As health data-sharing improves, but at a slow pace, the nation's largest EHR vendor, Epic, plays a big role in helping to create a clinically interoperable healthcare system. The company controls about a third of the U.S. hospital market share.
In the second part of a series on medical data exchange, Becker's talked to Matt Doyle, Epic's software development team lead for interoperability, about what the company is doing to ease the sharing of patient information among healthcare providers across health systems, specialties and geographies.
Note: His responses have been lightly edited for clarity and brevity.
Q: What will it take to get a level of EHR interoperability where if a patient winds up in the emergency department on vacation in another state, healthcare providers there have easy, digital access to their medical records from back home?
A: Epic's foray into interoperability happened a long time ago, in the early 2000s. It really started with a couple of events that happened, and our community coming to us and saying, "When patients come in, we need the full medical history, we need their current med list, etc., so we can help them get good care, so we can help them make the right decisions to get healthy and get back on their way."
There was one story that sticks out to me very early on from a pediatrician. There was a kid who had some chronic diseases, who just had a rough go at health, and their family was on vacation, and something happened, and they went to the emergency department. And because that doctor did not have the full medical record and didn't know what the history was with this child, they weren't able to make the right choices, and they didn't see how the healthcare they were delivering might have unforeseen consequences. And there was preventable harm, and the patient was injured.
And the pediatrician back home said several times, "If only they had seen this patient's medical record, if only they could have seen her history, they would have known how to take care of her, and it would have been easy. It would have been easy to make the right decision if only they had known."
So it was stories like that that really catalyzed us and our community into getting involved in interoperability very early, long before the discussion at the national level, long before the meaningful use program, long before a lot of the incentives that we talk about at the federal level today. It really began with stories like that, and people saying this is the right thing for patients. It's the right thing for care. How do we make this work?
Care Everywhere is the toolset we built to facilitate treatment-based interoperability. It has been around for a long time. We started work on it a long time ago, but the first customers came live in 2008. And since then, it's grown very rapidly. In addition to connecting to other healthcare providers using Epic, we can connect to others who use other platforms. At the time when Care Everywhere started, other vendors really weren't doing interoperability so it started just Epic to Epic. But as others got ready, we were able to connect to them. And it's been very important to us from the beginning that we follow national and international interoperability standards because we want to make sure that when a patient goes to another health system, their information can follow even if it's on a different toolset.
In addition to connecting together provider organizations — that could be clinics, that could be urgent care centers, that could be hospitals — our community has also been very successful connecting to federal partners. So beginning in 2011, our community connected to the SSA [Social Security Administration], to the VA [U.S. Department of Veterans Affairs], to the DOD [U.S. Defense Department].
Those last two have been really personally interesting to me, because we know that our nation's service members, both active and former, get care at the federal level, but also get care in the private sector. And it's been really important that we connect those together so they're getting the best care possible. And when they have to go to the private sector for certain advanced care, they know what's been happening for primary care at the federal level. So our community has led the way on that for well over a decade and has numerous success stories going on there.
In terms of how you quantify some of these successes, our community exchanges 10 million charts every single day, and half of that is with other vendors. So, unfortunately, there's this persistent rumor that interoperability is just within vendors, it's just Epic to Epic. But the data really doesn't support that story when a full 50 percent of the exchanges are with other vendors, which might be other EHRs. It's those federal partners: the VA, the SSA, the DOD. It's state HIEs [health information exchanges]. It's direct messaging.
Q: Can you explain direct messaging?
A: The core idea is pushing a message to the next caregiver. So maybe I'm an attending in the emergency department or ICU, and you fell off your bike and broke your arm, and I tell you to follow up with your PCP [primary care provider] in the next three days. I want to push a message to your PCP that says, "I saw your patient. This is what we did. Here are the imaging results. Here's the meds that I put them on. Here's why he's coming to see you in a couple of days."
And, that way, before you get to your PCP, they already know why you're coming. They've already got some context. So direct messaging is the standard that allows that. And an HISP — a Health Information Service Provider — acts as a routing go-between. So when I want to send that message to your PCP, really I send it to the HISP and the HISP relays it on to your PCP.
In some ways, it's similar to an internet service provider you have at home when you want to get a webpage to open in your browser. You might have AT&T, you might have Charter, whoever that company is, but they act as a go-between to help connect you to all those different resources online. HISP does a very similar job.
