Atrial fibrillation (AF) is the most common arrhythmia in adults, affecting between three and five million Americans. Looking ahead, experts say the prevalence of AF may double by 2030, and the healthcare system must be ready to adapt to the needs of a much larger, increasingly heterogeneous population of AF patients. A group of electrophysiologists who collectively have decades of experience treating AF joined Editorial Advisor Matthew R. Reynolds, MD, SM, for a roundtable discussion of how they will manage to treat more and more AF patients even as the U.S. healthcare system turns its focus from volume to value.
CVB Moderator: Matthew R. Reynolds, MD, SM, Cardiac Electrophysiologist at Lahey Hospital & Medical Center (Burlington, Mass.) and Associate Professor of Medicine at Tufts University School of Medicine (Boston)
Martin C. Burke, DO, Professor of Medicine at the University of Chicago
Eric N. Prystowsky, MD, Director of the Cardiac Arrhythmia Service at St. Vincent Hospital (Indianapolis), Consulting Professor of Medicine at Duke University Medical Center (Raleigh-Durham, N.C.), and Editor-in-Chief of the Journal of Cardiovascular Electrophysiology
Andrea M. Russo, MD, Professor of Medicine at Cooper Medical School of Rowan University and Director of Cardiac Electrophysiology and Arrhythmia Services and of CCEP Fellowship at Cooper University Hospital (Camden, N.J.)
Edward J. Schloss, MD, Medical Director of Cardiac Electrophysiology at The Christ Hospital (Cincinnati)
Paul D. Varosy, MD, Director of Cardiac Electrophysiology at the Veterans Affairs Eastern Colorado Health Care System and Associate Professor of Medicine at the University of Colorado (Aurora)
“We have embraced this quest”
CVB (Reynolds): Fifteen years ago, when I was training, it was the beginning of the catheter ablation era for AF. Much of the growth in electrophysiology (EP) procedures had to do with implantable defibrillators and CRT. Today, in my practice, a larger and larger proportion of my time is tied up in managing AF patients and doing AF procedures. What has been the impact of AF on EP programs?
Prystowsky: Atrial fibrillation is the unconquered land. It was considered for many years this little orphan in the corner. Just give the patients warfarin and go home. But it’s taken on a real life of its own. Most EP labs are now fueled by AF ablation, to the point, unfortunately, that modern-day EP trainees may not be learning enough about ablation of AV node reentry and WPW [Wolff-Parkinson-White] reentry.
Electrophysiologists are key to the appropriate care of AF patients, and many of us have broadened our programs. There’s been an avalanche of new research in the area. In the end, it will benefit the populations, certainly in Western society, that we have embraced this quest.
Russo: Atrial fibrillation has become an epidemic in our emergency rooms, and our offices are filled with patients needing to be seen for AF. The amount of monitoring, follow-up and education has increased. Shared decision making has increased the time we spend educating patients and their families. Trying to keep people out of the hospital is a big key here, not in the emergency room for every episode of AF. We’ve put a heavier emphasis on outpatient care, when it’s needed, to follow response to therapies. We’re working on a better system and algorithm to manage and increase cost-effectiveness.
CVB (Reynolds): There’s a movement to transition EP training from a one-year requirement to two years of dedicated clinical time, partially reflecting the complexity of mastering procedures. Meanwhile, in the Northeast at least, there are many programs where we train fellows who have a hard time getting jobs locally. Do we have enough electrophysiologists, or should we be training more? Does this affect how we’re deploying allied professionals?
Prystowsky: The two-year EP training requirement will apply 100 percent of the time starting in 2017. There is a group of elecrophysiologists who will be phasing themselves out of the EP lab because my generation and the one right behind me are not as involved in performing very complex ablation
procedures. Having said that, fewer people are applying to EP fellowship programs. It’s a bit perplexing. I think it’s because everybody would like to practice in Boston, San Diego or San Francisco; there is a maldistribution. We clearly need more sophisticated electrophysiologists, but the application pool has decreased.
Burke: A major advance in the overall clinic management of AF has been the new oral anticoagulants that have decreased workflow in coumadin clinics. A big impediment to managing a lot of AF patients and having a program wrapped around it was getting a hospital to invest or give resources for the management of coumadin. [The new anticoagulants] have helped my allied professionals and nurses not be so hampered with the management of INRs.
Training is a double-edged sword, because less people apply, but those people who do apply are very interested in complex things. We’re going to have to figure out a way to attract people. Technology is always improving procedures, but there needs to be a concerted effort from the Heart Rhythm Society (HRS) leadership.
