Alice G. Gosfield, JD

To create significant leaps in the quality of care by recognizing and rewarding healthcare providers who demonstrate that they have implemented comprehensive solutions in the management of patients, and deliver safe, timely, effective, efficient, equitable, and patient-centered care.

— Bridges to Excellence Mission Statement

The phenomenon of pay for performance (P4P) in provider payment is taking American healthcare by storm. The Bridges to Excellence (BTE) mission statement reflects the common impetus for these programs, to accelerate quality improvement by recognizing excellent provider performance with increased payment. This presentation looked at the evolution of quality initiatives over the last 30 years; offered a snapshot of the P4P world in its most common iterations; considered the relationship to existing financial incentives in healthcare; and evaluated how disconnected P4P is from the traditional provider contract context onto which it is grafted. Taking these factors into account, this presentation assessed the pitfalls and potentials for P4P, speculated on impending issues, and alluded to a different way of proceeding.

The Evolution of a Quality Imperative
To assess the implications of P4P programs, it is important to reflect on the glacial pace of market and regulatory imperatives to improve quality in the delivery of healthcare. Over the last 30 years, we can see five different eras of quality initiatives.

The IOM has noted that there were 100 definitions of quality in play between the mid 1960s and the early 1980s (IOM, 1990). Even with a gradual shift from post-payment claims review to more activist, concurrent utilization management, rising healthcare costs forced employers to reconsider how they were paying for healthcare. As companies like GE, Xerox, and GM were facing competitive challenges from Japanese companies, they had to consider value purchasing in their approach to offering healthcare benefits. To buy for value meant they would offer fewer choices in favor of more valuable plans and providers. During this second phase, which was the beginning of the managed care era, purchasers placed more emphasis on performance measurement as the basis for selection.

That was followed by more order in the market, which eventually developed out of the chaos of the collapse of Clinton health reform. The rise of the National Committee for Quality Assurance (NCQA) and its accreditation programs, along with the advent of the Healthcare Employer Data and Information Set (HEDIS) to be used by employers in evaluating managed care plans, represented a significant change in evaluating performance. At the same time, Congress entered the fray by spawning the Agency for Healthcare Policy and Research clinical practice guidelines program related to physician payment reform. To contain the Medicare physician fee budget, it would be necessary to know what services the Medicare population ought appropriately to receive in order to create the congressionally mandated volume performance standard rate of increase (Gosfield, 1994). The significance of these developments was that reducing unexplained variability in healthcare services and providing evidence-based medicine would be bedrock concepts of healthcare.

In the fourth era of quality, the new values coalesced in both managed care backlash and redefinitions of quality problems. It was during this period that the American public erupted into spontaneous applause while watching the movie "As Good as It Gets" as the heroine lambasted her son's HMO. The president called together his National Quality Commission. The National Quality Roundtable and other significant studies were produced, assessing America's healthcare quality problems as reflecting significant misuse, overuse, and underuse. The IOM's report To Err Is Human (2000) focused attention on quality failures and patient safety, but its report Crossing the Quality Chasm (2001, pp. 5-6) articulated a common policy view of what healthcare in the 21st century should be — safe, timely, effective, efficient, equitable, and patient-centered. These values, known collectively as STEEP, are seen in the BTE mission statement, the mission of the Leapfrog Group, and in most P4P initiatives.

If the natural development of quality policy, provider performance, and payment systems had produced appropriate results, the Rand Corporation would not have found in 2003 that Americans are only getting 55% of what they need (McGlynn et al., 2003), and P4P programs would not have emerged. The point of P4P is to propel healthcare delivery to more science, more safety, and more patient-centeredness through data made known with more transparency. The theory has been that application of purchasing power — paying for better results — will compel healthcare processes and systems to change for the better. It is believed that P4P will produce change faster than incremental change has produced to date.

Typical Forms of P4P
Pay-for-performance programs typically reflect three principal approaches: 1) threshold bonuses; 2) tiering bonuses; and 3) tiering plus sharing a pool generated by cost savings against a benchmark. Bridges to Excellence typifies the first phenomenon. Physicians who achieve certain process and outcomes measures, as determined by NCQA, are paid a threshold bonus (e.g., $100 per diabetic) for every patient with the diagnosis in their practice. The BTE program is concentrating initially on diabetic and cardiac care. There is an additional component for maintenance of infrastructure (e.g., information technology) that makes these initiatives easier.

Tiering bonuses are typified by the Integrated Healthcare Association in California and the Central Florida HealthCare Coalition. Under these programs, the universe of providers who seek to be paid on this basis is aggregated, and their performance is arrayed normatively. Providers in the top tier get additional payments or administrative burden reduction. Second-tier providers get somewhat lower payments, and third-tier providers likely get nothing. Under these tiering mechanisms, whether or not providers will ever receive any money is speculative because they have no idea what the other players will do, nor can they control the behavior of others in the risk pool.