So it's successes like that, especially the federal partners — the VA, the DOD, the SSA for benefits determination — that have made interoperability successful so far. And those numbers — 10 million a day, half with other vendor platforms — really tell a story that there has been success and that things have been progressing and we really have a lot of good things to point to.
One of the other really big success stories is Carequality. Carequality came together many years ago, and there's been a lot of success at interoperability within networks. Some of those networks are geography-based. Some are state government-based. Some are vendor-based.
The challenge that was remaining is, how do we crosswalk all these networks? We really want interoperability to work nationwide, where anywhere you go for care, even if you cross one of those geographical boundaries, your provider can still get the right information. So Carequality didn't invent new technology, they did not invent new standards. Instead, what they invented was policies and a governance structure that allow crosswalking these networks and foster the trust that's necessary for me to share confidential protected health information with actors who are in another network. So we're a strong proponent of it. We were one of the founding members.
If you look at the participants, you're going to find there's a lot of big names in there. CommonWell is a big name in there. Athena is a big name in there. They're not just vendors. You'll also find HIEs [health information exchanges] are there. Technology developers are there. There's all sorts of folks who have really come together to make this platform successful as a way to connect together networks so that you can get information even when it spans across different geographies and vendors.
Their community exchanges 300 million documents every single month. They connect together over 600,000 providers across the country. So the reason I bring all these up is not just the history lesson on how did we get here, but also to say there are a lot of successes out there. There are a lot of things that have gone right. But there's still work to do.
Q: How close are we to interoperability now, and what will it take to make the final push?
A: So numbers are difficult because the way you measure can of course influence the outcome number that you get. I can tell you that the Care Everywhere community — which is the Epic community, the folks that we work with — all of them are connected to Care Everywhere in the United States and are capable of doing interoperability. And I think that's why we've seen the success that we've had. And we bake interoperability right into our product. It's not a separate toolset where you have to think about: Am I going to license Care Everywhere or not? Care Everywhere is just baked into the software, and it's a fundamental assumption of using Epic. So at least among our community, we've seen wild success.
One of the other big successes that we've had is Share Everywhere. So Care Everywhere is provider-to-provider interoperability, and that's all based on open standards. We found that there's a last-mile problem. Sometimes you just go to a place that doesn't have an interoperable EHR, for whatever reason, but you as a patient still want that provider to have your care. A great example of this might be physical therapy. So if you go to a standalone physical therapy clinic, they were not covered by the promoting interoperability program. They had no incentives to get an interoperable EHR, so they may not have one. With Share Everywhere, you can log into MyChart, which is your patient portal. Basically on your phone, you can pull up a copy of your record, and you can generate a one-time-use share code, and then tell your physical therapist, "Go to Share Everywhere, punch in this one-time-use code," and then he or she can actually see your medical record, they can see what care you've received in the past, what meds you're taking, your allergies, your problem list, imaging results, whatever the case may be.
What's really exciting about that, I think, is twofold. One, it solves the last-mile problem. In a perfect world, this would be baked into every technology system, and providers would get it right there. But there is this last-mile problem, and this is a good way to still make sure everyone that you as a patient choose can still have access to your record. So that's really big.
And then second is it gives you as the consumer, you as the patient, choice. You can share your record through Share Everywhere really with anyone you want. There are some folks I was working with recently who said it's not just about providers like a physical therapist, maybe I want to share this with my social worker. My social worker is helping me work through not just healthcare problems, but other social dynamics. And I want them to be aware of what's going on in my medical life so that they can help me navigate both together. And so it gives you a lot of autonomy as a patient to choose who do I want to share my information with.
Q: What are some of the other challenges to getting to interoperability?
A: So I think one of the best opportunities is to make sure that vendors are adopting national and international standards toward interoperability. What that does is it ensures that we have the same language. If you and I want to communicate today, that works because we both speak English very effectively. But if we're not speaking the same language, if we don't have the same assumptions about the information we're going to share, then it's not surprising things might break down.
So there's lots of standards out there. The most successful by volume that we see right now for treatment-based interoperability is the CDA document. So that was written into many of the meaningful use programs. It's still very high-volume exchange, very effective. And it's one of the workhorses of interoperability.
