Reporting for Heart Rhythm TV, I'm Mehek Dande, and I'm joined by Dr. Camille Frazier-Mills, professor of medicine and cardiac electrophysiologist at Duke University. Dr. Frazier-Mills, thank you for joining us. Of course. I'm so excited to sit here with you. Absolutely. We're going to dive right into the discussion about cardiac remote monitoring for CIEDs and really the importance of data analytics in that space. Tell us a little bit about your background, your academic interest, and what you think of as the scope of data analytics when we talk about remote monitoring of devices. Yeah, that's a great question. I am the ambulatory director of electrophysiology, and so I run everything in our outpatient setting. Additionally, I run everything related to devices at Duke, and so at our main hospital and in our ancillary hospitals around. So I think it gives me a nice perspective in order to think about this. I think specifically my passion has always been devices within electrophysiology, so kind of marries all of the things that I really enjoy and brings me to the space that I get to talk about. So I think as I think about data and analytics within CIEDs specifically, I think as one component of our patients, right? So we take in their history, right? We get all of their clinical information, and that's another point of data. We combine that with maybe some testing that we do or the lab data that comes back, and that adds the other pieces to it. And then as we pull from their implanted devices, we're able to really understand more what's going on with our patients. So it's a very unique and important point that we have that actually helps us tell the whole story of our patient, and in a way that we can bring all of these stuff together for our patients. And so it's really about, as we hear the phrase, actionable data a lot. So it's really about getting the information, get rid of the noise, find the signal, find the actionable data to prevent any bad outcomes from happening, or even when an alert does come through, to get it to the provider in a timely fashion in a way that actually helps the patient decrease anxiety, decrease actual rough clinical outcomes as well. Exactly. Yeah, I think it's getting all of the pieces of information and then having systems that we have in place in order to filter that appropriately. And when we can use AI and make it more efficient for our patients and for our team, I manage the whole teams. And so I have to think about, are all of my people functioning to the highest capacity that they can, and am I having them doing clicks, clicks, clicks, which everyone gets frustrated with, as you know, and so getting people to do it. So when systems have more of an integrated fashion, then they're able to be more effective and do things that they really enjoy doing as opposed to being frustrated with hitting more clicks. Absolutely. And the role of ML, like you mentioned, of software of third parties that could help with this streamlining of data, data integration, there is a lot of technology out there. How do you think would certain vendors differentiate themselves from others or certain companies that providers find more trustworthy or that improve efficiencies or workflows? Yeah, I think the first line is safety, right? And so you have to have a system that has been well vetted and that you know will pass all of your IT security issues, right? Because you can't have maybe some smaller programs that are coming out, which just haven't been vetted well. The next thing I started thinking about is the degree of integration, right? And so when you have to log into this system for this part of it, and I have to log into this system for that part of it, then it makes it very frustrating for the provider who's trying to go through all of the information. So a system that's already integrated into your, like for us, we have EMR as Epic, then it makes it easier when you have one button that you push and go into the system. When you have the data coming in, then being able to look in real time in the patient's chart I think is really important and not having another window and having the double computer wide open all the time, right? So I think it's important for us to be able to pull admissions, pull laboratory data, pull medications, pull the problem list over into that system. And so I think that there are very few systems that have that degree of integration within the EMR that would afford us to have less time spent bouncing between systems but actually have it all in one. And what system do you use at Duke for remote monitoring? Yeah, so currently, initially we had our own that one of my colleagues had created. And then when technology continued to evolve, we had to go to a different system. And so we initially went to ScottCare. But then just this summer we converted over to PaceMate because I found that it was a much more integrated and comprehensive system and it allowed us to do things in a more efficient pattern. Yeah, we had a lovely webinar yesterday about PaceMate and the technology and it sounds like the workflows are a lot more efficient and it's very well integrated into the system and that the alerts are sent to the right person at the right time to be able to get actionable data and the pursuing actions in a timely fashion to help our patients. Yeah, I think that the integration is really nice. I think things come through like you would in your primary systems before where you'd had to look for different colors of alerts or things, but then you can self-select it. And so instead of having to log into CareLink and all of the separate systems, you're able to go into one and your same settings are held true across all of the vendors. And so something that's agnostic of vendor is really important and it allows you to have all those things that really are alert conditions for you to be steadfast because in some sites you have to log into and Latitude may have different settings for what they think is important for something