Let's Talk

Harnessing Real-World Data to Transform Pediatric Care with Epic

About this episode.

Let’s Navigate...

Epic is one of the largest electronic health record systems in America. On this week’s episode, Jackie Gerhart, MD, Chief Medical Officer, and Phil Lindemann, Vice President of Data and Research, join us to discuss how real-world data is transforming pediatric care through Epic’s Cosmos and EpicShare platforms. The conversation also highlights how data-sharing collaborations across health systems are driving pediatric research and innovation, with a special focus on the groundbreaking work of Dr. Kevin Dysart at Nemours Children’s Health.

Guests: 
Jackie Gerhart, MD, Chief Medical Officer, Epic
Phil Lindemann, Vice President of Data and Research, Epic

Host/Producer: Carol Vassar

TRANSCRIPT

Announcer:

Welcome to Well Beyond Medicine, the world’s top-ranked children’s health podcast, produced by Nemours Children’s Health. Subscribe on any platform at nemourswellbeyond.org or find us on YouTube.

Carol Vassar, host/producer:

Each week, we’ll be joined by innovators and experts from around the world, exploring anything and everything related to the 85% of child health impacts that occur outside the doctor’s office.

I’m your host, Carol Vassar. And now that you are here, let’s go.

MUSIC:

Let’s go-o-o well beyond medicine.

Carol Vassar, host/producer:

Welcome, everyone, to the Nemours Well Beyond Medicine podcast. I’m joined today by Dr. Jackie Gerhart. Dr. Gerhart is the chief medical officer for Epic, and with her is her colleague Phil Lindemann. He is Epic’s Vice President for Data and Research.

Epic, of course, is a big name in healthcare. They’re known for creating, supporting products that enhance the user experience. The user could be the patient, the user could be a researcher. The user could definitely be a clinician, with such things as electronic health records, practice management, clinical decision support, data analytics, patient engagement, MyChart is a big way to engage your patients, research and development. I’m probably just scratching the surface.

So I want to welcome both of you to the podcast.

Jackie Gerhart, MD, Chief Medical Officer, Epic:

Thank you.

Phil Lindemann, Vice President, Data and Research, Epic:

Good to be here.

Carol Vassar, host/producer:

So, one of the topics we had a myriad of topics that we could have considered for today, but I was really fascinated by something that’s called Cosmos, and it is the use of real-world data to improve health outcomes. You can detect trends, and you can also help researchers as they create the next generation of healthcare treatments.

Let’s talk about Cosmos from a very high level. What is it? What does it do? How does it benefit patients? And since we’re a pediatric podcast, let’s concentrate on how it benefits children.

Dr. Gerhart, I’d love for you to start.

Jackie Gerhart, MD, Chief Medical Officer, Epic:

Sounds great. So, Cosmos benefits clinicians and children by allowing a large body of resources that come from some of the most amazing treatment centers in the world to be available to clinicians at their fingertips. And so, Phil will talk a little bit about the history of Cosmos as a data set and so forth, and then I’ll get into a bit about what are the clinician-facing and patient-facing ways that you can import that data to actually help with decision-making at the point of care.

Carol Vassar, host/producer:

So, Phil, give us that overview of how Cosmos started.

Phil Lindemann, Vice President, Data and Research, Epic:

It’s one of these things I think people thought we were able to do since the very beginning, but it just simply that’s not how the world evolved. So really, what we wanted to do was create a collaboration of health systems that they could safely bring their data together to learn from each other. At this point, we’re at about almost 70% of health systems in the US have Cosmos live, and that means that they’re bringing their data together, and all told, it’s about 300 million unique de-duplicated patient journeys, and about 30 million of them are pediatric patients. Actually, a fun stat is: 3 million 5-to-18-year-olds, their entire life is documented in Cosmos, from a medical standpoint, from the research. So one of the things we had to do is ensure that the health systems and their patients knew they could safely do this. So we put together what we call the rules of the road, that has elected members from all different health systems that make sure that Cosmos is used correctly.

