Most people would be surprised to learn that UMass Memorial Healthcare was a late starter in the EHR game. But while the organization is behind the curve in some respects, it’s leading the way in others, boasting innovative initiatives such as a diabetes-focused portal and an eICU. In this interview, CIO George Brenckle talks about what’s like to play catch-up, how UMass become a best-of-breed shop by default, his three-pronged approach to ICD-10, and the momentous task of cleaning up data . He also discusses his social media strategy, and his “backyards without fences” leadership philosophy.
- 3 components of ICD-10 — “The technology piece is probably the easiest.”
- Setting the stage for analytics
- “Data is never useful if it’s not used”
- Cleaning up dirty data
- eICU bunker
- UMass’ groundbreaking diabetes portal
LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO SUBSCRIBE TO OUR iTUNES PODCAST FEED
You’re looking at this huge increase in the detail and the number of codes you’ve got, and while you may have a clinical note that would support appropriate coding in the ICD-9 world, it may not have the information you need and the detail you need.
We don’t view ICD-10 here as an IT project. There’s an IT component, but there are a lot of other clinical and operational components necessary for us to be successful in that space.
Data is never useful if it’s not used. You have to start using your data, and based on using it, you find out what’s wrong with it, and then based on that you improve the quality of it.
While in some areas we’ve been behind, in other areas there’s some real cutting edge stuff they’ve done here.
Here’s the number of diabetics we have in this country. Here is the number of endocrinologists you have. How do you extend care and spread your reach so that you’re really helping patients manage their disease?
Gamble: Having this type of focus on the project and having tight deadlines — how does that affect some of the other major initiatives like big data? I imagine that’s pretty challenging having all these big priorities at the same time.
Brenckle: Well, one big priority that’s also on top of us right now is being prepared for ICD-10, and so we are working with all of our vendors on that. I look at ICD-10 as having three components: there’s a structural component, a process component, and a content component. The structural component looks at whether all of your information systems up on the appropriate release level to support ICD-10. Does the system support the ICD-10 codes, and more importantly, will it support both ICD-9 and ICD-10 so you can make that transition? We’ve probably got an upgrade of some of our core systems happening just about every three weeks between now and the end of the year so that we will basically have ICD-10-compliant software.
Then you’ve got the process component, and that deals with coding. Do you have the processes in place, both on the ambulatory side and in the hospitals so that you’re doing your coding appropriately? Have you trained your abstractors and coders so that they can be efficient in coding in ICD-10?
And the third step is on the content side, does your clinical documentation support the level of granularity that you need with ICD-10? You’re looking at this huge increase in the detail and the number of codes you’ve got, and while you may have a clinical note that would support appropriate coding in the ICD-9 world, it may not have the information you need and the detail you need to appropriately code in the ICD-10 world. We’ve got a whole clinical education piece in front of us, and that all has to be in place by October 1, 2014, and so we’re in the process of going through that right now.
When you think about it, the technology pieces are probably the easiest in that whole puzzle. So we’ve got that work on top of the work we need to do for Meaningful Use and getting those pieces in place, plus we have the work we need to do to have the information and data we need for population health management and getting ready for accountable care organizations and things like that, and that touches on the big data piece.
Gamble: It can get pretty overwhelming, I’m sure, with the deadlines coming up next year. Is your approach with ICD-10 a divide-and-conquer-type thing where you have those three steps just to navigate through that a little easier?
Brenckle: It’s a partnership across the organization. Again, we don’t view ICD-10 here as an IT project. There’s an IT component, but there are a lot of other clinical and operational components necessary for us to be successful in that space. And so it’s definitely all hands on deck. Everyone is involved and we have to not so much divide and conquer, but make sure we’re taking care of all the pieces.
Gamble: You talked a little bit about Meaningful Use on the ambulatory side. How are you positioned in the hospitals?
Brenckle: Obviously I’ve got some big hurdles in the hospitals from the standpoint of until I get CPOE up and in place. I’ve got some issues with being able to meet Meaningful Use on that side. But again, I think once we’ve got all that in place, we’ll be in a position to start to attest for Meaningful Use.
