Pay for Performance: Here to Stay-for the Time Being

Improving health care quality is absolutely the right thing to do for our patients, and different approaches are being used by the various organizations involved in health care. For example, the member boards of the American Board of Medical Specialties are developing specialty-specific maintenance of certification (MOC) programs that are based primarily on a quality improvement approach, with a final exam at the end of the 10-year MOC cycle. In the waning days of the last legislative session, the Congress passed legislation expanding pay-for-performance to Medicare physician payments following the lead of private payers. Pay-for-performance intuitively sounds like a good idea and is consistent with our capitalist system in the United States. Furthermore, it certainly appears that pay-for-performance is here to stay for the foreseeable future, so we might as well get used to this approach. However, there are issues that must be addressed if pay-for-performance will achieve what is hoped.

How Should Quality Be Measured?

Perhaps the foremost challenge is a reliable, valid means of evaluating the quality of a physician’s practice. The thoracic surgeons have developed a system that measures the outcome of coronary artery bypass graft procedures, which is frequently held up as the gold standard for quality measurement programs. Indeed, it is a very good system that many believe has improved the quality of care in that specialty. However, thoracic surgeons perform a limited number of procedures and have very small ambulatory practices, which makes measurement of coronary artery bypass graft procedures a relatively easy and manageable approach for them.

Otolaryngologists, on the other hand, perform a wide variety of surgical procedures, but even these make up a small part of our practice, as most of our activity occurs in the ambulatory setting. Furthermore, many of our operations are performed in ambulatory surgical centers or even in some offices, which makes a thoracic surgery-like system difficult to apply to our specialty. Therefore, measuring quality in otolaryngology will be difficult, but it must be achieved and will likely require not only an in-hospital approach as in other surgical specialties, but also ambulatory measures like our nonsurgical specialty colleagues.

Also important in any system that measures outcomes of patient care is risk adjustment. An otherwise healthy patient who presents to the doctor with a given condition is more likely to have a good response to therapy than one who has a host of other medical problems, including diabetes, congestive heart failure, and other comorbidities that contribute to the patient’s overall well-being. This factor is particularly important for conditions requiring complex surgery, such as major head and neck resections with substantial reconstruction with flaps. Without taking comorbidities into consideration, a surgeon with a head and neck practice involving very sick patients will not look as good as one who performs smaller head and neck resections on otherwise healthy patients. The American College of Surgeons’ National Surgical Quality Improvement Program (NSQIP), which has been used successfully in the Veterans Administration hospital system to improve quality of care, takes comorbidities into consideration and could serve as a model, with modification, for the surgical subspecialties.

According to the Institute of Medicine’s reports and other studies, there are too many mistakes made in the management of patients. However, physicians are not directly responsible for most of these mistakes. Certainly we physicians must participate in a leadership role to improve the system, but the system is not made up just of physicians—it also includes nurses, pharmacists, and all the other myriad health team members. If pay-for-performance is to really work, then it may be best to focus on the systems (whether it be all the individuals involved in the care of a hospitalized patient or a doctor and a nurse in an office setting). However, if this approach is to be used, the financial incentive must be shared with all the parties to incentivize all those involved in the health care team.

Financial Incentives

The financial incentives must be, to use the accountant’s term, material. In other words, pay-for-performance incentives must be large enough to stimulate the desired changes. Many businesses motivate their employees with year-end bonuses or other financial incentives to stimulate the activities the management feels is desirable. Frequently these “bonuses” are in the range of 10% to 20% or more, so that the employees will see a meaningful financial return for their efforts. A token, small-percentage pay-for-performance incentive is not consistent with using money to stimulate a change in behavior.

Another concern that has been expressed is that we will all be required to practice “cookbook medicine.” There are many different approaches to managing a patient’s disease or disorder, but as evidence-based medicine expands (see related article, page 1), it will become clear that some approaches work better than others, which should improve quality if all physicians follow best practices. One advantage of a pay-for-performance system is that it will stimulate physicians to be aware of and follow the evidence-based medicine approach. However, as we all know, not all diseases have “read the textbook,” and variations in management are sometimes required. Any pay-for-performance system should take these unusual, but real, situations into account.

In addition to being valid, whatever system is used to measure quality must be as unobtrusive as possible relative to the physician’s time. If physicians are required to take time away from seeing patients to fill out forms and document this variation or that, or hire additional employees to do these tasks, then pay-for-performance could have the opposite impact than is envisioned.

In my opinion, most otolaryngologists treat most of their patients with good quality medicine most of the time; there are few “bad apples,” but they do exist. However, all of us can improve our practices if we assess what it is we do, determine why we do it, and implement ways of doing it better. Will pay-for-performance work? Will physicians respond to a financial incentive in the way it is intended? Only time will tell. Those who promote the pay-for-performance approach have an obligation to monitor the results of this experiment. It may turn out, for all of the reasons cited above as well as others, that what is intuitive does not work. It certainly has happened before in many areas of our lives. Indeed, it may turn out that the unintended consequences of pay-for-performance may result in totally undesirable effects that are unknown at this time. In the end, we are professionals, and I believe that the biggest stimulus is to do the right thing for our patients is our professionalism, our desire to help people in need, and our inherent need to do the right thing.

©2007 The Triological Society