New Marketplace
New Risk, New Business Models

The Road to Successful Bundled Payment Risk

Interview · September 22, 2016

In 2013, Cleveland Clinic piloted Medicare’s Bundled Payments for Care Improvement Initiative at Euclid Hospital. Bundling for total joint replacement under this program proved successful, prompting Cleveland Clinic to expand the initiative to nine of its other hospitals. What did it take to succeed? Monica Deadwiler, Senior Director of Financial Product Innovation at Cleveland Clinic, sat down with Leemore Dafny, Lead Advisor for the NEJM Catalyst New Marketplace theme, for an in-depth discussion about where they started, what they learned, and how they scaled up.

“I would say [the] number one important [thing] with any type of alternative payment model program is to have strong physician leadership,” says Deadwiler. “We had varying types of physician leaders, but I must say across all of the sites that we’ve rolled this out at thus far, [the physician leader] has been our key to success.”

Another key learning involved pre-discharge planning to help patients navigate post-discharge, and the role of the specialty care manager in this process. “Once we were able to start focusing on the process from the very beginning, the onset of the care path or journey,” explains Deadwiler, “we were able to then redesign the supports for the patient in that pre-period that then impacted their outcomes post-discharge.”

Are bundled payments a harbinger of broader change to come? “In this drive toward more accountable care, the payment model structures are definitely driving change,” says Deadwiler. “Do I think everything should be bundled and reimbursed via bundled payment methodology? Absolutely not. But I do think that it has its uses in specific instances.” Read or listen to the full interview below.

 

 

Leemore Dafny: This is Leemore Dafny for NEJM Catalyst. I’m speaking today with Monica Deadwiler of the Cleveland Clinic. Monica is Senior Director of Financial Product Innovation and the author of a recent case study for Catalyst on bundling for total joint replacement. Monica, thanks for joining me today.

Monica Deadwiler: Thank you for having me.

Dafny: Obviously our audience for Catalyst is incredibly interested in bundled payments and, in particular, in successful examples of reforming care in order to succeed under bundled payment programs. So, you’ve described what the Cleveland Clinic did at Euclid Hospital. Let me start by asking you, where did you start? How did you get a lay of the land on what was happening at Euclid and what you might need to do in order to succeed?

Deadwiler: Yes, Leemore, that’s a great question. We actually started with the data. As a part of entering into this program, CMMI [the Center for Medicare and Medicaid Innovation] shared with hospital systems detailed claims data that represented all of the care that surrounded the total joint replacement. When we received that data, we started to interrogate it and started to try to pull together a story: what was the patient’s journey when they were having a total joint replacement? And we found some very surprising results. We never realized that patients went through so many different variations of their journey, especially when it came to what happened to the patient after they were discharged from our care.

So, that’s where we started at — in looking at variations and then using that information to identify which patients seemed to have better outcomes, also looking at the spend data from Medicare [and] combining that information with our own internal data that we collect. Our orthopedics institute has been collecting patient-reported outcome data around joint-replacement patients for a number of years. To be able to combine that data with the claims data or financial data that we receive from Medicare really helped us to start to pinpoint areas of opportunity.

Dafny: Monica, you said a lot of fascinating things there and I’d like to ask some follow-up questions. You used the word journey to start with, the patient journey. Just so we’re all on the same page, can you explain where the journey begins and the alternative endings, if you will, so we have a sense of the scope for what the bundle is?

Deadwiler: Maybe if I could start off by saying the journey — or let’s say the care path — might have a different time frame than what we’re contractually obligated for from the bundle payment standpoint.

When we started to look at the care redesigns we started to look at the journey from the standpoint of when the patient was indicated for surgery. This could be 30, 60, 90 days — even longer than that before the patient actually has their actual surgery. And then taking out that journey through, of course, their pre-care, their surgery, their inpatient care, and then their post-acute care, which can go out 30, 60, 90 days.

So, we looked at the data at different time frames to be able to really try to capture some of the outcomes we were seeing, and then [we] also tried to look and see what could have been done differently, maybe earlier in that patient’s journey or care path.

