Phyllis Teater, Associate VP/CIO, Wexner Medical Center at The Ohio State University, Chapter 1

Phyllis Teater, Associate VP/CIO, Wexner Medical Center at The Ohio State University

When you achieve something accomplished by less than 100 out of more than 5,000 U.S. hospitals, you’ve hit the healthcare IT big time. What’s even more impressive for Wexner Medical Center at The Ohio State University is all four of its hospitals achieved HIMSS Analytics Stage 7 designation at the same time. To learn more about the organization’s IT journey, along with the career of its CIO Phyllis Teater, recently caught up with the Columbus-based executive.

Chapter 1

  • Garnering the HIMSS Analytics Stage 7 award
  • Wexner’s eHealth journey
  • Becoming an Epic shop, going big bang
  • Getting away from best of breed
  • Clearing the final Stage 7 hurdles
  • Interfacing with non-Epic health systems
  • Mega HIT investments and ROI



We really had some concerns with continuing to propagate our best-of-breed environment and felt that the maturity of Epic’s offering, as well as the future possibilities to expand our usage as a more integrated system, was the best way to go at that time.

I think any systems are hard to interface, and I wouldn’t characterize Epic as being any harder than the others we actually have had. The quality of the resources available from Epic that help to design and build the interfaces I would put second to none.

I don’t know how to measure easily without having to pour more resources into measuring than I did into implementing to tease out which of those medication error improvements I can attribute to IT.

Guerra:  Good morning Phyllis. I’m looking forward to chatting with you about your work at Wexner Medical Center at the Ohio State University.

Teater:  Good morning, it’s good to talk with you.

Guerra:  I think we will start with the news, which would be the HIMSS Analytics Stage 7 Award for four hospitals, which I assume comprise all the hospitals in your medical center?

Teater:  They do, yes.

Guerra:  Well, congratulations first off; all four hospitals at the same time, very impressive, congratulations.

Teater:  Thank you very much.

Guerra:  Tell me about it. I think we’re going to have to dovetail this with your career because you’ve been at the organization for over 20 years, correct?

Teater:  Yes, I have.

Guerra:  Over 20 years there, promoted to CIO in 2010, interim in early 2010 and then full time in late 2010. The evolution of IT at this organization dovetails with your career. Why don’t you just start with some background of the journey of this organization towards achieving one of the major milestones in the industry. I believe there are only 86 Stage Seven facilities out of 5,000 US hospitals — certainly rarefied air. Just take me a little bit through the evolution of your organization, and then we’ll get more specific from there.

Teater:  I think, like many institutions, starting in the late ’80s we had some homegrown systems that started to display results for clinicians and caregivers to have better access to information about their patients. Meanwhile, of course, the IT world was starting to change tremendously from our good old green screens towards PCs, communication devices were changing, everything was really starting, I think, to pick up speed in the healthcare system space and, again, like many organizations, we move forward in ’96 and signed our first contract with Siemens for their electronic medical record system and went live over the ensuing years with a number of different modules.

In 1998, we came up with our first CPOE site and had some success there and had some benefits that we saw in turn around time, and were able to publish of few articles that talked about that particular journey. In 1999, like many others, we were transitioning almost all of our systems from the remaining homegrown systems to vendor-based systems and then upgrading everything to account for the Y2K issue. We held for a bit there and then went hospital-wide live on CPOE with the rest of our inpatient areas.

In 2006, we signed our first contract with Epic Systems Corporation for our outpatient electronic medical record and, over the ensuing three years, went live in all of our outpatient sites and then, in 2009, we signed a contract for Epic’s Enterprise Systems and brought that in and began work on that.

All of that culminated live in October of last year when we did what’s termed the big bang. Like many places, we went live with all of our inpatient system, ED system, our revenue cycle system, patient scheduling, OR system, pretty much the whole shebang all on the same day in October of last year. We worked very hard to be ready to apply for Stage 7. Soon after that, we did need some of the functionality from installing Epic to be considered for Stage 7, and we were awarded Stage 7 this spring after the HIMSS folks came out and reviewed our usage.

Guerra:  So 2006, let’s go back to that point for a moment, 2006, you had Siemens as your inpatient system, correct?

Teater:  Yes.

