Oncology Care Pathways: Perspectives on Strategies and Best Practices

In this roundtable, part of a content partnership between Journal of Clinical Pathways and Flatiron Health, participants discuss how to develop and implement clinical pathways at cancer centers. This includes integrating both the latest clinical evidence and payer contract requirements to develop the most efficient clinical pathways and developing strategies to surface pathways at the point of care, all to allow clinicians to provide the best treatment to patients in a value-based care framework.

Gordon Kuntz, Kuntz Consulting, moderates this conversation between Jeremy Drainville, Moffitt Cancer Center; James Hamrick, MD, Flatiron Health; Larry Shulman, MD, University of Pennsylvania; and Bryan Loy, MD, Humana.

Read the transcript:

Gordon Kuntz: Welcome to this roundtable on oncology care pathways, where we'll be discussing a range of important topics around the implementation of pathways, and hearing perspectives from both the clinical side and the payer side of the health care continuum. I'm your host, Gordon Kuntz, and I'm a consultant with almost 20 years of experience in oncology clinical pathways in the business of oncology.

With us today are Jeremy Drainville from Moffitt Cancer Center, Dr James Hamrick from Flatiron, Dr Larry Shulman from UPenn, and Dr Brian Loy from Humana.

If I could ask each of you to please briefly introduce yourselves, then we'll get started with our discussion. Jeremy, let me start with you.

Jeremy Drainville: Sure. Good afternoon, I'm Jeremy Drainville. I am the director at Moffitt Cancer Center for the Clinical Pathways Program and Initiative. I have been involved with pathways for the last seven years at this point with Moffitt Cancer Center. And previously, my background is on the technology side and management consulting, where I got to really work with integrating different systems, everything from the health care large EMRs [electronic medical records] through to large payment systems and some of the other facets.

Kuntz: Great, thank you. James?

James Hamrick, MD: Thanks, Gordon, great to be here today. I'm Dr James Hamrick, I'm a medical oncologist, and I lead the clinical oncology team at Flatiron Health who focuses on our point-of-care products. So, we spend a lot of time making sure that clinical decision support tools and pathway tools that we understand the needs of the users and that we design solutions that make sense for busy oncologists in their care teams trying to deliver the best quality of care they can.

Kuntz: Excellent, thank you so much. Larry?

Larry Shulman, MD: Yeah, hi Gordon. Thanks for having me today. I'm a practicing medical oncologist at the University of Pennsylvania. I still see patients and I have for now 40 some odd years. But aside from my clinical activities, I had a long-standing interest in technology and how technology would help us to practice through EHRs [electronic health records] and through other facilitators. And I've been very interested in pathways for a long time for a variety of reasons, and I'm helping to lead the pathway efforts at the University of Pennsylvania throughout our health system, both in academic centers and community centers.

Kuntz: Excellent, thank you. And Bryan Loy?

Bryan Loy, MD: Well, good afternoon, everyone. I'm Bryan Loy. I'm a hematopathologist by trade, but I've spent the majority of my career in the payer industry. I've worked for a wide variety of payer types, third-party administrators, large Medicare contractor for traditional fee-for-service Medicare. I currently work for a Medicare Advantage Plan.

I've had a variety of roles in oncology, and at the core of all that, it's been my responsibility to make sure that our members who are fighting cancer get a shot at the very best care that they can get while they're going through that journey. So, the role the clinical pathways plays in our business is to try to optimize that as best we can. So, thanks for letting me be here today.

Kuntz: Excellent, thank you very much. Let's get started on our conversation, and let's start with the clinical perspective. Jeremy, I'm going to go to you first. A couple of questions here. For the oncology care center representatives, what are the key clinical considerations and challenges in developing and implementing oncology care pathways? And second part, how do you ensure that these pathways remain dynamic and adaptable to evolving research and treatment protocols? Jeremy?

Drainville: Well, we certainly don't start with small questions, do we? I think, you know, the big challenge, first of all, is really understanding the scope of work that's required for a clinical pathways program. I think most organizations significantly underestimate the effort that's required and really where you fit into the strategic goals within the organization.

You know, first and foremost, as anyone who is implementing pathways, whether you're doing it internally, like we do at Moffitt Cancer Center, or you're using a vendor as a partner and looking at it from that perspective, you really need to understand first, define what are those clear goals for the pathways? What's your main driver? Is it standardization of care? Is it quality? Is it integration or taking advantage of it from a payment and payer cost savings perspective, cost management? Or is it to support expansion of your facilities?

