Episode 10 : AI and Virtual Nursing:

Smart from the Start_ Karen Murphy: Audio automatically transcribed by Sonix

Smart from the Start_ Karen Murphy: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.

Intro/Outro:
Welcome to Smart from the Start, presented by Care.ai, the Smart Care Facility platform company and leader in AI and ambient intelligence for healthcare. Join Steve Lieber, former CEO of HIMSS, as he interviews the brightest minds in the health provider space on truly transformative technologies that are modernizing healthcare.

Steve Lieber:
Hello, and welcome to Smart from the Start. I'm your host, Steve Lieber, and it's my pleasure to bring to you a series of conversations with some of the sharpest minds in health information technology. We'll discuss the smart directions healthcare companies and providers are pursuing to create smart care teams. Today, I'm joined by Karen Murphy. At Geisinger, Karen is an Executive Vice President and serves as Chief Innovation and Digital Transformation Officer, as well as founding director of the Steele Institute for Healthcare Innovation. In these roles, she's developed Geisinger's Digital Transformation Office, which includes Geisinger's AI lab. She's also the architect of the Intelligent Automation Hub, designed to optimize operations and produce significant cost savings. Prior to Geisinger, Karen served as Pennsylvania's Secretary of Health, and she also served as president and CEO of Moses Taylor Health System. Welcome, Karen.

Karen Murphy:
Thanks, Dave. Thanks for having me. Thrilled to be here today.

Steve Lieber:
You know, I've been really thrilled as well and really looking forward to getting your perspective on a number of issues. But where I'd like to start out first is two areas that, identified during your intro, the AI lab and the Intelligent Automation Hub. Tell us a little bit about how those came about, what sort of objectives they have, and maybe a couple of examples of what you're focused on.

Karen Murphy:
Sure, Steve. So the Steel Institute was started five years ago this past July, so we got our five-year birthday, and the goal of the Steel Institute was really to produce transformation innovation that had quantifiable outcomes and solved problems. That's a tall order, to be able, I always say this job is the hardest job I ever had in my career, and I had some tough ones, but the reason this job was difficult is because the healthcare delivery system is very complex, and it's easy to create a bunch of shiny new nickels that you say is innovation, but to move the needle on quality or cost in a meaningful way is really different. So I start out by saying that is, as we built this innovation center, we were always going to solve problems. We were always going to do it in a fundamentally different way, and we were always going to have quantifiable outcome. So about five years ago, AI really started to healthcare, and some companies were coming in offering us to work with them. And I say we started with six of us in a row to evaluate companies, understand what they were offering. We started out with a couple of pilot programs. One turned out to be one that is still active today, and that was ..., a startup Israeli company who worked with us to develop an algal marker that identified patients who are at higher risk for colon cancer. So very successful, we've saved lives through that, but that started our evolution into AI. We started an AI lab within Geisinger. We have our own team that does two things, assesses AI opportunity and then also develops AI applications. We're doing one of our own risk profile. So risk stratification what patients are going to be admitted to the hospital or frequent ED visits, that's one example. We have another program going on for lung nodules to be sure that we're picking up the ... and have follow-through. So we really are all in. Flash forward to last, this week we had our AI committee, and there were 61 people at the meeting, 61 people on the committee, where we have some really high-level discussions about AI, where it could be helpful, where are the gaps, and then, of course, now with Gen AI, if you don't say ChatGPT five times a day, you're really not innovative. So we've developed the way to test in a safe way, generative AI within the organization, put together a governance structure, and I think we'll be talking about generative AI, really increasing the conversation in the very short term.

Steve Lieber:
Yeah, your examples are really striking in that you're very focused on clinical issues in terms of your examples anyway. And I'm going to contrast that with something that came out of ... in August, where Judy announced that their focus is on, let me look at my notes here, in baskets, clinical notes explaining medical bills, notes, summaries, coding, billing, more back office sort of things. Can you give us some insights in terms of where you're going, where they're going, where AI is going?

Karen Murphy:
And I would say to you, we're going in the very same direction, because many of those things that you mentioned, they are going to be our early AI projects. I think Gen AI is more appropriate for those types of, and I think those types of use cases, and I think it's going to save us tremendously over time. I mean, let's not get hyped up here, like it's going to take us time to develop, but if the only thing you could do is stand still, and that's not acceptable. So it's tough. I don't say that as a plug-and-play solution, that we're going to be able to answer all of our problems. That's just not the case. But is it another tool, the toolbox, to help us increase the quality of life for providers as well as help us be more efficient?

Steve Lieber:
Yeah, you mentioned just a moment ago standing still is not an option, obviously. So how are you feeling about the pace of change and the pace of innovation, pilots and full scale and scaling up and that sort of thing? How do you feel about that? How do you feel where you folks are situated as it relates to getting in there, identifying the problems, and driving solutions on a scale?

