On February 14, the Oakland, Calif.-based Integrated Healthcare Association announced that it had been awarded a three-year, $2.9-million grant by the federal Agency for Heathcare Research and Quality (AHRQ) to support a demonstration project around bundled episode payments for physicians, hospitals, and other providers, in the private health insurance market in California.
The demonstration project has garnered commitments to participation from several hospitals and medical groups in Northern and Southern California, along with one ambulatory surgery center organization, and four major health plans operating in the state.
Though the demonstration project is still relatively early in its development, much is already being learned, says Weslie Kary, program director, medical technology, for IHA. Kary spoke recently with HCI Editor-in-Chief Mark Hagland regarding the project, and her perceptions of the implications of the lessons learned so far, for bundled-payment arrangements in healthcare more generally. Below are excerpts from that interview.
Healthcare Informatics: How did the program come about? And can you explain how a bundled-payment demonstration project differs from the development of an accountable care organization (ACO)?
Weslie Kary: We are working on procedure-based bundled payments, not on any kind of ACO organization. While bundled payments might be a part of an ACO, we’re working on bundled payments around procedures. So in terms of what’s coming out of healthcare reform, per CMS’s ACE demonstration [the Medicare Acute Care Episode (ACE) Demonstration Project initiated in 2009 by the Centers for Medicare and Medicaid Services, and involving providers from Texas, Oklahoma, New Mexico, and Colorado], we’re working forward on the same concept. About two years ago, CMS started episode payments for something like nine orthopedic procedures and 20-something cardiac procedures, in four states. We proposed working forward on the same general principles.
HCI: What has been learned so far from the CMS ACE Demonstration Project that you want to build on?
Kary: Well, that project is still in process, so there are no results yet. But CMS had found in an earlier cardiac bypass demonstration project that if you could align hospital and physician incentives, there are probably some efficiencies and quality improvement that can occur in the system. So we started a little bit after the ACE demo started; we started in late 2009, to see how this would work in the private sector.
We’re starting with knee and hip replacement procedures. They have discrete beginnings and endings, so they lend themselves to episode-based payment. We’re still working on going live with all of our participants with knee and hip, and we’re starting to look at some cardiac procedures and outpatient knee procedures.
HCI: So not all are live yet?
Kary: That’s correct. One hospital-physician team—that’s [the Newport Beach-based] Hoag Hospital [now an integrated health system, with two hospital facilities and affiliated physician groups], and they have executed at least one contract with a health plan, and possibly two. The contract is with Blue Shield of California, and encompasses Hoag’s affiliated group of orthopedic surgeons.
HCI: What key goals did you set for yourselves when you launched this demonstration project?
Kary: We wanted to see whether it could be done in the private sector, with a real, prospective transfer of risk. This is an idea that has been around for a very long time, but that involves some very significant implementation challenges. If you’re going to take a single payment across a group of providers, there need to be contractual arrangements, and the building of a culture of working together, and then on the health plan side, they have to figure out how to make a single payment for a bundle of services.
And there are a lot of IT issues. And on the provider side, they have to figure out how to bill on a bundled basis, and when they receive the payment, how do they figure out how to distribute the payments to the physicians? When we started this, there were no software products out there to support this; so the model was to do this the same way everybody does transplants, manually, on both the provider and payer sides. I know McKesson and the Trizetto Group now have products that they’ve been developing. It really makes a big difference in terms of the ability to think about episode payments, for there to be software programs that can support this kind of work.
HCI: What has been learned so far?
Kary: That it’s hard. It’s hard, because it involves a culture change on the provider side. In our definition of what an episode is [for purposes of bundling payment], we are saying that related readmissions within 90 days are included within the episode payment. So the idea that you are accountable [for events] within that period, and that you would not charge separately for complications or readmissions; and the idea that you are jointly accountable, among the hospital and physicians, and that they actually need to have a relationship that allows that to occur—in some places, that exists, and in some places, that’s new for the hospitals and physicians.
On the health plan side, we are a facilitative convening organization, as the grant funds our ability to develop things across multiple stakeholder groups; everybody has to fund their own internal development expense. And this is a significant administrative challenge. And the way it operates is that each health plan is executing new contracts or contract amendments with each of these providers; so you look at this and ask how you can bring this to scale on the health plan side.
HCI: In other words, part of the challenge going forward is building such contracts on a broader basis, rather than just executing them one at a time?
Kary: Absolutely. And we only operate in California; but it’s a big market.
HCI: What has the value been of past experience with managed care and capitation in California?
Kary: I think that has been of real value, because we’re talking about the transfer of financial and health risk; so with regard to the ability to even think in terms of accepting that risk, I think we have, if not a unique, then at least an unusual, experience in California. It doesn’t mean that providers don’t nevertheless have to look carefully at what risk they can accept, and what risk can be transferred, what risk should be a part of this contract or not, what risk should be a part of hospital action. But there has been considerable experience with risk.
One of the things we’ve learned is that these procedures are done in many hospitals, so even with multiple health plans participating, the volume is not huge. That’s a positive in terms of experience, but a negative in terms of spreading risk. They can get their toe in the water with a manageable group of patients. But because the numbers are small and they’re looking at how they spread this risk, it’s been a conundrum.
HCI: Have they figured it out yet?
Kary: Well, working to figure it out now. And the number of patients we’re talking about is not going to break any hospital. But we do need to understand what the risks are and how we can manage them, before we move into higher-volume procedures.
HCI: Presumably, they’re using business intelligence tools?
Kary: I know that at least some of them are, and the ones who haven’t come as far as they feel they ought to so far, are looking at how to do that in this context, too, and are looking at which physician behaviors are driving costs in which directions. As part of our project, we’re not insisting that they do gainsharing with the physicians and hospitals, but we are assuming that they will; and that itself is an information systems issue. If they set up gainsharing, how do they set up the metrics to track thing? That’s another whole information infrastructure that they may or may not have contemplated.
HCI: What should readers think when they hear about these developments?
Kary: I think when you’re looking at the hospital-physician community, they should be thinking about how they’re going to do this, because Medicare has been very clear that this is where they’re going with procedure reimbursement. And what we’re doing here in California is an opportunity to test this out within a demonstration project environment. And that’s where the AHRQ grant we’ve received has been very helpful. Our initial question was, could we even bring one up and running? Now we’re thinking about bringing this to scale.
Another issue that comes up for hospital leaders is with regard to which service line or lines they want to begin developing bundled-payment programs. Which service lines are best-positioned initially to do well under such arrangements? For example, perhaps orthopedics; or perhaps cardiac care, because you might have a better relationship with your interventional cardiologists, say, than your orthopedists. You might also be asking the question, in which service line can you learn the most by doing this?
HCI: Obviously, nationwide, people may be underestimating the complexity of this, as it emerges into the mainstream Medicare program.
Kary: That’s right. It’s not an easy thing to do; but what the participants are finding so far is that it does work for all those involved—and it has to be a three-way win—the hospital has to be able to maintain their margin, even though they had to bid at a reduced Medicare rate, in the Medicare demonstration; it has to have some kind of benefit for the surgeons, if they’re going to take on this additional risk and help the hospitals with their cost structure; and the health plans won’t do this unless they see that they’re getting some kind of savings out of it over time.
And hopefully, the patient will get something out of this, too. And what the patient would get, I would hope, would be improved quality from the clinicians working together. And also ultimately, patients should see some ability to compare costs across facilities and physician groups. Over time, for example, you’ll be able to compare a knee replacement in one place with a knee replacement in another.
HCI: In other words, it’s about value.
Kary: Yes, it is.