Figuring out where those priorities rank in terms of some of those, among other priorities that you may have, is really the first foremost piece where I think you need to understand that foundation. Because of the scope and magnitude of really the effort with pathways, I think a lot of organizations underestimate that coming in.

Once you've got that established, the other part of that is it has to be fully integrated into the strategy for your organization. That long-term strategy for all of your initiatives, whether it's your digital initiatives and your digital vision, your data integration or other efforts that you're doing along your data, your payment goals—how do those tie into what you're going to be doing with pathways, because many of those touch a lot of the pieces that you're developing with a pathways initiative.

With that, the leadership commitment, and underestimating leadership commitment. Make sure that you really know that with this, this is a long-term commitment in perpetuity. This isn't a system that you just implement and you forget about. It's something that's going to be with you for the rest of your time as long as you're utilizing pathways. And they need to understand that that endeavor requires strong leadership commitment that says, "Yes, this is important for my organization, and we recognize how that fits into the workflow for our clinicians because of all the pressures they have on their time.”

And then you also have to realize the amount of time that you have from a resource perspective for all the different pieces of work. You know, this isn't just a small team behind the scenes. But really it engages pharmacists to a very large degree when you're dealing with order sets and order set mappings in collaboration, regardless of if it's internal resources working with the mappings or if it's external resources; the physicians themselves; the leadership; clinical systems teams as you're maintaining and dealing with things such as integration of perhaps clinical trials and some of the other factors; as well as a lot of ancillary parts from supportive care to personalized medicine and all the other folks that may tie into something you're doing with a pathways initiative.

So, I think, you know, when I talk about the challenges, it's first really understanding that broad scope that you have to truly understand and embrace to be able to be successful with pathways. You know, and how do you ensure that these pathways remain dynamic and adaptable? It's that commitment. It's that commitment of resources. You really need to develop that capability. And you need that organization support to say that this capability is strategic and it ties into a lot of the other strategic initiatives throughout the org. And that when it comes for that competition of resources internally, as we all know, everyone has as many help health care resources as you need, that really you've got to make sure that you've got the support for that so that you can gain towards those initiatives, whether it's on the reporting side, supporting payer strategies, supporting opening new facilities and ensuring standard of care, and some of those other facets that may be key priorities within your organization.

Kuntz: Great, thanks Jeremy. Larry, let's hear your perspective on this. Key clinical considerations and how do you keep them fresh?

Dr Shulman: Yeah. So, I of course would agree with everything that Jeremy has said. I would say that a lot of those goals are at the upper levels of leadership, right? I think about those things. I think about how care is provided through our health system which I'll come back to in a minute.

But a lot of the work is done by our colleagues in the trenches, the people who go to clinic every day, both at our academic centers and our community centers. At Penn, as I'm sure it is true everywhere, they're working their tails off. They don't have a lot of spare time. And this could be viewed, as they say, as yet another unfunded mandate. We're asking them to do something else on top of everything else that they're doing. And the benefits that Jeremy I think very well outlined, are not always going to be obvious to those people who go into clinic every day. You can tell them, but they may not feel it the same way.

On the other hand, I will say that I've been really pleasantly surprised by the reaction of our faculty to the whole process. And so, what we've done at Penn is that we've formed disease-based teams, a breast team, a pancreas team, a lung team. And as Jeremy said, each of those teams have physicians, they have nurses, they have pharmacists on them, and each one of them has one of us as the leaders of the pathway development group on it, because we understand a lot of the infrastructure and what's going on in the background.

And we've been very careful to include our colleagues at our community sites in these teams. The teams tend to be small—six, eight people. But there's good representation across our health system, which I think has been very important for buy in and for involvement.

And it's a lot of work. When we set up the initial breast cancer pathways, they met two hours every Monday for four months. And that's a big-time commitment for those people. But it was a good experience.

But there's another thing that Jeremy alluded to which is critical, which is that our world is changing all the time, you know, there's new data coming out and new drugs being approved. This is not a one and done type of process. This is a process that you do. And what you do today is good for today and tomorrow there may need to be modifications, and you need to have systems and commitment to make those modifications.

I will say that at Penn, and I'm sure this is the case elsewhere, we are very anxious to assure that there's the same quality of care on the same treatments being offered across our system. We have 6 hospitals, we have 11 ambulatory sites, and that's a lot of people in a lot of different places. And this allows us to, one, give them direction on what we think are the ideal ways to practice, but also for us to be able to monitor what they're doing.