Dafny: You said in the case study that you participated in Medicare’s BPCI Model 2, which I gather means that you are responsible for care beginning with the inpatient hospitalization for the joint replacement and concluding 30 days after. But one thing I’d be interested in is whether you are commercially expanding that range and/or whether you think that’s the optimal length of period for this episode.

Deadwiler: That’s a great question. Even with Medicare, we originally went into the Bundled Payments for Care Improvement program, the BPCI program, with a 30-day post-discharge window. We’re currently also participating in the Comprehensive Joint Replacement program — which is also a Medicare program — at one of our locations, and that program takes us out to 90 days. We do have commercial bundles that we’ll take out for 30 days up to even 90 days, but truly for a joint replacement we feel as though the real care that’s related to the joint replacement really occurs within the first 30 days of post-discharge.

Dafny: You mentioned outcomes, and that is of course critical to the success of this initiative. What did you learn about your outcomes at the start and what changes did you make in response to those learnings?

Deadwiler: What we realized was that it was really important to help the patient navigate post-discharge, and that in order to successfully achieve that, the planning needed to start far in advance of that patient being admitted to the hospital. It really needed to start at the beginning when they were indicated that they were a candidate for surgery.

So I would say that, that was probably one of our biggest takeaways, because then once we were able to start focusing on the process from the very beginning, the onset of the care path or journey, we were able to then redesign the supports for the patient in that pre-period that then impacted their outcomes post-discharge.

Dafny: Monica, I think that’s a very interesting insight that a big opportunity was in pre-discharge planning for post-discharge. And the needs you described, many would think of those as not traditionally the bailiwick of a hospital system, which means you may not have had the resources that you needed in place to execute on that need. So what can you tell me about the solution that you developed to satisfy the need?

Deadwiler: You’re spot on there, Leemore. You’re exactly right. We did not have all of the resources in place that were required to really operate in this longitudinal care world.

What we did was that once we developed the care redesign, we also worked with a variety of disciplines and at each hospital, calling together many of the different disciplines that support the patient in their pre-surgical period, their inpatient period, their surgical time frame, as well as their post-acute. So we started to talk through, what would the ideal care model look like?

So really laying that out and understanding what that was, and then from there being able to think through, do we have the right resources in place to actually do this? One of our physician leaders, who was really one of the architects of this program, volunteered up his own practice nurse to help the patients through this journey from the time that they’re indicated for surgery all the way through the time that they’re recovered (30 days out).

Dafny: You mentioned a physician leader who was engaged, and in your article you talk about that being the number one critical input into successful bundled payment program design. Can you share with me some best practices on engaging those physicians?

Deadwiler: I would say [the] number one important [thing] with any type of alternative payment model program is to have strong physician leadership. With the program we implemented at Euclid Hospital, it was led by a physician leader who was also the president of the hospital. Without his leadership we would not have been able to implement this program. And then when we wrote this out at our other hospitals, our first priority was identifying and getting the buy-in of a physician leader to lead the rollout at each one of those sites. So, it is very key to have physician leadership at the onset.

Dafny: And once you have that physician in place, I’m gathering that “my way or the highway” isn’t the way they’d go about working with their colleagues to develop a plan. Do you have some examples of how these leaders engage their colleagues?

Deadwiler: Great question. A key thing done with our rollouts is that once that physician leader is in place — we’re working through the plan, comfortable with the plan — there is a kickoff meeting that’s called. That kickoff meeting usually includes many of the practicing physicians and surgeons that are involved with the care redesign, [and] many other clinical staff that will be involved with the program as well. And that physician leader is the one who is giving the message, explaining the program, and taking the questions.

The physician leader is really owning the program. Many times the physician leader may be a hospital leader already. In other cases, this was an opportunity for some physician leaders at various sites to develop a deeper relationship in a leadership capacity at that site. We had varying types of physician leaders, but I must say across all of the sites that we’ve rolled this out at thus far, [the physician leader] has been our key to success.

Dafny: How did you scale up, and how do you plan to scale up, the role played by that practice nurse?

