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 one of the key places that we tend to see folks really excel is when they deeply understand the practice of medicine, when they understand how the workflows in the hospital are taking place and then integrate their solutions in a way that continues to inform and enhance that workflow experience. And worked with health systems to put a technology in place that solves a problem that they're experiencing today.
Thanks for joining us on this keynote episode, a this week health conference show. My name is Bill Russell. I'm a former CIO 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 keynote show. CDW, Rubrik, Sectra and Trellix for choosing to invest in our mission to develop the next generation of health leaders. Now onto 📍 our show.
All right, today we are joined by Seth Hain, SVP of r and d at Epic. Seth, welcome back to the show.
Good morning Bill. Fall is well down in Florida.
I'm excited about this series that we're doing, it's our five year anniversary. We're looking back and we took the top two most listened to shows over the first five years, and we're inviting those guests back and yours was one of the top two most listen to shows of all time on this, and we've produced well over a thousand episodes. So I'm looking forward to reprising this conversation and we're just gonna riff on a bunch of different topics. We're gonna take a look back at the world that you've been a part of 17 and a half years at EP and we're gonna take a look forward as well.
But before we get there, talk a little bit about your role at Epic. What do you do? What do you do? And then, and then obviously, we'll go back and talk a little bit about where you started.
Yeah, sure. So I'm in the research and development group here and I work with a variety of our applications that includes our outpatient applications, which is essentially anything you walk in for, right?
Might be a well child check might be a dental exam. And then I also, and for a number of years have worked with some of our foundational applications and in particular our analytics capabilities and the machine learning capabilities that are built into the core across our applications. And used in everything from our revenue and access products to obviously the hospitals and clinics, and as.
My day to day responsibilities really focus on developing out the roadmap for the code we need to write. And that's spending a fair amount of time with customers, understanding both the challenges that they're experiencing and also the dreams that they have, and sort of figuring out how to translate that into actionable code that can be written.
And then I also spend a fair amount of time understanding and diving into new technologies that are coming out, used across industries as an example, marrying that up into those roadmaps and then working with the teams on making sure we roll up our sleeves and execute on 'em. So that, that stuff shows up in the wild.
that's a phenomenal. Position enrolled me in I just interviewed David Baker from Pacific Dental at. At the health conference. And so you must have worked with him because they, they rolled, i, i the module's called Wisdom, right? The Dental, and they just rolled it out nationally.
And so they're, they're really excited about the integration of being integrated into healthcare because, For the longest time that that's been their vision. Go to Epic so that people recognize that dental care is healthcare. So David
and Pacific Dental more broadly, it's really interesting to hear and learn from them about the mouth body connection and the impact that can have broadly, but then, translating that into the types of technologies that are in the dental space.
be it a electronic smart toothbrush as an example, which is possible to integrate into my chart. And pull in to drive decision support. So it's a lot of fun chatting with David and his
team. Yeah, it's pretty cool. So, 17 and a half years at Epic. Somebody might be wondering how do you become the SVP of r and d at Epic?
So talk about the first 17 and a half years. What's the path that you took to get there?
Short answer to your question is there isn't one path as I look around at the My Peers a lot of folks have taken different routes there. Mine started really on the back end, so my background is in mathematics before I came to Epic was in Madison for that.
And I started on a, a technology that was used to scale out the use of Epic for very large organizations such as Kaiser. As an example. And quickly based on that, started working with our performance and database teams on both how to scale out the architecture and make it more resilient. It was a lot of fun spending time with that group on.
Those challenges might be an area we can dive into a lot of change over the years in regards to continuing to evolve that architecture as healthcare organizations have continued to grow and then brought that experience around architecture. Pulled back some of the math and had the opportunity to.
Work with a talented number of folks on building out our machine learning platform capabilities and data science more broadly. And then sort of stepped from there into those broader set of applications that I just described. So I think, I think the key is finding something that keeps your interest and then not being afraid to take that next step on, on a different type of challenge and a different type of opportunity, which has been a lot of fun here.lf years, we're going back to:
Yeah, I, I, well, and thinking about Epic at that point too, kind of com contrasting the two. Right? When I started at the company, we had roughly 90 customers. For those of you, you can see, I'm sitting in a conference room here on campus in Verona. There was a hole in the ground in Verona, which became the first parking garage at that point in time.
