EHR Standardization with Rusty Yeager - Episode 38

Hey John, looking forward to the conversation. Thanks for having us.

Yeah, because they have such great insights. But for those that don't know about you, tell us a little bit about yourself and Encompass Health.

Okay, I've been at Encompass Health for 21 years now. I separated from the Air Force right at the 2000 tech boom, wanting to try public healthcare. A friend of mine, many of you might know him, Randy Carpenter, asked me to come over here to Birmingham, Alabama, and said he had a job for me as Director of Infrastructure. I said, 'Randy, I'll come over there, but I'm just doing that for the interview practice. I don't really want to move over there.' When I moved over here, I fell in love with the mission, I fell in love with the community, and 21 years later, here we are. A little bit about Encompass Health: we're a little different. We own and operate 151 inpatient rehabilitation hospitals. 55 of those are joint ventures with folks like acute care hospitals, academic medical centers. Every year we discharge about 203,000 patients, with about $4.2 billion in revenue, and we serve about 31% of the Medicare inpatient rehabilitation patients across the United States. Recently, over the last two years, we got some great accolades: Modern Healthcare's Best Places to Work in 2021 and Fortune's Most Admired Company in 2022. So some really great stuff going on here. I got a great team; they keep me out of trouble. I was mentioning earlier, I don't do it, I just talk about it. So let's talk about it.

Yeah, it is different. It's probably good to talk about what an inpatient rehab hospital does, because a lot of people don't understand that. We do PT, OT, speech for patients that need inpatient rehabilitation post their acute care stay. Typically, these patients have had a stroke, brain or spinal injury, dysfunction, neurological condition, and they need care post their acute care stay. We get involved with those patients before they come to us. We evaluate their ability to do inpatient rehabilitation care because they've got to be able to do it for three hours a day while they're in our hospital, according to the guidelines. So we need to make sure they're good for the program. From an IT perspective, we have sales people, clinical liaisons that go in and evaluate the patient, capture that all electronically, and pull it into a physician view so the physician can evaluate and make sure the patient is eligible for inpatient rehabilitation care. So a lot is going on two or three days in advance before our admission. We are less complex; we have a single soul focus, which is to get that patient back home to the goals they want to have. We don't have EDs, we don't do surgeries, but we are 24/7/365 like any other hospital, and we have doctors, nurses, pharmacists, and the complement of that. We typically outsource lab and radiology services, but they interface back into our system. We have a full EMR, which we'll probably spend some time on. One of the things we do that is probably different from a lot of acute care hospitals is we're intensely centralized. We have a single database for our EMR and a single IT support organization for the company. The company has about 33,000 employees. At any one time, typically about 12,000 to 13,000 people are working in our EMR in a single day, and we do that from Birmingham, Alabama. So if you've got a problem in our hospital, you call us 1-800-CALL-ITG, and we fix their problem from here to the extent we can. We do everything with the exception of physical hardware changes or modifications from here. Our ability to scale that has been a great benefit for the actual users of IT as well as for the economics of the organization. IT has gotten so specialized that a lot of times our hospital CEOs will ask us, 'I need an IT person.' Well, the IT person you get today can't do all the things in that hospital that you would like them to be able to do because this thing is so scaled out that they wouldn't have the abilities to do that, because it's run here. Sometimes we will allow that, but typically all of it is run out of here. In fact, we have what we call IT liaisons out in those hospitals that are actually full-time employees, but they volunteer to serve as the go-between. Some of them are therapists, some are controllers of our hospitals, so they're our eyes and ears out in the hospital. We do that with about 198 IT people and about 56 clinicians on our clinical EMR team. I probably spend more time on technical challenges than the average acute care CIO would. One of the reasons is we can't buy anything off the shelf. We are the largest inpatient rehabilitation provider, and a lot of our other rehabilitation competitors aren't scaled up to be able to do the things we're doing from an IT perspective. So there's not a lot of business activity in inpatient rehabilitation. When people come to us to try and sell us something, we have to make sure we can configure it to our particular workflows and requirements for inpatient rehabilitation.

