Art Glasgow, VP/CIO, Duke Medicine, Chapter 2

Having spent most of his career on the vendor side, Art Glasgow knows about deadlines. Now the VP and CIO Duke Medicine, which includes Duke University Health System, he’s bringing all that experience to bear in handling the new, breakneck pace at which healthcare, and healthcare IT, are hurtling forward. To learn more about the organization’s massive Epic rollout, which Glasgow describes as almost stakes to play in today’s environment, healthsystemCIO.com recently chatted with the North Carolina-based executive.

Chapter 1

Chapter 2

• Building strong clinical partnerships, staffing the CMIO role
• WK Provation for order sets
• M*Modal for speech recognition
• Balancing the desire for data with the burden on front-line clinicians
• Meeting MU is “not in the top 3 things we’re focused on here at Duke”
• Thoughts on ICD-10

LISTEN NOW USING THE PLAYER BELOW OR CLICK HERE TO DOWNLOAD THIS PODCAST AND SUBSCRIBE TO OUR FEED AT iTUNES

Podcast: Play in new window | Download (15.6MB)

Subscribe: Apple Podcasts | Spotify | Android | Pandora | iHeartRadio | Podchaser | Podcast Index | Email | TuneIn | RSS

BOLD STATEMENTS

If you don’t understand how the periop and intraop environment works then you’re really relegated to just being the IT guy who needs to go deliver the stuff.

Making sure that both the CIO and the CMIO have the same vision for what the role should be is critical, and then it never hurts to have them genuinely like each other.

There’s a great deal of consternation as we roll out Epic with physicians who might not necessarily be used to charting electronically, and having a solid voice recognition solution that we can roll out with it is going to give us a great deal of comfort.

Guerra:  All the CIOs that I speak with, you can pretty much tell if their strength or their background lies predominantly on the technology side or the clinical side. Some come out of nursing, a few are MDs, most are more technically savvy, and you certainly sound like someone who has the stronger technical experience. My question to you is that usually you have to find support on the other side, in your position it might be through a strong CMIO, or just different ways to develop knowledge of the clinical practice so you can have rapport with an MD when they sit down in your office. Can you give me any thoughts you have around that?

Glasgow:  First of all, it is incumbent upon any CIO to become quite educated themselves in clinical processes and workflows and the basic vocabulary of the clinical side of the business. If you don’t understand how the periop and intraop environment works then you’re really relegated to just being the IT guy who needs to go deliver the stuff. But at the same time, for someone like me who’s not an MD or an RN, no matter how much clinical experience I have, you can never position yourself as being an equal in that environment. I never should or would do that. It’s very important to have strong partners who are MDs or RNs or other clinicians who can understand technology and who can drive that forward.

Duke did not have a CMIO when I came on board and one of the first things that I did was create the position and hire a CMIO. I was fortunate that there was a doctor here at Duke who was very qualified and has done a great job over the last year, and that’s Dr. Jeff Ferranti who is a neonatologist here at Duke. That’s been very critical to our success, but it can’t just be one person either. You have to start building out this network of physician champions and clinical partners. Some of which may be on the IT side of the house, but most of which should probably live in operations and within their clinical departments. It’s really about relationship building and creating a joint vision with them.

Guerra:  I would describe it as a luxury to be able to pick your own CMIO because, to me, I would think the CIO and CMIO having a good personal rapport is just about as critical as it could be in any relationship. You need to – I would imagine – just have a good rapport and be able to understand each other’s weaknesses and be honest about them.

Glasgow:  Yes, I think you’re absolutely right; it is one of the more critical relationships in a health system today. It was a luxury for me that I was able to craft the position and hire the position myself. The model we have here at Duke is the CMIO does report to me. I’ve seen other models where the CMIO does not, and I think that can work as well, but it is critical that the two leaders really be on the same page. I think there’s also some maturation that needs to occur just around the definition as to what a CMIO really is.

