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

Today on This Week Health.

I think a lot of practitioners out there from the front desk people through everybody in the healthcare continuum has sustained some degree of PTSD. And that's, true also of the patients. And I think patients... have this kind of level of they're on edge

Welcome to Newsday A this week Health Newsroom Show. My name is Bill Russell. I'm a former C I O for a 16 hospital system and creator of this week health, A set of channels dedicated to keeping health IT staff current and engaged. For five years we've been making podcasts that amplify great thinking to propel healthcare forward.

Special thanks to our Newsday show partners and we have a lot of 'em this year, which I am really excited about. Cedar Sinai Accelerator. Clearsense, CrowdStrike,. Digital scientists, Optimum Healthcare IT, Pure Storage, SureTest, Tausight,, Lumeon and VMware. We appreciate them investing in our mission to develop the next generation of health leaders.

Now onto the show.

 All right, it's Newsday, and I am excited to be joined by Dr. Kevin McNamara, with Optimum Health IT, Director of Provider Experience. Kevin, welcome to the show. Thanks, Bill. Good morning. I'm looking forward to the conversation. get to talk to doctors a fair amount, but not in this venue, not in the back and forth around the news.

So... This'll be interesting. And I'm curious, we've selected three stories clinician burnout a couple of AI stories. And so, these are actually great topics to talk to a physician about. Give us a little bit of, your, background and, and what you're doing with Optimum.

So, trained as an orthopedic surgeon, was in private practice and group practice for up until about 2001 when I got interested in healthcare IT and, no one in my group wanted to, uh, help with the implementation. Everybody stepped back and you were still standing forward. Is that what happened? And, yes, exactly.

Because I was I had gone to this group because I liked the model of a... and the unified patient medical record. And I thought electronic medical record, this is great. You'll be able to pull things up quickly and easily. And so I got involved in that and I have been doing work in healthcare IT for about 17 years now I've been with Optimum for approximately eight and a half years.

And in that time spent much of my time being part of implementation teams advising Physician executives, and then in the last several years have transitioned more to optimization. I'm called the Director of Provider Experience, but it's really about going into clients and working with, again, their teams to look at how physicians are using the system and are there opportunities to make their lives easier so that they have less screen time and more time at home.

I Thought the age of the big EHR implementation was done. And then last year, Epic wins, Intermountain Northwell that's just another big one, UPMC. And I'm like, I'm looking at this going, oh my gosh, there's still these massive implementations. I, although I agree, a lot of it has gone to optimization now, but there's still.

There's still some big ones out there.

and I think that, speaks to the underlying issue of interoperability. I mean we never really got to the point where it didn't matter what record you had or system that you had, you weren't. Getting that free flow of information into a unified patient medical record.

so You look at these larger organizations and certainly with the mergers and acquisitions and looking for size, hopefully gaining economies of scale and what is going to be the opportunity for the organization, or what is going to be the solution to help them really deliver on their patient care goals.

Yeah, I agree with you. It's the interoperability thing. Don't we still need to solve interoperability? I mean, you're still going to have the shops that can't afford EPIC, and they're going to stay on Meditech. You're still going to have the VA and the DOD and whatever on Cerner. There's still some Athena out there.

I guess there's some Allscripts and, I mean, all of these dwindling in size, but at the end of the day, we still have to solve this problem, don't we? Absolutely. I mean, my spouse just changed his, healthcare provider. And I was like, well, what medical record are they on?

Well, they're on some small thing. I said, well, You're split in, like, three or four different systems, so who's actually monitoring your medications, your, potential for interactions? I, fortunately, am on Epic with my primary provider, and they are attached to the hospital system that, when I need to use it, I can use it, so Epic talks to it.

So I'm a big believer in that. And then I look at these organizations, and even, within my experience, the clinical practice that my healthcare is taking care of has been bought three times. They've always stayed on Epic, but they've changed from one version of Epic or to another instance of Epic, and Even that isn't as seamless as you would hope that it would be.

Not unless they're on foundation, for the most part, it's a new implementation almost. Right. And so I go in and they'll say we've got to review all of this information with you and I'm thinking, and it's funny because I go, I'm like, well, I do this for a living. So do you have this?

And can you go here in Epic? And can you go there in Epic and, walk them through some of the things. to speed up my business. Yeah, you would think they could almost go from care everywhere, like pull the record in and do a migration that way. feels to me like still a problem very much that we need to address.