Things like the data can’t be sold, the data can’t be sold by Epic, data can’t leave the system, it really ensures that researchers can enter that and safely produce new studies. They can make new discoveries. In fact, one of the most prolific users of Cosmos is a physician at Nemours. Dr. Kevin Dysart has published, I think, four or five papers now, and he does a lot on neonatal and maternal health, and has actually done some pretty cool things we can talk about later. But I think the research aspect of Cosmos is one thing, but being able to derive insights off a data set like that, and then return them to the physicians is where, I think, Jackie can talk about some of the things we’re working on there.

Carol Vassar, host/producer:

Dr. Gerhart, do talk about that.

Jackie Gerhart, MD, Chief Medical Officer, Epic:

Oh, well, love to.

So I’ll give three examples. The first one, we call lookalikes. And the idea is, if I’m seeing a patient that has a rare disease, or a constellation of symptoms that I’m not exactly sure what’s going on, I can actually use, in the point of care with my patient next to me, those specific symptoms, or even in the future, some of their genetics, to list them in the portal, and then have Epic go and find other people within the Cosmos population that also have those same traits or diseases. And then, it connects me, as the physician, to other physicians that have seen similar patients before. And so, in the past, I might’ve gotten to my physician lounge at my hospital, or I might’ve called up a friend and said, “Hey, have you ever seen this sort of interesting case?” But now, I have some experts at my fingertips that I otherwise wouldn’t have, and that allows me potentially to have a faster diagnosis, potentially to help that patient on the journey of treatment, and really connect them even with a community of others through just that clinician and clinician connection.

So that’s the first one, and I think about that as patients with, again, rare diseases or mystery diseases. Then, on a little bit of the other end of the spectrum, I think of more common diseases that we may not have as much research around in specific cases. For example, let’s say that I had a patient with hypertension, and I know how to treat hypertension pretty well, my first and second line therapies, but I might not be up to date with the most recent literature on how do I treat that in kidney disease patients. And so, the tool that I would use is called Best Care Choices for My Patient. And in Epic, same thing, I would turn the screen towards my patient and say, “Hey, I’ve got your demographic information, your comorbidities, the treatments that you’ve tried in the past, and I can see that either it’s not working, or that we need some more information on what outcomes we’re actually trying to optimize for.”

So maybe in one patient I want to get a really good blood pressure, and in another patient, I might be worried about stroke risk. And so, I really want to figure out which medicines are the right ones for that patient in front of me. And so, Best Care Choices for My Patient essentially offers up information from the Cosmos community in patients that are like the one in front of me to give me an idea of how to best care for that patient.

And then, the third example I’ll give you are Growth Charts. You might know that the CDC and Growth Charts from a long time ago were on a very specific population, and as we’ve learned more about rare diseases, specifically in kids, how can we as parents understand how that child is tracking along with other children with a similar rare disease? So, single cell, for example, or Marfan’s, how can we understand, both as parents, but also as clinicians that are caring for those patients, whether someone’s on track with their developmental milestones? So really advancing the information that’s in front of the clinician, and then therefore allowing for this robust decision-making with them and their patient.

Carol Vassar, host/producer:

They’re doing this right at the point of care, aren’t they?

Jackie Gerhart, MD, Chief Medical Officer, Epic:

Yes, yeah. That’s the part that’s most exciting, for me. Right now is one of the coolest parts, in my opinion, the coolest times to be a part of medicine because we for so long have had multiple different sources of information that we can use in our tool belt, but Cosmos allows us to just have one more tool that we can put in our tool belt at the point of care to give this information right there, whereas we would otherwise have to maybe look it up, or try to figure it out in advance, or do a lot of studying afterwards. And this really allows that conversation in real time.

Carol Vassar, host/producer:

You talked about physicians getting access to the data, researchers getting access to the data. I want to combine that with something that Phil emphasized, and that was safety. I think for many hearing this for the first time, they’re probably wondering about those rules of the road, those privacy parameters, that governance structure. Expand on that a little bit, Phil.