And we had about 90 customer. And the largest organizations were folks like Kaiser who was implementing, as an example. We were working with Cleveland Clinic and on the back end we were very back end databases. And the architecture we were very focused on continuing to scale to meet the largest organization's needs is they were rapidly bringing more integrated applications.m me at that point in time in:
We didn't know exactly if that was gonna take off. Judy had a vision that folks would be able to show up in the ED when they were traveling and have their medical record right there with them. So those were the types of things we were working on and on the back end, on the data we were starting, we were pretty proud.
We had pushed a million database updates a second at that point. And we could, we could scale up to that. Yeah, what would happen? Big number. It felt like at that point I was
gonna say, what would happen in healthcare right now if we were still at a million?
Well, I mean, a number of things have shifted, right?
At that point in time, that activity was primarily being driven by the doctors and nurses and schedulers directly using the system. And so what's happened over that 17 years is we have seen that activity and the system shift, both continue to scale with new folks on the system, but increasingly, The backend architectures need to support the volume of analytics in machine learning registries, driving population health, the number of patients connecting in through MyChart and other apps. And so you've, we've really needed to make that shift to continue to accommodate all of that additional. Wow. Where
Where are we at today in terms of transactions? ? Well
There's been a number of changes from an architecture perspective on the, the backend databases. Now we have organizations doing an excess of 50 million updates a second re From a database perspective and in the lab we're over a hundred.
And I think in parallel to that, there are new technologies we continue to take advantage of over that time, right? We build out the data warehouse architecture kaboodle for purposes of self-service analytics. Similarly, we build out the cognitive computing platform, which is cloud based for deploying machine learning models at scale directly into workflows.
And so continuing to evolve both. The reliability and scalability of the systems behind the scenes and using these new technologies, languages, et cetera to enhance the workflows and the outcomes around
A million, 50 million, a hundred million. So the challenges continue to get greater, the speeds, M&A driving this, population health, different business models, the pandemic, a lot of things driving. How does a company like Epic stay ahead of what the future demands are gonna be on the system?
I think the first thing is to listen. We spend a lot of time talking with the folks across the healthcare industry, our customers and others about what are the challenges. So you can get a sense of what might be next down the road.
And then I think the other piece is just a discipline. So on the database side, as an example, and on the architecture side more broadly, it has been a discipline of with every release, how do you continue to push those limits ahead? How do you always have two or three next layers to the onion, if you will? That's the analogy I tend to think of. You gotta peel off the outside layer and make sure you understand what's next and work through those from a technology perspective to push it ahead. And there's really a discipline to that, building out the tool sets and continuing to invest in it. And I think that is the key. You just kind of can't let your eye off of it, otherwise it sneaks away.📍 📍 In:
📍 📍 Yeah. Let me let talk to you about two different things that I, I think are really interesting from from ethics, sort of growth and whatnot. know, One of the things we saw was a lot of customization. When people put their EHRs in, they did a ton of customization. There's this drive now back towards foundation. And why is that? Why is that important? Why is it important to get back to foundation for these health systems?
I think a key around that is operational success and. We're seeing this in the industry right now by having a consistent understanding, kind of at that core level of a foundation.
It means that you could benchmark your organization against others, understanding how your OR utilization looks compared to others, understanding how the connections you're making with patients, be that through a remote monitoring program or telehealth compare, and that benchmarking gives you a real opportunity as an organization to continue to improve both the quality of care you give and the timeliness of it. And so I think that is a real key. It also gives you, as within, I mean as is embedded in the word, right? With a foundation, you can then reach for new opportunities. And so folks are using that foundation system and that build is a way to then building their own machine learning models and deploying 'em.
We have over a hundred organizations that deploy their own machine learning models into the system. As an example auctioner has as an example, moved really heavily into using MyChart to help reduce The number of folks that need to come back after a significant cardiovascular event seen a 45% decrease in readmissions around that. So those types of programs are where they're now doing the innovation, using the foundation as something to build
on. Is there any aspect of it that one of the things is release cycles. Healthcare's changing so rapidly, we have to keep releasing code and updates and those kinds of things.