Guerra:  I don’t know if they had the relationship they now have with NextGen in 2006 or if you looked at – I know Siemens is still coming out with an ambulatory version for Soarian. But just take me through that decision process of going with Epic as the outpatient system in 2006.

Teater:  I think we had a number of considerations. One of them is we wanted to really look at an enterprise-level outpatient system that had future capabilities to round out our functionality. We really had some concerns with continuing to propagate our best-of-breed environment and felt that the maturity of Epic’s offering, as well as the future possibilities to expand our usage as a more integrated system, was the best way to go at that time.

At the time, we issued our RFP, Siemens did not have the relationship with NextGen, so we did not consider that. We did consider a number of other systems but really felt that Epic positioned us the best for future integration. We had begun, of course, at that time talking a little bit about the need to replace our inpatient systems, and so we were thinking ahead and thinking about the best way to get a single integrated record for our patients and moving more way from the best-of-breed environment.

Guerra:  Was Soarian ever considered?

Teater:  Yes. We did work with Soarian for a number of years. We were an early partner with Siemens on some of the Soarian development and provided some input to their early development of that product. But later by the mid 2000’s, we were looking at the outpatient world, and really that’s when we created a vision to move towards a more integrated set of systems than we thought we could achieve by continuing with Soarian.

Guerra:  You mentioned that you needed some of the Epic functionality to achieve the final pieces of Stage 7. Can you tell me what those pieces were?

Teater:  Many of those things were things that we had not done with our legacy system because there were some pretty big challenges. One of them was medication barcoding with us — in the previous world having one system for order entry, a different vendor for pharmacy, and we would have been looking at a third vendor for the actual medication barcoding functionality. We really felt the patient safety issues of trying to interface all of that asynchronous medication information. That encouraged us to wait until we brought up Epic. When we came up with Epic is when we went live with medication barcoding.

That was a big piece for us. We had also never automated some pretty important portions of the nursing workflows in our old system and so decided that we would bring that up fresh, so all of the I’s and O’s and the vitals, and as well as a big portion of nursing documentation itself we did not bring up until we came live with Epic in October. We also needed to utilize, and are very happy utilizing, Epic’s Care Everywhere Module to provide us with the functionality for the interoperability. Also, that came up actually a week before the big go live in October, right about the same time.

Guerra:  Have you looked at, or investigated, interfacing with non Epic systems?

Teater:  Yes, we’ve had some discussions with some institutions that are here in our local market where we do have patients in common. We are also, though, looking at a state HIE option with those same institutions, so we’re sort of evaluating our different options here and being sure that we are headed down the right lane for the long term.

Guerra:  One of the big criticisms of Epic, whether it’s valid or not I’ll ask you, is that it’s difficult to interface with non Epic systems. Does that have any validity in your eyes?

Teater:  I don’t know that it’s any more difficult with Epic than it is with any other set of systems. It’s hard, particularly with information that is involved when you’re maintaining asynchronous lists between the systems. If you’re looking at a medication list in one system that you’re trying to keep up-to-date to the minute with any medication and then you’re trying to maintain that same list in a secondary system, it usually ends up being a combination of some manual stuff and some interface stuff because no two systems interpret the information exactly the same way.

Our capability to get that right 100% of the time, I think, in the healthcare industry is somewhat challenging. There’s lots of double checking and syncing back and forth and looking at the other system and making sure that every single piece of information is correct, and there is no place probably more important to get it right than in that medication history and the current medication list. I think any systems are hard to interface, and I wouldn’t characterize Epic as being any harder than the others we actually have had. The quality of the resources available from Epic that help to design and build the interfaces I would put second to none.

Guerra:  One of the big debates is do organizations make investments of the type you’ve made — the large investments to get on one system? You speak to CIOs at organizations that have not done it and they say, ‘Well, we could never spend that kind of money, and you’re never not going to have to interface. Eventually, there’s always going to be that perimeter where you have to interface with another system, so what’s the point.’ I would imagine you would say something to the effect of, ‘At least I’m pushing that perimeter outside of my four walls, so to speak, or outside of my health system so I don’t have to deal with all of these interfaces inside. I still have to interface but it’s a different degree of pain.’ Does that make sense?  How would you describe it?