Phil Lindemann, Vice President, Data and Research, Epic:

Yeah. It starts with people having the courage to say, “We’re going to put our data into Cosmos, and we’re going to trust each other that we’re going to use it.” So, in order to build that trust, every single health system has signed identical paperwork that we call the rules of the road. And then, almost like a constitution of a country, that paperwork is managed by an elected group of individuals. Basically, health systems elect their peers at other systems to say, you’ll represent us, and then anytime there’s a new use case that comes to Cosmos, that’s kind of… Things evolve. Things are not sort of cut and dried all the time. We bring it to the council and we say, “Is this in the spirit of Cosmos? Is this the right thing for patients? Is this the right thing for everybody involved?” And basically, making sure we’re good stewards of it.

So things evolve, new use cases come along, we want to make sure to do that, but the rules of the road. It has teeth. It allows that if someone were to incorrectly use Cosmos, someone could bring a grievance, and they’d get kicked out of Cosmos. So there’s really severe penalties for how this is used. But my favorite ones are, this is not for selling data. It’s for actually producing new medical research, and all these rules still apply to Epic, and the data can’t leave, because then we wouldn’t know what happens to it. So, there’s a lot of protections, and there’s a lot of protections at a local level. I think a lot of people will talk about HIPAA and some of the standard privacy rules. We said, “That’s great, but we’ll obviously meet HIPAA, but we needed to go beyond that because we needed to support a global community.”

So we have to look at all the data-sharing rules across the globe and say, “How do we make sure that we’re meeting them as well?” So what we did is when really early on, when we designed Cosmos, an individual health system will say, “Well, in the colony that we have that hospital, they’re not allowed to release those types of lab results to a data set like this.” So maybe HIV, things like that. So every health system is essentially making sure that they adhere to their local rules and regulations, whether that be at a country level, a state, a province; however you’re thinking about it. So that’s been a big part of it, too, is giving the sites the autonomy to make sure they’re meeting their local regulatory guidelines, but also making sure they’re still participating in the spirit of Cosmos that can bring that data together. And in the end, the idea is that any researcher or physician can learn from the thousands and thousands of researchers and patients before that up to that moment.

So they basically have the newest information to say, “What are people doing to care for their patients?” So we obviously had to put some massive protections around that. And I’d say we’re almost six years in, and have not had a single grievance. So it’s been working well, and elections are underway right now. So this is actually a hot topic right now.

Carol Vassar, host/producer:

And I think it pretty much goes without saying, but I do want to say it, this is de-identified data, is that correct?

Phil Lindemann, Vice President, Data and Research, Epic:

That’s right.

Carol Vassar, host/producer:

So I’m curious, from the public health view, is there a way to leverage this data to start looking at public health concerns that might be of note to a local community, or across the nation? Dr. Gerhart.

Jackie Gerhart, MD, Chief Medical Officer, Epic:

Yes. The first resource I’d point you to is epicresearch.org. And on there, we have things called data trackers. There is one for cancer and cancer incidents, and that’s actually one of my favorites because, as you can remember, with the pandemic and COVID-19, we weren’t able to get some of those cancer screenings done early on. People weren’t necessarily able to get into their clinic. And so, we had the question, if we had missed cancer screenings, how many cancers would be found after we actually got those people back into screenings? Would it be a higher amount? Would they be more advanced cancers? So that’s a great public health question that the nation was wondering, and we’re able to use the Cosmos data. Because it’s a longitudinal record of the individual patient, we have the data point of when was their last screening, when that screening happened, what were the results? Did they get put on a treatment? And a progression of how that treatment went.

And so, you can ask these really deep questions of the data set within the community that you might not otherwise, from a single public health institution, for example. And then, on that website as well, we also will do studies that are looking at general public health questions. So, on the top, you can choose to submit an idea. And so, if you have a question in general that you’re curious, say, about measles, and you want to know what the latest rates are, you can submit that question, and the data scientists can take a look at it. But another way is really to communicate with the organization that you get care at. So if you’re questioning or have a query for Cosmos, and you happen to be seen at Nemours, you could absolutely go in and learn. Does your clinician, or does some of the folks in their department, have access to ask these queries in Cosmos? And if so, you can interact directly with them.

Carol Vassar, host/producer:

When researchers approach Cosmos, what’s the advantage of using the data that’s there? Phil.

