Bundled Payments, Barely Hatched, Go the Way of the Dodo
Last month, I wrote about Accountable Care Organizations (ACOs), medical groups accountable to the federal government for management of healthy populations. Even Zeke Emanuel recognizes that they are failing. Dr. Emanuel advised Medicare should “lump together” all the services associated with a procedure, such as a hip replacement, and pay one fee for the entire services.
As I noted, Medicare already does this via its Bundled Payments for Care Initiative (BPCI) which launched in 2013. At the time, hospitals and other providers were offered voluntary participation. Just a few weeks ago, the Medicare decided to make bundled payments mandatory for some procedures in some areas. Now we know why: Providers are learning that the bundles don’t work.
Whether we call them “lumps” or “bundles,” the results of the voluntary initiative are coming in and they tell pretty much the same story as the ACO experience:
Medicare’s voluntary test of bundled payments added new contracts in July, but about two-thirds of the hospitals, medical groups, nursing homes and other providers that had initially enrolled instead dropped out.
The initiative, known as the Bundled Payments for Care Improvement initiative and launched under the Affordable Care Act, initially attracted nearly 7,000 providers that agreed to formally review how they could enter bundled-payment contracts with Medicare. The CMS announced on Thursday that 2,100 providers finished that review and entered contracts under which Medicare will bundle the costs of treating various conditions—heart failure, joint replacement, stroke, heart attacks—into a single payment.
The reason for the failure of both initiatives is the same. ACOs are accountable to the federal government instead of their patients. Similarly, the “bundles” are bundled by the federal government. The only way to figure out which services should be bundled together in one payment is to let entrepreneurs try different bundles and let patients decide which to choose.
We can all agree that we want providers held accountable. But, to get there we need to get the incentives right. If we align the incentives of seniors, health plans and doctors/hospitals, then we can expect better outcomes.
Under current law, seniors are assigned to a ACO retrospectively. The ACO has no way of knowing who their members are. Furthermore, seniors have no incentive to work with their ACO if they don’t want to. They can use whatever care they want to without following the guidance of a care coordinator.
A hospital-centric ACO will not look for ways to save money on inpatient care, when doing so will reduce their income more than the performance bonus payments.
Finally, bundled payments are a characteristic of a system that competes on price. By contrast, the convoluted CPT codes are an attribute of a system that is bureaucratic and maximizes reimbursements against a formula. Because bundled payments are common in competitive industries, policy wonks thought they could impose them and good things happen. Bundled payments (and price transparency) are correlated with competition, but not the cause of competition (they’re a byproduct of competition).
And when patients move to Florida or Arizona for winter, that change can flip their ACO to a different provider!
“The only way to figure out which services should be bundled together in one payment is to let entrepreneurs try different bundles and let patients decide which to choose.”
The BPCI bundle that I work on does not pick and choose which services are included in the “bundle”. All Medicare services within a 90 day period after a particular episode (in this case an orthopedic surgery) are included. It’s all inclusive. Whether you get a colonoscopy or a follow up to your surgery. If it’s billed to Medicare, it goes in the bundle. There are many types of BPCI iterations going on right now. I was interested to read your article, but from someone who is actually working on this project, it doesn’t sound like you know all the particulars. Ours is going pretty successfully and we’re cutting a lot of wasteful spending.
I realize that sounds a little snarky – didn’t mean it to. Just trying to point out that this is a very intricate program and it’s hard to generalize.
Thank you. I am glad that your program is gong well. I cannot imagine a program – even a government program – where the failure rate would be 100 percent. Nevertheless, if two thirds of participants are already dropping out, that means it is not going well overall.
I think you may have misunderstood my point. I understand the bundle bundles all Medicare-eligible services within the 90-day period. That is the problem. In a functioning market, entrepreneurs – not the government – decide what to bundle.
Look at your car: Some items are standard and some are extra. The government does not decide what is in the standard “bundle” we call a car.
Thanks for the response. It’s interesting how different perspectives can be. There are plenty of commercial bundles out there being implemented by for-profit insurance companies and providers – including birth bundles, orthopedics, and more. When you say that bundled payments are going the way of the dodo – do you mean all bundled payments, or just government implemented ones? I don’t think bundled payments as a concept are going anywhere. Commercial insurance is VERY interested in bundled payments, because there is profit to be made from them.
Also, BPCI is a pilot program, and by the very nature of a pilot program, there will be successes and failures that CMS is trying to learn from. Your assumption that BPCI is a failure due to so many providers dropping out before implementing the program is interesting to me. I don’t see it as a failure, I see it as a litmus test that proves that care redesign is very challenging and something that many providers are unable to master in time to make the BPCI program profitable for them. But it’s got them thinking about bundled payments, I’m sure of it. CMS doesn’t have any obligation to “teach” them how to make this program profitable. And since it’s optional now, they have a choice as to whether to participate. When it becomes mandatory, say with something like CCJR, they are going to have to redesign care – cut waste, and provide value rather than quantity.
The entry only discusses the Medicare pilot, so is only addressing the federal government’s role.
I appreciate your statement that a pilot is to learn from. In the private market, we also accept trial and error. The difference is that trial and error in the private market leads to learning and improvement. When the government does it, that tends not to happen.