Amol Navathe, a professor of medicine at Penn, discusses new research on how to cut medical bills using bundled payments.

The U.S. spends more per person on health care than any other major industrialized nation. In fact, costs now reach about 18% of GDP and so anything that can slow the climb – or reverse it – tends to get attention. One promising idea for blunting costs comes from new research on how Medicare experimented with bundled payments for knee and hip replacements. That saved about 5% of total costs. Beyond the monetary gain, that approach also appeared to raise the overall quality of care. Amol Navathe, a professor of medicine and health policy at Penn’s Perelman School of Medicine, was part of the research team. He joined the Knowledge at Wharton show, which airs on Wharton Business Radio, SiriusXM channel 111, to discuss the findings. (Listen to the full podcast using the player at the top of this page). 

Following is an edited transcript of the conversation.

Knowledge at Wharton: Can you explain this model under Medicare?

Amol Navathe: Medicare started this program called bundled payments, which they had done before, where you take a hip or knee surgery and look at it as an entire product or service. The hospitalization, the surgery itself, plus all of the physical therapy and recovery gets bundled into one payment. The crazy idea here is they had done this not on a voluntary basis, where hospitals or physicians raise their hand and say, “yeah, I want to try this,” and instead they mandated it. They picked about a quarter of the urban markets in the U.S. and said every hospital in that market has to do a bundled payment for hip and knee surgery.

The reaction was really mixed. There were some hospitals that thought, “We can win doing this. There’s a lot of excess waste or spending that happens after a hospitalization. If we can get quality up and do some coordination, we can save money and share in that savings. We never got a chance to share in those savings in the past.” Other hospitals were terrified. “All of the sudden, I’m at financial risk for what happens outside of my hospital’s four walls?” That’s scary.

Knowledge at Wharton: What were the results?

Navathe: About half the hospitals that were forced into this program achieved savings. When we started to unpack a little bit more, the savings was about 5% of the overall cost of the whole joint replacement episode. Out of the gates, that’s good news. A 5% savings is pretty significant if we think about year over year. When we look at who’s actually achieving those savings, we add a little bit more texture. Some hospitals are winning; some types of hospitals are losing.

“Patients are the ultimate beneficiaries because if the cost goes down, their premiums go down.”

Knowledge at Wharton: How is this research viewed in terms of making policy decisions?

Navathe: I think this is viewed as a proof of concept. Hip and knee surgery is viewed as the best place to try this because it’s a very well-defined procedure and recovery. We know in general that 90 days after a surgery, people should be up and moving. They should be essentially fully recovered.

If you think of heart bypass surgery, for example, that’s a very different beast. Trying this in heart bypass surgery is perhaps a bigger affair. By the way, Medicare wanted to do that, but then it got canceled with the change of administrations. There’s more risk associated with heart bypass, particularly for the hospitals. So, I think right now largely we view this as proof of concept. It looks like it can work in this condition. Now we need to try it out in a number of different conditions and see if it works for other surgeries and non-surgeries.

Knowledge at Wharton: Are there other areas suited for this type of idea?

Navathe: Absolutely, to the extent that there are other well-defined surgeries or procedures. A great example would be somebody who has a heart attack and then comes to the hospital and needs a stent put in. We expect within 30 or 60 or 90 days, they’re going to be fully recovered from that whole episode. We know that the services they get afterwards — in terms of cardiac rehab, and follow-up from their primary care doctor and their cardiologist — all really matter in how well they do in the long run. That’s exactly the constellation of pieces that we need for this bundled payment to work.

Knowledge at Wharton: Tell us more about the savings with bundled payments.

Navathe: Just to give you a sense, there are about 450,000 hip and knee surgeries each year in the U.S. that are covered by Medicare. That accounts for about 5% of all hospitalizations for Medicare. It’s a really big deal. The other beauty of this is when Medicare starts to do this, commercial insurance companies like the Blues and the Cignas and the Uniteds of the world get the idea that they can do this, too. We need some more evidence, but we think this translates into the commercial sector.