The third type of P4P program is best seen in the Centers for Medicare and Medicaid Services (CMS) demonstration programs, both at the hospital level and in the physician group practice demonstration. In these mechanisms, providers are tiered but their behavior also is evaluated against a benchmark of cost savings. If providers do not achieve cost savings above the comparison pool, no bonuses will be available. Uncertainty about return for the effort exists in these programs, too, because of the tiering phenomenon.

P4P payment to physicians usually entails a per-patient payment, capitation enhancement, or some administrative burden reduction (e.g., no need to remain on formulary, no need for prior authorization). Hospital P4P payment is somewhat different, reflecting stipend awards, a shared bonus pool, or administrative burden reduction. These are not the only types of P4P programs. There are others that include quality score card bonuses, increased fee-for-service with a withhold, and the like (Leapfrog, 2004; Rosenthal, Fernandopulle, Song, & Landon, 2004).

Financial Contextual Considerations
To evaluate the potential impact of P4P, it is important to recognize that these bonuses are add-ons to an existing payment system that has its own incentives. Fee-for-service payment, at its most perverse, encourages overuse since the provider gets paid an additional fee for each service performed. In capitation and diagnosis related group (DRG) payments, in contrast, the perverse incentive is to underuse, since the fewer services performed, the more the economic return. Global capitation includes all of the pitfalls of basic physician capitation and finds a similar analog in percent of premium payments. These mechanisms are usually found with integrated delivery systems where the hospital and physician payments are combined. Case rates, which are far less common, pay for a broader continuum of care in accordance with a defined rate over a period of time for the specific diagnosis or procedure at issue.

The fundamental problem with these financial incentives is that many of them are predicated on actuarial rates. Actuarially determined insurance rates have nothing to do with quality nor evidence-based medicine. Rather, they freeze at a moment in time whatever the quality manifestations were and project them forward.

Contractual Contextual Considerations
In addition to the financial incentives, there are managed care contract provisions to which providers must adhere, which also influence quality. These pertain to credentialing requirements, network configuration, record keeping and plan access to data, communication clauses such as anti-disparagement provisions, and business confidentiality clauses.

Medical management provisions in contracts are where most P4P contradictions arise. Gatekeeping is the primary medical management provision that has fallen into disfavor. It is interesting to note that much of what is being touted in the world of the new phenomenon of "concierge care" reflects what gatekeeping was supposed to do in its most positive sense — establish the gatekeeper as the physician advocate for the patient who would have knowledge of everything happening to that patient and would facilitate his or her access to the rest of the system. Other medical management provisions in these contracts include the obligation to comply both with the plan's quality assurance and improvement programs as well as the plan's efforts to achieve NCQA accreditation and good HEDIS scores.

Providers have contractual obligations to cooperate with the utilization management programs that profile physicians, deny payment in the fee-for-service context, deny admissions in prior authorizations, or suggest discharges in concurrent review. Subscriber grievance programs are related to medical management because when subscribers are unhappy with their doctor-patient or provider-patient relationship, they take their troubles to these appeals mechanisms with which the provider must also cooperate. Formularies were intended to be scientifically based medical management systems by restricting the ability to prescribe drugs that are not the most effective. Few people regard formularies in this frame of reference today. At the same time that P4P programs have been developing, though, there has been a burgeoning focus on disease management and demand management in managed care plans. Some managed care entities have developed their own programs, and others outsource for these systems to produce better outcomes. There is little consideration of the implications of having a disease management program operating in the same context as a P4P program.

P4P Pitfalls
There are eight principal criticisms of the P4P programs that exist today. Most P4P programs are focused exclusively on chronic care, and while they may generate changed behavior for the diagnosis at issue (e.g., diabetes), it is not clear what will happen once many of the providers have moved to this level of performance. Will additional monies be made available for new diagnoses? Will money shift from one diagnosis to another?

This raises the fundamental second question: Where does the money come from? In the Bridges to Excellence program, atypically, there is actually new money in the system coming directly from the employers and not from the plans. Providers are extremely suspicious that much of the plan payments in P4P are illusory enhancements because plans are simply shifting money around without adding new money into the system.

The third and surprising phenomenon in P4P is that most of these programs are not contractually based; in other words, there is no contract establishing an obligation on the part of the payer to actually pay money. There is simply the expectation that providers will respond to information that the program is available.

This raises yet another problem: because P4P programs are add-ons to existing contracts, there is little attention to the intersection with contract obligations that are in place. For example, if a physician who is being paid fee-for-service believes that he needs to do more laboratory studies and have additional office visits to get the results required by BTE for a bonus, what will the managed care organization do when he is profiled against his peers? This is essentially unaddressed.

The fifth critical issue is that there is virtually no analysis by providers or payers whether these increased revenues will, in fact, increase provider margins. For example, it has been estimated that it takes 15 minutes to cull one patient's medical record for the data required by the BTE diabetes program. There are staff costs associated with producing that data. In addition, there are costs associated with checking the plan's data upon which payment will be made. Still further, there may be additional expenses beyond staff time, including supplies and equipment to produce the required results. As mentioned above, there is almost no consideration given to the impact of any related disease management program; in fact, in none of the information available to date has there even been an assessment of whether P4P and disease management programs co-exist, nor what happens if they do.

