College of American Pathologists
Printable Version

  President’s Desk Column


Think now about pay for performance

April 2005
Mary E. Kass, MD

The news this month is all about the recently released Medicare Payment Advisory Commission annual report. MedPAC, which counsels Congress on Medicare reimbursement policy, has recommended a "pay for performance" approach to physician reimbursement.

Pay for performance has a nice alliterative ring, but that’s largely what we know about it now and, as always, the devil is in the details. The concept is premised on the development of validated, evidence-based clinical performance measures that apply to all specialties. Reaching a consensus on these indicators will be a challenge for the medical community, no doubt, but it beats the alternative. MedPAC is presenting pay for performance as a way to control costs without use of the sustainable growth rate, a formula Congress enacted to limit Medicare spending. If the SGR remains in place, physician fee updates will be cut by 5.2 percent next year and by a similar amount annually through 2012, when inflation-adjusted physician reimbursement will have dropped to 50 percent of current levels.

Pay for performance has been a buzzword on the Hill for a while now, and Congress may be ready to admit that the sustainable growth rate was a misstep that needs to step out. Still, the eventual shape of any Medicare physician fee restructuring—if this Congress chooses to address it at all—is anyone’s surmise.

The MedPAC pay-for-performance brainchild is a concept in its infancy, which is all right with me if the aunts and uncles (physicians and other caregivers, patient advocates and the like) will have a hand in its upbringing. Cost-containment is important, but the final version needs to be patient-centered and structured to focus on continuous improvement in safety and effectiveness.

The clinical laboratory community may be concerned about MedPAC’s assertion that all Medicare claims should include laboratory values, submitted electronically in a standard vocabulary and format. This will be a burden, although I suspect we will find it less objectionable as the benefits emerge. Standardized vocabularies and information technologies in all laboratory reports will greatly enhance the usefulness of laboratory data.

MedPAC envisions transitional benchmarks to be employed while databases on clinical indicators are developed. Use of information technology is one; patient perception of care is another. While the former would be pathology-friendly, I am concerned about the latter. The limited patient contact of our specialty puts us at a disadvantage in this regard. What if Congress decides to go with pay for performance and few of the benchmarks apply to pathology? We need to be thinking about next steps.

The CAP cancer protocols would probably be a good place to start. A pilot study in Canada recently revealed a clear link between use of the CAP checklists (synoptic format) and completeness. More studies of this kind would enable us to establish the use of synoptic reports as a performance indicator. We might also be able to identify quality indicators through the CAP Laboratory Accreditation Program. Perhaps we could add checklist questions about how often pathologists sit down with patients to explain their test results, what pheresis is, for example, or what is to be gained from further testing.

Budget neutrality is the most controversial provision in the pay-for-performance plan. MedPAC believes that Medicare could retain up to two percent of funds for physician fees, then award those funds to providers who met quality benchmarks. The CAP and dozens of other physician groups have urged that MedPAC avoid at all costs rewarding some physicians at the expense of others. Pay for performance must be funded properly.

When the House Committee on Ways and Means subcommittee on health held public hearings on the MedPAC report, Jeffery Rich, MD, who testified for the Society for Thoracic Surgeons task force on pay for performance, argued persuasively against a budget-neutral approach. Dr. Rich proposed that incentives to encourage Medicare providers to collaborate in condition-specific databases and continuous quality improvement would translate to durable cost savings.

The thoracic surgeons have been collecting clinical data on open-heart surgery since 1989, and using those data to identify best practices. Their database covers 2.7 million surgeries at 600 institutions. Between 1990 and 1999, observed and risk-adjusted mortality for cardiac bypass graft surgery in their study population, projected to increase by 35 percent, instead decreased by 30 percent. They have used a carefully structured and validated standardized clinical database, partnered with thoughtful outcomes measurement and proper feedback, to fuel continuous quality improvement. And they have realized dramatic cost savings.

I would like to take that one step further and urge that any clinical database created under Medicare be SNOMED-encoded. Individual pathologists could contribute in dozens of ways, advising their hospital IT departments on SNOMED and working with colleagues to analyze data and prescribe adaptive behaviors. This could well be the silver lining peeking out from behind the perennial Medicare budget crises.

The late president John F. Kennedy liked to say that the time to repair the roof is when the sun is shining. We have a window, now, to develop our thinking on how such a system might work and what performance indicators we would propose for pathology. Whether this year or five years from now, pay for performance is likely to be on the table when physician compensation under Medicare is discussed. Let’s prepare to contribute to the dialogue.

Dr. Kass welcomes communication from CAP members. Send your letters to her at