One of the other ones I'm sure you've run across is FHIR. It's a newer standard, and it's very effective at certain things. It's effective at exchanging small, bite-size information, where CDA is very effective at bigger datasets. So if I want to exchange your entire medical record, CDA is one of the most efficient ways to do that. If I just want a little bite size, like one note out of your medical record, FIHR might be a better fit.
And so it's not so much that there is one standard to rule them all. We as an industry have gravitated toward these standards, and we need to make sure that all the vendors have implemented them. So I think that that's one of the biggest things that we need to make sure is in place for interoperability to be successful.
Q: How else can interoperability improve patient care?
A: One of the areas that I personally find really exciting is that interoperability is not just transactional, move this data from here to there, move it from me to you. The most effective interoperability is when it actually connects to a workflow and helps a provider or a patient get better care.
So I'll give you some concrete examples. Imagine a patient comes into the emergency department, she's having the worst migraine of her life, and you're the attending so you decide, I'm going to order a head CT and see what's going on. The Epic system can actually alert you as you're placing that radiology order: "Hey, your patient just had a head CT at the hospital down the street two days ago. Before you repeat that test, would you like to read that radiologist's report because maybe you don't need to do this again?"
So that's good because it saves time. And time matters. If you've ever had a medical emergency, you care about speed. It saves cost, and as an industry, we're always talking about how do we improve costs. And it's better for safety. You don't necessarily think about it, but irradiating patients, to do imaging over your lifetime, can add up to something significant. And so it's better for safety to not radiate people. And so that's a concrete example where interoperability can be dovetailed into the workflow at the right moment to help providers and patients have a much better healthcare experience.
Another concrete example is care gaps. So care gaps are when there are certain milestones that you should be doing on a regular basis. Maybe you're a diabetic, you're supposed to get an HbA1C [hemoglobin A1c test] and an eye exam every year. Once you're over a certain age, you should be getting a colonoscopy every few years. So there'll be certain milestones that need to be done. And your PCP might be tracking all of these and then calling you to say, "Hey, it looks like you're overdue for this test. Can we get you in to take yours?"
The reality of our healthcare system is that patients sometimes get that care in other venues. And so it might be you already got that diabetic eye exam somewhere else, and your PCP doesn't know about it. And it's a waste of time and resources for your PCP to be calling people trying to schedule duplicative care.
Care Everywhere can actually not just synchronize data, but then check off that list of tasks that need to be done so that duplicative care is cleared out of the way and providers and support staff are only spending their time calling patients for the care that is actually needed. So that saves provider and staff time, so that's good. It reduces unnecessary testing, which saves money. So that's good. And you as a patient are only getting your time consumed for the things that are actually valuable to your healthcare, so it benefits you as well. So these are examples where interoperability for treatment is more than just — in the moment can I move data from point A to point B? But it's ways that Epic and our community have baked it right into the workflow that healthcare professionals do.
Q: How else is Epic working through any obstacles that are preventing us from getting to interoperability?
A: There are definitely areas that we can work on with our community to improve. How do we get you, as an individual, access to your own information and to be a participant in the healthcare that you're receiving? Because if you're actively participating and engaging in your care, you're more likely to actually follow through on medical instructions, follow through on the things that need to be done, and hopefully, have better outcomes.
So we've done a lot of work on MyChart, which is our patient portal, and that focuses directly on not just for you to see your data, but for you to do things. If you want to schedule an appointment to go in and see your doctor, you do that on your phone and MyChart. If you need to refill a prescription, you do that right on MyChart. If you actually have two patient portals, because maybe you get your acute care here and your ambulatory care somewhere else, you can connect those together so that you have a unified view across the multiple venues where you've received healthcare. You can also get health coaching tools in MyChart as well. We call this Care Companion. Coming back to the diabetic example, it's important that you're exercising regularly, you're taking your medications, you're eating healthy, and so MyChart can actually prompt you to be doing the right things at the right time. And so it's more than just how you see information. It's how you interact and stay on top of your healthcare.
So there's continual improvements. And I don't think there's a moment where you say it's done. There is no moment where you say healthcare interoperability is done or patient engagement is done. Instead, it's every time we make progress, you then turn a corner, and you can see a little bit further down the road, and you see and get ideas about how do we unlock the next step of improvement and the next step of high performance.
So I don't think it's ever going to be done. I think instead what's going to happen is it will become more sophisticated. We'll find new ways to be innovative. We'll find new ways to weave it into the moment where it's really going to impact the health of you and all the people that we care about.