Most Medicare Payment Innovations Don’t Save Money – But We Shouldn’t Give up Yet

Could Medicare save money if it paid providers differently? The program has spent the past 20 years trying to figure that out.

For example, would doctors and hospitals provide care more efficiently if they were paid to better coordinate with each other? What if they received bonuses when patients experienced better outcomes?

To see if approaches like these would work, the Centers for Medicare & Medicaid Services (CMS), the agency that runs Medicare, has launched numerous demonstration programs in recent years. These are experiments in which new payment arrangements are tested on a small scale.

Most of the experiments have fallen into one of two categories. In the first, physicians use disease management programs to coordinate care. These are systems in which nurses outside of a practice serve as care managers. They educate patients on how to manage a chronic illness and then monitor their health and track their treatment.

The second category includes various arrangements in which physicians and hospitals are paid extra if they achieve higher scores on quality measures or can demonstrate cost savings. Among them is reimbursement through “bundled payments.” These are allotments that cover all inpatient and physician services for a designated procedure such as heart bypass surgery. The various providers involved then divvy up the payments among themselves.

The nonpartisan Congressional Budget Office reviewed ten of these experiments to see if they worked. Unfortunately, the results were not encouraging. CBO’s report, which was issued this month, found that most of them did little to reduce spending.

But there is a silver lining. One of the programs did have an effect. That was the one that used bundled payments for heart bypass operations. It reduced spending by 5% to 10% at five hospitals and by 20% at another two.

Why did that experiment succeed while the others did not? Perhaps the reason is that it was the only program that changed the underlying financial incentives for the doctors and hospitals involved.

Medicare pays most doctors on a fee-for-service basis. That means they earn extra money every time they perform a service. Hospitals are paid a set fee based on each patient’s diagnosis, which means they gain revenue each time a patient is admitted.

Most of the experiments continued to pay doctors and hospitals the same way. They could earn bonuses for better care coordination or better outcomes, but they still stood to gain by doing more, which generated more in fees.

The one experiment that actually changed the underlying reimbursement arrangement was the one that used bundled payments. In that program, doctors and hospitals received a single negotiated rate that was agreed to in advance, so they all faced a common incentive to coordinate care. CBO found that in competitive markets, providers agreed to rates that included steep discounts.

The lesson is that incentives can work if they are meaningful. As long as providers are paid on a fee-for-service basis, they will find ways to render more care. Added bonuses and extra layers of care coordination seem to have little effect.

However, if the reimbursement paradigm is truly changed, for instance by setting total payments for each patient in advance, health care providers can, indeed, learn to be more efficient.

The health reform law encourages more trials of Medicare innovations. The CBO report shows that some of them might work, it they involve more than just tinkering.