Doesn't it make that a little easier to know that a majority of your clients are on foundation and therefore the testing cycles become easier. The iteration cycles become a little shorter.
No, you're spot on there. And it both has the opportunity on the operational side to sort of reach for the stars, and then it has an important impact on the IT side.
Both these rapid releases as an example, give you the chance to be more agile as an IT team and be more consistent with the foundation system. It also gives you the opportunity. To use things like Nebula, which is a platform where we provide applications as a service. The rapid rollout of telehealth during the pandemic was enabled through that cloud-based capability.
And you were confident you had a, a firm foundation as an IT team to roll that out to your patients.
the other one I found interesting is this move to hyperdrive, right? I think the next two years will be marked by a significant move. In that direction. Give us a little insight into what it looked like as you were talking about Hyperdrive and why it becomes important as a foundational technology for Epic clients.
Sure. And for folks context, hyper Drive is an update to the UI and ux of our primary application deployed in clinics and hospitals. And that's just part of the discipline I was talking about before. Where you continue to evolve and update the infrastructure across, and it opens up new opportunities with those architectures.
This, from an IT perspective, it opens up opportunities to deploy the application in new ways. It also, the performance of it, notably, and it's excite, we're, we're getting data back from the early sites that are using. Already today. And it's notable the impact that it's having, so you know, it, it's part of that discipline that I referenced earlier.
All right, so there's, there's the changes that are going on into healthcare, and we're gonna look through the future in just a minute. I know people are like, oh, you gotta, you gotta ask 'em about the futures and we'll get there. A lot of technology changes over the last 17 and a half years. I think.
2005 was marked by data center virtual. Workstation virtualization. we now see a move to cloud. We see AI and ml. So it's, it's, it's evolving. Some would say it's evolving even more rapidly now than it, than it's been over the last couple of decades. So how will each of these evolutions and technologies, and maybe you could talk about these specific ones how are they gonna impact healthcare and how will they be utilized moving forward?
So we could start with virtualization, cloud AI and ml, but if there's others we could talk about those as well.
I think it's interesting because spending time behind the scenes on this stuff, you tend to think about it from the technology perspective and how it impacts the IT team. But it has, but what's more important is the impact that it has on the end users.
And so as we have seen. Things move from virtualization technologies, which you know, often time, which many folks are using across the industry. I mean, it's the standard way that this stuff is deployed these days, whether that's OnPrem, which is where it was historically or now in the cloud where we have folks using public cloud to apply our entire stack.
To applications provided directly as a service. It in it decreases the IT challenge in regards to deploying them, but it opens up new opportunities for end users. And I think that's the real key here. Those technology shifts have made the applications more accessible so that as a doctor, I've got h.
In my pocket when I'm out at the, on the soccer field watching a child play their game. And I can still, no, don't do
it. Watch the kid play the game. Don't pull up your .
I'm sorry. My wife is a physician. I appreciate that she's there with me watching the game together and that she doesn't need to leave.
Right? It gives you that opportunity if that's how we wanna do it, but I think it also opens up a number of doors. Around machine learning where the data scale continues to increase. As we look at and have folks analyzing genomic data as part of the medical record, as an example, bringing observational research through things like Cosmos, deeper into the workflow and so.
Technology change behind the scenes creates so much more opportunity on the application side to inform physicians and share more of the of that information directly in workflow and ultimately improve patient care. So I think we, we don't wanna lose why we're doing this as we make those technology changes.
We're sort of gonna transition slowly here into the futures. But so talking with Dr. Halamka over at Mayo and he was talking about how they have integrated a, a bunch of the machine learning in the clinical workflows. Which is we've always been slow to move in this direction just because of the the requirement of being perfect.
Right? You, you can't fail fast with clinical data because that's just not an option. So we have to validate those models. We have to get those models. But now we're seeing up to, I think he was saying like 10 to 15 models actually integrated into the workflow. Are you seeing the use of those kinds of models? To offload some of the burden to the clinicians and some of the cognitive load and other things.
Certainly we see folks using machine learning very broadly. Kind of similar to, as John was talking about, I think a key to this is the user experience as they walk through it. The purpose of these models is to help inform the care and decision physicians are making and others across the clinical realm. And how you present them in the workflow is a key aspect of that. And so we provide health systems with tools to embed their own models and to use the user experience, be it a hover bubble, to get more details on what the model is, even all the way down.