Patients are the ultimate beneficiaries because if the cost goes down, their premiums go down. And clearly, it’s their quality of care. It’s their pain levels that go down. It’s their ability to walk after this hip and knee surgery that goes up. This is supposed to be about the patients. We shouldn’t forget that.

Knowledge at Wharton: You’re talking about millions of dollars being saved each year on health care in a time where we know the spending is expansive.

Navathe: That’s absolutely right. Medicare has been testing a variety of models. I think largely we’ve seen pretty modest effects, and this is one place where I think we’ve seen pretty consistent effects. These hip and knee bundles have been tried since 2013. The mandatory program started in 2016, but there’s been a lot of experience with this. I think we have more than proof of concept at this point. In fact, we can show that if you mandate hospitals to do this, they can succeed on average. About half of hospitals succeed. There are systematic patterns. We do find that bigger,  higher volume hospitals, hospitals that tend not to be in the safety net, are the ones that are really succeeding here. I don’t want to overstate this. There are some reasons to be cautious or careful about this.

Knowledge at Wharton: But as this idea moves forward, could you figure out which hospitals would be able to do these type of payments?

“This is supposed to be about the patients. We shouldn’t forget that.”

Navathe: There are 67 markets that started out in this Medicare model. The Trump administration has shifted that now. They’ve actually cut it in half, so 34 markets are continuing with this mandatory model. But what I’d like to do is just paint a picture here as to why these things might be working. Imagine you’re a hip replacement patient and you’re deciding what to do after the hospital. I could go to a rehab facility or a nursing facility to get my physical therapy. In the normal system, the hospital would say, “Where do you want to go? Go ahead. Doesn’t really matter where you go because we’re not on the hook for the quality. We’re not on the hook for the cost.” When this model gets put in place, all of the sudden the orthopedic surgeon and nurses and care managers start to tell you, “This is a higher quality facility. The physical therapy is better there. You’re likely to get out faster and feel better faster.” This hospital is now communicating and measuring what’s happening outside of their four walls. That’s magic, right? All of the sudden, we have this incentive coordination.

We have other work under way that’s looking at the markets that participated in this mandatory experiment and how close are they to hospitals in other markets. Interestingly, we find that about 70% of Medicare beneficiaries reside in markets that look really similar to the ones that this test happened in. So, even if we scaled it just to those markets, we could scale the impact by at least three times, it seems.

Knowledge at Wharton: Because of the savings, will this model continue to be applied even though it has been reduced under the current administration?

Navathe: I think that’s right, and with maybe a little bit of a pivot. Tom Price, the prior secretary of Health and Human Services, wasn’t a big fan of this concept of mandating hospitals or mandating physicians to be in it. Understandably, there was push back. If you’re a lower volume hospital or  a smaller hospital — the other hospitals that didn’t do as well here — you might push back and say, “This is really hard on us.” Hospital margins on average in the U.S. are not that high. They might be 3% to 5%, so this is a lot of risk for a small hospital.

So, the Trump administration actually pivoted. They’ve expanded the concept of the bundled payment, including for hip and knee surgery, but they’ve done it on a voluntary basis. They still want hospitals and physician practices to raise their hand and say, “I want to do this.” What they’ve tried to do is create an environment around participation to be more favorable.

Knowledge at Wharton: Is there data showing that more hospitals and physician are voluntarily participating?

Navathe: The new program starts in October, so we don’t know yet. What we do know is there was a precursor voluntary program that started in 2013 called Bundled Payments for Care Improvement. That program had pretty rapidly expanding participation from 2013 through 2016. If we look back and say that over 1,000 hospitals and physician practices decided to participate in the voluntary bundle, it certainly suggests that there’s even more reasons to do it now than there were in the past. So, we’ll probably see pretty good participation.

