For Drex DeFord, Seattle Children’s Hospital offered quite a few selling points. The organization, which includes a 250-bed teaching hospital, research institute, and foundation, is among the nation’s top-ranked children’s hospitals, and is located in the scenic Pacific Northwest. But perhaps the biggest boon was the organization’s belief and involvement in lean methodologies. At Seattle Children’s, continuous performance improvement is a core philosophy that bleeds into every area and is guiding the way through Seattle’s five-year IT transformation. In this interview, DeFord talks about the role of change management in a successful implementation; the importance of not just training physicians, but helping them to see the big picture; what policymakers need to keep in mind when it comes to Meaningful Use; and why it’s critical to taking time away and unplug every once in a while.
- About the organization
- Running a Cerner shop (acute/ambulatory)
- Crafting the next five-year plan
- Engaging the docs
- TPS for change management “Going go Gemba“
- Keeping the troops upbeat
- Leveraging the best of a military career
- Continuous improvement processes
- Managing the MU requirements
- Maslow’s hierarchy of needs (the IS version)
There’s some governance in terms of acceleration control—the speed with which you can do things, based on not only financial resources and human resources, but the reality that we have planned a lot of projects over the next five years. And it’s really difficult to do all of them at the same time.
One of the things I’ve really tried to foster here is the spirit of transparency, and this helps in a lot of different ways. Transparency allows different areas of the department to better understand what other areas are doing.
I tell the guys all the time that realistically, you probably spend more time with the people you work with than you do your family, if you think about how your day is really constructed. And so you have to have fun with those people.
We look at processes and say, ‘How do I eliminate this process all together, or integrate it into another process,’ even though you know it has to exist. That really drives you to try to make it as minimal and unobtrusive as possible, and that really should be your goal.
There are definitely situations with Meaningful Use where when the flag is waved, we have organizations that have already finished the fourth stage of the tour, and we have some that are just getting ready to start. And the guys who really need the help—the guys at the beginning of the race, are the ones that are least likely to get it.
If you don’t make the investments that you need and have people with the skills that you need to run that enterprise architectural layer, and run it well and reliably and rock solid, then you can spend as much money as you want on the middle layer—the applications and electronic medical record
Guerra: Good morning, Drex. I look forward to chatting with you about your work at Seattle Children’s Hospital and Research Institute.
DeFord: Good morning, Anthony. How are you? I’m glad to be here.
Guerra: I’m great, thanks for asking. So why don’t you give us an overview about the organization. Certainly the Research Institute part makes you a little different than most community hospitals—more of an academic setting, I would imagine. But give us a little information on the organization.
DeFord: Sure. So Seattle Children’s is really three pieces: the hospital, the research institute, and the foundation. The overall mission is that we believe all children have unique needs and should grow up without illness or injury. With the support of the community and through our spirit of inquiry, we will prevent, treat, and eliminate pediatric disease. That’s really where our focus is across the board. We have a 250-bed hospital that supports that mission and sees a little more than 300,000 outpatient visits and 14,000 admissions. There are 5,000 employees, and there are probably another 2,000 volunteers that work for us regularly.
We’re the department of pediatrics for the University of Washington Medical School, so we definitely have the academic mission. All of the providers here are faculty at the University of Washington. And the Research Institute is a great place to partner with, in terms of their work in that mission. They are fifth in NIH funding for pediatric institutions and they do great work. We really spend a lot of time and effort integrating or doing our best to integrate the bench to bedside creations that come out of the Research Institute.
And then of course the foundation is a very important part of the organization. When you’re providing more than 100 million dollars per year in uncompensated care to the kids from the Northwest—because the organization really covers Washington, Alaska, Montana, and Idaho, and even some others who come from outside the area to be treated here—you need a strong foundation to raise the kind of funds that you need to be able to run the operation. So it’s a great place to work. I love Seattle. I’ve been here a little over three years and so far, so good.
Guerra: Are there any owned clinics or ambulatory sites?
DeFord: Yes. Actually, and I should know that number right off the top of my head, but I’m going to say there are about 10, and they are spread out from Alaska to all over Washington. And they range from the other side of the lake here in Seattle—where you would find the Bellevue clinic, which is one of our largest clinics that also has surgery capabilities, and then throughout the region, you would find other clinics that are purely ambulatory and serve the patients in those parts of the community.
Guerra: So there are no independent physicians that can refer into the hospital?
