Our guest Dr. Barry takes us through the evolution of Electronic Health Records (EHRs), once mere paper charts, now a crucial player in lab tests, radiology, and medication safety. As we navigate through the maze of healthcare tech, we touch upon an often overlooked aspect - the essence of individual uniqueness in medicine. Closing the episode, Dr. B leaves our young innovators with some priceless tips to test fresh ideas. So tune in, and let's explore how AI is reshaping healthcare and the profound implications it holds for our future.
As always, feel free to send us questions, comments, or suggestions to our email: nicupodcast@gmail.com. You can also contact the show through Instagram or Twitter, @nicupodcast. Or contact Ben and Daphna directly via their Twitter profiles: @drnicu and @doctordaphnamd. The papers discussed in today's episode are listed and timestamped on the webpage linked below.
Our guest Dr. Barry takes us through the evolution of Electronic Health Records (EHRs), once mere paper charts, now a crucial player in lab tests, radiology, and medication safety. As we navigate through the maze of healthcare tech, we touch upon an often overlooked aspect - the essence of individual uniqueness in medicine. Closing the episode, Dr. B leaves our young innovators with some priceless tips to test fresh ideas. So tune in, and let's explore how AI is reshaping healthcare and the profound implications it holds for our future.
As always, feel free to send us questions, comments, or suggestions to our email: nicupodcast@gmail.com. You can also contact the show through Instagram or Twitter, @nicupodcast. Or contact Ben and Daphna directly via their Twitter profiles: @drnicu and @doctordaphnamd. The papers discussed in today's episode are listed and timestamped on the webpage linked below.
Enjoy!
Speaker 1:
Hello everybody, welcome back to the podcast. We are at the CHNC it's the last day we are wrapping up and we are joined by Dr Jim Berry. Jim, how's it going? Great? Come closer to the mic. Yeah, it's been great. So you're uh, you're fortunate that the symposium is happening in your hometown.
Speaker 2:
Yeah, it's very nice and actually I don't always get to come to this conference, but since it's in my hometown I was able to yeah. I've been going to this conference since its inception, so it's really cool that when I was about 2010, 2009,. When they have the idea of this CHNC, I was afforded the benefit by my colleague to be invited to it, and it was a room of about 25 people, envisioning what CHNC could be, and so I've gone probably five year or five different times throughout the years, and each time it's neat to see how it's growing and evolving over the years.
Speaker 3:
Absolutely, and it really has grown to, I mean, quite an amazing scale.
Speaker 1:
In what? In what aspect do you think it has grown more significantly? I'm just curious about, since you have that perspective.
Speaker 2:
Well, it's just involvement. So when I first went to that meeting of like 25 people sitting around a room they were talking about the vision of what CHNC could be of. They all came together, they left their egos at the door and there are a lot of them that were very high profile, that could have been pretty egocentric and discussions, and they're like OK, we have to do what's best for babies. We don't have any benchmarking for children's hospital. Nikkus and Vaughn doesn't work for us, so we need to develop something that we can internally benchmark and then make sure that we're providing better care for our patients over time. So every year that I've gone, I've seen more focus groups, different, different aspects of quality improvement work and it just has grown and grown and grown.
Speaker 1:
Yeah, I think that's the one thing to me that I'm still struck by. It's the granularity of right. I mean you could say, oh, we have this data set and we're looking at. You know, like sometimes the bigger the data set, the more shallow it becomes, because how are you going to collect all this information? But all the people we've spoken to throughout the three days have projects that are so granular in the details and the patients themselves that they're looking at that it makes the endeavor even so much so more impressive.
Speaker 3:
Yeah, it seems like I mean, it's really this network of focus groups, of people who are really dedicated to the focus group we're saying, meeting at least on a monthly basis. How has your involvement with the CHNC changed over the years?
Speaker 2:
Well, I've been more peripheral, but I'm involved kind of indirectly because Teresa Grover is a colleague of mine in Colorado and she really has been instrumental in CHNC all along the way. She's on the board of directors and is executive, so I've been more. I talk with her about what's going on and then kind of evaluate. And so what I would say is that I do still think that CHNC is quite descriptive in what they're putting out there and publishing and they'll be taking the next steps, which are interventional, to say, okay, if we put these three things in place, are we going to improve outcomes? And that's what's exciting for me is to see that they're taking the next step. And it's also right now, in a period of time, as you know, that data management, data analysis and data output is exploding with.
Speaker 2:
AI and other things in their position, I think pretty well to really take the next step. The next step is kind of scary, but I think they're there and it should be multimodal, pulling in not just diagnosis and some of the data that they're recording, but we need to start getting images downloaded into their database, laboratory tests and other things to make a more comprehensive picture, and I think they're positioned well for that. But it's hard because neonatology we're always going to be on the very bottom of kind of the AI development and innovation. It's going to start in the adult world first, and then maybe pediatrics and then maybe NICU. So they're going to have to go out and kind of lead the way, which I think they have the right people in place to do that.
