More On Pay for Performance

More On Pay for Performance

There’s a lot more to say about the burgeoning pay-for-performance (P4P) movement than I mentioned in my recent column about the burden it places on primary care physicians and how reluctant we are to say anything that may be construed as antiquality.

American employers are determined to seek better quality for the money they put into paying for their employees’ health insurance. The Health Plan Employer Data Information Set (HEDIS) was the first such major effort. It is done with the National Committee for Quality Assurance (NCQA). Selected outpatient charts are reviewed for quality and the use of data derived from their networks of physicians has allowed some insurers to portray themselves as “higher quality” than their competitors.

Very likely, such claims are specious because the same pool of doctors overlaps in the treatment of the same group of patients and it is beyond credibility that all doctors (or even any doctors) would treat patients differently based on who provides their health insurance. (I, personally, cannot imagine Aetna asthma, Blue Cross backache, or Cigna circumcisions.)

In addition, it is expensive to evaluate charts.

The latest quality efforts, that will probably soon involve most physicians, look to computerized administrative data to provide the indices of quality upon which bonuses will be based. For example, diabetics are first identified either by diagnoses on claim forms or by the kind of drugs they are taking. Then the level of control is judged by the glycohemoglobin level. This year the medical group with which I am associated will get a bonus if enough of us have ordered the test at least once for each diabetic patient. Next year, the bonus will depend on the actual value that is reported, set relatively high initially, but sure to be ratcheted down in time.

Obtaining accurate data is just the first of many difficulties with what’s become known at P4P. My 2 freshman University of California San Francisco medical student preceptees readily identified 2 more. When asked what they thought of the intuitively appealing idea of paying a doctor more when his patient does better, the first student recognized a trick question and responded, “For just how many diseases do you get a bonus?” The second student chimed in, “And what happens to patients with other diseases?

That was a good start and I included their comments in my assigned talk on “Potential Pitfalls With Economic Incentives Linked to Diabetes Outcomes,” at the American Diabetes Association’s Annual Scientific Meeting in Orlando, Fla, on June 8, 2004.

Better diabetes care is one of many areas of P4P interest, and it stands to create the greatest value for patients, health plans, and employers by decreasing future complications. At the same time, tracking and implementing better diabetes care is the most expensive activity for physicians because it involves so much more than simply checking something like a Pap smear.

So a real danger of P4P is allowing talk of bonuses (currently in the range of 5%-10% of base compensation) to obscure underfunding in the first place. And with health expenditures running $1.7 trillion in 2003, concerns about money are sure to run in the background in every discussion about paying for quality. Indeed, there are no promises of new money for better quality. Consequently, there is widespread skepticism in medical organizations about P4P becoming a shell game in which the same money will simply be redistributed.

That could actually harm care if the highest performers amongst us (by current criteria, expected to be large clinics) are awarded bonuses that actually come out of the pockets of physicians with less computerization and whose statistics are poorer or not available. These doctors will then be even less able to afford the extra administrative costs, computers, software, and services that P4P requires.

Ease of measurement may count for more than other, less tangible criteria that may be more important. Some critics have gone so far as to say that P4P will mainly reward skill in data collection. And that ethics will take a beating, as noncompliant patients become direct financial liabilities to their doctors. Patients may also lose trust in physicians and physicians will have to wrestle with another blow to self-image created by being subject to Pavlovian behavior control.

There is a total lack of objective evidence that P4P, as a way to promote evidence-based medicine, actually works. In fact, Donald Berwick, one of its leading current proponents and a thought leader in medical quality, less than a decade ago said quite the opposite. Speaking of “superficial logic” and “toxic features to systemic improvement,” Berwick blasted P4P as a “naïve understanding of the complexity of human motivation” (Qual Mang Health Care. 1995 Fall;4:27-33). There is a large literature in the human resources field expressing similar views.

Even when taken at face value, the major players in P4P bring hidden agendas to the table. Employers want quality, but given the chance, would jettison their responsibility for health care altogether. Health plans are still smarting over their loss of reputation in the backlash against managed care. Pushing the idea of “paying for quality” makes for perfect rehabilitative public relations. More importantly, some thought leaders who give a good name to the entire P4P concept may have bigger fish to fry—nothing short of using P4P as a means to reshape medicine. Consider the illustrative comments of Dr Arnold Milstein, Medical Director of the powerful Pacific Business Group on Health in Medical Economics Magazine (Sept 9, 2002):

“Pay for performance is designed to give groups the money to ‘reengineer’ their practices. We’ve decided to reinforce performance excellence, rather then to ensure that every medical group survives. To the customer, the first is a lot more important than the second.”

Milstein is willing to sacrifice medical groups (meaning large groups). But an even bigger impact may be on physicians who are not in such groups. According to the American Medical Association, only 20% of US doctors practice in groups larger than 8 and fully 40% of the primary care physicians to whom P4P is addressed, are in solo practice

I am appalled that this industry spokesman (who formerly was with a firm that brought us hospital length of stay guidelines), assumes a mandate, is willing to exert economic power to actually destroy practices and presumes to speak for all patients everywhere. His apocalyptic language, applied to the evolution of quality in American medicine, brings to mind Marx and Engels overblown claim that “Communism will inevitably emerge out of the ashes of capitalism.” (Communist Manifesto)

Paying for performance is very much a loaded gun that can be dangerous. As currently conceived, it is just as likely to be a passing fad.