Advancing Atrial Fibrillation Care Through Clinical Performance and Quality Measures

© 2026 HMP Global. All Rights Reserved.

Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of EP Lab Digest or HMP Global, their employees, and affiliates. 

Interview by Jodie Elrod

In this interview, Emelia J. Benjamin, MD, ScM, discusses the new 2026 American Heart Association (AHA)/American College of Cardiology (ACC) Clinical Performance and Quality Measures for atrial fibrillation (AF) and how they translate evidence into measurable standards for clinical practice.

Could you start with a brief introduction about yourself?
My name is Emelia Benjamin, and I had the great honor of chairing the ACC/AHA performance measures document set. It was a process that took place over a year and really tried to help us think about how we implement AF guidelines into clinical practice. 

In terms of my background, I’m a clinical cardiologist at Boston Medical Center, which is an urban safety-net hospital. I’m also vice chair for faculty affairs in the Department of Medicine and interim associate dean for faculty development at the Chobanian & Avedisian School of Medicine at Boston University and Boston Medical Center.

What are the most important changes in the 2026 AF performance and quality measures, and what should clinicians prioritize when incorporating them into practice?
Well, I confess that I do have a bit of a bias as an epidemiologist as well as a cardiologist, but we did come up with a section in the performance measures—sort of the top 10 measures that we thought were really important. One of the first was really embracing and providing comprehensive care and secondary prevention to prevent the progression and complications of AF at all stages.

When you think about the stages of AF, there is the period before someone is diagnosed; pre-AF, which includes people who have evidence of remodeling; intermittent or paroxysmal AF, and persistent but not permanent AF; and then permanent AF. So, stages 1 through 4. At every single stage, secondary prevention should be implemented. That includes weight loss, physical activity, smoking cessation, alcohol moderation or cessation, and optimal blood pressure control, if indicated.

When you think about it, one of the reasons prevention was elevated to a performance measure—and this is a new measure—is that secondary prevention is standard of care in coronary heart disease. I mean, you would be really out of step with modern practice if you didn’t talk to people about smoking cessation or optimal blood pressure control after they had experienced a coronary heart disease event. So, why is it that we haven’t reached the same point with AF?

This has been documented in multiple studies. If you look at the international GARFIELD-AF Registry, something like 85% of patients with AF have comorbidities, and among those patients, less than half received all eligible indicated treatments. Less than half! I mean, it’s really shocking. That wasn’t just a performance measure; it also had implications for outcomes because people who received all guideline-directed medical therapies had an 11% reduction in all-cause mortality at 2 years. So, that’s big.

This was one of the most important aspects and why we elevated it to a performance measure and made it a new measure. There was a little bit in the prior guidelines about weight loss, but many of these other interventions have now been studied either in randomized controlled trials or, when randomization would be unethical, through robust observational data. You can’t randomize people to continue smoking or not smoke. That would be unethical. There’s no equipoise on that; the evidence is very strong.

The other measures that we thought were really important involve ensuring that health systems measure the receipt of guideline-directed oral anticoagulation, rate-control strategies, and lifestyle and risk factor management. We want them to really examine this in important demographic groups because we have marked inequities across all aspects of the AF care continuum—whether it’s diagnosis, management, ablation, anticoagulation, you name it. There are real inequities, and whenever we have inequities, we have poor outcomes.

Another quality measure that we thought was very important was the need for patient-centered discussions and shared decision-making about the benefits and risks of rhythm-control versus rate-control strategies. That was added as a quality measure.

The final one, which we’ll discuss a little later, is the importance of rhythm control in people with heart failure with reduced ejection fraction (HFrEF). That was a new quality measure that we thought was very important.

The performance measures may be used for public reporting or pay-for-performance, whereas the quality measures are intended primarily for internal improvement. How should clinicians and health systems approach implementing each set?
So, from my perspective, the performance measures are the floor. We all need to be doing this, and health systems need to look very carefully for documentation that all clinicians and health systems are implementing the measures—or documenting why they are not.