Phil Lindemann, Vice President, Data and Research, Epic:

There’s a couple things that are pretty, I would say, one of a kind that you can’t really get anywhere else. The first one is that the data is de-duplicated. Most patients have charts at multiple health institutions. So, at the point we’re going right now in Cosmos, almost one in two patients, their chart has been built by multiple organizations that bring their data into Cosmos. So that’s one is, they’re getting this ability to see longitudinal records. Number two is, it looks like the US population. Other large data sets in the US might be only patients that have insurance. So you’re missing a very vulnerable group, or there may be only older patients that are on Medicare. Well, there’s no kids in there. So basically, Cosmos is the best attempt for everyone in the United States who is seeking healthcare; it’s a representative sample of that. And if you look at where Epic is installed, we are in FQHCs, we’re in rural critical access hospitals, and all of that gets pulled together.

So you actually get an appropriate sampling looking across race, age, ethnicity when you start in Cosmos. So those,  I think, longitudinality, and the representativeness, but the other one, for me, it’s a little more nerdy, is that you actually can just begin querying the data instantly, whereas if you look at some of these larger contraptions, you need to bring in a bunch of different software, wire it together, and design some pipelines. Essentially, we’ve done all that hard work for our research community, and they can just literally log in and begin their query. And we even make tools. One of them is called SlicerDicer, which is intended to be very approachable for clinicians and other people who maybe don’t know how to code from a researcher perspective, but they know the clinical questions to ask, and we try and connect those two things with that tool as easily as possible. So yeah, I think that’s the other big thing is you get to go in there and start getting answers as quick as possible.

Jackie Gerhart, MD, Chief Medical Officer, Epic:

Yeah. I’ll add one other thing that I love about the dataset. It’s what we call Mom-Baby Link. Actually, the researcher he was mentioning before made use of this, which is, how do we figure out the things that happen during a pregnancy, and during the sort of before birth, how does that impact after? And a lot of data sets don’t link the mom and the child because they’re two separate clinical records. And being able to see how those are integrated, and how they affect each other, is a really big point of this data set, and of our community. It allows us to study things that maybe would’ve taken a lot longer to study before, or maybe couldn’t have been studied across institutions. So I think that’s a big one. And then, I mentioned a little before about the breadth and depth. Not only does it look like the US population, and not only do you have the ability to jump in and query really quickly into an individual chart, but there’s also this sense of, a lot of researchers can’t have access to all of the different points throughout the care journey.

It’s one thing to not have access to a certain age group, or, say, you’re only a certain insurance, but it’s another thing to say, can we study whether someone got hospitalized after something? Can we study if they had a lab that was done somewhere in Phoenix, and then they went to Connecticut? How do we know how that integrates? And so, the integration of the records that Phil was mentioning at first is really key.

Phil Lindemann, Vice President, Data and Research, Epic:

And an example of that, a mom and baby is a stroke during pregnancy, is actually a very rare event. And a researcher actually contacted Dr. Dysart and said, “Hey, I want to use Cosmos and do this.” So now they’re engaging in that research. So it’s types of things like that that individual data sets will never be able to be large enough to be studying that type of event.

Carol Vassar, host/producer:

And I think a good follow-up to what you’ve just laid out in that regard is the age-old issue of clinical trials finding a representative cohort. This can really make that happen, can’t it?

Phil Lindemann, Vice President, Data and Research, Epic:

Yeah, we’ve actually got a couple researchers that are going to be publishing some papers on that where they’re demonstrating the Epic community, the health systems that use Epic are doing a better job of recruiting diverse populations. So we’re hoping that there’s going to be some data published that there is some hope that we can find representative populations for clinical trials, and start to improve some of the current statistics that you hear about, where it’s usually shifted towards white males, unfortunately.

Carol Vassar, host/producer:

So let’s talk a little bit about that work that Dr. Dysart is doing. Tell us more about that.