I think a lot of health systems have gone to interoperability by default, and I've heard CIOs actually say this, well, once everybody's on Epic, we'll have the interoperability problem solved. I'm like, well, you do realize that will never happen. Like if Epic approaches 60 to 70 percent market share, then the government will step in and really screw things up and they'll say, Hey, that's too much, they'll break it up, and then we'll have. another problem to deal with. You can't, in today's day and age, you can't get to 60, 70, 80 percent market share. So there will always be a need. For interoperability, regardless. We still haven't talked about a news story. Would you like to talk about a news story, ? Well, but I think that that leads into it so you look within what system are you using and we're gonna look at, these large health care systems, Intermountain, UPMC Northwell, or any system that is using Epic or Cerner and, they are.

Implementing they all have their email channels, their messaging channels wearables, and this is all feeding into the electronic medical record, and somebody has to see this information. how do you filter it? How do you direct it?

How do you respond to it? you look at providers and burnout and it's this, tidal wave of information coming at them and trying to manage it and manage their patients and finding less and less time to actually provide clinical care, which is... Kind of soul crushing, because that's really why you went into health care, is went into health care to help people, and I look at, when we first started with InBasket InBasket was a little simpler, there weren't as many files and now it's become so complex and you ask yourself the question, why does it need to be so complex? where are all these messages coming and how do you actually manage them without just saying addressed and moving on from it?

you, as a physician, let me ask you this about AI. So, AI is really good at summarizing. I mean, large language models are really good at really crafting a response to something. Ask a question, give a response. We know that the accuracy is still in question, but that too will be solved not too distant future.

We'll have multi level models that, sort of like you send this out for the math problem, you send this out for the drug drug interaction problem, you send this out for the coding problem, that kind of stuff. But it'll all come back, and so we're going to solve these things. And I'm just... I was reading an article the other day that over 50 percent of people are now using large language models every day in the execution of their job, in the, delivery of their job.

And now, it wasn't specific to healthcare, but it I think it's indicative of, This is sweeping. it's sweeping the, nation. I talked to some CIOs, doctors were starting to use it for like prior authorizations and writing letters and all this other stuff.

Like before IT even knew what was going on, they're like, oh my gosh, this is being used. And it's that kind of tool that if you give really creative, smart people who understand healthcare enough time, they're going to sit there and go. Oh, I could use this for this, this, this, this. maybe not diagnosing just yet, but for a lot of things.

are doctors seeing it as that kind of tool that's going to take a lot of the burden off their shoulders? Or are they worried that someday it's actually going to take revenue off their plate? I think physicians are hopeful that it is going to ease the Burden on them, frees up some time to allow them to provide patient care.

I think they are cautious about Is this going to be another technology thing that I have to learn that is going to add another layer of responsibility to me, or is it going to solve a problem? And then from the financial standpoint, I think that I would say physicians aren't even at that point yet because they're just trying to get through their day.

And I think one of the points that. Zach made in his article was that with AI and the ability to take some of these tasks or address some of these tasks for physicians, it will allow the physicians more time to see more patients. Well, don't know if that's true. I think what it will allow is the physician to have more time to provide better care for the patient and focus on the patient.

If one of the side benefits is that it does open time in their schedule so that, with AI assisting and getting patients with certain diagnoses that, AI assesses, this is a 5 minute visit, this is a 15 minute visit, can this allow for more effective or efficient scheduling then yeah, there is that possibility, but I think, really the first thing that physicians are looking for is help with their work and getting through their day and simplifying things.

So where can AI help first? And I think that starts with Kind of assessing what are the issues out here? What, let's get a team together. Let's talk about this. what is crushing you during your day? what are the tasks that you think aren't really your tasks to do?

Or is it possible that AI could help? And, when you talk about a team, you're, you're talking about, you have a Data scientists or computer guys in the room are really smart about these and say, well, here are some of the things that we could potentially utilize for this and let's explore it and let's identify what the problem is.

We want to solve. Let's identify what the outcome is that we hope to get from this. And let's monitor it and, Adjust and see if that is really working. the article we're talking about is Unite. AI. How can AI help reduce cost of healthcare? And it has a handful of things.

Streamlining medical visits, lowering administrative costs, aiding diagnosing and treatment, improving writing and keeping, creating symptom checkers, monitoring global health. Do any of those really jump out at you? Given your experience and background. think one that really jumped out at me was first, symptom checkers.

Because I think one of the things that is overwhelming providers is the messages to their offices, either through InBasket or to their staff. So, if a patient could... Utilize an online bot or call into a system and, be prompted for symptoms and maybe it pulls up their health record from the system and notes what their pre existing conditions are, current medical problems are, and help them navigate this to say, this is a problem.

