
Myra Davis, SVP & CIO, Texas Children’s Hospital
If you ask Myra Davis, there’s a big component many leaders are missing when it comes to leveraging data: education. An organization can have all the coolest tools and technologies, but if clinicians don’t understand what exactly is available and how they can interpret it, the data just isn’t worth much. In this interview, the CIO of Texas Children’s Hospital talks about how her team has dealt with clinician expectations when it comes to data, and how they’re utilizing education and dashboards to help them get the most out of it. She also talks about the work her organization has done to implement an EDW and their plans going forward, how breaking down silos between IS and clinical has helped empower users, her strategy when it comes to fostering innovation, and why still thinks the industry is “a lot of fun.”
Chapter 1
- About Texas Children’s Hospital
- Epic in hospitals & clinics
- Finding the right balance with data access
- “Don’t underestimate the level you have to go through to the explain data.”
- Population health
- Future EDW plans — “The concept of data is never-ending.”
LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO SUBSCRIBE TO OUR iTUNES PODCAST FEED
Podcast: Play in new window | Download (Duration: 11:22 — 10.4MB)
Subscribe: Apple Podcasts | Spotify | Android | Pandora | iHeartRadio | Podchaser | Podcast Index | Email | TuneIn | RSS
Bold Statements
We spent a lot of time just scratching our heads on how do we help the organization get what they need, because it was becoming apparent that they were going to get it one way or the other, whether it was entering it off the screen into a spreadsheet, or asking for data that may or may not have been actually valid and not really understanding how the data was actually born in the system.
We had to actually go to that level of detail to explain in some cases and to say that you don’t want a hard stop or required field on everything that you want, because now the system is not very user-friendly. So there’s a balance, and I needed to determine exactly what that balance was.
The concept of data is just never ending. Now that we are 10 years into Epic, we have lots of data — data is an asset for us — but we still have work to do with educating the organization on what’s available to them and how they can get it and interpret it.
Now you’re getting into predictive analytics and all of that sits on the foundation of data, and understanding my data, just starting with basic reporting. How do I just get a report? And then you take it from there, so don’t underestimate the education that’s needed.
Gamble: Hi Myra, thank you so much for taking some time to speak with us today. Can you provide a brief overview of Texas Children’s Hospital — what you have in terms of the bed size, some of the ambulatory facilities, things like that?
Davis: Sure. We are licensed for approximately 650 beds. We have three hospitals — we focus on adult care and OBGYN as well. We have approximately 50 primary care/pediatric practices that are located throughout the Houston metropolitan area. We have in excess of 40 subspecialty care providers as well. We also have about four urgent cares, and we’re targeting almost 10 by next year.
Gamble: Okay, that’s a pretty good view and obviously you’re located in Texas. Where exactly are you?
Davis: We’re in Houston, in the Texas Medical Center.
Gamble: How long have you been with the organization?
Davis: I’ve been with Texas Children’s for 12 years.
Gamble: And how long have you been in the CIO role?
Davis: I have been in the CIO role for five years.
Gamble: So I want to talk about some of the efforts that are taking place to really use data to drive quality improvement, care quality and to do that, I think we should kind of lay a little groundwork. So can you talk first about the clinical application environment, what type of EHR system is being used?
Davis: We’re on Epic. We’ve been on Epic now since contract signature for 10 years and Epic is everywhere in our environment — each hospital and every ambulatory care and primary care practice is using Epic. We also have a health plan, I failed to mention that before, but they’re not using Epic.
Gamble: And so, you were with the organization through the whole transformation to Epic?
Davis: Yes.
Gamble: So obviously that played a huge role for laying the groundwork for everything that was going to be done with data, and was that a fairly long process getting everything up on Epic?
Davis: Yes, it was fairly long getting everything up. I will say that once we got everything up, data became the hot topic. I think it was initially thought you’d have lots of data in there the very next day. When we were bringing it up, we did minimal migrations. We went from paper to electronic on the inpatient side. On the ambulatory/subspecialty side, we went from electronic to electronic. We did some level of migration, but not a lot, so what we found ourselves doing once Epic was up was having to explain to everyone the actual amount of data that was in Epic versus not being in Epic, particularly from an inpatient standpoint. So we did a lot of education. We spent a great deal of time educating on Epic’s reporting tools, but that probably lasted only for the length of time that it was needed, meaning that if you don’t use it, you lose it.