And then the last thing I would say is that whatever process or vendor product that you're using, it has to be well integrated into your EHR. We all live in our EHR, I go to clinic, I live in my EHR, all right, not even sure I had another home anywhere, and anything that we're doing really needs to be in that EHR to help guide us. And we've been very careful about that as well.

Kuntz: Great, thank you very much.

Dr Hamrick: Larry, that resonates a ton with things that we've seen as we've talked to different health systems and individual practices. You know, on the one level, Jeremy, you alluded to, you got to have the executive level buy-in because it's very resource intensive, it's a multi-year effort. People may come and go; you have to have that culture built.

On the second hand, as you said, Larry, you've really got to win the hearts and minds of the front line as well. And you do that by respecting their time and how busy they are. So, build it into the EMR, think about it, build a tool that comes from the physician's workflow and not a tool that dictates the physician's workflow.

The other piece is, I really like what you said, Larry, about having a system. So, I practiced as a very much general oncologist until very recently, led a little department. What we've heard as we've spent a lot of time at the elbow of oncologists is it's really hard to keep up these days. And so especially as a generalist oncologist where I can't go to every meeting and read every journal article. Everywhere, people are trying to scale the knowledge of single disease experts, whether they're within their institution system or other ones, you know, key opinion leaders, scale those out so that as was just mentioned, we can have a system designed to provide high quality care to everyone.

We've invested heavily in a clinical team with pharmacists and doctors and mid-levels who try to provide updates about, “Hey, here's the newest stuff that's coming out from NCCN,” for example, to inform, to keep doctors apprised of best evidence. So, we designed our tool, and this is just sort of our approach to the market, as one that would provide the service of helping them keep up and scale not knowledge and best practices from the single disease experts to the people seeing all kinds of different cancers coming through their office.

Kuntz: Great, thank you. And Bryan, I know your perspective is a little bit different, but I think it's a similar kind of challenge. You know, as you're thinking about pathways, what are the clinical considerations and how do you make sure that the pathways that you're using, or the doctors who are using them, how do those reflect the most current clinical issues?

Dr Loy: Well, I think at the core there's a couple things that I might preface with. I'm going to borrow from Jeremy. I heard the word standardization, and Larry said it's a process, and if you're standardizing that means somebody has to change, as you’ve got a process. And if you're bringing doctors together to try to weigh in and you can get them to agree, you probably ought to be in a higher leadership role than you are today.

With that in mind, you know us as a payer, we're outside the system, so we're not viewed in the same way that you all would be as providers trying to collaborate within a practice to figure out what the standardization needs to look like in order for us to reach consensus. We find ourselves in a place where we have to defend ourselves, our positions, from time to time when we create what we think are legitimate pathways.

So, getting grounded to what would you want to be treated like if you were a patient? Getting grounded to what is the evidence point us to in terms of a preference of effectiveness, the clinical trial. And, secondly, what does the toxicity look like, whether it be a Part B or an oral drug, what is the patient going to have to live with long term? Is there a convenience factor to be considered? And then, and only then, if there's a tie up—it goes through those filters in that order, by the way—you know, what is the least costly alternative look like? We find ourselves to be on fairly solid ground. And, in some instances, we find ourselves covering or seeing in our pathways a more preferred agent because of the things that preceded even the cost analysis.

But in the end, it's our belief that if we can get the right drug to the right patient at the first time—and manage to that—we can do whatever we need to, to make sure that the economics get to where everyone can provide that care.

The other guiding principle for us is setting expectations, because I think more and more I hear from not only medical oncologists but specialists as well as primary care docs, “Don't tell me how to practice medicine.” But the process that Larry was describing to me brings out in the full view the evidence underneath alternatives, the evolving evidence base, et cetera. And those are time-consuming and mentally consuming tasks—and, as he said, they are not funded very well. I think he used the word “unfunded mandates,” and that resonates quite highly with me because there are a number of triggering events, whether it be an FDA label indication for the change or NCCN changes. We receive many of our marching orders from CMS. So, it doesn't really matter what we think; in some instances, coverage is dictated by whatever the NCCN panelists begin to think and many of the situations that we're asked to opine on or provide coverage for.

Lastly, I would say it is no easy task to maintain that. We get the privilege of monitoring all those panels, those changes, those triggering events that I just tried to describe, and try to stay in lockstep with that evolving evidence base or those key opinion leaders reaching consensus in order for us to be compliant with the coverage that is expected of us as a payer.