Russo: Medicine is a team sport, and certainly AF treatment is part of that. In our practice, we’re working more with physician assistants, nurse practitioners and nurses to help us. Physicians do education, but allied professionals help us by reiterating the importance of anticoagulation, the choices between anticoagulants and with management of remote monitoring for AF. There’s no way in the world I could do my job without my PA.
Schloss: By creating general local standards and then extending them into the nursing and nurse practitioner workforce, we can be much more efficient and still provide high-quality care. Without that we would be under water.
CVB (Reynolds): Maybe a model for efficient AF care is to have electrophysiologists in charge of the treatment pathways and some of the decision-making, and extenders doing the routine work, such as education, making sure patients have the right rate control, that patients on anticoagulation are on the right drugs, patients on antiarrhythmic drugs are getting routine follow-ups, checking EKGs, and so on.
“What could you actually measure?”
CVB (Reynolds): We’re in a time of significant structural change in healthcare. Whether or not all of the elements of the Affordable Care Act are implemented over the next several years, it’s clear there’s a strong desire to move the U.S. healthcare system, at least in terms of reimbursement, away from long-term dependence on volume, toward quality. With AF, do we even know how to measure quality? Is there any possible way that payers will be able to identify and appropriately evaluate the quality of the care we deliver to AF patients?
Schloss: The critical issue is how to measure. The value equation is quality divided by cost. We don’t really know how to measure either very well, which will put us in a position where, if we do try to measure it, we may end up with numbers that aren’t accurate. The value concept is relatively new. When I was in training, there was “fee for service”—you do more, you get paid more. And there was “capitated”—you do less, you get paid more. I made a standard for myself that I would practice identically in either system. I should be held accountable for that, so that ultimately every patient gets high-quality care, and it’s delivered in a cost-efficient manner. If I’m doing my job right, nothing really changes, except the way I’m being measured. The frustration for me is that “value-based care” is kind of more metric-based care.
In healthcare, we’re a few years behind what has happened in the education system. “No Child Left Behind” is their standard and it created a practice of teaching to the test. Educators tell you about the frustration of trying to meet national standards, sometimes at the expense of their students. We’ve replaced “teaching to the test" with “treating to the metric.” We have been living in this world of high attention to metrics that may not have the nuance required to really deliver care.
This hasn’t really hit AF yet, but the leaders in AF have to take control of how measurement is being done and arrive at some standards that we can all agree on, while not trying to bite off more than we can chew.
Varosy: I could not agree more that we have a problem. Whether we like it or not, we are moving to a system across all of healthcare that is driven by performance measures that will be publicly reported. Based on those measures, we are going to be reimbursed based on the value we can demonstrate. So, it’s not a question of whether we choose to accept that we’re moving into a measurement-based system of healthcare. We are.
Make no mistake, if we choose to abdicate the role of measuring ourselves, someone else will do it for us. The cardiothoracic surgeons woke up and saw this in the late ’80s. When the first CABG outcome reporting was happening with them, they realized quickly that if they didn’t do it, someone was going to do it for them and the results would be far worse than if they took it on themselves. That was the birth of the Society of Thoracic Surgeons database and the robust registry that has been enormously successful for them, precisely because they, as a profession, have chosen to engage in the process.
We need to do something to ensure that, when we are measured, we are measured by the right things. That process is going to take making sure that enough electrophysiologists who are actually doing the work and taking care of patients are engaged in the process. This is part of the reason I’ve chosen to get engaged in the processes that are going on with measurement, including the ACC/AHA Task Force on Performance Measures and the AF ablation and left atrial appendage closure registries—to ensure that we have the right people at the table.
Prystowsky: I’ve been a bit of a contrarian. I’ve been very anti-AF ablation registry. I was approached very early on about it and chose not to give my support. First, AF is such an unsettled field. We’ve had a difficult time just figuring out what to do with a registry on a simple thing like putting in a device. You take one of three types, then you put it in. You can measure a few outcomes. There have been scores of papers written, but what has it really given us? Very little, in my opinion.
With the AF registry … what are the measurements going to be? The initial [data collection form] proposed was over 20 pages long. The most recent one is multiple pages. The EP field can’t even decide what’s the best approach to AF ablation or what’s the best measure of outcomes for AF ablation. We as a field need to come to grips with appropriate training, candidate selection, techniques of ablation and measures of outcomes before a detailed registry is started. We are clearly not there yet. However, prospective collection of some key features of AF ablation may be worthwhile—for example, patient characteristics, ablation techniques and acute complications might provide useful data for the EP field.
I totally agree we have to figure out appropriate outcomes, not just for ablation, but for care of patients with AF. It’s very difficult, because it’s not a monolithic problem. It’s a disease state where there are thousands of papers written every year. I doubt even in your own group everyone does things the same way. Before you start settling on outcomes, you’ve got to figure out what the standard is, and there is really not a standard in ablation at this point.