DeFord: Oh no, we do have independents that refer into the hospital. I mean, the hospital is primarily subspecialty, so it’s tertiary/quaternary care. We have a strong relationship with our community and the pediatricians and general practitioners in our community who refer kids to us for specialty care.
Guerra: So you have the mix. You have a lot of employed physicians that are on the faculty, and then you also have independents that are sending in their patients and coming into the hospital to see them.
DeFord: Yes, absolutely.
Guerra: And you’ve got the inpatient and you’ve got clinics. So you’ve got it all.
DeFord: We have it all. One of the reasons I came here was because of how broad the mission was; that it included a research institute, and I thought it would be a great place to be able to help and to be able to learn at the same time.
Guerra: Right. Let’s talk about your application environment, starting with the hospital. I’m mostly interested in the clinical applications, so let’s start with the hospital and then we’ll talk about the clinics.
DeFord: Sure. So in the hospital, we run Cerner as our core electronic medical record, and today we have results reviewing with lab and RAD transcribed notes. In the hospital we have an ED, so we have the ED package running. We had been doing CPOE for several years before I arrived. We’re doing CPOE on the ambulatory side also. We have prescription writing, surgical scheduling, and an ED tracking board to manage patient flow. The EMR is integrated with lab, RAD, and pharmacy. We’ve got some direct entry documentation of notes, but it is limited—there is still lot of transcribing going on here. We have GE for radiology PACS. On the ambulatory side, it’s essentially the same. It’s not a separate application environment; they’re running Cerner also.
Guerra: Right. So it sounds like one of those elusive pieces is the physician documentation. Is that one of the last steps in this whole process?
DeFord: Well, we’ve just finished our five-year strategic plan in conjunction with the organization. And so there are things that will come over the next five years and hopefully sooner, but certainly there’s some governance in terms of acceleration control—the speed with which you can do things, based on not only financial resources and human resources, but the reality that we have planned a lot of projects over the next five years. And it’s really difficult to do all of them at the same time, as you can imagine. But coming up, we have bar code med admin, lab specimen collection, and bar coding. We will definitely do e-prescribing in the next probably year or so, and then surgery/anesthesia, although we have surgery scheduling already on board. We really don’t have the entire suite completed, so we’ll go down that road. We’ll also do ED physician documentation and then clinical documentation from a nursing perspective. That’s some of the work that’s on our agenda, including care plans and then also physician documentation. As it stands today, it’s something that we would hope to start later in the cycle, but we’ll see how things go with speed and timing.
Guerra: Do you find there is a big difference in how you need to work with the employed physicians versus the independent physicians when it comes to getting them to enter their orders and do documentation? Do you have to have a whole different approach?
DeFord: Since most of the folks that are practicing in the organization are part of the faculty, they adhere to all of the staff rules. We really have a great chief medical information officer and chief medical officer, and they’re very good at helping us make sure that the work that we’re trying to do is implemented well by the physicians. So while I think we’ve always got some work to do around training physicians—helping them understand why we’re doing what we’re doing, and giving them the big picture on why what we’re asking them to do is important, a very important part of this is also having good leadership support to push us forward on things like order entry.
Guerra: Do you think you’ve got a good rapport with physicians? And I wonder, do you think it’s important for CIO’s to have that rapport or can you use and leverage the CMIO and CMO as an intermediary? How critical is it for you to just be able to connect with them on some level?
DeFord: I think it’s pretty important. And I’d like to believe that I have a good rapport with the physicians, but there are more than a thousand of them, so it’s hard for me to know all of them. But I’ll regularly attend their division meetings, and there are several of them that I’m actually just very friendly with. I meet with them regularly, and certainly I see both them and our researchers at fundraising events and other social situations like that, so I’ve gotten to know a lot of them pretty well.
When it comes to the implementation of the EMR and the use of the EMR, I do rely on my chief medical officer and my chief medical information officer quite a bit, and I rely on my CIS staff. My staff knows the physician users very well too, and that’s a very important relationship to build because the physicians need to know that when they call or when they have a problem or an issue, or if they need someone to help them with something, that the person that really knows the answer to that specific problem can come to them and help them. From my perspective, yes, it’s important that the CIO have that relationship. I think physicians need to understand and believe that their information services leader in the organization has a plan to help move them forward and move the entire organization forward when it comes down to sort of the tactical day-to-day stuff. They’re not really looking for the CIO to help them with those problems. They’re looking to the CMIO or the clinical information systems analyst or others on the staff to help.