I say all of that and I'd say we kind of brought ourselves to solving for this as a payer for over the last decade or so. And I think this standardization and what we could say would be, you know, this makes sense in terms of trying to reduce the irrational variation in practice pattern while still allowing for exceptions when there are exceptions needed, whether it be a patient can’t tolerate a drug, for example, for whatever reason, or there's a drug shortage, for example.

So, not setting things up to where someone would expect 100% compliance, but if there is some accountability, and that folks that have gone through the exercise would at least have an opinion, an informed opinion, or an appreciation for why, you know, this payer’s pathways look like they do. And opening ourselves up in the event that someone wants to challenge that or bring new information to us, that willingness to be able to listen and carefully evaluate what's being said.

Admittedly, the biggest gap that I see is that we've reached this environment of accelerated approvals to the extent that there are many instances where I don't think we really know, nor do we have the track record with real-world practice or evidence, on how many of these drugs are going to perform in the populations that we serve. Many of the exclusionary criteria for the clinical trials to bring drugs to market are younger populations with higher performance status, they don't have the comorbidities or the extent of disease or the disease burden that we see in the populations that we serve, and yet the labels compel us to cover for that.

And so, we're taking a deeper dive in terms of what is this likely to look like in the populations that we serve relative to the alternatives—if there are none, the conversation can stop there. But also, examining the evidence very carefully to see, you know, is this the way we would expect medicine in the control arm, for example, if there is one, to be practiced in the US, and is it really legitimate evidence in order for us to feel like, “Hey, this is something that we can really hang our hat on” in terms of being confident that this is what would be good quality care for our membership.

So, I point all that out to say, I think we live in an era where we're financing a certain amount of uncertainty with confirmatory trials, sometimes elusive, and sometimes delayed in coming forward. And I think that those are some of the challenges that we're going to look at going forward.

I think the other thing that we think about as a payer also is it's not just about the care path around the drug. We look at some of the biomarker testing and the quality of that biomarker testing is helping with the drug is really what we're trying to aim for. And those bring in a whole new set of issues in terms of just exactly what were the mutations and how were things evaluated in order to bring the drug to market to begin with. So, looking at those elements and trying to carefully evaluate the science that brought those particular agents to market are areas of interest that we have.

Kuntz: Great.

Dr Hamrick: You know, Bryan, you bring sort of one of our guiding principles has been that when you talk about standardization, there's never going to be one path to rule them all. Oncologists are very good at their jobs and they're highly trained. We designed Flatiron Assist, which is our tool embedded in the EHR, to allow local customization because of sort of exactly what you're talking about.

You're not going to come up with a one-size-fits-all national standard for every disease; Those of us that have spent time in the front lines know that it's way too complex. There are lots of reasons that at a site might want to do it their own way, and they wouldn't be wrong. Larry, I'd love for you to talk about—at risk of sounding like the moderator—I'd love to hear how you think about this sort of localization and being able to say, “Yes, we're standardized, but we also have some nuances.”

Dr Shulman: Yeah. So, you know, I think that's really a good point. And, you know, NCCN, which, you know, Penn is an NCCN Center; we believe, very, very, very strongly in NCCN's mission. We have representatives on every one of the NCCN guideline panels, which actually helps in our work.

But the NCCN guidelines, when you look at them, often give you three or four alternatives to the same clinical situation. And for many of our practicing oncologists to sort of, each time they sit with the patient that one of those nodes of their care, to go over those four alternatives and say, you know, “How am I going to weigh these four and which am I going to choose?” just sort of on the spot, if you will, while you're sitting in the exam room with the patient is probably not the best way to do it.

And so, in these teams that I described previously that are meeting on a very regular basis, they've sorted through those NCCN options and said, “At Penn, we think that either this option or these two options are what we prefer and these are the reasons, and we have good reasons for them.” And then in Flatiron Assist, which is what we use, they come up as what we call Penn Preferred pathways.

So they are, in fact, letting the clinician in our community site or anywhere else in the system know that this team who has developed the pathways for breast cancer or pancreatic cancer, whatever it is, have among themselves chosen these as what we would like to do at Penn, this is what we consider to be the best thing to do. That does not prevent that clinician from picking another NCCN concordant guideline. And I think that's really critical to understand.

But I can tell you that when we've started to look at our data, and one of the most important parts of using a product like Flatiron Assist is it allows you to look at what people are doing, in about 95% of the cases we're using Penn Preferred pathways, and over 95% within the NCCN guidelines. And in our work with the payers, that's something we can bring to the table and show them and start to try to work out arrangements that are better than I think what we have currently, which is a prior authorization system, which makes many of us want to slit our wrists.