Thanks. Something that you can manage at home, monitor for this and this. If something changes, then yes, you need to move on to a different level of care. I think one of the other things that stood out to me was streamlining medical visits. We have this wealth of information in the electronic medical record.

People are collecting and validating information at multiple steps in the process. Patients are coming in for visits, maybe they've had lab tests, maybe they've had imaging. it would be incredibly helpful, I think, if that is streamlined, summarized, so the physician comes in and knows at a glance here's the chief complaint, or patient's being seen for this, they've had these tests, AI says, we need to explore this.

And ask the patient if they have any additional questions, so that it's really teeing things up. You go into an EMR now, and the screens are, they're loaded with information. And it's somewhat daunting to go across the screen and say, What do I really need to focus on here?

And, there's a percentage of the provider and, the entire staff population that is really good at saying, that's what I need to focus on. but I think there's a large percentage that get distracted by everything that's there and it Does become a little bit complicated to am I really supposed to be focusing on here?

by the way, the next article we're going to look at Stanford Health uses AI to reduce clinical deterioration events. And it talks about detecting patient clinical deterioration early on, holds the promise to decrease mortality, improve outcomes. And so what they did is they applied AI to this problem, which is just looking at all this telemetry data and all this data that's just coming across.

And when you think about number of patients across all of Stanford, number of people that actually monitor that information, you almost need technology to do what it does, which is to sift through large amounts of information. and identify the signal from the noise to say, hey, this could be a deterioration event.

It's not necessarily 100 percent accurate, but it doesn't need to be. It just needs to prompt us to say, hey something might be going on here. And it's interesting to me, as we talk about these AI models, this whole concept of practicing at the top of your license. It's going to be even more important because AI is going to be able to step in there and practice a lot of this stuff at the lower end of the license.

a basic response, a basic this and creating a summary. I heard of nurses that were creating these, complex ICU cases, but they were just. Pulling all this information together. They were essentially highly trained clerks pulling all this information in from all these different things.

Cause you have to do that for these complex cases. And that's the kind of thing that AI does really well. I'll go out, gather all this information, pull it into one place. Then the nurse can really do what they were trained to do, which is care for the patient and to, look at that information and apply human thinking and logic to it to make sure that nothing was missed.

Right, and I think that when, in reading through that article they followed a very specific methodology. They gathered a team together of not only physicians, nurses, and computer folks, they worked to say, what are the. Challenges that we face. What are the issues? what are the numbers that we typically look at?

Then, what is the problem we're trying to solve? Which was these deterioration. And then using machine learning to look at these large data sets identifying key data points and saying what are the indicators what is going to alert us? And when it does alert us, what is our response?

And, recognizing that there's individual bias, people have different threat levels and saying, oh, well, I don't really think that this is. that, that patient's doing fine. or what is the actual response? so using AI and machine learning to develop the signal saying this patient is at risk.

And then also applying A thoughtful process of it is not going to be an individual response, but we're going to follow an actual action plan, if this, then this and we'll be watching for this response. And, they've already seen a significant reduction in those deterioration events.

So that in itself is great for the patients. Who don't suffer those incidents. It's saving money for healthcare and, it's saving a lot of grief for patients and their families. You point out identifying the use case, identifying the problem, and I've heard that reiterated a number of times now, which is the AI is another one of those tools that could be applied in a lot of different places, and it's gonna be important to identify.

The use case that you are trying to solve so that you can measure how are we doing against that use case instead of this sort of shotgun approach where it's, oh my gosh, it's everywhere. Is it really delivering value and those kinds of things? So these individual use cases are so, important. we're not going to have much time to touch on this, but I want to talk about clinician burnout.

And it's interesting, this article talks about, AI and some of the things that are coming about to help clinicians. it used to be, we were talking about burnout because of the EHR. I mean, literally we were just saying, oh, it's just overwhelming the amount of information that's coming at them.

And yes, technology can be applied there. But in some of the recent studies I've been looking at, it's not only. In fact, the technology now moved down the list. environment that they're operating in seems to be creating a fair amount of stress. In fact, that's one of the top two, typically.

Violence. Nurses are citing, just the work environment where they're berated and those kinds of things, and sometimes physically assaulted in those cases. And they're saying we're almost longing for the day where we could go back and talk about the struggles we had with the EHR because now the struggles seemed a little bit more, personally difficult for the people who are practicing.

I'm wondering as we're, pulling these things off, the AI stuff and the clinical the technology things, we're getting better at those things. Are we still going to see burnout as a result of these other things which seem to be escalating now? Absolutely. I think we were trending in that direction prior to the pandemic.