Russo: We do have some National Quality Forum–endorsed performance measures for anticoagulation in the setting of AF. All of us, I’m sure, have seen consults in people who have CHA2DS2-VASc scores of 4 or 5, or certainly higher than 1, that haven’t been on chronic anticoagulation, where no one’s ever had that discussion with them. Measurements for anticoagulation are more straightforward than for how to do an AF ablation [procedure]. For public reporting, we need to be sure that we’re measuring things that are pretty clearly measurable and we have outcomes data on what we should or shouldn’t do.
Prystowsky: There’s one other area where many of us will agree, and that’s drugs that are contraindicated in certain types of patients. Those are important, and we should get moving on them.
Schloss: I’d like to extend that. To get into the metric in the first place we’re going to need to decide how to define what constitutes AF. I’m not even sure that we actually know that, especially in an era of device-related AF. I see a lot of disparity of opinions and very little data, none of it good, about what to do with short bouts of AF in a device. Is it 30 seconds? Is it 5 minutes? Is it 5-1/2 hours? Until you can actually define what even is the population you’re going to measure, then we’re kind of lost at that point.
Burke: Stroke prevention and management with anticoagulation … [aren’t] done in a very uniform way, even though there are randomized clinical trials that suggest who should be anticoagulated from a risk/outcome ratio. It’s really the only thing you can measure. Atrial fibrillation itself—the management of rate vs. rhythm, ablation vs. medications—is a personal thing between a patient and a doctor. Certain patients shouldn’t go on certain meds, and that’s a metric that could be measured.
Schloss: Absolutely. Atrial fibrillation anticoagulation strategies are well worked out with a fair amount of data and uniformity of opinion, so that’s one we can measure. What do you think about adequacy of rate control? We have some data. Would a metric that evaluates appropriateness of rate control be one worth hanging onto?
Prystowsky: [When we wrote the guidelines], we struggled for weeks and came up with one. As soon as we did, the whole field started hammering us. Remember, we sit around the table. We talk about things. We have different approaches. But, when you set standards, you should be able to have hard-core data to fall back on. We do have that with anticoagulation. We don’t have it with rate control. Although I agree with your idea, I don’t know what I would propose that everyone needs to follow based on randomized clinical trials.
Burke: The very important point is that they made a judgment of what it should be. I thought it was extremely liberal, and they still got hammered. That’s an important point: What could you actually measure and how are our colleagues going to be considered around that?
CVB (Reynolds): If we have difficulty agreeing on a definition of what an episode of AF is, and we have difficulty agreeing on what is the right benchmark for sufficient rate control, and then we have lots of practitioners who may be spending 50 or 60 percent, or even more, of their time doing procedures, and the government and other payers are clamoring for some way to evaluate the quality of what they’re doing and tie that to payment, how do we satisfy that need? How do we move toward being able to provide the payers with something that’s measurable, that we believe is reliable and meaningful, that they could attach payment to? Are there procedural outcomes?
Varosy: The way things are headed, we can’t just say, “We don’t like the measures. We don’t like the idea of measurement. We just think you should pay us, because we’re the doctors. Trust us.” It’s not going to work in the model that is coming down the pike. Do I think we have it all worked out? No. The answer is more engagement. The stakes are high. We really need to make this happen.
What drove the National Cardiovascular Data Registry (NCDR) in this direction is that hospital systems, payers and administrators begged for a registry. They said, “We’re seeing patients coming back for fourth, fifth, sixth, seventh AF ablation procedures in some places.” Clearly that’s outside the norm, but it gets the idea across that we need to explain, and be able to answer, which patients are we taking and why? Are there guideline-concordant reasons for us to do that? I think that’s reasonable for us to do.
The outcomes that are worth measuring at this point are the ones that have been studied in the randomized trials, such as symptoms of AF. Can we demonstrate that patients feel better after an AF ablation? That’s not an unreasonable thing for us to get a sense of.
Russo: We’re being measured on administrative data, so we need to have a [better] way to measure what we’re doing, because we as EPs don’t want to be measured on lipid control, hypertension or even some process measures. We need to be at the table, and we need to create a reasonable way. It’s going to get more complicated, actually, if we go into the whole groupers and the condition-based episodes vs. procedure-based episodes.
You need to have a lot of practice to be able to learn to do these procedures; so, on one hand, we have this high-level training … but we also need well-rounded physicians who know about antiarrhythmic drugs and about options other than ablation. It’s complicated, but, unfortunately, we have to come up with some way to do it.