But I think that this whole concept of letting your physicians wrestle through this, it's intellectually good for them. It makes them think through everything; it gets buy-in from our whole system because we have representatives from everywhere. And then when they go to see that patient who has a particular stage of their disease, they know this is the way we thought it through; this is what we're going to do.

Kuntz: So, Larry, on that topic, we'll let you continue here. What are the common hurdles you face in daily practice? And how can cancer centers support providers in overcoming these challenges as they relate to adopting and adhering to clinical pathways? And then any specific strategies about collaboration, kind of as you were just mentioning, between clinicians and pathway developers?

Dr Shulman: Yeah, so I think those are really important questions. And, you know, I may be the oldest person on this call, and I spent much of my career in the days before electronic health records with paper charts, and oncology was certainly simpler back then. We had fewer drugs and fewer options.

Electronic health records have been a big boon in many respects, but they're complicated. They're not always that easy to use and behind them sits all the possible treatment options that we have at that particular center for all the diseases that our patients have.

And the pathway tools help to, you know, turn a very confusing forest into a much more focused view of being able to say, "Okay, this is my patient. Her tumor has these biomarkers, this stage of disease,” and bring us right to where we need to be: to think about what the best treatments are.

And I will say, and this goes back to the first questions, change is hard. Some of my colleagues have said, all change is to be resisted. People like to do it the way that they've done it forever and convincing them to do it a different way is always a challenge. And that's something I've dealt with as a clinical leader for decades. But once they actually start to use the tools, they realize that it's, in fact, making their lives much easier and bringing them to the right place much faster.

And I think this is important for our academic faculty at our cancer center, but it's even more important obviously for our community oncologist who often are generalists and seeing patients with multiple diseases and struggling to keep up with what would be the most beneficial, most recommended therapy in a particular situation.

So, I think that medicine overall has used technology poorly compared to other industries, but I think this is really using technology in a much better way in what's an ever-increasingly complicated field.

Kuntz: Thank you. James, are there a couple of things you might want to mention as well about, you know, what you're seeing as oncologists who started to use this, any challenges that you're hearing about in sort of their daily practice, and then how do you help foster that collaboration that's so important?

Dr Hamrick: Yeah. So, first of all, making a tool that is built, as I mentioned earlier, to reflect the physician workflow and not the other way around is very important because they're super busy and it has to add value. And most of the time, you know, the doctors on the front lines actually are really good at their jobs and know what they should do. But it really matters the, you know, 20-ish percent of the time that they might really lean on decision support, that it be there.

And as Larry said, you know, you don't want to be doing mental gymnastics around comparing three checkpoint inhibitors in first-line non–small cell lung cancer in your head or just using the one that you used in your fellowship. It's great if you know that the people that really think about this and have the time to look at the data and look at the at the site-specific factors that go into decisions, can signal to you, right in your workflow, that, “Hey, this is the one of those three equally effective, per NCCN, checkpoint inhibitors that we're going to align around here.”

And I think, as Larry said, the other thing is doctors want to know how they're doing. So, the reporting piece is nice. You know, I led a division for a number of years and it was interesting; you don't really know how your colleagues practice. I mean, you might curbside them, hopefully you're going to tumor board, but it's nice to actually have some insight into how are we managing certainly these common scenarios that come in.

And so, I think having the data insights, as Larry said, we're able to track in our tool—sites can track and service line leads can track concordance with NCCN, concordance with local preferences. We're seeing both in the mid to high 90s, which is great. And then we track adoption. So, when we partner with a center, we track adoption to make sure if doctors aren't using it, it's because the tool is not designed right and we're not demonstrating the value correctly.

So, I think it's like everything. You got to track metrics. You got to continually ask yourself, “How can we improve? What's working and what isn't?” And you've got to be willing to iterate.

Kuntz: Great. Jeremy, is there an item or two that you'd like to discuss briefly about any specific strategies for fostering collaboration between clinicians and you as a pathway developer within that organization?

Drainville: Yeah, I think, you know, just to touch on a couple pieces that James and Larry were talking about before too. I think that, you know, the big important takeaway is that you're making recommendations. You're simplifying the workflow by putting face-up recommendations, not restrictions. Right?

That's a key part of this. It's that here's what recommended for the 80% of the cases without the variance, the patient can tolerate it—here are the best practice recommendations for that. Providing that so that it adds or enhances the workflow for the physician or clinician rather than taking away or creating restrictions where it makes it harder for them to get to the pieces they need to do in a simplified, efficient way.

