Pay-for-performance programs: Ethical questions and unintended consequences

PFP may be more challenging than fee-for-service medicine

Special Advisor, American College of Physicians Washington, DC

Despite increasing use of payment policies to promote quality and efficiency in medical care, little has been written about the ethical implications of so-called pay-for-performance (PFP) programs. Neither the Institute of Medicine’s (IOM) report “Rewarding Provider Performance: Aligning Incentives in Medicare” nor the report to Congress of the Medicare Payment Advisory Commission on the subject in 2006 addressed ethical concerns.^1,2

Similarly, many believe that the rapid expansion of PFP programs among insurers and managed care companies has occurred with insufficient attention to the unintended consequences of such programs. These consequences range from a budget shortfall in funds necessary to cover the promised quality bonuses in the United Kingdom, to damaged reputations from data errors among private programs in the United States, to alterations in the physician/patient relationship.^3-5

  Just what is “quality”?

When considering programs that intend to promote quality improvement, questions arise about how quality is defined.

The IOM identified 6 quality aims to ensure that care is safe, effective, patient-centered, timely, efficient, and equitable. It recommended that performance be measured along 3 of those dimensions: clinical quality, efficiency, and patient centeredness.

It also recommended that differential payments reward physicians who exceed performance standards or who improve sufficiently above a floor for each of these dimensions. Medicare has chosen to apply the IOM definition of quality to performance measurement and PFP.

  Is rewarding clinical quality ethical?

Medical ethics in this context refers to the ethical constructs of beneficence and nonmaleficence—doing good and not doing harm. Arguably, good clinical care is worth more than substandard care, and several clinical quality measures are sufficiently mature and tested to distinguish among levels of quality. Obviously, improved data collection methods must be used to account for noncompliant patients and to eliminate confounding variables such as excluding previous hysterectomy from rates of women receiving Pap testing. Still, the core issue of whether there is ethical hazard in measuring and rewarding clinical quality must be answered in support of the imperative to use all reasonable means to improve patient care, which is part of our ethical duty.

  Is rewarding patient- centeredness ethical?

Similarly, it is difficult to imagine how providing care that is patient-centered would present ethical challenge. Preserving patient autonomy by treating patients with respect and compassion, answering their questions, considering their comfort, and being considerate of their time and wishes is part of our obligation toward humanism as caregivers. Measuring these qualities can provide important information to make care better. There should be no challenge to the autonomy of physicians or objection to rewarding those who seek to make the experience of care as good as it can be.

  Do payers have ethical and legal responsibilities?

Does publicizing information about clinical quality and patient centeredness create ethical problems? The answer should be “No” in the current era of transparency—with the important caveat that the data must be accurate and of sufficient quantity to meet tests of significance.

If payers collect and publicly release data that identify individual physicians’ performance on clinical quality and patient centeredness, they must accept the responsibility to assure patient privacy, accuracy of the data, and relevance of the measures used. Failure to do so not only opens ethical questions but may also prompt legal challenges from physicians who believe that their reputations have been wrongly damaged.⁴

  Rewarding efficiency: the greatest ethical concern

What is “efficiency”? The IOM simply defined efficiency as “avoiding waste.” The AQA Alliance (formerly the Ambulatory Care Quality Alliance) defined efficiency in more detail, and states: “Efficiency of care is a measure of the relationship of the cost of care associated with a specific level of performance measured with respect to the other 5 IOM aims of quality.”⁶

Because of legitimate concerns about the sustainability of the health care system if costs continue to rise, payers and physicians have an ethical duty to avoid wasteful health care spending. Each also has a fiduciary duty to the patient who, after all, ultimately pays the bill. And, as responsible citizens, both payers and physicians have a responsibility to challenge fraudulent or patently wasteful care.

Payers, with their growing market power, have additional moral and ethical responsibilities if they implement programs that reward efficiency, either through monetary rewards, preferential “tiering” of practitioners, or inducements to patients to select from more “efficient” physicians. These responsibilities include:

• Using fairness and accuracy in measuring and rewarding efficiency

• Revealing openly and clearly the measures used and how they are used

• Providing an open and fair appeals process.

