For most of its history, Medicare paid for health care in ways that encouraged more services — whether they improved health or not.
Critics called it an emphasis on volume, not value.
The Affordable Care Act was intended to change that, and Medicare started a number of programs to do so, including several new ones this year. Nearly a decade after passage of the A.C.A., is value-based payment working?
The Obama administration’s goal was for 90 percent of Medicare payments to hospitals and doctors to be tied to measures of quality by 2018. Although distancing itself from specific targets, the Trump administration has also committed to this goal.
According to the Health Care Payment Learning and Action Network, Medicare’s 90 percent value-based goal has been achieved. That sounds like mission accomplished. But it’s not as simple as that.
“It doesn’t mean that the vast majority of the care Medicare purchases is linked to actual value,” said Sherry Glied, a health economist, and dean and professor at the Wagner School of Public Service at N.Y.U. “We don’t even know what we mean by value. How do you pay for something when you don’t know what it is?”
Different programs have different notions of “value.” Medicare’s Hospital Readmissions Reduction Program, established in 2010, penalizes hospitals that have high rates of readmissions for certain illnesses. Although hospitals can lose only up to 3 percent of payment, 100 percent of their payment is considered “tied to value.”
The key is whether programs like this improve health care quality or reduce health care spending. The Hospital Readmissions Reduction Program has been extensively examined, with studies drawing different conclusions. Initial analyses of the program suggest it is responsible for reducing hospital readmissions, saving Medicare billions of dollars a year.
But some later studies found that these reports of success were overstated, among other concerns. Some studies found that reduced hospital readmissions were associated with increased risk of death, though not all studies agree on this point, and experts disagree on the value of the program. What is clear is that the program has had a smaller impact on hospital readmissions than originally thought — perhaps reducing them by as little as one-third of a percentage point.
Another Medicare value-based payment program is considerably more broad. The Hospital Value-Based Purchasing Program was introduced in 2011 and rewards or penalizes hospitals based on mortality; infection rates; patient experience and safety; cost; and other measures of quality — 20 in all. Typical bonuses or penalties are a fraction of 1 percent of a hospital’s total Medicare payments.
A study published in Health Services Research compared about 2,800 hospitals in the program with about 300 exempt from it. The study found no effect from financial incentives of the program on quality of care or patient satisfaction. But the study included data only up to nine months after the program’s start, and it may take hospitals longer to make measurable changes.
Another study, published in BMJ, looked at outcomes two and a half years after the start of the Hospital Value-Based Purchasing Program. It found no differences in changes in mortality rates between hospitals in the program versus those exempt from it. And another study, published in Health Affairs, found no evidence that the program improved patient satisfaction up to three years after implementation.
Jose Figueroa, a physician at Brigham and Women’s Hospital in Boston, and an assistant professor of medicine at Harvard Medical School, was an author of both of these longer studies. “So far, there’s no evidence the program has improved quality or patient satisfaction,” he said. He ticked off some possible reasons: “The financial incentives are too weak to drive any meaningful changes across hospitals. The program’s design, with numerous measures across different domains, makes it hard for hospitals to understand what to focus on.”
Medicare has achieved greater success with programs that have raised the stakes — ones that have put hospitals and health care organizations at greater risk of financial loss or have offered prospects for larger financial gain.
One popular approach, called “bundled payments,” pays health providers one amount for all of the care for a certain condition within a certain period — like 90 days for hip replacement care. There is evidence that some of these programs can save money without reductions in quality, although design details matter.
“Bundled payments are a straightforward way to make hospitals consider all the costs they are responsible for,” said Adam Sacarny, assistant professor with Columbia’s Mailman School of Public Health. “The evidence suggests they encourage hospitals to treat patients more efficiently, although the cost savings are at least partly offset by extra payments to hospitals to reward them for saving money in the first place.”
Another approach — accountable care organizations — also takes many forms. What they have in common is they offer health care organizations the chance to earn bonuses for accepting some financial risk, provided they meet a set of quality targets. Many studies of accountable care organizations (A.C.O.s) have found they reduce spending with no quality degradation.
“There is strong evidence that, on average, Medicare A.C.O.s save a modest amount of money,” said Alice Chen, assistant professor at the University of Southern California Sol Price School of Public Policy.
Although accountable care organizations have saved a few percent of Medicare spending, the amount varies by program design. “We’ve found that A.C.O.s that are physician groups as opposed to big hospital systems have produced more savings,” said Michael McWilliams, a professor at Harvard Medical School and a general internist with Brigham and Women’s Hospital. “That’s because physician groups don’t erode their own revenue when they keep their patients away from hospitals.”
So over all, is Medicare moving toward higher value? “There has been some progress, but even the most generous read of the evidence is very far below the projections made by fans of value-based payment before the A.C.A.,” Ms. Glied said.
Robert Berenson, a fellow at the Urban Institute, agreed: “Value payment overemphasizes performance measurement, but even so, it’s been disappointing. We simply lack good metrics that can’t be gamed or evaded by most targeted providers.”
But some are more optimistic. “The successes are more like singles than home runs,” said Michael Chernew, a health economist with Harvard Medical School. “Despite the modest results, I think some approaches, like A.C.O.s, are a foundation for future improvements.”
Paying for health care value is a popular slogan, but Medicare is still figuring out how to do it.