Physician payment is undergoing its largest reform since the introduction of Medicare's resource-based relative value scale. It is well recognized that fee-for-service, capitated, and salaried payment systems fall short in aligning financial incentives with clinical excellence. The Institute of Medicine1 has recommended that physician payment be altered to reward better quality and outcomes. Paying physicians for meeting quality and outcome targets has been called pay for performance (PFP). Pay for performance can be defined as "[t]he use of incentives to encourage and reinforce the delivery of evidence-based practices and health care system transformation that promote better outcomes as efficiently as possible."2(p5)
On a small scale, healthcare organizations have already begun reimbursing physicians for meeting quality standards.3 It is estimated that 1% to 2% of physician compensation among participants in PFP programs is from incentive pay for quality.4 However, the number of physicians and the amount of money that will be involved in some form of PFP in the near future are likely to increase substantially,5 although the effect on physician income remains unclear.6 A recent evaluation of PFP found no effect of PFP bonus payments on improving mammography rates or glycosylated hemoglobin testing and only a modest effect on improving cervical cancer screening.7
In the absence of a substantive empirical evidence base on the quality and cost effects of PFP, the specific contours of these programs will be guided by opinion. The practicing physician's perspective has been largely absent from these deliberations, although physicians are the intended target of PFP incentives. This commentary provides designers of PFP programs with the physician's perspective on how PFP programs should be developed.2 Our analysis is based on a consensus conference that convened 250 physicians and medical managers to discuss how physicians believe that PFP arrangements should be developed to align healthcare toward affordability, evidence-based medicine, and public accountability for how resources are used. We structured this commentary to address the following 6 core components of PFP programs that emerged from the consensus conference discussions: (1) payment structure, (2) transparency, (3) metrics, (4) evaluation, (5) community and patient participation, and (6) fairness.
Financial incentives can be paid to individual physicians or to physician organizations that directly interface with payers. Compared with individual physician incentives, payments given to physician organizations are more likely to alter the infrastructure of the practice milieu in ways that promote better care for multiple aspects of the care delivery process (eg, improved information technology). Another benefit of using the organization as the accountable entity rather than the individual physician is that the organization is more likely to have patient samples large enough to produce statistically meaningful results. Payment to physician organizations is also more likely to promote a shared sense of accountability for a patient population, whereas individual physician payment could promote less interphysician cooperation, resulting in more fragmented healthcare. For these reasons, it seems logical to disburse PFP payments to physician organizations rather than to individual physicians.
Pay for performance payments may be most effective if they are provided as incentives to organizations that meet certain target thresholds and to others that demonstrate a clear improvement over baseline performance levels. The threshold approach is necessary to reward excellence. However, if target thresholds are the sole criterion for payment, groups making the least improvement in quality (ie, the high performers) will garner the greatest share of the financial payments.7 Organizations showing improvement over baseline performance should also be rewarded for substantive improvements in quality. This approach offers incentives to all physician organizations, because low performers can be rewarded for improvements and high performer excellence is reinforced.
In the early phases of PFP, negative financial incentives should be avoided; PFP programs should be based on positive financial incentives. Developing adequate levels of provider buy-in to the process will be critical to program success. Negative incentives would likely discourage skeptical providers from participating.
Pay for performance programs do not need to be solely based on financial rewards. Public disclosure of results is a nonfinancial incentive that can alter a physician's reputation or an organization's prestige among peers or patients and may constitute a stronger incentive than bonus payments.3 Disclosure of physician participation in a PFP program sends the public the positive message that physicians are willing to be accountable for their performance.
The key questions for physicians about disclosure in PFP programs are how and when this information will be made available to the public. In the early phases of PFP, the list of physician organizations participating in a program and the quality and outcome metrics used should be publicly disclosed. The argument for disclosure is motivated in part by a general consumerist trend to make more information available to the public. Patients' right to know this information is counterbalanced by physicians' desire to keep information about their professional practice private and confidential.
As the validity of the PFP assessment is demonstrated, PFP programs should consider publicly disclosing lists of physician organizations that meet quality and outcome target thresholds and those that are demonstrating improvement over time. This disclosure should occur after a baseline period during which physicians are provided with opportunities to review, validate, and interpret their results. A process that permits physicians to validate their results and to express their agreement or disagreement with the findings should be established in all programs. Once the validity of the quality and outcome assessments has been substantiated, disclosure of results can proceed.
Physicians recognize that disclosure of the results of PFP may have negative effects. For example, if PFP uses only a limited number of measures, consumers choosing a practice will have incomplete information about the global quality of care delivered by that practice. In such cases, physician practices may be penalized or be rewarded inappropriately. Some physicians are concerned about the misuse of PFP findings by payers and purchasers or by lawyers in malpractice proceedings.