Speaker 1:
So I would be remiss if I didn't bring this up, but you are one of the leaders of the Neomind AI group and people who yeah.
Speaker 3:
so why can't you guys, why can't we start a neonatology?
Speaker 1:
Hold on. So if people are interested in, first of all, and people are interested in finding out more about Neomind AI, which is a group of neonatologists focused on the development of AI tools for the NICU, you can find out more about it at neomindaiorg.
Speaker 2:
Dot com. I think, we have both. It's both. We have both. Yeah right, it's bothcoregoncom.
Speaker 1:
And we have an interview on the incubator podcast that's lined up for 2024, where we're going to talk to some of you guys to try to bring awareness to this collaborative. I think it's like you said, it's probably the future of where our field is going. How do you see maybe this aspect integrating into what the work of the CHNC is doing? You've alluded to it in the fact that they have the data and now maybe, if we tack on artificial intelligence tool, maybe we'll be able to discover things that we did not anticipate.
Speaker 2:
Well, I think it's necessary. Right now, 90% of data that's in EHRs is not being used, so it's unstructured data that just sits there and goes unused, and we use 3% of it to make the decisions to care for our patients 3% 3% out of 100 is what we're using at EHRs on a daily basis.
Speaker 1:
And God knows, we document.
Speaker 3:
Yeah, exactly yeah. So I guess that's my question. Is it because the EMR and I interrupted you? I'm sorry, but is?
Speaker 2:
it, the EMRs.
Speaker 3:
Are we not looking at the data as a whole that we are contributing to putting in there? Why do we only use 3%?
Speaker 2:
Well, first of all, why were EHRs really created?
Speaker 3:
Billing, yeah right, they're a great that wasn't easy, yet so I got it right.
Speaker 2:
They're a great billing machine and they do what they're supposed to. The clinical care and data aggregation visualization utilization was a secondary thought by EHR companies. Now we're kind of stuck with the platform that we have, and especially Epic is now over 50% of the market. They're starting to work with Microsoft, so I do think that they're well not starting. They have been working with Microsoft, so I think that there's AI that will be coming down the pipeline.
Speaker 2:
There's certainly already the ambient documentation where they're listening on rounds, creating progress notes for the adult world at least, and that's being looked at currently. And then there's also something that I think just came out today, something called Microsoft Fabric. I don't know if you've heard about it, but I think it was. I just was reading about it first thing this morning where it's multimodal and it's supposed to help both patients and clinicians. I kind of understand the healthcare that's going on for different patients and be assistant, so I think it's evolving pretty dramatically. But the EHR was never intended to be what it's used for. It's a billing machine that works perfectly. I don't know if you've ever been able to look behind the scenes of Epic, but there's something called an Epic Impulse dashboard. What do you think about that?
Speaker 3:
Yeah, I mean, I think it gives you information in a way that actually seems more usable than what we have on the front end.
Speaker 2:
it seems like oh yeah, so there's. So I've recently gotten my MBA won because I knew that, in order to take some next steps in my career and help change things that I needed to have more of a finance and kind of accounting and business background to understand how to be involved in the change in healthcare. One of the things that afforded me was to be able to go work with some of our coding or billers or revenue cycle management director and she showed me what they have that they can look at on a daily basis with Epic and it's an impulse dashboard that has 22 different metrics that they're following.
Speaker 2:
They can drill all the way down to the reason that a bill was delayed getting out. There is because the secretary on the fourth floor did not register the patient correctly.
Speaker 3:
So they know that. They know that within a day.
Speaker 2:
Just think if we had that capability on the clinical side where we could drill down, and that's where I do think, with AI and predictive analytics and others, that's where we'll move. We're gonna be like two decades behind where we should be.
Speaker 1:
Yeah, that's crazy.
Speaker 3:
Do you think we were better at assessing clinical data before the electronic medical record?
Speaker 2:
I don't know. I do think that the electronic medical record has been helpful for something. So I do think so did you guys probably didn't live before there was an electronic medical record, did you so?
Speaker 1:
I grew up. My fellowship program involved me reviewing charts since the inception of my unit in 1974. So I actually requested those charts.
Speaker 3:
When I was a resident, we were still using paper charts.
Speaker 1:
I have this picture of my daughter in the room with all the charts.
Speaker 2:
So I used to document in paper. You couldn't read it, it wasn't legible. Medication orders were not very good. Sometimes they weren't legible, so getting away from paper was necessary.