Bundled payments are driving 5% savings, and that’s big news. But there are other models out there that Medicare is pushing, like the Accountable Care Organizations (ACO) model that acts on a very different population of patients. These might be the chronic disease patients. This might be your grandmother who has diabetes and a bad heart and bad kidneys. For them, you need a model that’s about total care. That’s what that ACO model is. We’re still figuring out how to piece this whole pie together in terms of bending the trend on spending. But I think we are finding that there are specific models, like bundled payment, that seem to work in focused areas.

I think it’s impossible to get a one-size-fits-all solution because of the variability in terms of community demographics, community needs, types of patients, disease burden, etc. That being said, I think some of the principles that we’re seeing here, creating some accountability on the hospitals and the physicians for the cost and quality of care, are common across these different models. Those are the pieces that states can say, “In our Medicaid program, we’re also going to use these principles just like the federal government’s going to use in Medicare.” That part, I think, is generalizable. It needs to be tailored to the specific population, but I think there is some real promise in that concept.

“We’re still figuring out how to piece this whole pie together in terms of bending the trend on spending.”

Knowledge at Wharton: Are states starting to buy into this?

Navathe: Absolutely. One of the interesting things that Medicaid programs did for the states in the past is they used traditional managed care contracts. This is like the Clinton-era health care management organization (HMO). They tried that in Medicaid and found they can cap the budget, but they’re not getting a lot of gains in quality. More recently, they’ve tried to reinvent what that looks like. They call them ACOs, or Accountable Care Organizations, just like Medicare does. They have a global budget, so they have a fixed budget for patients. But they’re deploying more incentives around quality. They’re saying, “We want you to coordinate care. We want you to tailor programs to the specific communities because a Hispanic community might be very different than a rural white community.” And they’re starting to see real results. We’ve seen it happen in Oregon. We’ve seen it happen in Colorado. I think there’s promise.

Knowledge at Wharton: How is the insurance industry reacting to this?

Navathe: The insurance industry largely has been favorable on this because without a big behemoth like Medicare changing the incentive structure, you could imagine sitting around a negotiating table for a Blue Cross plan in a health system like Penn and say, “We want you to take a risk.” The health system would say, “Why would we ever do that? That sounds crazy.”

If Medicare, which is the biggest payer for the majority of these different hospitals across the country, is saying, “You’re going to take risks,” which the hospitals can’t really opt out of, then the commercial payers are coming in and saying, “We’re going to create alignment for you. We’re going to make it so that the way you deliver care is consistent across your Medicare patients and your Blue Cross Blue Shield patients or your United Health Care patients or your Cigna patients.” That alignment is really valuable to these hospitals. That’s where Medicare has this opportunity to catalyze change in a way that I think any single commercial insurer would struggle.

Knowledge at Wharton: What’s next for this research?

Navathe: I think the next steps are, do these results persist? What we’re showing you are the year-one results of the program. Is this going to persist into year two and year three in those 34 markets that are continuing in the mandatory program? That’s number 1.

Number 2 is, are there unintended affects that we might really worry about, such as vulnerable populations that are either from low-income communities or black patients who traditionally had disparities in their ability to access hip and knee surgery? Are we exacerbating problems in access for these vulnerable populations? We really need to get a handle on this because while this could benefit some people, it could be really bad for others. And that wouldn’t be good for the federal government to push forward.

Knowledge at Wharton: Many patients have felt caught in between because of costs. Hospital systems have to feel much better about these types of results because, to a degree, they’re stuck in the middle.

Navathe: Absolutely. We’re talking about change here, and change is hard. There are some hospital systems that are in environments like Massachusetts or Rhode Island, where the states have put cost targets or controls on the prices. That has created a situation where they know they need tools and mechanisms to drive savings. That does create more alignment. You have other places where we don’t have those kind of restrictions and price controls. The hospitals there are kicking and screaming because change is hard and they have less skin in the game, so to speak, around that change.