During the six years he spent as an officer in the US Navy, Daniel Barchi endured quite a few “white-knuckle moments” that tested his skills in crisis management and leadership. But without that experience, he might not feel so comfortable guiding Yale New Haven Health System and Yale School of Medicine through a large-scale infrastructure project designed to unify IT operations across the system, facilitate better data management, and reduce costs. In this interview, Barchi talks about rolling out Epic, the importance of clinician buy-in and solid leadership, and how he is leveraging his dual CIO roles to bridge the gap between two organizations. He also discusses the value of IT rounding, the challenges of working in academics, and why he wouldn’t trade his job for anything.
- About Yale New Haven Health System
- Managing a dual CIO role
- Leveraging Epic to align the health system & school of medicine
- Going from 350 Epic users every day to 3500 users
- Starting in the physician practices – smaller, more controlled environments
- Importance of having a strong project manager
- “EMRs are really just tools that standardize the way we practice”
LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO DOWNLOAD THIS PODCAST AND SUBSCRIBE TO OUR FEED AT iTUNES
When it comes to clinical tools and working with patients, we’re trying to create as many integrated solutions as possible. So within both of them, we’ve been able to align quite well and even across both of them, primarily driven by our new EMR, we’ve been able to align.
It’s easier to start in physician practices because they’re smaller, they’re more controlled, and you can build some specialized content for them. And we know we wanted to get Epic up and running as quickly as possible.
The physicians and the physician practices that don’t make the investment in that time upfront really struggle; not only do they struggle, but it takes that much longer to get back to full productivity. It’s painful, it takes extra time, and quite frankly, it’s a resource drain on the team.
It’s allowed me to focus on the long-term vision; what we’re going to expect when we go live, what we’re going to see, how to prepare our physician workforce, how to prepare the executive’s response to issues that we’re going to have — because we always will have issues — and then more importantly, how we’re going to use that tool going forward.
Only when we start using those tools in integrated ways that start giving feedback about what’s working well or what’s not, or building best practice alerts that are based on best evidence medicine, can we really change what we’re doing in getting the information we need out of the system to drive accountable care and integrated management of our patients.
Gamble: Hi Daniel, thanks so much for taking the time to speak with us today.
Barchi: Absolutely, I’m happy too.
Gamble: Let’s start off by getting some information about Yale New Haven Health system. I know you have the three hospitals but why don’t you give us a little more information about what you have in terms of clinics, and tell us about the School of Medicine?
Barchi: Sure, it’s actually a really unique setup, although it’s not that unique to academic medicine. Interestingly, the Yale New Haven health System and the Yale School of Medicine, Yale University, are two entirely separate organizations although I am the chief information officer for both. In 2010, I was hired as the CIO for both the School of Medicine and the Yale New Haven Health System to bridge the gap between the two organizations because the clinicians practice in the health system and are faculty at the School of Medicine, and together we’re rolling out a $250 million electronic health record. So the administration in both organizations thought it was really important to have leadership that spanned both of them.
Within that setup, we’ve got the three hospitals of Yale New Haven Health System: the 1,000-bed Yale New Haven hospital, the 175-bed Greenwich Hospital, and the 500-bed Bridgeport Hospital. In addition to that, we have a wholly-owned medical group with about 200 physicians in it and within the School of Medicine we have the 800-physician Yale Medical Group. Together they make up about 1,500 employed physicians and the hospitals serve a faculty of about 7,000 physicians.
Gamble: Okay. So with having both of those roles, CIO of the health system and the school of medicine, do you find that it is like having two separate roles, or do you try to keep them as integrated as possible?
Barchi: Certainly they’re two different organizations with two different leadership structures and I sit on the executive side of both of them and have to meet the needs of them as individual organizations, but when it comes to clinical tools and working with patients, we’re trying to create as many integrated solutions as possible. So within both of them, we’ve been able to align quite well and even across both of them, primarily driven by our new EMR, we’ve been able to align as well.
Gamble: Okay, and I definitely want to get into that. But I was reading about that the big IT infrastructure project at the Health System and School of Medicine that includes the EHR, a clinical trial management system, and a clinical research data repository; so why don’t we start off with the EHR implementation. This is Epic, correct?
Barchi: It is.
Gamble: What phase is that in right now?
Barchi: We kicked off the Epic project in October of 2010. We went live with the first physician practice in October of 2011, which is about six months ago, so we’ve been live on Epic in our ambulatory practices for about six months. We have about 25 practices and about 180 physicians live and using Epic every day. We’ve had about 13,000 patient visits so far. The numbers are what you would expect them to be — about 90 percent physician order entry and about 75 percent of the encounters are being closed the same day. We’ve had pretty good numbers on dictation and we’ve got an embedded dictation system. Our physicians are able to get dictations back in the EMR where they dictated them within about 35 minutes.