Speaker 2:
So I don't think that we did a very good job. Then I do think the EHR has been helpful for laboratory testing. You can look at things over time. Radiology imaging is much better and then medication safety should be safer because our orders are computerized and so forth, so there's some guardrails run. So I do think there's some benefits for EHR, but in terms of me sitting at the bedside of a sick baby and helping me take care of that patient, I don't think it does that well, so I don't think that part of it has been benefited.
Speaker 3:
You asked us a very interesting question at the end of our talk what are the three tips that we would give young people who are starting out or who are interested in testing a new idea? What are your three tips?
Speaker 2:
Well, you guys actually described it was icky guy. I don't know if you did it on purpose, but that's what you described.
Speaker 2:
What you did that, what you love to do, what the world needs what you can get paid to do and that's what everybody should be doing, and I think everybody's icky guy changes over time. But I think the first thing is that you do like people that get admitted to medical school today. They are unique. I wouldn't get into medical school today because I'm not that unique. I got good grades, I did fine, I got accepted.
Speaker 1:
The depreciation of Jim on a constant basis. I don't like it.
Speaker 2:
No, but it's true. So if you look at the, the people, that get admitted to medical school are amazing. They come from all these unique backgrounds. Experiences like University of Colorado Medical Center always kind of advertises who they get. They get Olympians, they get somebody who is a Nobel laureate.
Speaker 1:
I mean the people that they get are amazing, but it's interesting how, in the past, they used to be non-traditional applicants. That's right and now they are the traditional applicants. They have these diverse backgrounds and they've accomplished so much. I teach at the medical school. Every time I see them, they're like yeah, I did that before med school.
Speaker 2:
But guess what happens to them when they go to med school Repressed, they're treated exactly the same. So the 115 med students that start get treated exactly the same and their uniqueness and individuality get suppressed medical school, residency, fellowship, and then, early in your junior faculty years, you're also being just trying to survive. So you go a decade, a decade and a half, suppressing your uniqueness, when we should actually be elevating people's uniqueness all along. So I had a medical student ask me, actually when we were in San Diego at the AI Med Global Summit, came up to me and said you know, I was involved in two startups when I finished my undergrad. I started medical school and I talked with the dean of our medical school and I asked him if I could still be involved in that and was told no, I know that you should just stick to medical school and that's it.
Speaker 2:
That's exactly the wrong answer. We need to be more flexible in our education of trainees coming up so that they have a growth mindset and they're willing to think outside the box, because that's what health care needs right now is and you guys brought it up being disruptive, right. I think disruptive. It gets kind of a negative term and I don't like disruptive that much.
Speaker 1:
The term, the term disruptive, it means it's kind of negative. You're disrupting the fabric of something.
Speaker 2:
Yeah, and I get where the term disruptive comes from and AI is going to be disruptive for health care but I think there could be a better term, which is helping us do things better and using what we have better and getting to a place where we're providing patient care better, and so I don't like the term disruptive too much, but I get where it comes from.
Speaker 1:
Absolutely.
Speaker 3:
But I mean it's interesting. We're all saying how can we individualize medicine? We're in the era of precision medicine, but we really haven't gotten to precision education, right.
Speaker 2:
Where are we? Individualized pathways and I did a period of time where I was. I did 18 months of learning about adult human learning theory and so forth. And one thing that I took home from all that and it was quite some time ago now is that we talk about the learning curve, but we don't talk about the forgetting curve.
Speaker 2:
Very much and the forgetting curve is probably the most important, and that's where, if you're able to use interventional, individualized education, you can improve somebody's knowledge by understanding their forgetting curve. And so they do it in art, they do it in languages, learning that once you get to a certain point, they can actually highlight. Ok, you need to re-study these five things because you're going to miss them on the next test.
Speaker 2:
And then it's iterative. On the next test You're evaluated, and so I think what you're bringing up is we have to model that, and I do think, probably with new digital technology, ai and other things, that we have the opportunity to individualize our teaching. But if you look at the admission criteria for medical schools, guess what it is today? Is it any different than when I check boxes? I had to do organic chemistry and all that? The check boxes to fulfill entry criteria for medical school applications is exactly the same today as it was when I did this 25, 30 years ago. And is that really? Has medicine not evolved?
Speaker 1:
100% and the medicine is being taught in the same manner and, being a faculty, we're having ongoing discussion at NOVA, for example, because students don't show up to class. They're like this is not how I learn and that's a big, big problem right now. So I'm looking forward to our Sunday episode with you and the Neon Mind AI team, because this is going to be fun. Very exciting yeah thanks for coming and being here. Oh, it's our pleasure and supporting neonatology CHNC.
Speaker 2:
They're doing great things.
Speaker 3:
Yeah, amazing. So thanks for supporting us. Of course, our pleasure. It was a pleasure. Thanks so much.