Yeah, we really had to. The ability to scale was challenging. It was faxing: you find the fax, get it to the physician, have them sign it, then scan it into the record. When I first took over as CIO, I was like, 'Can't we just digitize this?' It took us a little bit of time, but we were able to build an incredible—I call it the killer app. We've got it on an iPad, on PC, and our clinical liaisons out in the field are taking a lot of data that then begins the care if the physician is able to accept the patient. All of that data starts before we even admit the patient. From an interoperability perspective, I think we've all been there: if you build it specifically for a workflow, then you get a lot of value out of it. If we're moving ADT around, it's challenging to get that, and we'll get there over time. But it was a business and clinical necessity for us to build something specifically for these workflows.

Yeah, it's a really neat story. About 2009, the HITECH Act was in the offing, and we started thinking: 90% of our patients come from an acute care hospital, so they're going to have an expectation at some point that we're starting to do interoperability. This was before interoperability was really in the headlines, but we knew that would be something. The prescient leadership of the organization knew we needed to do some of that. Having done that in my past career in the Air Force, meaning rolling out EMRs, I was pretty scared. I was thinking, 'We're going to roll out an EMR to this many hospitals?' At that time we had 98. So I had some trepidation about that, but I also reflected it as something we needed to do. The organization did an RFP; we had several folks respond. We tabled it for a little while, then got back to it. There were three folks that were interested. Ultimately, we decided on Cerner because they had some experience in inpatient rehabilitation with what used to be called the Rehab Institute of Chicago. That was interesting to us because, as I mentioned earlier, things aren't off the shelf for us. We knew we were going to need to do a lot of customization and configuration to meet our particular requirements. So we went with Cerner, and we call it Advancing Clinical Excellence Through IT—ACET. That's been a really good rallying point for the team; we're acing it. We rolled out to a de novo hospital that we built in Northern Virginia in June of 2010, and it went pretty smooth. Then we rolled out to two more de novo hospitals; it went pretty smooth as well. We were feeling pretty sporty about it. This is a new hospital, a lot of our team members are new to inpatient rehabilitation, and it was working pretty good. Then about 18 months later, we went to our first existing hospital. We went out to a hospital in Phoenix, Arizona, that had a really great CEO and began our implementation there. We hit some challenges. We've got a hospital that's been in existence for 20-something years, and they know inpatient rehabilitation better than a de novo hospital. My small team of clinicians know, and they've got workflows that have been in existence for years. That is very typical on the acute care side but less typical for us at that particular time. So we began working with them, and I got a call one day: 'Rusty, you need to come out here.' The CEO of that hospital was a friend of mine. I said okay, I'm out there. When I got out there, we had some challenges. We got together, and I said, 'We've got to get together here.' I got the hospital clinical team and my clinical team together, and we came to an understanding that we're all in the boat together, and we're going to be listening and build this thing for you guys. We did that, and it went over well. Then about 18 months or two years later, we decided to go enterprise-wide. Over that time, we had 98 hospitals. We planned on a five-year rollout because that was our ability to pay for the EMR as well as the change that was happening to the whole organization. We went on schedule, on time, and on budget, and it was working out well. Then we started thinking, 'We've got this bolus of data that nobody else has. What can we do with that?' I sat in a meeting with one of our senior VPs, and he said, 'If we could just affect this acute care transfer ratio, that would be nice. It would be great for the patient and great for the company in terms of our outcomes. What do you think we can do?' I'm thinking to myself, 'We've got all this data. Why don't we mine this data and see what the things are that people can't see but the data can see?' We approached Cerner about it. They were interested; they had the data scientists. We didn't have data scientists; we had the clinicians that know rehab very well. Let's put them together. They ended up building an algorithm we call REACT—Reducing Acute Care Transfers. Acute care transfers are not a good thing for us or for the patient. We went after that mission. We spent probably 18 months crafting this, then rolled it out, and we've been running it ever since. It has had a great impact on those acute care transfers. The beautiful thing is, with the number of patients we discharge a year, if we can affect 1% of that, that's a lot of people. At the time, it was probably 1,500 people; now it would be 2,000 people. That was our first real great outcome from using the EMR. We put it in these hospitals that were completely paper, with the exception of an older pharmacy system. In 10 weeks, they went from paper to one morning they walked in and it was completely electronic. Then a year or two later, they were getting workflow observations about a patient's condition that would have been hard for them to see even in an EMR, much less a paper-based record. So it was really about process: standardize the process, which Encompass Health has done really well, then automate that process, then we can scale it. Once we scale it, we can get data at scale, and then we can use our clinical and business knowledge to infect process improvement. The best thing about that has been that we're able to build it into the workflow so it becomes natural. This is just how we do it, like that old disco song: 'This is how we do it,' and we do it that way everywhere. We've got a single database; all we really change is configuration issues. So we can report business and clinical measures every day across that whole continuum of hospitals and episodes of care, and drill into each of those with a system we built called Beacon, which is essentially a BI system that has worked well to mine the Cerner data to give us visibility into what is happening everywhere.

