This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.

Thanks for joining us. My name is bill Russell. I'm a former CIO for a 16 hospital system and creator of this week health, a channel dedicated to keeping health it staff current and engaged. Welcome to our series healthcare patient room. Next briefing campaign. I'm excited to get to our topic today. We have a great conversation with Dr. Stephanie LAR CIO CMIO at monument health. And today we're gonna talk about ambient clinical intelligence solutions for documentation. This podcast series. Is going to culminate with an excellent webinar panel discussion, and we're gonna have experts talking about patient room next and how it improves care efficiency. You're gonna hear some of that today from Stephanie, just a fantastic conversation. You can check out more information in the description box below. There's also gonna be a registration link there as well. You could also just hit our website this week called.com upcoming webinars in the top right hand corner. we wanna thank our sponsor for today. Nuance for giving us some time with Stephanie to discuss this great topic now onto 📍 the show,

all right. Today, we are joined by Dr. Stephanie Lahr CIO CMIO for monument health Stephanie, welcome back to the show.

Glad to be here again. Always good to see you, bill. Yep.

I'm looking forward to this conversation. We've been talking clinical effectiveness and the last time we talked about automation and AI, we talked Artisight and Neteera. Today. We're gonna talk a little bit about ambient clinical intelligence and solutions around that.

So tell me about your journey into ambient clinical intelligence. What does that look like and what have you done?

Yeah. So this is a journey we've been on with nuance for about a year and a half, I would say. And for me personally, a journey I've been on with them. I work really closely with them and their product development teams for a lot of years now.

Because we've known that this was something that we needed to do that we needed to be able to. Take the, the documentation tools where we've got voice to text, but take it to the next level. And so just level setting on what that looks like is that ambient clinical intelligence documentation, it is technology that right. It sits in the room and through capturing the native conversation between a patient and their clinician. On the back end, that information can be turned into the appropriate elements of a document. So that instead of, I mean, again, as good as drag a medical one, and in these other voice to text solutions have become, it's still a separate step, right.

I finish my interaction with the patient. Then I still have to go and sort of regurgitate it in medical speak and put it into a document or maybe leverage tools through, through voice to text as well as some of my EHR documentation. And so this journey that we've been on with them is, well, what does it look like?

Again, if, if that technology can be in the room and can turn that conversation into a note that then all I need to do as the clinician is review it and potentially make some minor edits. And of course that is only possible with the utilization of artificial intelligence, that through lots and lots, I mean, hundreds of probably thousands and maybe millions of different documents that have been assimilated. and recordings that have been compared, right. The system in the background can start to say, oh, wow. Okay. When I hear. Document this. And so when we started on the journey with nuance to incorporate this into our health system, we looked at it in a couple of ways. One we wanted to look at who were some of our clinicians that particularly on the specialists side, who were really hard to get into and maybe weeks and months out.

And if there was a way to make them more efficient in their day, that they could add one to two more visits a day. Then we could work on that backlog of appointments. And so maybe a way to get them two more appointments a day would be to take the documentation burden away and allow those notes to kind of create themselves.

And so. We started again in a pilot situation with a handful of physicians. This technology is super exciting, but AI and I, I think I, I took this from Dr. Gustine from art site who I think took it from. Dr fer in California. So it's, it's re reutilized, but it's a great comment cuz it's AI is local and I think that that's really a key concept to get your head around as we start incorporating AI into workflows, is that the local part of it means out of the gate. It's not perfect, right? The nuances for no pun intended of how I speak the room that I'm in, the kinds of things I talk to a patient about all of those things have to be. Worked on by the AI to truly get out of what we need to. And so, as we worked with nuance, they were still early in their journey on the AI side of this, and there was a human component to it as well, but it's been really fun to hear the comments we get back from physicians saying.

Had another couple of patients a day, just don't take this away. And that is the kind of comment that we're getting back from people who really are able to make the best use of this transitional kind of documentation.

Okay. You've beyond Peak's my interest. So I'm, I'm gonna ask questions like a former CIO that I am assuming I'm a CIO now the so the question becomes set up pilots. So these are room systems, right? So we're talking about microphones and cameras in rooms, or we think, is it different than that?