Deadwiler: With that we created a new role called a Specialty Care Manager, because before, of course, a lot of the discharge planning was performed by transitional care nurses in the inpatient setting, but we realized that care management needed to start far earlier than the time that the patient is discharged. So we created a new position, and at each one of our sites that has this care model, [it] is a key requirement for that position to be in place.

That position is really best akin to almost like a navigator for the patient: ensuring that the patient understands what type of surgery they’re about to go through, that they’re prepped for surgery, [and] ensuring that the patient is actually linked to the other providers —  because another key learning is that the patients needed to be optimized for surgery if they had any risk factors — [and] that those optimizations were occurring.

Then with the inpatient stay, visiting that patient on the floor, checking in again with the patient and the patient’s family as to what their plans were post-discharge, and then, of course, post-discharge, following up with that patient throughout the time frame that it takes for that patient to recover regardless of their care setting. If they’d go home and receive home health that nurse is following up with them. If they’re discharged to a skilled nursing facility then that nurse is working with that patient as well at the skilled nursing facility.

Dafny: Monica, the notion of having a manager to help the patient through the entire experience sounds long overdue and incredibly helpful. It also sounds awfully expensive given the personalized attention that is necessary, and yet you report pretty positive financial results. So help me understand where you offset those costs with savings elsewhere along the journey.

Deadwiler: Some key areas we focused in on were, first of all, is the patient optimized for surgery? Although the patient may be indicated for surgery, the patient may not be optimized. That patient may have significant risk factors. And all of the clinicians knew that the more risk factors that the patient had, the less likely that the patient was going to have a successful outcome, the more at risk they would be for readmissions, the more at risk they would be for having to go to a post-acute facility instead of going home.

All of those types of adverse outcomes have a cost to them. And so what we felt as though was that if we had this care manager navigating the patient through and helping that patient reduce some of those risk factors, be optimized for the surgery, be prepared for what’s going to happen post-discharge that, that would actually reduce the total cost of care for that patient for that episode and basically offset the expenses of having this new role for this program.

Dafny: Great answer. Now let me ask you the sort of tough questions back to that, [which are] of concern to those who would like to see bundles spread more broadly. Does that mean that providers such as Cleveland Clinic now don’t have an incentive to take on pretty complicated cases where there are higher risks associated with operating, but it might be best for the patient? Is it the case that the reimbursement scheme in place is going to disincentivize that to such a degree that we should worry about supplying care for these patients?

Deadwiler: That’s a really astute question there. What we found via this program was that there was a difference in risk levels in this patient population, which we were being paid one flat rate for. And one of the primary differentiations that we found was the difference between patients that had planned an elective knee, hip, or ankle surgery versus the emergent cases that might be a hip fracture — and those hip fracture cases required a different level of care. With that said, we gave significant feedback back to Medicare, and now Medicare has revised both the Bundled Payments for Care Improvement program as well as the Comprehensive Joint Replacement program, their mandatory program, to differentiate between those two.

But I would say at the clinic we are treating the same types of patients we treated before. And basically, we understand that you are going to have some population of patients where there is not a good way to optimize the patient or mitigate that risk because it’s an emergent case or maybe there are other risk factors that can’t be mitigated. That does not mean that, that patient does not receive surgery and does not receive care. So that’s been our philosophy for the length of the program since October 2013. Overall, taking in all of the patients regardless of what their risk factors are, we have still been able to be successful at essentially bending the trend on the Medicare inflation for these types of patients.

Dafny: Do you think that providers in some areas should be concerned that certain systems will try to avoid some of these more costly cases and send them across town, or is that a fear overblown?

Deadwiler: You know, it is a possibility. I can’t say whether that’s actually occurring right now, but I also do know that whenever you have a change in a business model — regardless of what industry you’re in — when you change the business model I think you also have to look at what constructs are part of that business model that may also need to change.

I think hospital leaders probably need to start looking at their financial models slightly differently and looking at admitting privileges as well, and say that in the past, when a difficult case may have come to their hospital from a physician or surgeon that had admitting privileges but did not normally practice there, that was seen as a win. In this type of program, if you do see those types of anomalies occurring, I think it’s important for hospital leadership to, A, recognize that they are occurring, and B, [to] have some type of protocol to be able to further investigate that trend.