One of the things about performance measures is that not everybody should receive a direct oral anticoagulant (DOAC), and there are specific reasons why. For instance, if you have mitral stenosis, you should not receive a DOAC. In the performance measures and quality measures, we provide guidance on when there are appropriate exceptions that should be taken into account, but those exceptions should be documented.

We also think it will be very important to consider comprehensive secondary prevention, as I already discussed, at all stages of AF. This includes documenting each patient’s validated stroke and systemic embolism risk score annually because it changes over time.

You can be 64 today without hypertension and turn 65 tomorrow—or a month from now—and start developing hypertension. So, even if you’ve determined that somebody is at low risk of stroke or systemic embolism, it is important to continually update that assessment. Their risk can change over time, partly because of age and partly because of the accumulation of other risk factors. We also discussed using DOACs when indicated.

Performance measures are measures for which we felt the reporting burden was not overly burdensome and that are so important to the quality of care that they really merit being used for pay for performance. Quality measures are not really ready for public reporting or pay for performance, but they can be very helpful to health systems as they consider how to deliver the best-quality care to all of their patients.

The document adds a quality measure emphasizing rhythm control in patients with AF and HFrEF. What evidence and gaps in current care prompted this measure, and how might it influence treatment decisions in clinical practice?
We know that AF and HF have a bidirectional relationship. AF can cause HF, and HF can cause AF. We also know that the morbidity and mortality associated with both conditions increase when you develop the second condition.

When you have AF and HFrEF, sometimes you don’t know whether the reduced EF is secondary to the AF itself. Even if you can control somebody’s heart rate, sometimes addressing the AF rhythm—including the lack of AV synchrony, etc.—can actually improve the EF.

There is fairly robust evidence that an aggressive approach to rhythm control can reduce the burden of AF, improve ventricular remodeling, and halt occult arrhythmia-induced cardiomyopathies. We know that AF is an important cause of arrhythmia-induced cardiomyopathy. For that reason, we felt it was very important to introduce this as a quality measure.

What practical challenges may clinicians and health systems face when implementing these measures, particularly in collecting data, integrating them into existing workflows, and identifying inequities in AF care?
Yes, that’s a very important question. Something we grappled with a lot as a committee was that there were some things we wanted to add, but then the quality improvement experts said, “That’s not practical. It’s not practical to implement this at this point as a performance measure because of documentation burdens, etc.”

For instance, I would have loved to see health equity become a performance measure rather than a quality measure. However, we didn’t make it a performance measure because health systems are still grappling with how to document all the potential sources of or contributors to inequities, such as race, ethnicity, and socioeconomic status. It’s easier to document whether somebody lives in a rural community, but factors such as health literacy, sex, and gender are more challenging. There are many different reasons or contributors to people not receiving equitable care, and the relevant information is sometimes not documented in the medical record. So, that’s one issue.

We also know that clinicians are busy, and we hope they are having shared decision-making discussions. We know that shared decision-making takes time. If you have 20 minutes with a patient and need to discuss their anticoagulation, risk factors, and blood pressure, it can become quite difficult.

We acknowledge that metrics more easily captured in the health record, such as whether somebody received an echocardiogram, are easier to document. You either have an echocardiogram or you don’t; there’s a code, or you can see the test in the medical record. Discussions about shared decision-making are more complex, as are the reasons or exceptions for why people are not following the guidelines.

I think people sometimes haven’t focused enough on documentation. Frankly, I think people have gotten the message that it’s important because of litigation. If somebody isn’t receiving an oral anticoagulant, why aren’t they? We need to document that because, otherwise, there is a significant liability risk. With some of the other factors, there’s less of an imperative in terms of liability, and busy clinicians may not get to them.

Again, we tried to weigh those potential barriers or challenges when determining what should be a performance measure used for pay for performance and what should be a quality measure.

