Xiaoxi Yao, PhD, MPH, on Guideline-Recommended Anticoagulation Rates

In this podcast, Xiaoxi Yao, PhD, MPH, answers our questions about her team's research that investigated the use of anticoagulation and antithrombotic therapies in patients with atrial fibrillation, including whether guideline-recommended therapy was actually being prescribed by physicians and taken by patients. She presented this research at the American Heart Association's Scientific Sessions 2021.

Additional Resource:

Xiaoxi Yao, PhD, MPH, is an associate professor of health services research and an associate professor of medicine at the Mayo Clinic in Rochester, Minnesota.



Amanda Balbi: Hello and welcome to another installment of Podcasts360—your go-to resource for medical news and clinical updates. I’m your moderator, Amanda Balbi with Consultant360.

Today my guest is Dr Xiaoxi Yao, who is an associate professor of health services research and an associate professor of medicine at the Mayo Clinic in Rochester, Minnesota. She’s speaking with us today about her team’s research, which she recently presented at the American Heart Association’s Scientific Sessions 2021.

Thank you for joining us, Dr Yao. To begin, can you give us a brief overview of your research and its findings?

Xiaoxi Yao: Sure, oral anticoagulation is recommended for patients with atrial fibrillation. Most patients with atrial fibrillation will need to take oral anticoagulants for the rest of their lives. So, we wanted to see how many patients are receiving guideline-recommended oral anticoagulation stratified by their stroke risk based on their CH2AD2-VASc score.

Because the guideline recommendations are based on the CH2AD2-VASc, we wanted to evaluate the utilization because DOACS, direct oral anticoagulants, have become available over the past decade. These new drugs are easier to use. They are fixed-dose drugs, and some of them are associated with a lower risk of bleeding.

Our hypothesis is that patients may be more likely to adhere to a guideline-recommended medication these days because these drugs are more convenient to use than warfarin. On the other hand, new drugs are more expensive. So, we are not sure. That's why we are examining the utilization risk.

What we did is we used a national administrative claims database called OptumLabs, which includes commercially insured patients, Medicare advantage patients, and Medicare fee-for-service patients. We looked at patients over the past 5 years, and then we enter on a visit with an atrial fibrillation diagnosis coded. Because a visit is an opportunity, even if the patient was not on anticoagulation, this could be an opportunity for the provider to start or restart medications.

We looked at, within 90 days of that visit with an atrial fibrillation diagnosis, how many patients were on anticoagulation. These patients could be new diagnoses, but they could be prevalent Afib patients, too, because they might just be a routine visit. We also excluded patients who might have had other indications for anticoagulation. For example. if they had a VTE or joint replacement procedure, some of the anticoagulation could be for short-term use; or mechanical valve replacement in patients who underwent watchband implantation, because when they have watchband, they might not need anticoagulation in the long term.

The results were interesting. First, we looked at how many patients all into the low-risk, medium-risk, and high-risk groups. We found that 4.4% of patients were in the low-risk group, which is CH2AD2-VASc of 0 for men and 1 for women, because women get 1 for gender. The medium-risk group included men with a CH2AD2-VASc score of 1 and women with a CH2AD2-VASc score of 2. So, 8% of patients fell into that group, but 88% were in high-risk group. 

The last group, according to the guidelines, should receive oral anticoagulation. Specifically, most should receive a NOAC because NOACs have been recommended in preference to warfarin for most patients who are eligible for NOACs. But interestingly, surprisingly or not surprisingly, nearly 40% in this high-risk group did not receive any treatment, including oral anticoagulants or antiplatelets— clopidogrel, ticagrelor, or prasugrel. Aspirin is a little bit hard to capture in claims data because aspirin can be filled over the counter. So, sometimes patients might not submit a claim to their insurance plan, so we might not have complete data on aspirin. So, 31% patients were on NOAC, and 16% were on warfarin.

This is both expected and unexpected, because in our analysis 5-6 years ago, we found half of the patients didn't adhere to anticoagulants. So, it is a little bit discouraging to see there was little improvement over the past 5 years, but the message is more that we still need to figure out how we can encourage or improve adherence to medication. I think that shows the gap in practice. It is interesting data to see.