Dafny: That makes sense. Since you’ve broadened it here, let me push further and ask you how you feel bundling is going to spill over into broader questions of hospital management and care redesign. Is this a harbinger of broader change to come? Do you feel we’ll see a lot of replication of bundling across a variety of episodes, but we don’t have a solution for the others? What are your thoughts?

Deadwiler: In this drive toward more accountable care, the payment model structures are definitely driving change. Do I think everything should be bundled and reimbursed via bundled payment methodology? Absolutely not. But I do think that it has its uses in specific instances. Where you’re not being paid underneath a bundle, I think using a bundle construct wherever you have a total cost of care agreement, and using that bundle construct to be able to interrogate your accountable care organization’s spend, is a useful tool.

I think bundling works well when you have specialty care that’s based upon a procedure or that’s very acute. I think that it does not work well in constructs where it is basically a chronic disease state. So, for instance, I wouldn’t recommend bundling for diabetes, but maybe for total knee and total hip replacements I think it is a great thing. And just to build that out a little bit more, I do think that it is a stepping stone into being able to assume more risk — for providers to be able to experiment with risk models on a very limited basis.

Just to summarize, I think that it’s not for everything. I do think it’s a good construct for providers to be able to experiment with risk. And if you do have a total cost of care arrangement, I think it is a good way to be used as an internal medical expense management tool to understand what is happening within that agreement. Maybe not paying based upon it, but just to be able to better understand your medical expense portfolio.

Dafny: Monica, thanks for that recap. Couldn’t have done it better myself. I appreciate all of the comments that you shared with us today and encourage anyone listening to follow up and take a look at some of the impressive results reported for Euclid Hospital.

Deadwiler: Thank you, Leemore.

New Call for Submissions ­to NEJM Catalyst

Connect

A weekly email newsletter featuring the latest actionable ideas and practical innovations from NEJM Catalyst.

Learn More »

More From New Marketplace
ED Telehealth Express Care Service Patient Room

Revolutionizing the Delivery of Care for ED Patients

How the NYP OnDemand Emergency Department Telehealth Express Care Service reduced stay times and revisit rates in one year.

Targeting Unconscionable Prescription-Drug Prices — Maryland’s Anti–Price-Gouging Law

Why, in the early 21st century, are so many drugs that were cheaply available in the 20th century becoming prohibitively expensive?

The Economics of Indication-Based Drug Pricing

What would indication-based drug pricing accomplish?

Controlling the Cost of Medicaid

Both political parties should support policies that focus on incentives as a mechanism for improving and sustaining their value.

Resetting the Nation’s Health Care Quality Agenda

Poor measures of care quality have consequences. The National Quality Forum is essential to creating measures that are valid and reliable.

Opportunities for Private-Sector Entrepreneurship in Health Care Transformation

Two veterans of public service say that government can do only so much — which creates attractive business opportunities for entrepreneurs.

Moving Past the EHR Interoperability Blame Game

Why can't EHRs talk to one another? We never created the right incentives, but we pretend that we did.

Emerging from EHR Purgatory — Moving from Process to Outcomes

What's the effect of the mode of physician payment when it comes to EHRs?

Reframing Analytics: Transforming Insights into Action

Centralizing clinical data for an integrated delivery system revealed a surprising lesson: sometimes predictive analytics are not enough.

Infographic: The Effects of Health Insurance on Health and Survival

There is strong evidence that expansions in health coverage have increased people’s use of health care across multiple domains of well-being and reduced deaths overall.

Connect

A weekly email newsletter featuring the latest actionable ideas and practical innovations from NEJM Catalyst.

Learn More »

Topics

How to Have a High-Performing Employed…

It’s a generally accepted view that all hospital-employed physician groups are constitutionally incapable of operating…

Value Based Care

127 Articles

Jumping into Value-Based Reimbursement with Two…

What payers would like to see when it comes to engagement in value-based reimbursement and…

Jumping into Value-Based Reimbursement with Two…

What payers would like to see when it comes to engagement in value-based reimbursement and…

Insights Council

Have a voice. Join other health care leaders effecting change, shaping tomorrow.

Apply Now