Yeah. So one of the great things that happened during the pandemic was nuance realized that. the elements in the room that we had originally been talking with them about, which you can do really multi array microphones and things like that, just wasn't gonna work.

I mean, we had physicians sitting at home talking to patients, but I couldn't put a multi array microphone in. So they retooled things and said, we're gonna use this. And so that's actually what we use, which we you're gonna use your phone. You use a phone to capture the voice. And so you set up your, your actually you've got your EHR up on your screen.

So if you wanna be looking at old notes or whatever you can, but the capturing of the recording of what's happening between the patient and the clinician. Is through an app on your phone going directly to their private cloud, which is then got all the AI and then their people in the background that are doing the QA on it.

And so that literally means I can put this in the hands of anyone in any location. Now that doesn't mean at some point down the road, we may not wanna install those kinds of devices in the room, not only for this, but for other purposes. And we have seen in a couple of scenarios. Where, especially we have, if you have a, let's say a surgeon who takes call and they happen to be on call and the device they were using was their own phone.

We've actually morphed to providing them or having their office buy a phone specific for this, cuz you don't wanna get a call. While you're all actually doing the visit as well, but in general, most of our providers are using even their own device. They put it on, do not disturb and roll with it. And then at the end of the session it's done which again means they can take it anywhere.

We can do it in any clinic. It also means because the reality is, and I know nuance would say this too, this tool, especially today, while we're still working through some of the learnings and growth. It's not gonna be the best tool for every single person, but it is going to be a game changer tool for a group of people who in particular struggle with technology and the, or struggle with documentation and the time management around that.

And so if all I need to do is. Hand them a device like this, or they use their own and they get a license. It means I could have seven rheumatologists and three of them might use this. And the other four might not because their workflows are varied and that's okay. I don't have to distribute it throughout the entire clinic cuz they just take the device into the next room and they're off and running.

So you really can step into this. you can step into this. It's not like we're retooling a whole room. Cause when I went and saw it at I think it was hys.

Hys. Yeah. Yeah. It was so cool.

It is really cool, but you're looking at it. You're going, oh my gosh, I gotta do this to these rooms. Not that we're not already, we're already putting cameras into rooms. We're putting microphones into rooms. And I think as we start to think about this into the future, that might be the standard for a lot of rooms. I would assume.

Yeah, absolutely agree. I think one of the other things that's been great though, about being able to do this for right now is I don't know when the last time was that you tried to order cameras and microphones,

supply chain

reports and all those kinds of things. So that's been hugely advantageous too, is we're using those kinds of things where we really need to, but when it's simple and straightforward, like. The phone works. Great.

Yeah. And I'm somebody who has three iPhones. People are like, what do you need three iPhones for? Well, one of them's a camera, one of them's a microphone. And one of them's my phone go figure. But that's the, that's the business I'm in, but I could see that being a clinical setup, you could almost have a phone on a tripod and essentially have the meeting, but I love how you can move around with this. So you said you set it up for a handful of practices.

Are there practices that it works better for than others? You talked about specialties. Are there certain specialties? It works in and not yet in others?

Yeah. So I would. So when we first started talking to nuance, they said, gosh, you might not wanna do primary care because when you think about primary care, what is primary care?

It's a little bit of everything. So when you try to teach AI we try to teach an algorithm. When you try and train an algorithm of any kind, you wanna constrain the variables as much as you. Well, primary care is no constraining of variables. I might talk about hip pain, my depression, and a skin condition all in one setting.

And that's really three different specialties now with that said how baker at WellSpan? Great friend of mine really phenomenal work in this space too. He started with primary care cuz he felt. Man, this is where I need this is where that's, where he needed access. That's where he was seeing the highest levels of burnout.

And so he has a whole bunch of primary care providers. I think maybe over a hundred at this point that are doing this in our setting. It has stayed more with specialists. But now we're rolling out to really it's more anybody. Who wants it. And a couple of keys to that for us at least have been, and I'd be happy to talk to anybody in more detail.