These P4P initiatives play into the increased emphasis on making more provider performance data available to consumers. The P4P providers who are being paid on a capitated basis, and who perform effectively, may be at risk for adverse selection when higher volumes of patients at high risk flock to their offices, which have demonstrated higher quality. This phenomenon has not been taken into account in the construction of capitation rates. Finally, most of the data in these systems is self-reported. Although some of it is based on claims data, other data, as in BTE, is presented by the physicians themselves. Although there may be some level of sampling (e.g., 5%) of some of the physicians who are participating, the validation process is very limited.

Assessments
The phenomenon of P4P is very new. Most of these programs are being rolled out with limited applications and often in pilot projects. Hardly bold and sweeping, there are few leaps afoot. This is not necessarily bad since there are so many unanswered questions. Because these programs are rolled out regionally and piloted with a limited focus on specific diagnoses — mostly chronic care — P4P can be transitional at best. Because of the add-on nature of the payment to a system already proven incapable of achieving the results that we want in quality, other models will inevitably emerge. The great value in P4P as it exists today is that it recognizes not all providers ought to be paid the same way. As a result, it offers significant opportunities for providers to acknowledge the STEEP values, and to develop more significant strategies that meet their own needs, to be worked out collaboratively with the payers to achieve improved quality and better margins.

A Different Way
Although the details are beyond the scope of this article, the potential shortcomings of P4P systems can be confronted through a different approach, which applies five essential principles that would change healthcare quality (Gosfield & Reinertsen, 2004; Gosfield & Reinertsen, 2003; Gosfield, 2004). We need to:

• standardize as much as possible what is clearly standardizable to the science;
  
  
• simplify the working context and systems that support clinical care delivery, particularly for physicians;
  
  
• make far broader aspects of the work environment clinically relevant, from documentation systems, to information technology design, to health manpower planning;
  
  
• engage patients around the science that will treat them; and
  
  
• fix accountability at the locus of control.

To use clinical practice guidelines as the foundation to apply these principles could lead to a payment system that would engage physicians the way they treat patients, improve outcomes for patients, and offer a payment alternative that would improve physician financial margins and offer the opportunity for greater consistency in clinical care between physicians and hospitals.

P4P is an important development because it recognizes that providers should be paid depending on the processes and outcomes they produce. While it is another incremental change in a long history that is short on dramatic improvement, the mere fact of its emergence offers the opportunity to develop better models.

Author's Notes
For a more comprehensive consideration of pay for performance, see Gosfield, A. G., "Contracting for Provider Quality: Then, Now and P4P," in Gosfield, (Ed.), Health Law Handbook, 2004 edition, West Group, pp. 103-183, which is available at http://www.gosfield.com/PDF/ch3PDF.pdf.

The Leapfrog Group publishes a compendium of P4P programs, which is available at http://ir.leapfroggroup.org/compendiumselect.cfm.

References
Gosfield, A. G. (1994). Clinical practice guidelines and the law: Applications and implications. In Health Law Handbook. A. G. Gosfield (Ed.). Clark Boardman Callaghan, pp. 65-120.

Gosfield, A. G. (2004). Contracting for provider quality: Then, now and P4P. In Health Law Handbook. A. G. Gosfield (Ed.). West Group, 103-183.

Gosfield, A. G. (2004). The doctor-patient relationship as the business case for quality. Journal of Health Law 37, 2, 197-223. Available at www.gosfield.com/PDF/DrPatientRelationship.pdf.

Gosfield, A. G. & Reinertsen, J. L. (2003). Doing well by doing good: Improving the business case for quality. Available at www.uft-a.com.

Gosfield, A. G. & Reinertsen, J. L. (2004, January 22). Paying physicians for high quality care. New England Journal Medicine. http://222.uft-a.com/PDF/NEJM%20Epstein%20letter.pdf.

Institute of Medicine, Committee on Quality of Health Care in America. (2001). Crossing the quality chasm: A new health system for the 21st century. Washington, DC: National Academy Press.

Institute of Medicine, Committee on Quality of Health Care in America. (2000) To err is human: Building a safer health system. L. T. Kohn, J. M. Corrigan, & M. S. Donaldson, (Eds.). Washington, DC: National Academy Press.

Institute of Medicine. (1990). Medicare: A strategy for quality assurance volume II. K. N. Lohr (Ed.). Washington, DC: National Academy Press, 116-139.

McGlynn, E.A., Asch, S. M., Adams, J., Keesey, J., Kick, J., DeCristofaro, A., & Kerr, E. A. (2003, June 26). The quality of health care delivered to adults in the United States. New England Journal Medicine 348, 2635-2645.

Rosenthal, M. B., Fernandopulle, R., Song, H. R., & Landon, B. (2004). Paying for quality: Providers' incentives for quality improvement. Health Affairs 23 (2): 127-141.

Proceedings from the Quality Colloquium at Harvard University:

• Introduction

• Patient Safety: Ethical Considerations in Policy Development

• How Online Patient Networks Can Enhance Quality and Reduce Errors