Gamble: Okay. So when we talk about big data, as an academic medical center, I imagine this is something that’s always is on the forefront, especially with the amount of data you have to deal with and the complexity of that data. Where are you in respect to that? Would you say you’re setting the stage for analytics at this point?
Brenckle: Yeah, I would say we’re pretty much in the mode of setting the stage for analytics. What we came from and what we were was primarily a paper-based organization. The way you manage data in a paper-based organization is very different from what you need to do going forward, so we’re working a lot right now with making sure we’re capturing the appropriate data. We’ve done a few efforts with natural language processing so that we can pull key pieces of information out of our narrative data. We are starting to look at it from the standpoint of, are we collecting the data at the right place? Are we collecting it appropriately? Are we collecting it completely and timely and starting to pull those pieces together? That’s a journey that you go on. Data is never useful if it’s not used. You have to start using your data, and based on using it, you find out what’s wrong with it, and then based on that, you improve the quality of it. So it’s a self-correcting system. But if you’re not actually using your data in that standpoint, it never really suffices for the value.
The other thing is trying to collect data once and use it many times. A lot of times people will say well your claims data and your billing data are dirty. The word ‘dirty’ doesn’t mean that there’s anything wrong with the data; the data was perfectly suited for generating your bills. But now you’re trying to use it for other purposes as well — is it complete enough and does it have the level of information and the accuracy you need for secondary and tertiary purposes? Starting to introduce those also helps you improve your data quality.
Gamble: Another topic I wanted to talk about was something called the ICU bunker you have at UMass. That sounds like something that’s really cool and I was just wondering if you could talk a little bit about the command center that’s patched into the ICUs.
Brenckle: This was actually in place when I got here. While in some areas we’ve been behind, in other areas there’s some real cutting edge stuff they’ve done here that’s been pretty good. In all of our ICUs right now, the patients are monitored. In addition to connecting up the equipment in the room, there is both video and voice capability, and we have intensivists who are another set of eyes. This doesn’t replace anything that’s part of traditional intensive care; we have full staffs in the ICU and you have intensivists in each of the ICUs, but in addition, you have another set of eyes that’s watching, looking for trends, looking for issues, and trying to head off problems before they occur. All of that information — the video, the voice, and the electronic data — goes back, and you have someone who is sitting there monitoring the patients in the ICU and alerting his colleagues onsite when he sees a potential problem starting to develop.
We’ve been able to show that this has been very successful in improving the quality of our care and improving the overall success of intensive care. The head of this group gives a very nice talk where he talks about the people who are walking around in Worcester right now that wouldn’t be if we didn’t have this program in place. We’re doing it right now for all of our member hospitals, plus we have a number of affiliated institutions in Massachusetts where we’re providing it to them as a service. They’re not part of the healthcare system, but they’re actually purchasing the service from us.
Gamble: Okay. That’s interesting that it’s branching out. And within UMass you have a level 3 NICU, right?
Brenckle: Yes, we do.
Gamble: And a Level 1 trauma center as well?
Gamble: Okay. Like you said, another pair of eyes is always a good thing. It’s interesting how you said the organization may be behind in some respects, but is cutting edge in others. That makes for an interesting contrast I would think, especially from your point of view.
Brenckle: Yeah, there are some basic things we have to get in place, and we’re working on that. Another example where I think we’ve pushed the envelope a little bit is with our diabetes center of excellence. We’ve introduced a diabetes-centric patient portal that basically allows our diabetics to upload readings from their glucose monitors. We worked with the vendor and with Allscripts so that data move into Allscripts so that their primary care doctor and their diabetologist can look at their data in their record. It has helped improve engagement between our diabetics and their caregivers. It has improved communication, and it has helped enable patients to better manage their condition.
It’s a rapidly growing program. It started in with our endocrinologists and diabetologists in our center of excellence, and it’s branching out. Right now we’re bringing up more and more primary care offices where they have the capability to monitor patients with diabetes. The nice thing for the physician is they don’t have to go to another application to do this. It’s connected with our ambulatory electronic medical record.
This is something that we think is going to have a major effect. The head of our Diabetes Center of Excellence will say, ‘here’s the number of diabetics we have in this country. Here is the number of endocrinologists you have.’ How do you extend care and spread your reach so that you’re really helping patients manage their disease?