There was a lot of frustration in the organization around not being able to get the data that they need, or the data that they were looking for was incorrect. So worked very closely and partnered with quality and the senior VP at that time. We spent a lot of time just scratching our heads on how do we help the organization get what they need, because it was becoming apparent that they were going to get it one way or the other, whether it was entering it off the screen into a spreadsheet, or asking for data that may or may not have been actually valid and not really understanding how the data was actually born in the system, meaning how the system was designed to capture the data. In some cases, it wasn’t designed to actually capture the data. What I mean by that is if you’re looking for specific fields and sometimes it’s in there, sometimes it’s not, that means it wasn’t the required field. And so we had to actually go to that level of detail to explain in some cases and to say that you don’t want a hard stop or required field on everything that you want, because now the system is not very user-friendly. So there’s a balance, and I needed to determine exactly what that balance was. You don’t underestimate the level of effort you need to go through to explain data. It’s pretty detailed.
Gamble: I can imagine. And so how did you work to approach that? Were people broken up into teams or how did this go just as far as the whole education process?
Davis: We did several things in the beginning. Then the senior VP had attended a conference and she ran across this group and came back in and said, ‘Myra, this is really cool. I think you need to see it.’ So we did. I took a look at it and brought some of my team to look at it. And what was really good about it is that we were able to stand up an enterprise data warehouse in record time, basically taking ETLs and porting data over into an EDW within a four-month time frame.
Now, this was before Epic came out with Cogito, so we actually started very early. What we liked about partnering with this vendor, though, was not only did they assist in accelerating our ability to stand up an enterprise data warehouse, but they also had a clinical aspect to it. They worked very closely with clinical teams — physicians, nurses, quality analysts — and they assisted them in actually understanding the data that was captured in the EMR, and understanding it by cohorts. They split teams up by care teams, so we have a diabetes team, we have an asthma team, and we have appendectomy team.
We have multiple teams, and what they basically do is look at the population of patients that we’re caring for by those particular cohorts. They’re looking at the data that’s captured in Epic, but also our enterprise data warehouse consists of 12 unique source systems that float in there and they’re able to correlate data from those other source systems. Cost accounting is a system, Press Ganey is one of those systems, and our financial systems are in there.
So they’re able to basically do a level of correlation by cohorts to determine how we’re actually providing care as a system for those patients. To some degree, we’ve been managing populations as defined by the term ‘population health.’ We’ve had our enterprise data warehouse up for close to five years, and we’ve been pretty much doing that since about six months into that five years.
That’s been very good, but I will say that recently we’ve had a hurdle where, because we have teams, we still have not done a great job with getting data and having data accessible to those that are not part of a team, so we still have providers frustrated because they’re not able to get data that they need, etc. So what we’re doing is now that Epic’s more mature, we are looking at Epic’s tools — Radar, Slicer Dicer, and Reporting Workbench. We’re looking to educate the providers on what those tools are that are available to them and we’ve also put some really cool dashboards in place from the enterprise data warehouse.
It’s a matter of we have the tools, but in terms of the maturity and understanding of the tools available to the organization, we still have some work to do. I think the concept of data is just never ending. Now that we are 10 years into Epic, we have lots of data — data is an asset for us — but we still have work to do with educating the organization on what’s available to them and how they can get it and interpret it.
Gamble: Right. And it’s pretty clear that the ultimate goal is to have this repository of data where you can have that consistent view, but it seems like it really is a never-ending struggle just to make that happen, between the data and then the tools and just trying to get that all together.
Davis: Yeah, it’s been kind of interesting. Since inception, I think we’ve gone through the terminology of population health, big data, and data as an asset. So call it what you want, but we still have to help the organization understand the data, right? So I always try to find humor in everything I do. If not, I’d probably pull my hair out. That’s just been an evolution. And so for us, now you’re getting into predictive analytics and all of that sits on the foundation of data, and understanding my data, just starting with basic reporting. How do I just get a report, right? And then you take it from there, so don’t underestimate the education that’s needed, more than anything.
Share Your Thoughts
You must be logged in to post a comment.