Yeah, I gotta give it to the team. They're just incredible. We've been working together for seven or eight years now on it, and it's really come together.

Unfortunately, we did not get any government money. We paid for this on our own, and we felt it was the thing we needed to do to support healthcare in the community. In fact, it has worked out well.

Yeah, so the six to ten hospitals that we build, and if we do joint ventures, part of our joint venture arrangement is that we're going to get our system so that we can manage every hospital in a standardized format. It works great. Almost three years ago, I got assigned the responsibility for design and construction in Chicago. How does that go together? Well, it's really processed too. We built a great team, we've standardized on the hospital rooms, etc., and we replicate that. This year we're going to open nine hospitals, and nine next year. In fact, we've gotten it so standardized that we're using some prefab construction that we just deliver on 18-wheelers, and there comes a room, a hallway, another room, and just chunk them, set them right out. They work really good for bed addition. In a lot of our communities, we're at capacity, and that gives us a very quick turnaround on being able to add capacity. The EMR is standardized as well, and we put it in from day one, so you don't know any different. By day two, we're able to reflect on the management and clinical things that are going on in that hospital with everybody else. So as a hospital comes on board, we can compare it to somebody in a similar position and then affect change where we need to.

Yeah, it's beautiful.

Yeah, we need to modernize a couple of things. Our patient revenue cycle system needs to be modernized. It is actually on 151 databases and built specifically for us, but it is time to modernize that. We are on a pilot project with Cerner to put in their revenue cycle system. We'll put it in at one hospital as a pilot, and we have a centralized business office as well. We'll carve out a little piece of that business office to evaluate that. If that is successful, which we think it will be, it will provide us some good opportunities in the integration between Millennium and, in this case, the pilot will be on Soarian for registration and other things. The hospital folks, I think, are going to get a really good benefit from having to log into two different systems to admit patients, so we're pretty excited about that. The ability of some of the workflowing we've seen in the system will help our business office as well. We're also begrudgingly going more to the cloud. We're working on Office 365. I say begrudgingly because it's challenging from an operating expense perspective, and that's really the only reason we've been slow to go in there. We like the capabilities, but it's costly, as any of the CIOs out there will know. We've got about a third of our hospitals on Office 365, so we'll be doing that the rest of the year. We're also looking at our ERP and supply chain systems in the future. So we're in a little bit of a rebuild mode to set us up for the next few years. On the EMR side, we're going to continue to work with Cerner in the post-acute innovation center. In fact, we did two more algorithms: a readmission algorithm that used our home health data from an organization we used to own, as well as our data, and built a readmission algorithm. One that is super great for our patients, and we just really are loving it. It's only been out for about six or eight months, and it's a fall prediction algorithm. We used to use a fall prediction that was probably acute care based. Well, our patients are different. We're getting them up three hours a day, and they have a propensity to fall because they're moving around. So we thought again, we've got all this data, so we mined that data and put together a fall prediction algorithm with Cerner's help, with the data scientists and our clinical teams, and we've seen a great reduction in fall prevention. The other thing that is super important to all that kind of activity is to build a program. You can't just throw that IT out there and expect it to work. It's really challenging with academic medical centers that really want to get to the bottom of it. So part of our impetus for building what we call the Clinical Beacon was to show here's the things that are driving this algorithm in the workflow. We took our ClickView environment, hooked it up into Cerner on SMART on FHIR. Most EMRs are challenged today to show trending and other things that are pertinent for that particular organization. That gave us the opportunity to really give the clinicians trending. When you walk in in the morning, here's all your patients in your hospital, and these top five are the ones that are sicker than the others. Concentrate on them first, and then you're in a better position for everybody's quality of care. So that relationship with our innovation center has really paid dividends.