Guerra: You mentioned that you attend division meetings with the physicians. That’s interesting, because you get out there into their world; you don’t hang back and just wait for them to bang on your door when there’s a problem. Is that part of your philosophy of engaging—to get out there as opposed to hanging back?
DeFord: Yes. I think that’s a big part of it. It’s not just me; I actually have my chief technology officer out doing the same thing, along with my director of enterprise architecture, and my CIS director. We spend time in our offices, but I joke around in an Ed Marx-ian way that my office is in my backpack. I have an office and I use it, but very often if you drop by my office at the hospital, what you’ll find is that there’s somebody else in my office having a meeting, because it’s open and available. I try to get out as much as I can.
One of the things we do at Seattle Children’s that’s very interesting and something I really love—and was another big reason that I came here—is that we use the Toyota Production System for all of our change management work in the hospital and the Research Institute. Going out there and seeing folks, which in Toyota production speak is called ‘going to gemba,’ was very valuable, and so I try to spend time in gemba as much as I can to see how things are working or not working so that we can provide some assistance on that stuff.
Guerra: I was just thinking that IT can be thought of as a service department to some degree, and many times people just call when there’s a problem. So for the CIO is it a priority to make the staff know about all the good things that are being done and the positive feedback you do get? Is it a danger that in the service department moral gets low because people are just calling with problems?
DeFord: Yes. I think you’re so right, and you have to be very measured on that. One of the things I’ve really tried to foster here is the spirit of transparency, and this helps in a lot of different ways. Transparency allows different areas of the department to better understand what other areas of the department are doing. If the enterprise architecture guys better understand what the business application guys were trying to do this week, it helps them understand how their work impacts the business apps folks. And this is carried over from my military days, but every week, the directors all submit—in varied formats, because another thing I want to happen is for their personalities to come across—weekly activity reports or WAR reports. We compile those WA reports into one big report, and we publish it every week to the entire staff so that everyone can read about here are the things we’ve accomplished, here’s where we have blocking issues, and here’s where there are other challenges.
And I always tell the directors, don’t let the first time that somebody finds out about a blocking issue be in the WAR report. You should already be working it. I don’t take offense if somebody says something about a problem in the WAR report. We need to do it in a very artful way, but if somebody has a problem and they bring it up there, that’s a good indicator that it’s big enough that they need some help with it. So we try to be transparent through that whole WAR reporting capability.
And then when we a have big go-live, certainly we try to have celebrations. When we have people leave the department, we try to have celebrations and say thank you for their work, and when new folks show up, we welcome them. I think that we do a pretty good job of celebrating our successes and not always being negative. Sometimes it’s one of those things where you have to stand back and really remind yourself to celebrate the successes because it is easy to get bogged down with all of the problems.
Guerra: You mentioned that you don’t want to see things for the first time in the WAR report. You seem like pretty fun guy to work for. I can’t picture you getting too mad.
DeFord: Well I guess the guys would say that when I get mad or I get upset about something, it’s got to be pretty severe, and they take it very seriously. I really don’t get mad. I’m trying to think back to the last time that I really got mad. Now I get a little upset; I’ll be disappointed about how certain things go, and I think the staff—especially my directors, my senior directors and Wes Wright, my vice president and chief technology officer—know me well enough to know when I’m not happy about how something is going.
Generally speaking, I think we laugh and joke a lot. We have a lot of characters on the staff, and so it’s easy to make this a fun place to work. I tell the guys all the time that realistically, you probably spend more time with the people you work with than you do your family, if you think about how your day is really constructed. And so you have to have fun with those people. You’ve got to like them and work with them, and so I think it’s a good place to work. I think we’ve built a good team over the last three years and we’re in a good spot to continue to accelerate the success we’ve had here.
Guerra: You’ve mentioned your military experience a few times and I definitely intended on talking to you about that. I’m wondering, some might think that if you bring over some of the dictatorial aspects that could come with the military—giving orders and having them be accepted without any push back, which is what you need in the military—that wouldn’t work in the environment that you’re in now. I’m sure you don’t do that, so how did you take the good from the military but not bring over the things that wouldn’t fit?