In some places they’re kind of the CIO for clinical applications and in other places they’re really more around informatics strategy and stuff like that. Making sure that both the CIO and the CMIO have the same vision for what the role should be is critical, and then it never hurts to have them genuinely like each other. Jeff and I spend an awful lot of time together, not just around work stuff, to ensure that we have a very strong relationship and that he can complete my sentences and I can go out and ensure that I’m driving his agenda as well.

Guerra:  You can walk into each other’s offices, not everything has to be super formal, that type of thing?

Glasgow:  Yes, absolutely.

Guerra:  Just poking around the internet, it looks like you’ve been very busy the beginning of this year. I see three deals M*Modal, iSirona, Wolters Kluwer for the ProVation Order Sets. I’m sure there’s a bunch of other stuff. Would you say it’s been a busy 2012?

Glasgow:  It has been a busy 2012. It’s likely not going to slow down throughout the rest of it into 2013. Those deals, I think, are just an extension of what I said previously where we have quite an extensive order set today in the case of ProVation, but ProVation provides a nice container and easy management framework for us to be able to ensure that order sets stay current, especially in a post-Epic implementation. Rather than kind of extend and recreate what we already had, we decided to just go third party. I think the M*Modal deal is definitely one that’s built around physician satisfaction. There’s a great deal of consternation as we roll out Epic with physicians who might not necessarily be used to charting electronically, and having a solid voice recognition solution that we can roll out with it is going to give us a great deal of comfort.

Guerra:  Let’s talk a little bit about the M*Modal deal just in the sense of your thoughts around balancing structured and unstructured data.

Glasgow:  Our goal, as we roll out Epic, is of course to get as much discrete data, structured data as we can right from the front of the process. A lot of that is in the clinical realm, but an equally large part is thinking about research down the line. If we don’t structure the data when it’s charted, then it’s going to become that much more difficult for us to extract it and support our research organization. M*Modal is a way for us to help that process and to convert some of the free text areas into structured text and to be able to help the clinicians speed up their workflow without sacrificing data quality.

Right now, we’re using it just as a front-end speech recognition solution, but some of M*Modal’s capabilities that we’re looking at as well, especially with their new announcement of Catalyst, are really about going in and structuring data from large piles of unstructured data. That’s something that we’re thinking about, as well as to how do we get into text analytics and speech analytics and some of the other things that we know we’re going to need to get to.

Guerra:  The challenge I would imagine with this stuff is it can’t work 85%, 90% of the time. It’s got to work 100% of the time in order for this data to be valid, especially for research purposes. It’s like if you have a transcriptionist and they get everything right most of the time, once in a while there’s – I have this situation, we have a transcriptionist we use, once in a while there’s just a ridiculous transcription error that totally changes the meaning of a sentence. You can’t have that with this kind of data, correct?

Glasgow:  You’re right and that even extends, not even to the research environment, but just within clinical care. That’s why it’s important, I think, to really do an implementation like this with an eye towards your clinical processes and your care re-design efforts. They should be part and parcel of your EHR implementation. If you’re expecting the software to do everything for you and work 100% of the time, then you’re probably going to get let down at some point in the future but if we roll this out in a way that it’s additive to the clinician’s workflow, but yet they understand that they’re still responsible for validating their note, then I think we get the best of both worlds.

Guerra:  This also makes me think that we have to be careful about overloading clinicians with documentation burdens on the front-end, at the point of care.

Glasgow:  That’s right, there is a balance. You can’t create a perfect crystal cathedral at the expense of the clinicians who really have to work within it; it’s always a balance, and the CMIO’s role is so critical to finding that right balance because, ideally, they should live in that world. They should practice, and they should understand first hand the implications of the technology we put in place.

Guerra:  Do you have independent physicians or is it a closed system?

Glasgow:  It varies by hospital. Our main academic hospital is a closed hospital but our two community hospitals are open where we do have community physicians who maintain privileges.