Pay for performance designers should include a sufficient number of metrics across a spectrum of health promotion activities and disease states to provide a balanced view of performance. At the outset, the number of metrics is likely to be limited, but over time the list needs to be reevaluated and expanded to be more comprehensive.
In their current state, PFP programs are experimental. It is unknown if their effects are positive, negative, or neutral. Therefore, we believe that every PFP program should have some level of evaluation. These evaluations should include periodic assessments of intended and unintended program effects on access, costs, quality, health outcomes, physician satisfaction, and patient satisfaction. A national database of PFP evaluation results could be established so that organizations implementing PFP programs can share and learn from the experiences of other organizations.
Although US private sector reforms tend to occur on an ad hoc basis, PFP programs will be most successful if employers, public purchasers, payers, and providers serving the same medical market coordinate their efforts to develop a common set of measurement procedures. If payers within the same market develop unique methods, PFP runs the risk of failing. Community-wide participation facilitates statistically valid evaluation of smaller physician practices by capturing a large share of their patient populations in quality assessments. Moreover, fewer resources among physician organizations are required for measurement if a common approach is used. Communities should consider developing common data sets that aggregate information across payers, purchasers, and providers to have a uniform method for assessing and reporting performance. A consortium of plans and purchasers in a community that cooperates on the design and implementation of PFP can affect a large enough share of physician income to produce real change. Metric sets that do not overlap across payers (or purchasers) using PFP will increase the level of confusion among providers about aspects of clinical care and practice organization on which they should focus change efforts. A common set of metrics and implementation procedures across payers within a community permits specific community priorities to be targeted.
As central actors in care processes, patients have a critical role in quality improvement, and physicians believe that patients should be involved in PFP program development and assessment. Quality and outcomes of care cannot improve without the active engagement of patients in healthcare processes. The importance of their role in improving quality is often overlooked, but it is sensible to include patient preferences in the design of PFP systems.
Pay for performance programs that are perceived by physicians as unfair will fail. It is incumbent on PFP designers to include methods to maximize fairness by addressing differences in patient health status, adherence with prescribed regimens, and social complexity. Patients do not randomly distribute themselves across providers. To promote fairness, quality assessments and payments based on those assessments should be adjusted for differences in patient mix across providers (ie, risk adjusted). Achieving quality and outcome targets will be more difficult for some providers than for others whose patients are healthier. Once differences in patient mix are accounted for, the quality rankings of some organizations can change substantially,8 with some previous "bad apples"looking good.
Positive and negative outcomes may result from PFP programs. Aligning payment to promote quality may have important beneficial effects on outcomes. On the other hand, PFP may change the holistic patient-oriented approach to patient care if healthcare is delivered by managing the metric rather than managing the patient. Quality of care for conditions not included in the incentive system could deteriorate because of opportunity costs within a practice (ie, addressing a small set of health conditions to the detriment of care for other problems). Without adequate risk adjustment, PFP may create unintended disincentives for physicians to practice in areas with patient populations that have high levels of healthcare needs or social complexity. Last, practice administrative costs may rise if additional funds are not provided to generate the PFP metrics. Pay for performance programs must be evaluated to monitor their effects on patient access, practice burden, quality, and outcomes for conditions not targeted by the PFP formula. The ongoing input and feedback of physicians will be critical to determining the future success or failure of PFP.
We gratefully acknowledge input from John E. Anderson, MD, Nashville; Charles H. Booras, MD, Jacksonville, Fla; Richard Chung, MD, Honolulu, Hawaii; Timothy Crimmins, MD, Minneapolis, Minn; Gregory B. Diette, MD, Baltimore; Linda Dunbar, MS, PhD, Glen Burnie, Md; James K. Geraughty, MD, Nashville; Michael J. Goran, MD, Tiburon, Calif; Mark S. Hackman, MD, Nashville; Douglas J. Hiza, MD, Eagan, Minn; Joel C. Hoffman, FCA, ASA, MAAA, Denver, Colo; Sid King, MD, Gallatin, Tenn; Patty M. Orr, RN, EdD, Nashville; Alan Sokolow, MD, New York, NY; and Sidna Tulledge-Scheitel, MD, MPH, Rochester, NY.
From the Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Md (CBF); Health Technology Vector, Farmington, Conn (VGV); and American Healthways, Inc, Nashville, Tenn (JEP).
This study was supported by American Healthways, Inc. The final content of the manuscript was completely at the discretion of the authors. Dr Villagra is a consultant for American Healthways, Inc, and Dr Pope is the chief medical officer.
Address correspondence to: Christopher B. Forrest, MD, PhD, Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, 624 N Broadway, Room 689, Baltimore, MD 21205. E-mail: email@example.com.