Varosy: You can bet that, with the existence of the new AF ablation code since 2013, it’s just a matter of time before CMS or AHRQ starts trying to tell us the quality of the care that we are delivering based on administrative data sets. The question for me is whether we want to be measured and reported and have papers presented about the quality of care we’re delivering based on administrative data or whether we want to actually define the data elements and the data that are being collected, analyze the data and have the reporting of those data under the control of professional societies.
Doing this by administrative data is like trying to do air traffic control by looking at jet fuel receipts. So, if the plane fills up in Omaha, then fills up in Newark, that plane flew from Omaha to Newark. That’s essentially what we’re doing. We’re using billing and coding data for a purpose that is not intended, and we need to be mindfully engaged in the process of collecting the right data so that we can actually be in control of what that’s going to be.
Burke: I’m okay with the registry. We’re all talking about the same things: We don’t want to be measured with things that are not accurate. I think the EP community, more than any in our field, has coalesced around randomized clinical trials for device therapy. That has given it a lot of legs and a lot of good data to talk to patients about. What we are lacking is randomized clinical trials in the management of AF for certain populations. That would organize it better. That’s what we need to call for.
Registries might help you guide what you measure and what you actually go through, but it should not lead to any type of concept of what quality is.
I straddle the divide about the role of registries and their success. I’ve been involved in the ICD registry from pretty close to the beginning until today. I’ve been involved in research on complications from ICD implantation and in research on quality of life improvement with AF ablation. I published a paper more than 10 years ago about complications of ICD implantation using Medicare claims data. I don’t believe the data in the NCDR registry with respect to complications from ICD procedures. We know from audits that people under-report complications to the ICD registry, and we know that the definitions are narrow so that they ignore certain things that happen that are still meaningful to patients, and we know that the timeframe is too limited, that if you only count complications that occur before the patient goes home that you’re missing an awful lot. After all these years, ironically, I only believe the papers [where the registry is] linked to claims data.
So, I have concerns about the ability of registries to accurately portray quality. The AF ablation registry will have to be a lot different than the ICD registry in terms of the completeness and the accuracy of the data for it to really give us a true picture of what’s going on.
“We need to do the right thing for the patient”
CVB (Reynolds): There certainly are challenges and opportunities for our specialty to improve the quality of care for AF patients. What is your advice for our colleagues around the country as to what we should be doing to move forward not only caring for AF patients, which we all work hard to do, but also addressing issues of cost and quality?
Burke: Number one, we need to focus any type of quality metric on anticoagulation. Number two, we need industry to invest in randomized clinical trials in selected management for specific populations that we as professionals have decided are the best way to go.
Schloss: Our regulators tend to work with population health and business practices that involve standardization and efficiencies. Most of those folks don’t have a lot of experience sitting in the room looking a patient in the eye. I don’t practice population health with a patient; I practice individual health. No amount of standardization is going to tell us what to do when we’re sitting in the room with a patient and not feel like we’re following a set of checkbox guidelines that restrict our ability to provide nuanced medical care at the individual level. We will need to fight as clinicians to maintain that level of control, because people would rather us be a black box that they can wind up and get predictable results. That’s just not how medicine works.
Prystowsky: First, we have data on anticoagulation, and I think that HRS, in particular, if this really is our wheelhouse, needs to figure out a way on a national level to be sure a follow-up, or at least a discussion, is had. Second, HRS needs to take control of who is qualified to do AF ablation. It’s a disgrace right now what’s going on now around the country. A couple of years ago, about 80 percent of ablations were being done by people who do less than 25 ablations a year. It’s up to HRS to get out in front of this, because we need to protect patients. It is inappropriate to allow patients to undergo a complex procedure by individuals not appropriately trained or maintaining their training.
Russo: We need to do the right thing for the patient. It’s population health we want to consider overall, but it’s a patient. We want to do the right thing and focus on the particular patient we’re caring for. We have to realize that there are gaps in the guidelines we haven’t studied. Although we would love to study more and have more ability to have more randomized trials, and hopefully we still will, there will be gaps in the guidelines. You’re not going to be able to address everything in any national coverage determination nor in every guideline, so you need to be able to treat the individual patient. For performance measures that are used for measuring quality of care, stick to ones that are really based on randomized clinical trials right now, like anticoagulation.
Varosy: We’re in a complex world, and the healthcare environment is increasingly going to demand data as a way of paying us. I wish that we could just say, “I’m the doctor, and you should do what I think is the right thing, working it out with my individual patient.” We need to do both. We need to do what’s right for the patient individually, and we need to be able to demonstrate value based on the changes in the way the healthcare system’s designed. I worry that, if we’re not at the table, we’re going to be on the menu.