, and for various specialty societies about procedures such as coronary bypass.13 Similarly, existing public reports of provider performance sponsored by the CMS, JCAHO, the Leapfrog Group, and others could be used as low cost data sources for pay for performance.14
Antitrust Issues: A Consultation With Federal Trade Commission Staff
Joint decisions between or among otherwise competing payers that relate to what/how they will pay providers under a P4P (or any other) program always raise an antitrust issue, though not necessarily an antitrust problem. How these types of agreements among competitors are analyzed under the antitrust laws is discussed at considerable length in the joint Federal Trade Commission/Department of Justice Antitrust Guidelines for Collaborations Among Competitors (go to: http://www.ftc.gov/os/2000/04/ftcdojguidelines.pdf; Plugin Software Help) and in the Commission’s decision in Polygram Holding, Inc., 5 Trade Reg. Rep. (CCH) 15,453 (FTC 2003) (go to:http://www.ftc.gov/os/2003/07/polygramopinion.pdf); Plugin Software Help) (affirmed by the Court of Appeals, Polygram Holding, Inc. v. FTC, 416 F.3d 29 (D.C. Cir. 2005). In addition to consulting these resources for further information, seeking advice from competent antitrust counsel would be advisable if payers are contemplating acting jointly, particularly regarding payment/provider contracting issues.
Development or adoption of uniform P4P quality/performance standards and data reporting requirements—rather than actual payment terms by payers under such programs—would appear to be less problematic and easier to justify as pro-competitive. There are potential justifications for such joint activity that relate to the inability of small, individual payers to do this effectively on their own. Competitors working together to make P4P programs possible, more efficient, or less onerous to providers may be more likely to be considered a legitimate justification for joint behavior under antitrust law because of potentially pro-competitive effects in the marketplace.
Overall, from an antitrust perspective it is far less risky to develop or adopt a set of "best practice" standards for P4P programs and allow individual payers to decide whether or not to use the "gold standard" in their programs, rather than agreeing to do so (particularly regarding terms of paying providers). Similarly, adopting uniform standards for data compilation and reporting, so that all programs would have access to a broader data source to evaluate their respective programs, would appear to have far less potential for raising anti-competitive concerns. It is important to note, however, that even otherwise potentially efficient and arguably pro-competitive agreements among competitors regarding this type of activity still could raise antitrust concerns where the competitors together have market power and thus are able to compel acceptance of their standards by providers in the market.
Note: Views expressed by members of the FTC staff do not necessarily represent those of the Federal Trade Commission or of any individual commissioner.
ii. This means without coordinating with other purchasers in the market. Go to Question 3 for issues in collaborating with providers.
Question 3. When and how should we engage providers in P4P discussions?
Basic options for provider-engagement include information only, advice, and shared decisionmaking. Providers are both a potential ally and a potential source of resistance to P4P. Understandably, providers may have particular concerns about the quality of the data and the validity of measures created using the data.15,16 Historically, providers have been very skeptical about data produced by outside stakeholders such as government agencies or employer coalitions.16-18 Physicians also have expressed concern over their ability to influence many outcomes measures of quality—such as smoking cessation or control of high blood pressure—because of the substantial role played by patient actions and preferences.
Attempting to meet provider concerns in the development of a P4P program could improve the effectiveness of the effort and its long-term chances but may also slow down the initiative if there is substantial resistance. The existence of provider-sponsored quality measurement programs—e.g., the Society of Thoracic Surgeons’ coronary bypass database—suggests that at least some providers value performance measurement for internal quality improvement purposes. They may not as a whole, however, advocate payments in which some are winners and some are losers. Further, they may not trust groups that are not clinical to develop valid metrics and truly focus on quality (rather than cost). 12,19 Involving providers early could help purchasers identify performance indicators or measurement systems that meet providers’ standards for validity and could facilitate cooperative relations needed to maintain provider participation.
In several communities, public reporting and P4P programs have been successfully developed using a multi-stakeholder approach that involved key providers. The California IHA program is a leading example of this approach. Individual health plans also have successfully involved providers in the development of their P4P programs, despite the sometimes contentious environment that surrounds contracting. In Washington State, for example, the Premera Blue Cross health plan worked closely with the major clinics that provide care to its enrollees to develop performance reporting on both cost and quality measures, and then to use the same performance data to support financial rewards. In the development of the program, the participating physicians often supported tougher standards than the plan initially proposed.
"Allowing physicians to be involved at every step of the program’s development has been critical to our ability to set meaningful performance goals and truly engage the clinics in quality improvement."
—Mark Sollek, MD, Medical Director, Premera Blue Cross plan of Washington State