So generally it’s going quite well in the ambulatory side. We are about 20 days from going live in the first hospital. We’re starting with Greenwich Hospital, and once we take Greenwich Hospital live, it will go from about 350 or 400 consecutive Epic users every day up to about 3,500 consecutive users. So within the next month, we’ll make a major step forward. We’ll have the 1,000-bed Yale New Haven hospital go live in February of 2013, and then the final hospital, Bridgeport Hospital, will go live on June 2013. And sprinkled throughout there will be go-lives for Yale Medical Group practices, Yale New Haven Health System’s Northeast Medical Group practices, and physicians in the community who are purchasing Epic from us.
Gamble: As far as starting on the ambulatory side, what was the strategy there?
Barchi: With most of these large enterprise go-lives, the build is about the same, but it’s easier to start in physician practices because they’re smaller, they’re more controlled, and you can build some specialized content for them. And we know we wanted to get Epic up and running as quickly as possible. We thought that the timeline for our first hospital would be about 18 months and we could probably get our first physician practice up in about 14 months, so that’s why we started with the physicians.
As it turned out, we were able to do it in about 12 months and it’s been good. Our Epic team is about 150 people who came from all walks of life — mostly from the university and from the health system, but they were finance people, some IT people, some revenue cycle people, and some clinical people. Not all of them were builders and developers, so the process of building Epic on the ambulatory side and maintaining it as an integrated 24/7 system was good for many of these people because they’re much better prepared going into the hospital environment.
Gamble: Yeah, that makes a lot of sense. What has the experience been like for the users so far — have you heard pretty positive feedback as far as their experience with Epic?
Barchi: I’ve turned up a large enterprise EMR and implemented Epic in another health system on a slightly smaller scale — it was about 1,000 beds and 700 physicians total, but I’ve seen many of the same things. The physicians and the physician practices that invest their time upfront — not simply going through eight hours of training but really doing development, some planning on their own about how their office is going to use it, building some of the workflows and templates and practicing ahead of time — when go-live happens, they do pretty well. It’s a tough couple of days to get in the swing of it relatively quickly, and within four to six weeks they’re back to full productivity and they’re starting to see the benefits and being able to order and get results in the system and share data with their colleagues across the health system.
The physicians and the physician practices that don’t make the investment in that time upfront really struggle; not only do they struggle, but it takes that much longer to get back to full productivity. It’s painful, it takes extra time, and quite frankly, it’s a resource drain on the team because they need more people sitting with them shoulder-to-shoulder showing them how to use the tool or asking questions that got covered in training, or that they could have figured out by doing more practice. And so I would tell anybody going live with a large enterprise EMR that the investment of time upfront by physicians is definitely worth it.
Gamble: The experience you had — this was when you were with Carilion, right? In Virginia?
Barchi: That’s right.
Gamble: That’s also a large system. How did the implementation compare and how did the experience benefit you in your role, just having that experience with Epic?
Barchi: I’ll tell you what, in both of these go-lives I worked with very talented people and had very, very sharp vice presidents who were the Epic project directors who really were the project managers and the leaders of the project itself. In both cases, as the CIO, I served as the executive responsible for the project, working on the budget, on the rollout plans, on the coordination with hospitals and with physicians. Doing it within Carilion gave me good experience because I saw what works well, the kind of attitude you need, the kind of leadership — both from the hospital and at the board level — and was able to take that experience and bring it to the Yale New Haven Health System and School of Medicine. With Lisa Stump, who is the vice president and Epic project director, running the project and making sure we’re meeting the day-to-day build, testing, and rollout goals, it’s allowed me to focus on the long-term vision; what we’re going to expect when we go live, what we’re going to see, how to prepare our physician workforce, how to prepare the executive’s response to issues that we’re going to have — because we always will have issues — and then more importantly, how we’re going to use that tool going forward.
I’m a big believer that these EMRs are very valuable, but at the end of the day, they’re really just tools that standardized the way we practice. Only when we start using those tools in integrated ways that start giving feedback about what’s working well or what’s not, or building best practice alerts that are based on best evidence medicine, can we really change what we’re doing in getting the information we need out of the system to drive accountable care and integrated management of our patients. So the experience I got at Carilion has really put me in a spot where I can see not only the day-to-day but where we need to look forward as well.
Gamble: And then in terms of putting together the Epic team, you were talking about 150 people — is that something that’s put in place as you move on to the hospital rollout just to make sure you have all the support that is necessary?
Barchi: That’s right; 150 people sounds like a lot of people, but when you break it down and you realize that you’ve got a team focused on physician practice go-live and a team that’s building inpatient content, and we have about 30 trainers and we have an integration testing and workflow team, you really realize that it’s not that many people. The reason why we have 150 as well is because we’ve got practices that are going live even while the hospitals are going live. So it’s an integrated team focused on both inpatient and outpatient. But it’s really a team that we have around for the three or three-and-a-half years of the project, and then we’ll work those skilled people and those rolls into our larger organization.