Amanda Balbi: Yeah, and you mentioned before that there might be some limitations to prescribing these medications—cost and availability, that sort of thing. Can you talk a little bit more about that, and how those factors might have affected the numbers?

Xiaoxi Yao: There can be a number of reasons for nonadherence or not prescribing. For the prescribing power, some patients may have some bleeding risk factors, like older age, prior bleeding, or impair kidney function. Actually, one of our prior studies showed that bleeding risk factors are highly correlated with stroke risk factors. 

When patients are at high risk of bleeding, they're also at high risk of stroke. Some studies have shown that patients still benefit from oral anticoagulation, even if they have some prior bleeding risk factors. One intervention could be to improve the education to providers, especially primary care providers or providers outside cardiology, that these patients should receive anticoagulation. 

That’s the provider part, but the other part is also patient, because the anticoagulation could have a monthly copay for some patients, which can be substantial. Especially for older adults, who may have many other drugs or other medical expenditures. Even if we say, like, $50 may be a lot for them, considering their other costs. Hopefully some of the drugs might become generic over the next decade, so we'll see. 

Overall, medication adherence—for NOACs or even for statins, almost all of them have become generic, but adherence is still difficult because these drugs are used to prevent an event—it's not like a drug you use to, for example, control your heart rhythm so that patients can feel better. Sometimes patient might not know what this drug is for. They might think, “If I feel some symptoms, that's when I take the drug.” But this is not how it works. 

Or they might think, “I took this drug for a few months, nothing happened.” But that's the point, right? That's the prevention aspect. So, there might be some interventions we could do for both patients and providers, but that's a long-term task. We just need to figure out which is the best way to do it.

Amanda Balbi: Yeah, absolutely. I think you bring up some great points for both the patient and the provider. So, with this new data that you just found, how can clinicians take that into practice and better patient care?

Xiaoxi Yao: I think one is just to have more discussion with patients about the benefits, potential risk, and causes. And then if the patients are concerned with some aspects, we could consider alternative treatments, like left atrial appendage closure with watchband. If it is so difficult for patients to adhere to the medications, could we use a watchband or device to prevent stroke?

So, just more shared decision-making or discussion with the patient, because sometimes providers may think, “I prescribed medications,” but patients might not feel the medications. There might be a gap there. I think that could be something, but it could be hard, too. I understand some providers have a time pressure. They have so many other things, to discuss with their patients.

Amanda Balbi: Absolutely, I think that's a good point as well, definitely something to keep in mind during their 15- or 20-minute visit, right? Okay, so, then what would be the next step in this research for you?

Xiaoxi Yao: We actually have an RO1 grant from the NIH to look at left atrial appendage closure. We developed an AI algorithm to identify who might benefit more from the procedure, because the procedure could be beneficial for patients at high-risk for bleeding or have difficulties adhering to the medications, but they also have an upfront risk of complications and upfront cost. So, it's not for everyone.

We want to identify a subset of patients who might benefit more and send an alert to their primary care provider, so their primary care provider can refer to cardiology for further assessment and discussions. That’s one of the solutions we can think of.

But another line of research is just to see how we can improve medication adherence, but that's a more challenging topic. Same as statin and hypertension drugs; it’s just hard.

Another line of research is we are trying to see whether we can diagnose more atrial fibrillation earlier, because atrial fibrillation could be asymptomatic. Sometimes when we diagnose Afib, it’s after the patient already had a stroke, and then we diagnose Afib. So, we’re trying to see can we diagnose Afib early and start anticoagulation early.

We developed some AI algorithms based on sinus rhythm ECGs to detect Afib, and we also developed some AI algorithm based on trying to use, like, Apple watch or other devices. So, there's another line of research we are trying to do, hopefully also to prevent stroke.

Amanda Balbi: Great! Sounds like you have your next couple of years planned out. Thank you so much for speaking with me today and answering my questions.

Xiaoxi Yao: Yes, thanks Amanda.