You Get What You Pay For—Or Do You?

The Affordable Care Act, as it turns out, isn’t just about providing health insurance coverage for the 40 million previously uninsured Americans. It’s also about reforming Medicare, in part to pay for some of the costs of providing health insurance for everyone, in part to keep Medicare from going bust, and in part to improve the quality of care provided by Medicare. The favorite strategy for modifying Medicare is “value-based purchasing,” which is another name for pay-for-performance. The idea is simple: don’t just pay whatever doctors or hospitals ask for and don’t pay per service (the original fee-for-service model); instead, pay based on results. After all, physicians aren’t supposed to perform tests and procedures just for the sake of doing something; they are supposed to do things in order to improve health. So why not pay physicians only if they make people better? 

The problem, of course, is that not everyone will get better, no matter how state of the art their treatment, and some of them will get better but along the way they will also suffer from all kinds of complications. To deal with the realities of taking care of people who are old and sick, Medicare has adopted a policy that rewards—or penalizes—hospitals based on their performance on a combination of measures: the processes of care, the outcomes of care (specifically 30-day mortality), the patient’s satisfaction, and whether or not patients are readmitted to the hospital within a month of discharge. The big question is, does this approach work?

Previous studies have failed to show any benefit on clinical processes or patient satisfaction. Now, a new study in BMJ suggests that it doesn’t improve mortality either. The authors examined mortality among patients with heart attacks, heart failure, or pneumonia (the 3 conditions for which Medicare “incentivizes” hospitals using its value-based reimbursement scheme). They compared mortality rates for these conditions before and after the introduction of Hospital Value-Based Purchasing. They studied whether changes in mortality in the target conditions differed from changes in a comparable group of patients with other medical conditions. They tested whether the trends were any different at hospitals that didn’t participate in the HVBP system. And to look for trends, they determined mortality rates over a 3-year period before the introduction of Hospital Value-Based Purchasing and over the 3 years after its introduction. The result: nothing changed. 

Not everyone will be satisfied with the authors' choice of the comparison group—either of patients with different medical conditions or of hospitals that participated in a different reimbursement scheme. The risk adjustment process might be flawed. Maybe 3 years wasn’t long enough to see an effect, especially since the incentives have been changing—initially, hospitals were rewarded if they did well, now they are penalized if they do poorly, and the magnitude of the penalty increases annually. So it would be premature to conclude that value-based purchasing is a failure. But surely it isn’t a great success, either, if no one has been able to prove that it does what it’s supposed to.

Medicare has the potential to shape geriatric care in the U.S. There’s no question that strategies invoked in the past such as the introduction of prospective payment for hospital care (ie paying a fixed amount for a given condition, rather than a fixed amount per day in the hospital) have made a huge difference in both costs and outcomes. But it’s not at all clear that the prevailing enthusiasm for pay-for-performance is the answer to providing better, more cost-effective care to older people. 

Maybe we need to go back to the drawing board and analyze the weaknesses of our current system. Perhaps what we will find is that the weaknesses are not just fragmentation, lack of coordination, and the triumph of high tech over high touch, although these are all important. Perhaps what we will find is that the weaknesses include a focus on disease rather than function, on individuals rather than families, and on the values of physicians rather than patients.