What are the features or inputs and their importance in regards to this prediction as an example. And so helping gain confidence and understanding is a key that I think has sort of flown under the radar as this adoption has taken place. And I think it's something that we also as an industry, need to continue to evolve as we think about how ultimately that helps inform the decisions made by physicians and.so we, we talked about:
Which has been the norm this year. It's been a very challenging financial year. We have new competitors coming into the space, so healthcare providers need to be more nimble. Present new offerings to the marketplace, be more responsive to the consumers. We have again, changing business models, more value based care contracts, more health systems taking on risk.what does:
Well, I hit on one of them already, right. There is an important focus on using data to inform operations and to identify those efficiencies that sort of needs to be the base layer of this. Yeah. On a more hopeful note, I think, and looking, kind of looking forward with you here a little. We're seeing more and more sort of digital native uses of the technology that I think provide real opportunities to the health systems as well, to patients to help solve some of these challenges.
So an example of this that I heard about recently and was. It's just great. It's so much fun to see these health systems put this stuff into practice and then to hear about it after the fact when you're the, when you're working on developing it. Sutter build out in my chart a care companion application.
This is a, a patient journey application, if you will, for pregnant couple. To help the woman walk through this situation and manage through her pregnancy. And the first thing they found was that by just having it natively inside of my chart, they were able, they had natural uptake with very little marketing or other advertising.
Two-thirds of folks immediately enrolled in the program. And so we were excited. This obvious. Both improves the patient experience. But the thing that surprised us and surprised them that they shared was they saw a 16% decrease in the number of messages to physicians. As a result of getting it in place, they put it in place to, to build that connection with the patient and actually saw the amount of time physicians needed to spend with them, decrease as a result of it.
So those type of kind of digital first approaches to solutions have a real impact at the health system level in regards to staffing is one example. The other trend that I think looking forward, We've had a, we've had some seen some excesses successes as an industry over the last five plus years of this.
But I think it's going to only accelerate is the connecting of the dig of the healthcare ecosystem and doing that in digitally and doing it in a way that reduces documentation and facilitates more communication, not just between health systems. Today, you. Through Care everywhere. There's 11 million exchanges daily.
Health records and half of those are between, just to be clear, half of those are between sites that use Epic with somebody that doesn't. But increasingly what we're seeing built on that foundation is new members of the healthcare ecosystem coming in working with pairs as an example, to automate prior auths and close care gaps automatically.
That saves time and money for folks across the ecosystem. Being able to rapidly connect to specialty diagnostic groups as an example, bringing in those types of advanced care in labs such as genomics easily into workflow, and then having the discrete results file back so that you can easily make those connections as a health system.
So I think They're exciting cuz as a patient you have new opportunities and access to labs, et cetera. But as a health system, it's a decreased lift to put this stuff in place and it ultimately saves time and energy in regards to running the operations. Wow.
Any other cool stories like that?
I feel like, I feel like I'm sitting, sitting here with grandpa and saying so you build these foundations and you see these health systems taken in different directions and do different things. Are there any other of those kinds of stories that you could share? Like where you looked at it and went, wow, that's really cool how they're using this platform?
one of the ones that has always stood out to me, so during Covid NYU build out a machine learning algorithm to predict folks that were likely to have favorable, well to not have severe outcomes from Covid. These are folks that didn't need to occupy a bed in the hospital as an example, and they built that out, put it in the workflow.
As you talked about rapidly, this is over a matter of a, a few weeks, they were able to. Build the models, validate it, and then put it into workflow. But then they took the ne next step. And they shared it across the community. So as you talked about with that foundational system, these types of both technologies, but then workflows are shared across the community.
And Oxford down in Louisiana put it the machine learning model created up in New York City into place. So that type of collaboration is one of the things that I think really shines a bright light on on the future possibilities.
I, I saw an article recently that talked about the distinction between digitization and digital transformation, I think, which is obvious to you and I we've digitized all this information in healthcare and more and more, in fact, I've, I've talked to some people about the amount of data that's being stored in healthcare and it's just growing exponentially.