Guerra:  Is there a big difference in the dynamic as far as the CIO role is concerned when dealing with independent physicians versus staff physicians?

Glasgow:  I think there is. It’s more on the business side than just the pure technology side. Cost is always a concern for independent physicians, both from the standpoint of real cost — what’s it going to cost me in order to have to go to Epic or have to adopt some of the other systems and policies in place — but also the cost of both their work productivity and their referral flow; that’s always something that you have to consider as well. We’re taking the tact definitely that the platform we’re putting in place is going to be a platform for Duke and those that want to practice with Duke. We’re not really going to customize the solution quite extensively to have different variants of it, but we’re looking at ways to make it easier for our independent physicians and community physicians to partner with us in improving care by using the platform.

Guerra:  Someone once told me to keep in mind that just because a doc is employed doesn’t mean he or she isn’t under pressure to be productive.

Glasgow:  Absolutely. Being an employed physician doesn’t give them a pass on their RVU outputs that they have to hit and, in many cases, their compensation is tied to that. They are under the same pressures. It think the financing mechanisms for how you get the solutions in place is different between the employed physician and the community physician, but there isn’t a doctor out there who isn’t under pressure to continue to increase their productivity.

Guerra:  How has your organization done with Meaningful Use?  I would imagine you attested to Stage One?

Glasgow:  We are in the process of attesting to Stage One. Meaningful Use is important, and I don’t mean to say otherwise here, but it is not probably in the top three or four things that we’re focused on here at Duke I think for a couple of reasons. One, our decision to move towards Epic probably came a little late in the game if we were going to be completely focused on Meaningful Use. Rather than risk a rather large implementation just to hit some numbers, we’re making the decision to make the implementation successful and then pick up as much Meaningful Use dollars as we can. That means, for a certain percentage of our enterprise, we’re making the decision not to attest to MU level 1.

We fully expect to catch up by MU level 2 but, again, that has less to do with the technology than it does with the use of the technology. We’re working very hard with our clinical partners right now to get them ready to be able to attest and to do the things they need to do with the technology once it’s ready.

Guerra:  Does it make sense to say that the MU dollars won’t mean much if you disrupt your revenue cycle?

Glasgow:  I think that’s a fair argument to make. If you look at just the scale of the Meaningful Use dollars compared to the monthly billing at a place like Duke, there’s no comparison.

Guerra:  Right. What were your thoughts on the ICD-10 postponement, is it much appreciated?

Glasgow:  I think, of course, any time that we get is valuable time, but we were on path to hit the original date. My personal feeling was if the delay was coming it would probably have been better to delay it longer than a year or, if you’re going to do less than a year, just don’t delay it at all. A year is almost like a perfect middle ground where you don’t feel great about it on either side.

Guerra:  Right {laughing} that’s a good point.

Glasgow:  The implication it brings up for us is our original Epic timeline was built around a pause in late 2013 in order to remediate the two community hospitals on their existing systems for ICD-10, and now that that’s pushed out a year, so do we feel comfortable enough pulling in those two community hospitals and making Epic our ICD-10 plan? There’s, of course, a cost savings for doing that, but it doesn’t leave you a whole lot of time in case you’re not successful. For us, not yet having gone live on the main hospital and not proving to ourselves that we can do this on time brings up a little bit of an interesting dilemma.

Guerra:  After the delay, I was frustrated with the government. I thought it was just ridiculous, because now you have to go back to the drawing board and try to marshal those same resources who may not take you as seriously this time.

Glasgow:  {Laughing}. Yeah, I think it’s important to note that right now what we have is a proposed delay. We don’t have a final approved delay. There are loud voices that say we should stop now and every CIO, if they’re responsible, should be saying we can’t stop now. We don’t know exactly what’s going to happen. Your personal credibility is a little bit on the line, but I think that’s not limited to just ICD-10. To be a successful CIO I think you have to be comfortable putting that personal credibility on the line and then delivering against it.

Chapter 3