DeFord: Well, I wouldn’t say I had a successful military career by being a rather nontraditional military officer. The person that I am today is the person that I was in the military. I didn’t throw things and get mad in the military and somehow figure out how to not have that behavior once I retired and became a civilian, so I’m essentially the same guy. As far as giving orders and not expecting any conversation, even when I was in the service, there are times where you have to—especially in urgent situations—tell someone, ‘I need this done right now.’ And they need to do it because they understand the urgency of the situation. But when that situation isn’t like that—when there’s time to explain to you why this particular action has a positive effect on the entire system, it’s okay to have those conversations, if you’ve got time for them. And I think if you work with a team that you’ve known for a while and they start to understand how you think and how you work and what your motives are, and if you’re transparent and they’re transparent, the need for the conversation certainly lessens at the point that you ask somebody to do something, because they already understand where you’re coming from. That’s really how I was in my military career, and that’s how I’ve been since I’ve retired, so it’s worked out very well.
Now the other thing, to go back to the Toyota Production Systems stuff, we call it Continuous Performance Improvement—that’s sort of our branding of it at Seattle’s Children’s. I’d always been a big fan of the system and tried to use it in various points in my career, even when I was in the military. I had read W. Edwards Deming and others as a young military officer, and so wherever I could create a situation where we would put in place reliable methods or standard work. The military was really built around that concept in a lot of ways—checklists before you take off in an airplane and before you land an airplane. All of that is the creation of a standard of work, and then the constant review of that work to see, ‘Can I take one more of these steps out? Can I reduce the cycle of this particular part of the standard work from 30 seconds to 10 seconds?’ That’s what it was all about, so that was a lot of my work in the military and that’s been a lot of my work here at Seattle Children’s.
Guerra: That’s interesting. So TPS has a lot to do with acceleration control; the speed with which you can do things. You said that before, and now we’re talking about TPS and pace and always trying to be more efficient. So you’re very in tune with doing things quickly, but I would imagine you also agree that there is a limit to the amount of minutes you can take out of a process, right?
DeFord: I think that’s true, but I think saying it like that might frustrate someone’s attempt to continually try to figure out how to do a process better. We look at processes and say, ‘How do I eliminate this process all together, or integrate it into another process,’ even though you know it has to exist. That really drives you to try to make it as minimal and unobtrusive as possible, and that really should be your goal. How do I just make this part of the process almost disappear if I can? I think that relentless pursuit of excellence, that relentless pursuit of making things better, is very important.
Now, you can’t improve everything at the same time. For example, you can’t do all of your projects all at the same time because very often, they draw in the same resources. Toyota Production System allows for a concept called strategy deployment, which I think is very important, and is something that we’ll really just starting to get into here. It’s the idea that you can really take your strategies and drive them down to your daily work, and how those things are connected. But one of the most important parts of strategy deployment to me is laying out the projects and the work that you’re doing, and making sure that in a giant timeline, you’re not stepping on your own toes too many times. And that’s why I think when you look at our work over the next five years with the electronic medical record, the reason we’re not saying we’re going to try to slam the rest of it in within the next year is that realistically, we need to understand that there’s a lot of workflow reengineering involved and there’s a lot of discussion around everything from end user devices to how the application itself gets built. All of that takes time, and it takes time from people that are also doing other daily operations that are working on other projects. And they have a life, and we want that to occur too. They need to have time off and time to go get their education and do other things. So that part of strategy deployment, to me, is very valuable. It really helps us make sure we’re doing things in a very organized fashion.
Guerra: With my previous question, my thinking was around Meaningful Use and all the change that’s being asked of organizations. We can even look at ICD-10 coming down the road. There are a lot things being asked of hospitals and providers in a pretty short timeline, and there is a lot of change. Do you think what’s being asked is reasonable, and what are your thoughts overall on Meaningful Use, ICD-10, and ACOs?
DeFord: Wow, there’s a lot to talk about there. Okay so you did something recently with Marc Probst. Is it okay if I just say something like, ‘What Marc said is my answer to Meaningful Use?’ I mean, I’ve met Mark. We’re professional acquaintances. I’m not sure if he would really know me, but he’s one of the smartest guys in our business. To sort of extend his analogy there, since the Tour De France is going on and I’m a big bike race fan, I think he’s exactly right. There are definitely situations with Meaningful Use where when the flag is waved, we have organizations that have already finished the fourth stage of the tour, and we have some that are just getting ready to start. And the guys who really need the help—the guys at the beginning of the race—are the ones that are least likely to get it, because the other ones have already made investments and they’re already down the road. So there’s definitely a fairness issue there.