Phil Lindemann, Vice President, Data and Research, Epic:

So what he did is his focus was really, like Jackie said, studying those mom-baby linkages. And he hit it right at the right time during COVID. The big areas of his work was during COVID, you had to have the baby. That was an event that was happening no matter what. So these moms would come to the hospital to have the baby, but they didn’t want them in an area where they could be susceptible to COVID during that time. So the goal was get them home as soon as possible, but that came with a potential risk that if you send a patient home too early, what’s the likelihood that they’re going to come back because they weren’t ready to leave? He was able to fairly rapidly demonstrate that these moms and babies, during COVID, were getting sent home earlier, but there was no negative effect on them, that they were staying home happy and healthy, just like they were before.

And he was able to get that research out very quickly so that as moms were coming in to deliver, they, “No, it’s okay. I’m going to go home a little early, but that’s all right.” So that was some of the early work that he did, but he’s basically built this kind of sub-dataset within Cosmos that researchers are now attracted to and are saying, “Hey, can I use the work that you’ve done?” They even will have code snippets that they share back and forth so they can build on each other’s work to rapidly get new papers out.

Jackie Gerhart, MD, Chief Medical Officer, Epic:

I’m going to add that one thing that’s really neat about that work is, Phil mentioned the word early, and I think in medicine we often have an idea of what everyone’s done before, or what should be done, but not necessarily, when we have a crisis, we have the opportunity to learn. And I think in this particular case, we learned, okay, there’s an opportunity to get people home faster. Could that now become more of a standard of practice? What should we study after the fact, that is not just specific to COVID, that actually can advance medicine? And that’s, I think, what Cosmos does is, as we think to advance medicine and how we’re practicing it is more real-world evidence and data around why we do what we do, and how we could do it.

Carol Vassar, host/producer:

And to pull on that string of sharing that Phil had mentioned, and Dr. Gerhart, there’s a related effort that Epic has put out there. It’s called EpicShare, and that’s designed to facilitate knowledge-sharing and collaboration across clinicians, researchers so that members of the Epic community can really learn from one another. Explain how that works, and explain who it benefits.

Jackie Gerhart, MD, Chief Medical Officer, Epic:

Going onto that website, you can go to a few different areas. The first one, which I think some people really like, is called Hey Judy, which actually just talks stories from our CEO and founder who founded Epic in 1979 and grew it from, how she describes it, one-and-a-half employees in a basement in Madison, Wisconsin, and grew to the company that it is today, and the people that it’s serving today. And so, I think find that section particularly fun, but other ones that are really meaningful to organizations and to individuals, so first, there’s a CEO perspective, or there’s a perspective section where it goes through organizations that use Epic, and people that use Epic, their experience with the software. There’s also another section that goes through successes. So it’ll describe a healthcare organization’s, let’s say, success on MyChart, or maybe even success with Cosmos, and outline almost a recipe or a plan, including outcomes, as to how other healthcare organizations can do something similar. Do you want to add to that?

Phil Lindemann, Vice President, Data and Research, Epic:

Yeah. I think I like it because sometimes it’ll just give the ground truth to some things that maybe we’ve been talking to the community about. And we’ll get the CEO of a health system, and they’ll get into the details. They’re getting their hands dirty about something that is very, very real, and very changeable at these health systems. And they’ll read that and say, “We can do that, too.” And what we like to do is pull together these stories from around the community, and then essentially have a to-do list, a build guide to say, “Here it is, this is how they did it, and now you all can learn from it as well.” So it really facilitates that, and we supercharge it with some content on the topic.

Jackie Gerhart, MD, Chief Medical Officer, Epic:

It reminds me: we do a lot of development, but the development is only as good as the people that turn it on. And so, if we don’t have success stories from healthcare organizations, or from customers to explain, “You should use this, and here’s why,” then oftentimes it’s sitting in a silo, and this really allows people to use what they’ve purchased, and actually get some benefits and outcomes.

Carol Vassar, host/producer:

Do any of those stories pop into your mind right now from EpicShare users who have really had good experiences with the tool?