Karen Murphy:
Well, Steve, you can probably imagine I never move fast enough. I mean, that, that's somebody once told me, I was speaking frustratedly about, we're just not moving fast enough. And my friend ... said, if the world moved as fast as you want it to, I can't imagine what an awful place it would be. So I guess that tells you that I like to move fast. I think we're going into, I think we're going into a phase of innovation that everybody realizes that a million pilots are not the way to go. That you can be taking, you can be a mile long and an inch deep, and never really get anything that's meaningful. So, what we have done is we have begun to not only look at the pilot phase, but we design the implementation at the same time as the pilot. So when we're designing the pilot, we're saying, okay, understanding we have to iterate, but here's where we're going to spread it, and here's how we can spread it. And I think we're going to see healthcare, because of all the pressures on healthcare, not only financial workforce, aging population, there are a lot of pressures for us to get it right. So I think you're going to see faster moving industry, and I also think these tools that we're talking about in our toolbox, you brought up intelligent automation, how we started intelligent automation probably about eight months before the pandemic. And what we do is build bots to really do low-level tasks so that we can actually maximize the, maximize our human intelligence wherever possible. So it started out, and we thought that this is a, perfect to your point, back office kind of functions that would really be helpful. I can't even describe how excited everyone has been because, of course, during the pandemic, when we were getting 36,000 calls a day, we didn't have people to answer 36,000 calls. So we built a bot to do that. Testing results of the height of the pandemic, we were reporting 500, 600 results a day. We build about ..., so it saved us millions and millions of dollars. And it wasn't like it saved us millions and millions of dollars, or we would have hired the people. There were no people to hire. It was, we had to do it. So now, post-pandemic, we've set into a strategic plan where we're really trying to give workers relief, our workers relief in every area that we can. So let's take, you can't find four people to work and do the job. Let's identify the tasks that could be automated of those workers that you can't find. Number two, where is the automation then that can raise revenue? So can we do claims review more effectively so we can identify missing pieces that missing information that would ordinarily be then kicked back to the revenue cycle department to be completed? Can we have a bot that's actually going to proofread that claim before it goes? So it goes on and on and on, but it is a technology that is really, it saves time, saves money, and also provides greater efficiency, and sometimes even more quality of work than what a human does.

Steve Lieber:
You mentioned the challenge, both during the pandemic, and I think it continues today in terms of staffing and in really on a couple of levels. Availability is one, but also burnout and just fatigue, and exhaustion. At times, we hear people raise concerns about is, are AI tools going to replace people. And as you sort of indicated, well, the first response is there are so many vacancies, we're not replacing people with this, we're filling gaps. What's the tone of the workforce as these tools are coming in? What are you hearing, sensing, in terms of, particularly clinicians on that side of the House, reaction to the tools?

Karen Murphy:
I think what realistically, and Steve, I'm a registered nurse, I started out my career as an ICU nurse for a decade, so it really laid a foundation for me that allows me to understand what clinical, what everyone, I worked, I always say, in every department, except the Lord, in a hospital, so I can relate to the issues of short standings, short staffing stress, and burnout. If we're going to bring AI tools to the clinical environment for the providers, we have to make sure they make their life easier, not harder. So that's my, you know, if you're going to go in there and you're going to say, hey, this is a great thing, shiny new nickel, but it's going to add another ten minutes onto your day, that's just not. And I think we have to keep, we have to be realistic, and we have to bring the clinicians in, the nurses, the doctors, ... techs. Whoever you're working with has to be a part of the design process, because if you build the design process in a box and you don't consider or really have a good understanding of what the workflow is going to be, it doesn't, it just doesn't work. I'll give you a good example of that, and we talk about AI. Well, a data scientist can build an algorithm that pretty much can produce all the data you want, depending on what the question is. And early on, and this was a great lesson, early on, we developed an algorithm that looked at prediabetes. It looked at the signals for prediabetes. And then, obviously, we looked at pre-diabetic care gaps. And when we went to put it in, great information, first of all, it was 30,000 in the population. Second of all, when we went to put it in the workforce to say, hey, here's your 30,000 patients, every doctor, maybe 1000, here's your thousands of patients that are pre-diabetic and have six care gaps. And the clinicians look just, ... wireless, what do we do with that? Like, well, do you get on the phone and call all 60, do you, or a thousand? My point here is we did not think it through to say, how is this going to go in the clinical work? So we had it since refined. We want to make sure that the information produced by the AI in the clinical area is clinically relevant. And number two, that we can actually get that, we can share that information in a meaningful way and integrate the information in the clinical workflow, and it doesn't put more pressure on the physician, or nurse, or support staff.

Steve Lieber:
Sticking with the nursing area for just a moment, given your background as well as your current roles, certainly, the pandemic drove us into a higher level and quantity of virtual care had to. So what are you seeing today, and what has been the trend? I'm going to make the assumption, maybe there was some drop off in the amount of virtual care from the peak of the pandemic, but it probably is going to plateau and then probably move back up again. A little outlook in terms of your thinking around virtual nursing.

