P4P: Juggling Performance Rewards & Clinical Judgment

From the Institute of Medicine

Interview with Steven A. Schroeder, MD, Distinguished Professor of Health and Health Care, UCSF

Medicare and other health insurance programs are "sick," and many believe that pay for performance (P4P) is the "cure." P4P is probably going to affect you soon, if it isn't already, so you had better be prepared.

The number of health insurance plans in the United States offering P4P programs has ballooned from 35 in 2003 to well over 100 today. And it will likely be a part of Medicare during the next several years. The Institute of Medicine (IOM) has just completed its third and final report on "Rewarding Provider Performance: Aligning Incentives in Medicare," in response to a request from Congress. The IOM recommendations include:

? Phase in Medicare P4P over the next 3 to 5 years

? Derive initial funding largely from existing Medicare funds

? Reward physicians who improve or excel in providing high-quality, patient-centered, efficient care

? Ask physicians, initially, to participate on a voluntary basis, using financial incentives sufficient to ensure wide participation, setting standards that can be met in small practices

? Promote, recognize, and reward improved coordination of team care throughout the entire illness

? Help physicians implement electronic data collection/reporting systems.

The IOM envisions a P4P Medicare program that would reward physicians who promote 6 quality aims: safety, effectiveness, patient-centeredness, timeliness, efficiency, and equity.

Will P4P Work?

IMWR

"P4P programs reward physicians who achieve prespecified targets, such as managing chronic conditions according to established guidelines," Steven A. Schroeder, Distinguished Professor of Health and Health Care, Division of General Internal Medicine, University of California at San Francisco, and chair of the committee that produced the IOM reports, tells .

JAMA

However, management-by-guideline can sometimes be bad for a patient. A recent study (. 2005; 294:716-724) showed that following management guidelines for a hypothetical 79-year-old woman with multiple comorbidities would expose her to a complex nonpharmacologic regimen and 12 prescription drugs that have the potential for drug/disease interactions.

"Performance standards are no substitute for good clinical judgment," Dr Schroeder cautions. "But it's hard to think of diabetic patients whose glucose control should not be good, even if they have heart failure and other things. It's hard to think of someone who has had a heart attack and is discharged from the hospital not getting a beta-blocker and aspirin."

The IOM's P4P standards do not conflict with one another, he adds. "If you are caring for someone with a very long problem list, and you only have a short amount of time, you may not be able to cover everything during a single visit. But the idea is that over episodes of care, the best possible quality is going to be delivered. That's what you would want, and that's what you would want for your family."

The IOM recommendations include a starter set of 26 ambulatory care and 20 acute care measures (Table), the Health Plan Employer Data and Information Set (HEDIS) measures, and the Minimum Data Set nursing home quality measures.

"These are pretty cut and dried," Dr Schroeder says. "For example, what percentage of people with hypertension are having satisfactory blood pressure control, what is the glycohemoglobin level of people with diabetes, what percentage of eligible women get mammograms and Pap smears?" Noting that there are a lot of medical conditions for which no standards exist, he says, "It's going to require health professionals to get together and say, ?What standards do we want to sign off for?'"

Rather than imposing guidelines on physicians, Dr Schroeder believes that it will be "the practitioners themselves saying, ?These are the things we think are good medicine, and let's see if we can rise to that level.'"

The IOM advocates a combined approach that would reward both significant improvements and excellence, enabling all physicians to achieve at least one of these goals.

The California Example

The largest P4P program was begun in California in 2003 and includes about 35,000 physicians from 211 organizations that encompass >6 million patients.

The 7 participating health plans reward physicians based on their group's performance in:

1. Clinical quality (50%), including 4 preventive measures as well as 6 chronic care measures

2. Patient satisfaction (30%), based on patient ratings of communication with their doctor; overall care, specialty care, and timely access to care; and coordination of care

3. Investment in information technology (IT) (20%).

"Score cards" showing how well each physician group performs are available to all at www.opa.ca.gov.

The most recent results released by the Integrated Healthcare Association (IHA; www.iha.org/071306.htm), a statewide nonprofit group that spearheaded the California P4P initiative, are encouraging. Compared with 2004, participating physicians in 2005 screened 60,000 more women for cervical cancer, tested 12,000 more patients for diabetes, and immunized 30,000 more children.

The IHA's report shows that between the first and second years of the program, 87% of physician groups improved their overall clinical scores by an average of 5.3 percentage points. During these first 2 years, health plans distributed more than $90 million to physician groups who met P4P quality measures.

Nevertheless, Dr Schroeder describes the experience with P4P to date as "a glass half empty, glass half full area." While physicians have said that P4P worked, one physician group felt that the cost of compliance exceeded the rewards they received. "But this group still said, ?this is worth it, because this plays into our sense of being seasoned professionals and trying to do the right thing.'"

Challenges in Primary Care

As Dr Schroeder sees it, P4P presents 2 major challenges to primary care physicians. "One is the patient with multiple chronic conditions who is being managed by multiple people. Is there a primary doctor who is coordinating all of that care, or are they independent specialists who are doing their own thing but not doing any coordination?" Without sufficient coordination, "there are opportunities to drop the ball or have medicines in conflict with each other."

The second challenge is that "much of P4P is predicated on an electronic health record and, particularly for small doctors' practices, those are expensive to install and to maintain." Dr Schroeder concedes that "we haven't figured out how to accelerate the use of IT, particularly in small practices where an awful lot of medical care in this country is practiced." (The IOM reports are available at www.nap.edu.)