Teater:  Yeah, I think that would make some sense. I do think that as we look at some of the care environments that a patient can transition through quickly, those are the areas that I think you see the most benefit from an integrated system. For instance, a patient that is seen in one of our outpatient clinics here on campus who has some significant event is immediately transitioned to our emergency department and then immediately transitioned from the ED to the OR onto the floor. In our old world, they would have passed through four systems in the scope maybe of 24 hours.

When I think about the speed, the need for information, for decision making and those kind of things between those kinds of care environments, and yeah it’s nice to not have to worry about it within my four walls from an IT perspective. But when I think about the impact to the patient to have that feel seamless to the clinicians that are involved to those complex transition in care, I think it’s pretty important to the patient.

Guerra:  When you look at the challenges that will come with using an HIE to integrate with other systems, I assume you’re happy you won’t have to face those in-house.

Teater:  I think that when a patient transitions from the ER upstairs, most of that happens on the same physical property, so in many cases there are opportunities to impact our patient population by having a more integrated record. Every time you broaden that circle to make it more seamless, you impact that many more patient care episodes or events.

Guerra:  How would you describe the ROI or lack of ROI for these systems? Or is it not even valid to talk about ROI when it comes to patient-related systems?

Teater:  Oh, you’ve got all the hard issues. J ROI is a difficult discussion, I think, for healthcare IT. I think that when you look at the kinds of metrics that we know intellectually are impacted by the usage of the system — take, for example, the decrease in medication errors. Certainly, I believe — perhaps a biased view — but I believe that we are impacting that by having a better, more seamless, integrated system between care areas. What is hard to do is for me and IT, or us as an organization, is to pretend that there aren’t many other amazing initiatives that are also impacting that number so I don’t know how to measure easily without having to pour more resources into measuring than I did into implementing to tease out which of those medication error improvements I can attribute to IT. I think you have a whole set of metrics that you are hoping are, and frankly that we believe are, being impacted by the use of integrated systems, that I don’t know how to claim credit for because, again, of all the things that impact those. I think that we work closely with our partners to understand that, to try to understand the kinds of things we are impacting and then to measure, at an organizational perspective, if we are seeing a decrease in medication errors. That is a wonderful thing for our patients, no matter what the reason.

Having said that, there are also a number of things that you can track directly back to system use. For example, we transitioned much of our environment where we were doing dictation and transcription into an environment where we are either documenting through discreet fields or we’re using Dragon as a way to do voice recognition in the documentation. We have cut our transcription cost by over $1 million dollars a year just since October if you analyze the numbers. There are things that you can look at that we can attribute to the use of the system directly that have financial realization behind them, but I don’t know that those things alone will ever offset the entire cost of the system. It will certainly help us to offset a major portion of the investment, and we are measuring a number of those post-live after October, but it’s a sticky question which I think, of course, all of us wrestle with, the question of the real quality metrics that you’re impacting. IT is only one piece, and so to tease out the impact to the system is a pretty daunting task.

Guerra:  You make a great point that you can get lost in trying to analyze the impact.

Teater:  Another goal for the outcome of system installation is length of stay reduction. I would use that as an example. We have a number of initiatives that are hoping to reduce the amount of time our patients spend in the hospital for lots of good reasons, one because we want to make them well faster and get them home so they can heal in that environment. One of those is increasing our ability to turn around tests and to communicate the results among clinicians. All of those things may be byproducts of this system but, at the same time, we are changing staffing in different areas to address the parts of the discharge process. Some institutions are creating a discharge lounge where the patients can go while they’re, perhaps, waiting on a ride or waiting on something to be able to leave.

You see all these organizational initiatives happening to impact length of stay. If length of stay goes down 0.5, it would take a team of very smart people to try to figure out what percentage of that 0.5 reduction IT was responsible for. What portion of it is that 10 minutes I saved in communication between caregiver one and caregiver two. What portion is because I got the meds a half an hour early? What portion is because I got them down to the discharge location and was able to fill the bed? Just trying to break down into those elements and tie it back to our overall organizational metrics is a task that can sometimes be done, but it’s frankly hard to do on a consistent basis for all the metrics of an organization.

Chapter 2


Email Newsletter

Sign up to receive our latest updates delivered straight to your inbox.

Share Your Thoughts

To register, click here.