Pay for performance medicine

Pay for performance medicine… Roger Lowenstein, who wrote a justly-praised analysis of Social Security reform for the New York Times Magazine a while back, has another penetrating piece of policy journalism in the same publication today. This time Lowenstein’s subject is health care–specifically, the increasingly influential ideas of David Cutler, a Harvard health care economist. Cutler believes that previous health care reform efforts, such as Hillary Clinton’s (on whose task force he served) failed because they explicitly tried to control costs. The way forward, he argues, is to forget about consciously trying to contain costs (if society wants to spend more on health care, why is that wrong?) and focus instead on improving quality. He would rearrange the incentive structure of the health care market to reward hospitals, HMOs and other medical enterprises for improving the actual health outcomes of patients, rather than just for delivering the service, as is the case now. Such changes, he says, would not only extend people’s lives, but would have the effect of containing cost increases. Lowenstein, who in this piece basically channels (and presumably agrees with) Cutler, writes:

America’s fee-for-service system does not require doctors to measure. It rewards them for each instance of delivered ”care,” Cutler notes, but not necessarily for the end result — for ”health.” This is especially true for chronic patients, whose well being depends on following a long-term regimen of care. Diabetics, for instance, should receive yearly eye exams, regularly monitor blood sugar and cholesterol and take other steps to avoid problematic (and expensive) complications. ”Doctors say, ‘You really should get your eyes examined,”’ Cutler notes. ”There is no follow-up. Every doctor you talk to says: ‘I know we don’t do a good job on that. We don’t get paid for it.’ My way, we would pay them.”

How would ”paying for performance” work? In the late 90’s, HealthPartners, a not-for-profit health plan in Minneapolis with 630,000 members, instituted a bonus system to providers. It paid doctors extra if their diabetic patients got blood sugar and cholesterol below certain levels, ceased smoking and took aspirin daily. In 1996, 5 percent of patients met all criteria. By 2003, 17 percent did. Similar gains were registered with heart patients. ”These clinics are trying to provide quality care,” says Dr. George Isham, the plan’s medical director. ”What we’re doing is putting a measurement on it.” In 2003, the plan awarded a total of $9 million to doctors on merit. But was it worth it?

At Isham’s request, Cutler and a team of colleagues analyzed the economic payoff. They found that the program reaped huge rewards. It cost $330 a patient and was expected to save roughly $30,000 over each patient’s life. While the analysis is not precise, Cutler wrote, it ”illustrates a general point that professionals in health care have known intuitively for some time: . . . comprehensive disease-management programs are clearly worth the investment.”

So far, so good. But why, you might ask, are health care firms not already implementing such pay-for-performance systems? Lowenstein provides the answer:

The rub is that the investment was only marginally worth it for HealthPartners. The gains went to the patients, in the form of better health, and to their employers, who were expected to suffer less absenteeism. HealthPartners did recoup some of its investment, as members were hospitalized less frequently. But some of those people would change jobs and change insurers, so the benefits were largely reaped by someone else. Ultimately, as patients retire, they will be reaped by Medicare.

This ”exemplifies some of the problems inherent in our current system,” Cutler wrote. A program with huge benefits for society (and for patients) offered only a marginal incentive to the health plan to create it. HealthPartners did not have the option of simply raising its rates, because healthy patients would have departed for a cheaper plan. This is why closed-loop systems — systems in which patients made healthy don’t leave — tend to work best.

The two examples of such “closed-loop systems” that Cutler and Lowenstein point to are Kaiser Permanente, the giant nonprofit HMO, and GE, with its hundreds of thousand of lifetime employees. Curiously, however, they don’t mention the biggest and most successful example of such a system: the VA. As Phil Longman reported in the last issue of The Washington Monthly, veteran?s hospitals, once dangerous and scandal-ridden, have turned around dramatically and are now providing arguably the highest-quality care in America. They’ve done this by implementing precisely the kind of careful, computerized case-management systems that Culter advocates. And they’ve been successful at this implementation because they have what Cutler and Lowenstein acknowledge is the key to making the whole quality regime work: patients who remain in the system. Vets tend to stay with the VA for decades, and so the VA reaps the long-term economic rewards of its investment in quality control. That is not the case in private medicine, where patients jump from one system to another as they change jobs. (Slate’s Timothy Noah has smart thoughts on Lomgman’s piece here.)

Why didn’t Lowenstein mention the VA? Perhaps he simply hadn’t read Longman’s article. But another possible explanation is that the VA’s success screws up Cutler’s argument. He’s an economist and a pragmatist. He thinks government-run health care is a political nonstarter, believes market competition in health care creates valuable incentives for innovation, and advocates a form of vouchers as the best way to achieve universal health care. Count me as someone who shares Cutler’s general view of the world. But the fact is that a government agency, the VA, has achieved what no private-sector health care provider so far has. And reforms that make the health care system even more market-driven are likely to lead to patients jumping around even more among health care providers, further undermining what little incentive there is for investments in quality. I wish Lowenstein had dealth with these fly-in-the-ointment facts in his piece. I would honestly like to know how Cutler would respond to them.

Paul Glastris

Paul Glastris is the editor in chief of the Washington Monthly.