What, what kind of digital transformation do you envision over the. I mean, I could put you in the futurist chair and say 10 years, but that's, that's a little tough. But what, what kind of transformation do you see in terms of how we deliver care over the next maybe five to 10 years? And, and what things are going to impact that?
Is it sensors? Is it just the being able to process the data and put it into models? Is it consumers starting to take responsibility for their own care and giving them the tools that they, they can not care but their own. Giving them the tools. I mean, what, what kind of transitions?
I'm answering my own question. I hate when I do this, but, and so, so does my listeners, so I'll open it up to you. What's your vision of what the next five to 10 years might look like?
I, there's a couple of things that really stand out to me. The first one is that I think care is going to continue to be more informed.
oftentimes we talk about machine learning in these sort of technical topics. But at the end of the day, it's about putting more information at the fingertips of both patients and providers. And that's happened in through a couple of ways and I think it's gonna continue to accelerate. The first one is integrating in new members of the kind of healthcare ecosystem.
This might be companies building digital therapeutics as an example. The ongoing advances around specialty diagnostics that I referenced earlier. . The other way that this happens is through deep analysis of data that is personalized. Cosmos, which is an effort across the Epic community, has over 160 million patients in it, and so that provides an opportunity to put observational research directly into work.
And I think as we're thinking about that sort of connected ecosystem, right? You got new members joining the healthcare grid, you have deeper analysis of data informing the point of care. You also have patients playing a new role in that. They're more connected through both devices that I've got, I've got something that can take an ECG in my wallet just with two leads on my thumb.
And that type of information can flow directly back into the system to help provide care, as an example, be that at home or on the go. And that opens up real new opportunities. So when you look forward and start thinking about, I, I grew up in Nebraska small little town outside of Lincoln of 5,000 people, and you don't have as much access to clinical trials as an example there.
And so this type of connected e. Opens the door to new types of discoveries by working closely across the life sciences communities as well, enabling decentralized studies to happen at multiple different sites, as an example, and I think so there's a real possibility that we will over the next five years, see both a speed up to discovery.
but then through this connected ecosystem, a speed up to deployment and use and putting it into practice. And I think both of those steps are key. And, and it's encouraging and exciting to be working on that stuff.
Let's transition a little bit to talk about the partner ecosystem and so, Epic has a very robust partner ecosystem. And so I'd like to talk about, let's see. I think the question is, talk about the, some of the things that partners have done maybe over the last couple of years and what what has made the ones that are successful, successful, what has given them staying power in that community of innovators on the Epic platform?
I think one of the key places that we tend to see folks really excel is when they deeply understand the practice of medicine, when they understand how the workflows in the hospital are taking place and then integrate their solutions in a way that continues to inform and enhance that workflow experience.
And so the folks that we have seen really excel, have appreciated that and worked with health systems to put a technology in place that solves a problem that they're experiencing today. Those have been the ones that have really stood out to me.
And you don't want to call anyone out as, as I really like what they've done with the platform.
Cause if you call one out, then that becomes your favorite child kinda thing.
It, it does create a little bit of a challenge. But I do think that my favorites really understand how to have an impact and help put their technology into practice.
Well, fantastic. Seth, I, I, I love these conversations and we should make 'em more often, not every three years or something like that, because the pace of change in healthcare is pretty rapid. Plus, I just, I love your perspective on things and I, I appreciate you taking the time to share your wisdom and experience with the community. So thanks again for your time.
I really enjoyed the conversation, bill. Yeah, let's not make it three years next 📍 time.
I love the chance to have these conversations. I think if I were a CIO today, I would have every team member listen to a show like this one. I believe it's conference level value every week. If you wanna support this week health, tell someone about our channels that would really benefit us. We have a mission of getting our content into as many hands as possible, and if you're listening to it, hopefully you find value and if you could tell somebody else about it, it helps us to achieve our mission. We have two channels. We have the conference channel, which you're listening. And this week, health Newsroom. Check them out today. You can find them wherever you listen to podcasts. Apple, Google, overcast. You get the picture. We are everywhere. We wanna thank our keynote partners, CDW, Rubrik, Sectra and Trellix, who 📍 invest in our mission to develop the next generation of health leaders. Thanks for listening. That's all for now.