Another Failed Government Program: Pay For Performance

Pay for performance is a health insurance program wherein financial incentives are made available to physicians and hospitals in return for achieving “better” outcomes. Under this system providers of health care services are rewarded financially for achieving certain goals or performance measures for quality and efficiency. This model represents a fundamental change in how doctors are compensated from fee for service payment. Although some more recent programs have attempted to analyze cost savings, as well as quality, the main focus to date of pay for performance has been improvement in the quality of medical care.¹

Two recent reports from England, where pay for performance has existed for ten years, cast doubt on the assumption that pay for performance improves quality. Both reports appeared in recent issues of the New England Journal of Medicine.^{2, 3}

One large study analyzed the 30-day in-hospital mortality for three medical diagnoses among 1,825,518 hospital admissions to 24 hospitals. Mortality for these conditions was compared to 137 hospitals not in the pay for performance program. The authors of the study concluded that although short-term relative reductions in mortality were achieved, there was no long-term reduction in mortality in the hospitals using pay for performance compared to the control hospitals not using pay for performance.²

The second article attempts to assess the effects of pay for performance quality measures on physicians treating patients in their offices. Most physicians used electronic health records and had to meet specific goals using metrics just as is done in the United States today. After analyzing more than 20 systemic studies over a ten \-year period the authors conclude that the pay for performance program has many problems and will have to be “reined in.”³

They list five lessons to be learned from the United Kingdom experience with pay for performance:

1. Pay for performance is not a “magic bullet” and needs to be combined with other quality improvement initiatives.

2. Financial incentives may increase the risk of unintended consequences such as gaming or “cherry picking” healthier patients and avoiding sicker poorer patients. Financial incentives must be aligned with professional values.

3. Pay for performance administrators need to recognize that large parts of clinical practice cannot currently be measured.

4. There are substantial problems with linking patient experience scores (rating of doctors by patients) directly to physicians’ pay and this unpopular indicator was dropped in 2011.

5. Attaching 25% of income to pay for performance, which increased revenue for individual physicians (by as much as $40,000) resulted in doctors ‘focusing on limited areas of patients’ problems. Reduction of pay for performance’s role in physician income was enacted and was welcomed by physicians.

To summarize, this study shows that pay for performance as it directly affects clinicians in the United Kingdom is at best a work in progress if not an outright failure.

In the U.S. there have also been studies of pay for performance. The largest such study using Medicare data compared outcomes between 252 participating and 3,363 control hospitals between 2003 and 2009. The authors compared 30-day mortality among more than six million patients who had acute myocardial infarction, congestive heart failure, pneumonia, or who underwent coronary bypass grafting. They concluded: “We found no evidence that in the largest hospital-based pay for performance program that led to a decrease in 30-day mortality.”⁴ Other studies have shown similar results.

Studies of pay for performance have been undertaken in low income minority populations. Hospitals that care for these patients performed worse on both quality and cost metrics. There are fears that pay for performance could exacerbate racial and ethnic health disparities if providers avoid patients that are likely to lower their performance scores.

Despite the absence of clinical benefit, policy makers continue to promote pay for performance. Some investigators have challenged the basic concept that quality measures as determined by experts can improve the quality of care. They point out how the metrics used to define quality care can change. One example they cite was the recommendation for tight control of blood sugars in ICUs. This recommendation was later contradicted by new data. They also point out how experts in a particular field can disagree on optimum treatment of patients. Their conclusion is that “Medicine is an imperfect science… Information evolves and changes… Flexibility is appropriate… A good doctor exercises sound clinical judgment by consulting expert guidelines and assessing ongoing research, but then decides what quality care is for the individual patient. And what is best sometimes deviates from the norms.”⁵

In another article titled “Rise of the Medical Expertocracy” the same investigators elaborate on this point by challenging governmental panels of experts who offer “sweeping recommendations on clinical care.” They note that in most of the 20th century a form of paternalism existed where the physician dictated what was to be done and the patient complied. This has been replaced by a new form of paternalism “based on the assumption that Americans are not receiving quality care. A new lucrative industry has grown up to generate ever more medical metrics, to give report cards to doctors and hospitals, and to base payments on compliance with ‘best practices’. Yet beyond safety protocols, there is scant evidence that such measures improve health.”⁶

Despite the lack of evidence that pay for performance improves quality it has become popular among policy makers and private and public payers including Medicaid and Medicare. It is included in the Affordable Care Act. A number of physician organizations have expressed concern about the validity of quality indicators, patient and physician autonomy and privacy, and increased administrative burden. These organizations include the American Academy of Family Physicians, American College of Physicians, American Geriatric Society, American Academy of Neurology, and the Endocrine Society. The American Medical Association has developed guidelines for pay for performance.⁷

All of the major objective independent studies to date in England and the United States have shown that pay for performance does not improve the quality or efficiency of patient care. Preliminary studies also indicate that it does not reduce costs. It is simply an added costly burden that physicians and hospitals have to cope with. The government, including all of its policy makers and experts, must now review the data, accept the facts and cut back on their cherished but failed pay for performance programs.

Biography

Arthur H. Gale, MD, MSMA member since 1976, is a Missouri Medicine Contributing Editor. He practices Internal Medicine in St. Louis.

Contact: agalemd@yahoo.com