Yeah, there's two things: the IT staff and the clinical staff. On the IT side, we start with a mission. The mission is: if you take care of the caregivers, they'll take care of the patient, and everything else will take care of itself. When you get the mission, you can build culture, and that culture will breed self-accountability. This team that I've been working with is incredible. This year, we've got about 5% turnover in IT. Typically, we average about 4.75%, and our average tenure is 13 years. That includes a lot of support organization people that will kind of come and go, so we've got a long tenure in the back office of IT, which really enables us. We know this business, so we're pretty compelling there. In the hospitals, of course, everybody's doing everything they can to chase this limited resource. We've been using some cloud technologies that enable a much smoother interface to our potential employees, as well as moving all of their data into the system as soon as we get it. We're doing role-based stuff in our EMR so that if somebody shows up to work, they can start working that day with their training. So everybody's running at that one pretty hard. We're finally starting to see a little bit of a turn down in it, so we're happy about that.

So I'll say it and then I'll figure it out later. Learn to ask for help. That's kind of hard to do. As I came up through my career, as many of you guys and gals out there did, I got a lot of advice, and it was good advice. Some of the best advice was, 'What do I do in this particular situation?' That's good because if it's good advice, it really moves the organization forward, especially as you're working up the ranks. Well, as you get up to the pinnacle of an IT person's existence, you're at the CIO level. You're not getting much advice because you're the guy or gal that's supposed to know everything. So what do you do? You've got to learn how to ask, and it's counter to what you've done most of your career. You've gotten good advice, but it's counter to asking. You gotta ask. That was the best advice I've got. I struggled with it for a long time. I'm a lot more comfortable with it now, utilizing my CHIME network and other networks, using the vendors for things because they see a lot of things that you don't see. You see your stuff; they see all kinds of other stuff. So learning how to ask for help, ask for advice, has been the most compelling thing that I've gotten advice about. It took me a long time to internalize that, so I wholeheartedly recommend it. I'm a pretty shy guy, so that was challenging.

Yeah, and we use it a little more discreetly too. I call it, 'We got to get everybody in the boat.' So we come into a room together and say, 'Okay, here's what we're going to do. What are your thoughts? What are your thoughts?' 'Oh yeah, I didn't think about that.' It really circles up some of the threads that, if somebody were to pull on them, could get the project to come apart. So we say, 'Okay, everybody in the boat,' and we kind of look around, 'Yep, we're all on the boat, so we're in it together.' Just like we talked about in Phoenix: we're all in this together. It really helps bind the team, and we build mutual respect that way, and ultimately the accountability goes along with that.

We hire remote, no problem for IT. In search of a couple security people. I hear the coast is great in Bama though.

Thank you.