versus Medtronic versus whatever. And so in order for you to be able to truly get the data for all of your patients to have the comprehensive level of care that you want, it's nice to have it in one system. And beyond the operational optimization and having data in one spot and not having to open multiple windows, as you said, and being vendor agnostic, which I think is absolutely key, once there's enough data, once there's enough reliable data, do you think that it could have other implications from a predictive standpoint down the line? And if so, what are you most excited about in that space? Yeah, this is an area that we're really passionate about and we're looking forward to. So as I said, we transitioned, it was the end of June when we started and I think through the first few months and we got all of the, I mean, the integration was seamless. I have to applaud their team and applaud our system that we really worked well together early on in order to make it flow. And I was a little nervous how it would go, but it worked out really well. But the things that we really were interested in is one, obviously, the normal day-to-day patient care is our highest priority, but secondly, the research potential and some of the QI initiatives, right? And so how can we do things better? Can we look for early predictors of AFib, someone having so many PACs that they're going to have AFib? Are they having runs of non-sustained VT that then leads to their ventricular tachycardia and therapy? Should we be intervening earlier in these patients? We need to think about ablations or medication optimization and those type of things. Of course, kind of the heart failure diagnostic is important and we can integrate that with our heart failure team as they're doing their whole remote monitoring of heart failure patients. The stroke risk and making sure, like it's really nice to be able to say, yes, these patients are on anticoagulation, they have AFib and you can query and say, how many of my patients who have AFib are not on anticoagulation and do I need to figure out why they're not on it? Maybe they're a high risk of fall patient and those ones seem to be Watchman candidates. So I think it's really the query ability of the system is really nice. The other part I think is really appealing for us as a major academic center is that you have de-identified data. And so certain centers have agreed to share their data. And so it's not just my center at Duke, but it's also centers across the country that we have some de-identified data that we can really delve deep and look for some really nice predictive models that we can create. And so there's many aspects of it. I think once we, you know, I think we're at the point of integration, you know, the ordering, all of that stuff is working out really nicely, but we can start thinking towards those next steps and really looking at the research potential. And then I think the other part of it is efficiencies of your system. So from the kind of business aspect of things, there's all these analytics in it and like who's doing what, how long is it taking them to do this and that and the other thing. And so being able to make sure your people who are working for you are functioning at the capacity that you really need them to work. And if they're not, then figuring out how can you help them and get them to whatever capacity that you need. And so before it was just, you know, so-and-so was assigned this many reports to me today. Did they get that done? I can better see how people are spending their time and I can give other tasks or do things that I can help our system better. We've been able to shift a whole FTE of a admin by converting over to PaceMate and I think that that's something that was really exciting for us. So we had a person who was arriving orders in the system every day and you had a schedule of 80 patients who were supposed to transmit this day and she's typing in and arriving these orders and it takes her five minutes per order to do the things that she needed to do with all the clicks and this, that and the other thing, right? So it was really nice to have her be able to make sure our connectivity has improved because she can allocate her time to calling and talking to patients and helping them stay connected and making sure they're doing what they need. That improves our total patient care. And so she's much more satisfied in her job than what she was when she was pointing and clicking all day because she gets to engage with people and, you know, those type of people are really, they're people people, you know, and they want to do that. So it's been rewarding to see some of the other effects of changing over systems. You never know how much you were stifling some of your staff when you had them doing kind of menial tasks that really didn't bring, it brought, it was needed, right? We had to do it. Somebody had to do it. But it was, it's nice to see the change that I've seen in her. Yeah, overall sounds like quite a win-win situation from easy integration to easy to use platform, shifting FT to a more satisfactory job description and not only the current analytics that are that are easier to access, but also the future potential of research and of improving patient outcomes. It sounds like it's been quite a success and I applaud you for for taking the leap because whenever someone says change of software or upgrade, you're exactly right. The first question in every user's mind is, is this going to add work and is this going to add complexity? So this is a really nice example of how technology can really help us move forward and improve outcomes for our workforce and our patients. So thank you so much for your insight, Dr. Frazier-Mills and for joining us today. It's been a pleasure talking to you. Thank you. And for more hot off the press content from HRX 2024, follow HeartRhythmTV on YouTube, LinkedIn and X. Keep watching.

cardiac remote monitoring

CIEDs

data analytics

predictive analytics

clinical decisions