The other part is also about that dichotomy of workflows. If you have a tool that's creating a dichotomy of workflows that, oh, we treat this disease this way, but I'm sorry, I don't have coverage in this other disease, you need to go to your other screens or interface to navigate to be able to interact with the ordering process, the documentation process, then you're creating more challenges working with the clinicians. And really partnering with the clinicians so that you've got an interface that is streamlined, that doesn't interrupt their current workflow, replacing their primary interface in a lot of cases or creating duplicate work. First of all, it is appreciated.

I think the second piece where James was touching on about the data is extremely important. I think, you know, just to highlight one thing that really from the first question about that too though, is going through pathways is going to force organizations to really understand more about their data quality within their EMRs as well. And do not underestimate the challenges that may appear with the discrete data quality relative to staging, relative to this visit reasons, relative to the granularity of data that you may not have below ICD-10 [International Classification of Diseases-10] code when you're trying to differentiate between anaplastic thyroid, papillary thyroid, medullary thyroid, that's all under one ICD-10 code. And trying to get to the right recommendations face up, there's a lot of challenges of that. So, if you're able to get down to that level of granularity with your data, provide more value with some of the pieces you're able to generate from smart, intuitive pathways solutions, you're starting to provide a lot of power to the clinicians.

And then you can get into all different kinds of solutions, whether it's around pathways, helping them to understand, as I think my peers or colleagues talked about here, as case mix or is it variance? Is it getting into time to treatment? Going into the outcome side? There's a lot of capabilities once you've actually got a well-implemented tool and solution.

Kuntz: Great. So, let's switch gears just a little bit. And I want to start with Bryan to talk about the payer perspective. How do payer representatives, like yourself, view the balance between cost containment and delivering value-based care within oncology care pathways? And what strategies can cancer centers employ to demonstrate the value of their pathways in a way that aligns with your expectations and reimbursement models. Bryan?

Dr Loy: Yeah, I would tell you in large part, obviously I can't speak for all payers, but the roles that I've been assigned as it relates to development of any of these types of initiatives is to put patient interests first and to evaluate the evidence as best we can. And as I was listening to the others comment on previous questions, the processes that we look at, the standardization, resist all change, all of those point to, what is it that we can do to create that standardization and also use the same percentage of those that use 80%.

If there is a way, though, that we can all walk away from this part of this segment and time and say we're evaluating evidence more thoughtfully, we’re giving critical thought to how it applies not only at the population level but at the individual patient level, I would see that as progress compared to conversations that I had nearly a decade ago where I saw practice leaders arguing amongst each other at the leadership level around breast cancer regimens, and it was largely focused around what was said earlier. And that is, “This is the way that I was trained. This is what I'm used to being used.”

So, in order to move ourselves away from that, there's got to be a recognition that we've got to build trust, which is what I'm hearing in this conversation. And I don't think you get but one attempt at that to build that trust. And it's easy to just completely destroy the trust. And I think it's a recognition that there have been many minds carefully thinking through what the evidence is telling us. But I also think there's a recognition now where we're starting to see that we don't know everything.

And in order to trust those care paths and the minds that came together to build consensus around what a care path looks like, it has to be a recognition of what we don't know. So, for example, in the world of immuno-oncology, what happens when you get out past two years? We've really never had to answer a lot of the questions in terms of, is the patient really getting benefit? Are they getting benefit at this level? Are they getting benefit at this dosage, as well as this schedule? And when you start to get into that, well, what do you do with those uncertainties? I think those are some of the questions that are outstanding right now that I don't think we have had good answers to.

So, I think knowing what we know and being able to communicate that credibly in order to try to get the physician behaviors and standardizations everyone's talking about that has to be counterbalanced with what we don't know and a recognition that perhaps we have to look at things with a skeptic’s eyes until we get more information, which quite frankly may never come.

Kuntz: Great, thank you. Jeremy, I know Moffitt's very involved in value-based care and managed care contracts. What do you do? What strategies do you use to demonstrate the value of your pathways to align with payer expectations and their reimbursement models?

Drainville: Overall, I think the first thing is we've got a direct relationship with the payer strategies team on our organization. They're ultimately driving the search for alternative payment contracts, shared savings agreements, or whatever it might be. Our pathways, you know, through measurement, we know, as Larry said, about adherence. And I think with, you know, the academic centers, you find very high alignment to best practice standard care, you know, that 90-95% concordance in those cases, I think, is something that you see throughout most of those organizations as well as Moffitt.