Looking ahead, how might emerging evidence on wearables, pulse field ablation, stroke prevention, and individualized AF management shape future measures?
Yes, one of the things that’s really important about guidelines is that they’re not static. The evidence base changes over time. In creating the performance measures and quality measures, we focused on Class 1 or Class 3 recommendations. Class 1 means you need to do it. Class 3 means it is either neutral, offers no benefit, or has the potential for harm. So, we really tried to focus on the highest classes of recommendation.

For wearables, there aren’t robust data about their use. As you’re probably familiar, there has been an enormous amount of controversy. We know there is a lot of unrecognized AF. We know that unrecognized AF is associated with a higher risk of stroke and HF. Sometimes, people present with AF and HF at the same time, but the AF didn’t start that day. Sometimes, people present with a stroke and AF, and typically, the AF didn’t start on the day their stroke occurred. The idea is that if we can get upstream of that, we can potentially prevent downstream complications.

Most of the screening literature has not provided robust evidence that screening actually improves outcomes. There is robust evidence that screening detects a lot more AF, but there isn’t robust evidence that detecting it changes outcomes. So, we need more precision about whom we screen, and then we need more precision and more data about what we should do once people screen positive for AF.

Stay tuned. That is definitely an emerging area of research because an ounce of prevention is worth a pound of cure. If we can get upstream, when people are just having little bursts of AF but aren’t yet in clinically detected AF, ideally, we could prevent the progression of AF and some of its complications.

In terms of prevention, we have virtually no randomized controlled data to elucidate primary prevention strategies, and we really need to get there. We need to be with AFib where we are with coronary heart disease. You don’t wait for somebody’s first heart attack to try to prevent it. I’ve been in medicine since the 1980s, and you would be considered an extreme outlier if you didn’t try to prevent coronary heart disease. But we really haven’t thought about AF that way.

We need to get better at stratifying who should receive primary prevention, and then we need rigorous randomized controlled trials to determine which strategies are most efficacious. For secondary prevention, it isn’t rocket science. I mean, we know we need lifestyle modification to prevent the progression and complications of AFib. Why aren’t we doing that? How can we lower the barriers? Is that going to involve more team-based care? Is it going to involve GLP-1 receptor agonists and SGLT2 inhibitors? We don’t yet know how to do a better job of implementing secondary prevention in practice. We need more implementation science to determine how to lower the barriers to doing what we need to do.

For stroke prevention, one of the things I didn’t mention is that, in the prior guidelines, CHA₂DS₂-VASc was the stroke prevention risk stratifier. Now, we’re saying to use whichever validated score you prefer, probably because this is a rapidly changing area. I think that, with the advent of artificial intelligence, more wearables, and genetic and genomic data, we’re going to increasingly refine our ability to predict who will develop a stroke and who, if anticoagulated, will have a major bleeding complication or an intracerebral hemorrhage.

That’s what you want to do: provide the right treatment at the right time to the right patient. Our ability to do that really needs to improve so that we can minimize thromboembolic risk, reduce bleeding risk, and prioritize the people who would benefit most from anticoagulation.

Another area that is really confusing is what to do for people with stage 5 chronic kidney disease. It’s very controversial. We need more research, and that’s another important area in stroke prevention.

You also asked about ablation and what needs to happen in that area. The technology is changing rapidly. We know that people are increasingly using pulsed field ablation, but we need ongoing research to optimize patient selection and procedural methods. Every time you open up a medical journal, there is another trial or study looking at ways to refine and improve ablation technologies. 

The final thing I wanted to discuss is that we really need to determine how to reduce inequities in the implementation of evidence-based care. We’ve talked about race, ethnicity, and social determinants. We haven’t even discussed insurance status, and that’s a huge issue as well. In Massachusetts, where I live, something like 95% of people are insured. In other states, not so much.

There are also huge gaps in care based on insurance status. We need to determine how to make sure that everybody is receiving evidence-based therapies that can improve their quality of life, longevity, health span, and lifespan.

Thank you so much for your time today!
Thank you. I really appreciate the opportunity and your interest in the performance measures.

The transcripts were edited for clarity and length.