I think the stresses of the pandemic and changes to practice And changes to our daily lives were incredibly stressful. I think a lot of practitioners out there from the front desk people through everybody in the healthcare continuum has sustained some degree of PTSD. And that's, true also of the patients.

And I think patients... have this kind of level of they're on edge. You see it in, not just in healthcare, but you just drive down the street. I know, but everybody's like, it's now becoming a common joke, a common part of our vernacular to use it as a punchline to say, oh, it's almost like healthcare where you have to wait 40 days before you see a physician.

And it doesn't matter if you say, hey, look. I think I'm having a heart attack. It's like, well, either go to the ER or your next appointment is in 45 days. And that's creating angst, by the way, on both sides. I mean, the clinicians don't want to be saying to people, Hey, it's 45 days till I can see you any more than the patients want to hear it's 45 days till I can see you.

Absolutely. we have to take a serious look at, how we're going to deliver health care and what are the advanced practice providers that can take a different share of the load that we can actually get people into C providers. it's the aging population in America that There are more needs there are more needs, unfortunately, I consider myself in that group, at the age of 66, it's like, Oh, okay. I feel great. I'm still working. I have some health problems and, I'm on a regular schedule with my doctors, so I know that things are happening, but that's not good for everybody.

Are you technically part of the baby boom generation then? I am. I was born in 57. Yeah, so you were, tail end of the baby boom generation. So you're following this big. of people that have changed everything like that. Baby boom changed everything in America at every stage, right? the number of, of baby carriages sold in the years between the baby boom generation was huge and then suburbia grew as a result and then, at every era, it's just sort of changed overloaded the system and changed it.

Absolutely. And now you have. We've got the senior living centers which I don't know what they're doing about healthcare within the senior centers, but, the seniors talk to each other and they share their experiences and they share their experiences. Thank you.

Symptoms, and they share their diagnosis, and that leads to questions, and maybe I should talk to my doctor. You think about that, and you think, okay, AI and a symptom checker might be a good thing and I'll tell you, it used to be that We had this larger population that wasn't tech savvy.

That's no longer true. And, as we start to move into other care models, where you know, hospital at home and the other opportunities to care for people at home that have less severe conditions that you want to monitor, whether it's diabetes or renal disease or what have you, and they're using wearables, and have a scale at home that communicates with the medical And that is going to add to the workload.

So you have to think, okay, if we're going to do this and we're going to add these things to the system one, can AI help with filtering out all this information so that it is alerting the right people at the right time? So, People don't get to a point where they're going to require a greater level of care when a simple intervention could have worked.

What kind of staff do we need to have to support those things? organizations across the country are looking at this and, dealing with this or addressing this and trying to come up with solutions to this. And hopefully they're ahead of the tidal wave and not going to get taken out by it. these are really interesting times to live. I think we're going to see AI make significant progress with regard to the mundane tasks that a lot of clinicians had to deal with, and I think it's going to create some capacity to hopefully address some of these access issues that we have.

And then it's going to be incumbent upon us to use that time wisely, like, how do we, how do we even leverage it further because there are areas like rural healthcare, we didn't even start talking about rural healthcare we're seeing. The number of at risk hospitals continue to climb, and then technology is going to be really important to get people in front of clinician that can help them who's potentially Hundreds of miles away.

yeah, it's going to be really interesting. Kevin, I want to thank you. I want to thank you for coming on the show. I want to thank you for the conversation and hopefully we can do it again. Well, thank you very much, Bill. It was great to talk to you. I love to talk and discuss ideas.

So, I don't get to do it enough, so look forward to perhaps future opportunities.   📍 And that is the news. If I were a CIO today, I think what I would do is I'd have every team member listening to a show just like this one, and trying to have conversations with them after the show about what they've learned.

and what we can apply to our health system. If you wanna support this week Health, one of the ways you can do that is you can recommend our channels to a peer or to one of your staff members. We have two channels this week, health Newsroom, and this week Health Conference. You can check them out anywhere you listen to podcasts, which is a lot of places apple, Google, , overcast, Spotify, you name it, you could find it there. You could also find us on. And of course you could go to our website this week, health.com, and we want to thank our new state partners again, a lot of 'em, and we appreciate their participation in this show.

Cedar Sinai Accelerator Clearsense, CrowdStrike, digital Scientists, optimum, Pure Storage, Suretest, tausight, Lumeon, and VMware who have 📍 invested in our mission to develop the next generation of health leaders. Thanks for listening. That's all for now.