But at the same time, I would say I’m not sure I’ve got a better plan. I think that’s a little bit of the frustration for me when I stand back and look at it not really from my own hospital’s perspective, but from the perspective of health care in general. And then I would think Stage 2 might just exacerbate that situation. We’re a Medicaid hospital; almost 50% of our revenue comes from Medicaid, and Medicaid has different rules than Medicare hospitals. In fact, each of the 50 states can do some things in addition to Meaningful Use that are different even from what the federal government will do. So that makes the situation more complicated.
And I think, to go back to Marc’s comments, in the end, it’s not just about getting an electronic medical record in and trying to get Meaningful Use from that electronic medical record. There’s a whole bunch of other things that you have to do really well to be able to make the EMR work properly—security, networking, service desk support, and all the management engineering and workflow adjustments that have to go into the design and the implementation. There’s a lot there, and I don’t know that that’s well considered or well understood necessarily by the folks who’ve put the rules into place and continue to push really hard for them to move forward quickly.
Guerra: I feel like you really understand change management and process. If you were able to give a message or explain how much time these things take and what is involved with any aspect of any project you’re doing right now—if you got to sit and educate some of the people on the policy committee or ONC or CMS, what would you want them to know about why these things take time? How would you best explain that to them?
DeFord: I think I would probably go back to fundamentals. One of the concepts I use here and at every place that I’ve been even from time I was in the military is Maslow’s hierarchy of needs. But you have to think about an information services version of Maslow’s hierarchy of needs. So we’re going to cut that pyramid into three layers. The top layer is knowledge management—the things that you really need to know and want to know to make better business and clinical decisions. The middle layer of the pyramid would be something like applications. So it would be business applications and clinical applications, just to group everything into two big groups. Those are the tools that you use to get your job done. And the base layer of the pyramid would be enterprise architecture, and in that, I would include everything from phones to networks to running the data center to storage—all of those basic and fundamental types of things.
Realistically, if you’re going to run a good information services organization, you have to have the reliability of operations of enterprise architecture or that base layer. Because if you don’t make the investments that you need and have people with the skills that you need to run that enterprise architectural layer, and run it well and run it well and reliably and rock solid, then you can spend as much money as you want on the middle layer— the applications and electronic medical record, and it won’t matter. Because in the end, if the end users can’t get to the electronic medical record reliably, then it’s not worth anything. So you have to really focus on the basics: the infrastructure that you need to run and support an application, and the work that goes into the application—the design, the deployment, and the work flow enhancements and changes.
And those are things where again, in the culture of medicine, you really have to have good relationships with doctors and nurses and allied support so that they understand what you’re doing, and so that they will change their practices to adopt an electronic medical record. That includes a lot of training that is necessary in order to make that happen. And then once you have those pieces in place, there are certainly times where the knowledge management piece gets integrated into the application; for example, to make sure that if you’ve got drug-drug interactions, you get notified about those things. In the end, you’ve got applications in place to help you deliver the work that you need to do. But a secondary use of the data that you put into those systems will help you make better business decisions and clinical decisions and strategic decisions about where to make your next investments and where to put your next piece of work.
And that’s where the knowledge management block of the pyramid comes in. I think having a view of the world that, ‘If you can just get this application in, things will be better,’ is a relatively narrow view, and that change management and the overall Maslow’s hierarchy of needs support of a system like an EMR, all has to be taken into consideration.
And again, to go back to Marc Probst’s point, not everyone is starting from the same place when you look at small physician practices or many rural hospitals. Really basic enterprise architecture infrastructure is not really in place in some of those organizations. Or maybe it’s running, but it’s not running to the level of professionalism that you really need it to run to be able to make sure you put an EMR in place and that it’s secure and private and all the things that it needs to be. So there’s a lot of work surrounding the implementation of an EMR that I think gets overlooked. I think some people take it for granted because that’s the background they come from. They come from organizations that have that good infrastructure and that good support in place, but a lot of places don’t, and I think we have to take a step back and think about how many of those organizations there really are out there in the United States.
Guerra: Because as you were saying the first time, if that application takes a few extra seconds to come up, that physician is gone.
DeFord: Well, yeah and I think that the enterprise architecture piece of this—the underlying foundation of this, if it doesn’t work well, can cause problems in the initial adaptation of an EMR if it doesn’t work reliably and it’s intermittent. Even after you’re using the EMR, it encourages things like unnecessary printing and creation of shadow charts, because people are innovative and if they can’t rely on what you’re providing to them, they’ll figure out ways to set up their own backup systems. And that’s not the intention of Meaningful Use or any of the work we’ve been trying to do.