Supporting that from the payer strategy side is really first, showing that you can prove that through the measurement in metrics is really that entry point for a discussion where I think a lot of the payers are interested in engaging with your organization because they know you follow a standard. You know, they may not be using your pathways actually for the measurement, for the shared savings goals—they might be running against their own internal engines in some cases—but the fact that you have those and you've got measurable results to show that is really the first starting point of the conversation. The relevance of that also comes into things such as bundled payments and some other agreements that you may end up working through with the payer strategy side for the benefit of both organizations in a win-win.

Kuntz: Great, thank you. Larry, anything different at UPenn?

Dr Shulman: Well, you know, I think we're actually blessed with pretty good relationships with our payers. And I think we have a lot of common goals. You know, we've grown up with this mentality that it's, you know, we're on different sides of the fence and we're fighting. But actually, I think that we both want the best treatments for our patients. We think high-quality care is actually the most efficient, cost-wise, care that there is. Bad care leads to bad financial outcomes, no less bad outcomes for the patient.

So, we also work with our contracting team very, very closely. We've sat in rooms with our payers now as we get away from COVID actually in person, where we can sit there and talk about what our common goals are. The prior authorization process is expensive and, frankly, frustrating and torturous for both sides. It's frustrating on the provider side, it delays care. We have a whole team that does it that I have to pay for. But on the payer side, they have a whole team that they have to pay for too, you know, who are on the other end of the phone call.

And Bryan, not to throw stones, but sometimes the people on the payer side really don't have the sophistication in the cancer area that they might, and they know that. So, that if we can come up with an agreement about how we're going to do things, how we're going to measure it, how we're going to share those data in a very transparent way. I think we're going to get to a much better place. And we've started to be able to do that. And I think that's really, in many respects, the future.

I did want to make one other comment quickly. I mean, you know, as I mentioned, in what we've measured at Penn, we have very high concordance rates. But a lot of the work that I do nationally, which I do a fair amount of, shows that throughout the US, there's a lot of deviance, if you will, from guideline-concordant care. It's not provided at high levels everywhere, and I think the worst part of that is that there are inferior patient outcomes. But these are ways for us to help our colleagues to, in fact, do the right thing. And I think they want to do the right thing. It's just getting to be harder to do that.

Kuntz: Great, thank you. James—

Dr Loy: Gordon, would you mind if I responded to Larry?

Kuntz: Absolutely, please do.

Dr Loy: So, I think, you know, I understand the delays in care. I understand the mismatch. That's been a question that's been raised over the years. What we've tried to do is to make sure that the appropriate specialist on the front end, not through an appeals process, a board-certified oncologist, medical oncologist, is on the phone to be able to hash out anything that needs to be discussed with someone, especially if there's disagreement. So, I think you're specifically talking about human processes, or prior authorization, that present delays in care.

So, what we attempted to do is to try to find out what's agreeable, right, and coverable and get out of the way. So, make that care easy to get through the process with some assurance that on the other end that you're going to get paid for it. And there have been many conversations around, can we somehow use pathways, transparency, and if we all agree on what is good quality care, get out of the way of the prior authorization process? And the answer has been, yeah, I think we can if we're willing to share, I'll call it bi-directional visibility to each other's systems. And I don't think we're there yet. I still think that's possible, and I'm hopeful of that, but I would also be quick to tell you, too, that when folks say, “I need some real assurance that I'm going to get paid, and that someone's not going to come back and ask for recoupment for whatever reason,” there's a certain amount of comfort knowing that there's an authorization number that's been issued.

So, I would just tell you that, at least from where I sit, there's been much done in the near recent times to be able to try to find out what good quality care looks like and minimize the administrative burden associated with it. So anyway, back to you, Gordon, thank you.

Kuntz: Thank you. James, I want to give you the last word on this question. So, from your perspective at Flatiron, how are you helping cancer centers to demonstrate the value of the pathways, that you helped put together, to help align them with their payers?

Dr Hamrick: Yeah, so a couple of things. First of all, we don't make pathways at Flatiron. We leave that to NCCN and to the local experts. We have built a tool that enables the latest evidence to be surfaced for the doctor, and that's an important distinction because that's not our strong suit.

Now, what our strong suit is, is building technology that enables things like was just touched on by Bryan and Larry. How can we exchange data and use data in real time to understand, you know, how are people practicing? So, on the front end with a doctor at a health system, understanding what are my concordance rates? Am I using, starting with the basics, am I adopting and using the tool? What are the concordance rates? We think adoption is a measure of value. Concordance with expert guidelines and local preferences is a measure that all the effort folks like Jeremy, Larry, and their teams are putting into building this custom local content is worthwhile and is paying off. And then you can use that to get people to reduce variation in care.