Karen Murphy:
I think virtual nursing will only accelerate, and for two reasons. We talked about the one, Steve, about the shortage of nurses, but it is also protecting the quality of life of those that are at the bedside. And those nurses, if you're working short-staffed, have so much to do that if, and I will say this, inpatient nursing is an early career game. You're not going to find, so the dynamics are changing, and that, it's a lot of work, very hard work seven days a week, 24 hours a day. I think we have to understand it's a very demanding profession, and heretofore, I don't think anybody appreciated that. I think that, other than nursing, I think the second piece is that we're not going to be able to recruit adequate numbers to be at the bedside, but we can recruit nurses who want to be virtually and just do certain tasks that are non-physical that are, you know, that aren't difficult. I'm a nurse. I can do a discharge with the patient right here, and I can use my skills as a nurse, but I don't have to have the physical burden of being at the hospital, at the bedside. And if that's two, I think three, we were at a conference yesterday and they talked about the, I heard about the health system in the Midwest that is taking the nurses off the floor and giving them breaks to be able to go do their virtual nursing. So I think virtual sitting, virtual nursing checks, or changing the care model. I think virtual nursing is only going to go, it's only going to accelerate.

Steve Lieber:
And would it be fair to say that utilizing virtual tools, remote monitoring, and that sort of thing, can take away also some sort of below the highest level of your licensure task so that you do operate at the peak of your licensure?

Karen Murphy:
Yeah, I definitely think about a patient, I mean, if you think of the, patients in the hospital today are really, really sick because we've shifted so much to the outpatient. So patients are sick, they're complicated to take care of, they're time-consuming to do a good job. So yes, I think it allows, and when you're a nurse in the ICU, you're not going to, I mean, you have to be there next to the bed, really holding that patient in your hand, and you're never going to replace that. So what we have to do is take the tasks away from everyone that can be done by, to your point, someone a lower license level or at the same license that doesn't desire to work in the field but will be willing to do other functions. So I'm very excited about the future of virtual nursing and all that will, all the benefits that will come.

Steve Lieber:
Right. In your background, as we mentioned in the introduction, Pennsylvania Secretary of Health, and I think there was a focus that you had, and it probably exists given Geisinger's location in a largely rural area. In terms of helping rural healthcare, in terms of supporting it, and ensuring it's there for patients, what do you see as what we can do for rural healthcare settings?

Karen Murphy:
So thanks for bringing that up, Steve; it is a passion of mine. I spent a great deal of time studying rural health and actually developing, when I was the Secretary of Health, developing a payment model, payment and delivery model, working the federal government with Sam and Mike, because I realized they were so challenged. So as we talk about this technology, these are tools that will help. Rural hospitals are understaffed, and both clinicians and physicians, they are, sometimes it takes them 2.5 hours to get to a place where they can receive care that we can get down the street. They're underfunded, so they cannot reinvest in these digital technologies that we're talking about that could be beneficial for them. How I see AI and really all the tools that we talked about today, virtual care for rural communities, I see there has to be investment on the federal side and the state side to really be able to deliver the tools that we have in urban areas for rural communities. And I think that all of these tools could be beneficial, but we have to find a way to connect them and have the rural communities truly understand what the value is and how we can get the value to them.

Steve Lieber:
Excellent. So to wrap up here as a final note and comment to our listeners, we have folks like you and as well as CNIOs, CMIOs, CIOs that will be listening. What's a practical piece of advice? What's the big takeaway you'd like to share with the audience?

Karen Murphy:
So I think that we didn't realize that the world, and including healthcare, the world was going to experience post-pandemic problem. We thought getting rid of the virus was going to take care of everything. I think it's clear now that we are going to go through a period of challenges that were developed during the pandemic or exacerbated during the pandemic, and are now even more exacerbated. I think what's really important now, Steve, is to build the resilience that you have in yourself. And it's not about, you know, it's not just organizational resilience, it's really, we've got to get out of that mindset. We can slow down a little bit. There's, you can't put every fire out all the time, but try to get back. You are resilient. You, everyone will be lifted. But I think we just have to be a little more patient, and I think we have to be thinking positively of the future, and we will get through this. And I've never been in healthcare for the decades that I've been in healthcare, that it was easy, so I'm not going to say it's going to be easy, but I think we got to turn the temperature down a little bit.

Steve Lieber:
I think that's a great Karen in terms of, yeah, we get all focused, especially those of us in the technology space, in technology and innovation. Let's not forget the people. That's what I'm hearing. Karen, thank you very much. This has been a wonderful conversation. I really appreciate your time today.

Karen Murphy:
Oh, it's my pleasure, Steve. Thanks so much for having me.

Steve Lieber:
You bet. And to our listeners, thank you for joining us. I hope this series helps you make healthcare smarter and move at the speed of tech. Be well.

Intro/Outro:
Thanks for listening to Smart from the Start. For best practices in AI and ambient intelligence, and ways your organization can help lead the era of smart hospitals, visit us at SmartHospital.ai, and for information on the leading Smart Care Facility platform, visit Care.ai.

Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.

Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.

Sonix has many features that you'd love including automated subtitles, collaboration tools, generate automated summaries powered by AI, advanced search, and easily transcribe your Zoom meetings. Try Sonix for free today.