Jackie Gerhart, MD, Chief Medical Officer, Epic:

I mean, I have a couple that come to mind. The first one I remember about Mayo Clinic talking about nursing, and their nurses being able to leverage Epic technology to try to decrease nursing burden and burnout. We’ve also seen a couple of others specifically to clinician burnout, and how do we get to decrease the administrative burden on clinicians, and therefore increase the amount of time that they can spend with patients, or be able to care for them? So I’ve seen those as a good clinician-side example.

Phil Lindemann, Vice President, Data and Research, Epic:

Yeah, I’m obviously thinking about data all the time. So one of the ones is, we had a CEO of Emory talk about their use of executive dashboarding. So basically, imagine rolling up the most important things across the health system on one dashboard, and talked about how we use them, and how his executive team works through this, and how it cascades down to the individual frontline nursing staff so that they all know we’re all going towards these common goals. And it just helps saying, this is an organization that’s doing it effectively, and then others can learn and say, “Hey, I want to do that, too.” Those are some of the ones that I think stick out to us.

Carol Vassar, host/producer:

As we come to a close here today and the Nemours Well Beyond Medicine podcast, I want to ask you how you see Cosmos and EpicShare impacting healthcare, public health, and research moving forward?

Jackie Gerhart, MD, Chief Medical Officer, Epic:

I would say that it all comes down to what the patient and what the human is needing from their health system, and how we can get that best informed in real time. So it’s MyChart being able to let a patient know that it has a rare disease, that maybe they’re available for a clinical trial because they have done searches within Cosmos. It’s also the Growth Charts in front of the clinician at the point of care. So the advancing medicine piece, and the ease of being able to get a better diagnosis and treatment piece, I think, are really key.

Phil Lindemann, Vice President, Data and Research, Epic:

I had a physician tell me the other day, he said, “You know, Phil, patients are going to ask two questions when they go to their doctor in the future. First, they’re going to ask if they have MyChart, and then they’re going to ask, ‘And do you have Cosmos?'” for the reasons of the tools that Jackie’s talking about. So I think that’s our vision, is that this would become a household name that people would probably learn to want to have and form their care.

Carol Vassar, host/producer:

This is exciting stuff. We’d love to have you back sometime in the future to see how all of this is advancing for the betterment of patients, clinicians, and the public health overall, moving forward.

Dr. Jackie Gerhart is the Chief Medical Officer for Epic. She was joined today by Phil Lindemann, who’s Epic’s vice president for data and research. Thanks to both of you.

MUSIC:

Well Beyond Medicine.

Carol Vassar, host/producer:

It’s clear that the future of medicine is being shaped in part by the power of real-time data, collaborative research, and AI, of course, and we’re excited to see where this digital journey leads.

To our listeners, thank you for tuning in. A quick reminder to visit our website, nemourswellbeyond.org, to listen to previous episodes of the podcast, subscribe to it, and leave a review. That’s nemourswellbeyond.org. You’ll be glad you did.

Our production team for this episode includes Cheryl Munn, Susan Masucci, Lauren Teta, and Steve Savino.

Join us next time as we talk with a national leader in raising awareness about achondroplasia and skeletal dysplasia. Her name is Chandler Crews, and she just happens to be one of Nemours’ newest associates.

I’m Carol Vassar. Until then, remember, we can change children’s health for good well beyond medicine.

MUSIC:

Let’s go-o-o well beyond medicine.

Listen on:

Put a face to it.

Meet Today's Guests

Carol Vassar

Host
Carol Vassar is the award-winning host and producer of the Well Beyond Medicine podcast for Nemours Children’s Health. She is a communications and media professional with over three decades of experience in radio/audio production, public relations, communications, social media, and digital marketing. Audio production, writing, and singing are her passions, and podcasting is a natural extension of her experience and enthusiasm for storytelling.

Jackie Gerhart, MD, Chief Medical Officer, Epic

Dr. Gerhart is a family medicine physician and clinical informaticist passionate about using technology to reduce barriers to care, improve clinician well-being and advance patient-centered innovation through research and data-driven insights.

Phil Lindemann, Vice President of Data and Research, Epic

Lindemann brings two decades of experience in data, analytics and health care technology. He leads teams focused on unlocking insights from electronic health record data to drive clinical research, innovation and better patient outcomes.

Subscribe to the Show