I think when, you know, we envision a world, and I think this is what Bryan just alluded to, where sites that repetitively are able to demonstrate high-quality care and concordance with NCCN and other markers, whether those be payer pathways or local pathways, but some reasonable standard of good, scientific-based practice, that streamlining of prior auth is absolutely possible. And that will be done, as you said, through bi-directional exchange of data that's efficient, that eventually gets clunky, you've seen one, you've seen one, tech intermediaries, and fax machines, and humans on the phone, you know, minimizes that and enables us to spend less time, you know, litigating the finer points of an individual clinical decision and more time allowing doctors to take care of patients.

Because the last thing I'll say here is, you know, we don't have a system to handle the capacity of all the cancer patients in the United States if we can't get every doctor to have access to the tools to take advantage of the best science. Not everyone can go to ASCO [American Society of Clinical Oncology] every year; not everyone can read every journal article because they're taking care of 25, 30 cancer patients a day or more. And we've got to have tools available to give them the data they need to make the best decision.

Kuntz: Excellent, thank you. And I appreciate the fact that you already answered the next question I'm going to get into. So, thank you very much. Jeremy, I'm going to go to you, and we just have a few minutes left. But if you could talk very briefly about the metrics that you all look at from your standpoint, as James was talking about, there's utilization of the tool, there's compliance with the tool, or concordance, what do you all look at, at Moffitt?

Drainville: I think there are several different facets to that. You know, certainly we're looking at, you know, utilization. How is the tool being utilized, first of all, from a tool utilization? How is the data being utilized for decision support? Where is the variance and what is the adoption of the tool across the platform?

And then we go into a few different areas outside of that as well, because the one thing that pathways provides that you don't have discretely in your systems otherwise is reason for treatment change. You know, that's essentially where you're documenting an inflection point within care to understand why are you changing treatment or why are you providing treatment at that point. That data is not discrete in the system elsewhere. And that provides a lot of opportunities where you can extend that data into also looking at, eventually, going into the outcome. You can also look at time to treatment, and inflection points between different times to treatment. There are opportunities for melding that with some of your case mix and the volumes and some of those other components. So, we actually, we do look at measurement really for the usage adoption utilization variants and all of those just as I think everyone does.

But then we look at ways to try to take advantage of that data further downstream for things that ultimately benefit the physicians and practices directly, and that's something that engages them more as we can provide that data back about their practice and some of the patterns.

Kuntz: Great. And Larry, I'm going to give you last word on this topic and for our panel. If you could just chat maybe very briefly about the feedback loop that you see that allows the insights and the metrics, that I know you've already talked about, to continually improve and refine the pathways.

Dr Shulman: Yeah, so, you know, I think for much of my career, we've practiced in a data vacuum. You know, we don't have a lot of information about how we practice. Jeremy's talked a little bit about that just now.

And the other sort of twist I wanted to put on this was what Bryan said earlier, which is, you know, how are these treatments being used and working in the real world? Clinical trials are very restrictive. The patients on clinical trials don't look like all the patients we're seeing in clinic every day. And so, I think measuring what we're doing in the real-world patients, if you will, who we need to make decisions on and we need to provide treatment for to better understand that is really critical. And, you know, I think our oncologists want to do the right things and they want to know they're doing the right things, but we've never had a way to tell them that. And now I think we have a much better way of all of us being introspective and understanding how we're thinking things through, how we're doing it, and, in fact, what the outcomes are.

So, you know, when we use our tool—and we do all things Jeremy said, you know, we do utilization, we do concordance with UPenn preferred and NCCN pathways, and that's pretty straightforward. But the pathway tools allow you to cohort patients that are very specific in either their biomarkers or their stage of disease, and to look at those patients and look at what we're doing and how the patients are doing.

And I think this is an incredible opportunity that we should have had 15 years ago and we haven't. But this is the opportunity to have that now and really to elevate the level of care that we provide in the outcomes for our patients.

Kuntz: Excellent. Thank you all so much for your time today and your perspectives. This has been a really robust discussion, and I'm glad we had it. Jeremy, Larry, James, and Bryan again, thank you very much. I also want to thank the Journal of Clinical Pathways for pulling this group together. And with that, I will sign off. Again, I'm Gordon Kuntz. Thank you very much.