  Challenges to physicians if payers reward efficiency

By identifying “efficient” physicians and rewarding them for using fewer resources, PFP programs introduce considerable potential for ethical lapses that exceed those in the traditional fee-for-service system.

Physicians may be tempted to stint on appropriate services if the higher use of services places them in a higher-use “tier.” Such tiering may impact their reputation and place them at a financial disadvantage in the plans’ payment policies. In addition, a system of rewards or penalties based on resource use may tempt physicians to select patients with a lesser burden of illness—so-called “cherry picking.” Finally, experience with PFP systems in the UK suggests that physicians “perform to the test” and may devote less attention to services that, while appropriate and needed, are not part of a system of measurement and rewards.⁷ For example, anecdotal evidence suggests that patients may have difficulty making advance appointments for comprehensive exams because the system rewards physicians who keep open appointments for those who wish to be seen within a certain number of hours after seeking an appointment.

  The difficulty in attributing services as part of measuring efficiency

One of the most difficult aspects of rewarding efficiency is the imprecision in assigning responsibility for resource use to an individual physician in the care of a patient. This is particularly vexing with elderly patients, who may have multiple illnesses and are cared for by a number of physicians. While grouping software can attribute services to a particular spell of illness and assign services to an individual physician, it is often unable to determine who, among multiple caregivers, is primarily responsible for the decision to use particular medical interventions. This is especially true now, when the care of a patient is often a team effort, and decisions around that care are shared among multiple members of the team and the patients themselves.

In trying to mitigate this problem, some payers attribute the responsibility for resource use to an assigned primary care physician. Unless patient self-referral to various specialists is prohibited, this can be patently unfair to the physician. And, if such referrals do require prior authorization by the primary care practitioner, they add to the administrative burden, which is one of the factors already detracting from the attractiveness of primary care.

  Ethical concerns about PFP vs fee-for-service medicine

The potential for conflict of interest in the practice of fee-for-service medicine arises when physicians benefit financially from providing more services, some of which may be of marginal value. Some critics have referred to this method of reimbursement as “toxic to quality.” Others assert that the relationship between physician and patient is more valued if the patient is paying something for the services they receive, assuming that they have the means to pay for it. Certainly, if the clinical judgment of physicians is influenced by the profit motive so that there is confusion about whether the services provided are intended to benefit the physician rather than the patient, one can infer that beneficence and commitment to service may be compromised by a desire for gain.

Because the AMA’s Principles of Medical Ethics allow physicians to choose whom to serve, except in an emergency, economically driven selection of patients and performing to the test are already potential problems—with or without PFP. The issue is whether this moral hazard is sufficiently enhanced by PFP so that these issues emerge as ethical unintended consequences and what, in these circumstances, should be done to minimize these consequences.

References

1.  committee on redesigning Health insurance performance Measures. payment and performance improvement programs. Rewarding Provider Performance: Aligning Incentives in Medicare. Washington, dc: national academies press; 2007.
2.  Medicare payment advisory commission. Report to the Congress: Medicare Payment Policy, 2005.http://www.medpac.gov/documents/Mar05_entirereport.pdf. Accessed oct 10, 2007.
3. Galvin R. Pay-for-performance: too much of a good thing? a conversation with Martin roland. Health Aff (Millwood). 2006;25:w412–w419.
4. Nakashima E. Doctors rated but can’t get a second opinion. Inaccurate data about physicians’ performance can harm reputations. Washington Post. July 25, 2007:a01.
5. Werner RM, Asch DA. The unintended consequences of publicly reporting quality information. JAMA. 2005;293(10):1239–1244.
6.  AQA. AQA Principles of Efficiency Measures. apr 2006. http://www.aqaalliance.org/performancewg.htm. Accessed oct 10, 2007.
7. Campbell S, Reeves D, Kontopantelis E, et al. Quality of primary care in england with the introduction of pay for performance. N Engl J Med. 2007;357:181–190.