I don't wanna get into too much of the nuts and bolts of this, but we have each cost center pay for this themselves. This doesn't sit in my budget. It's super easy to do that. Cuz it's a per license an individual license basis, which then means again, if cardiology has 15 cardiologists and this works really well for three of them.

Then they build that into their cost center. And it's not a cost that I'm distributing across all of the org the entirety of the organization. It also makes them a little more accountable for the tool. And does it work and does making sure that they're using it cuz it's coming out of their cost center and then that's part of what we're doing is. People will come to us now and say, Hey, so, and so is using this super cool thing on their phone, where it writes their note for them. Like, I wanna do that. And then the first thing we actually do is we go in, we look at leveraging epic as our EHR. We look at their signal data and see how they perform as a documenter.

And in epic, if you're super efficient, it doesn't matter what specialty you're in. If you're super efficient and you're turning notes around real time, this is probably not the tool for. But if you're doing documentation after that patient walks out of the room, if at five o'clock, that door is being shut and you spend another two hours on documentation, there's a lot of opportunity for this to help.

The reason it doesn't help on the people that are really quick turnaround is, is the cognitive element of this. Cuz the, no matter what, I'm gonna have to look at this later. Right. So that means I have to rewind my brain to that patient experience from a couple hours ago or 30 minutes ago, or whenever it comes back into my queue.

And review it and say, yep. That pretty much captures it. We have some providers that are super efficient with front end tools that write their note while the patient is there still have good patient experience. It doesn't overly burden them to do it. They walk outta the room, the note's done. The patient's happy.

They're happy. Great. This is probably not the tool for them, but a lot of our providers that isn't the case. So we do that analysis as well, because there is that piece of it takes a while for it to come back. And that's one of the other pieces that we work through in setting expectations is. Early on it's gonna probably take a little bit longer, cuz the AI is still learning and more QA is happening and then it will get faster and shorter over time.

And that's the exit question on this episode, which is all, AI is local. And there is a, there's a learning curve. There's a learning curve on the AI side.

There's also a learning curve, obviously on the clinician side, but on the AI side for it to truly understand you and to build an effective note, let's say one that's only needs a couple of minor edits. How long does it take for it to really, I, I don't know, learn you so that it's effecti.

Yeah. So I think where nuance is at today, cuz they want this, they think that this will get better and better over time. But the numbers that we're kind of using generally is the AI gets turned on or really leveraged after about a hundred encounters. And then it's probably 400 encounters or more give or take specialty in some other things.

Before you would start to Let it ride and go into self-edit and not do QA and, and those kinds of things. So it's it's a commitment and that's another piece that we look for is if this is a provider that really thinks they're only going to use the tool. Twice a week when they have long consults or something that come in, they might be a better candidate for it down the road.

But it does take some time that number is coming down further and further, and it's getting faster and faster. And again, some of it is provider dependent. If you have a really strong template in how you do your work and you follow that, well, the AI will follow that and learn from that. And that will be much faster than than somebody else. So it is a little bit individualized, but those are some rough numbers

that I right. And so you could have people on dragon medical one, you could have people on DAX and. And yeah, the right tool for the right experience. Stephanie always fantastic to catch up with you. I really appreciate the time and love what you're doing up there at monument health.

Thanks so much. Good to talk to 📍 you, bill.

I love talking to Stephanie, so many great things happening up there at monument health and another great discussion with her. I wanna thank our sponsor for today. Nuance for investing in our mission to develop the next generation of health leaders. Don't forget this whole series ends with a great webinar. We have four magnificent healthcare leaders, Anna Baker, Garber, former C for HCA healthcare, Billy pro. Emerging technologies and innovation at Intermountain, Fred Holson director of healthcare at Sirius healthcare at CDW company. And of course, doctor Stephanie LAR CIO, C M at mind human health. And we're gonna talk about patient room next, what the future holds for the patient room. We're gonna talk. All these really cool technologies that you've been hearing about in this campaign, you can check out more information. Our description blocks below. There's also a registration link. Of course, you can hear our website this week. health.com upcoming webinars in the top right hand corner. Love to have you join us September 29th at one o'clock Eastern time. That's it for today. Thanks for listening. That's all for now.