HIStalk Interviews Frank Clark PhD, VP/CIO, Medical University of South Carolina

  

A reader heard that MUSC was implementing the Oacis clinical data repository/EMR along with McKesson’s Horizon clinical systems and asked me to find out more. CIO Frank Clark was quick to offer to let HIStalk’s readers know what’s going on in his organization. Some interesting projects, as it turns out, of which the Oacis implementation is certainly one. Thanks to Frank for letting me call him at home to chat.

Tell me a little bit about your organization and what kind of projects are happening at MUSC.

We are a freestanding academic medical center. Our mission is education, research, and patient care. We have about 800 physicians and another 300 residents and fellows. We operate three hospitals – two adult facilities, a children’s hospital, and a psychiatric hospital, for a total of about 850 beds.

Of course, we have the big outpatient clinics where the College of Medicine faculty hone their clinical skills and stay current. It’s probably a $1.6 billion operation enterprise-wide.

Tell me about your IT department.

IT has 250 FTEs. Most of those are centralized within the Office of the CIO. The combined clinical and academic operating budget is probably $35 million. The capital budget ebbs and flows, depending on the projects. We’re in the throes of a big clinical implementation of advanced point-of-care clinical systems within the hospitals, about two years into that and another year to go.

That’s all McKesson, right?

Pretty much. We’ve been a best-of-breed organization. When I came in about five years ago, we had a lot of stuff in place. We couldn’t rip it out and start things over from ground zero. We’ve got systems like Cerner lab, IDX radiology, and Agfa PACS. 

We made a decision for McKesson as a preferred vendor. In fact, on Wednesday, we signed a large revenue cycle contract with McKesson . They have a new product, a new ADT, registration, and patient accounting product. We’re implementing their document imaging technology in the business office and also in medical records.

You have a PhD in mathematics. That’s got to be a difficult academic accomplishment. Do you find that’s a good fit for what you’re doing today?

I think I do. I’ve always been involved in computing, very early on as a user and more recently as a provider of services. I’ve always used computing in my academic career and in teaching. My PhD was in applied mathematics, so it had a nice fit for computational analysis.

I started out in an academic career and sort of drifted over into providing IT services. Then the CIO role become an established C-level position, so I moved into that.

How do you run things with regard to IT governance?

Governance is essential to good IT service and a good IT environment. It was one of the first things I set about to establish when I came in here. With any organization, the first thing I’d put in place is a well-defined IT governance structure.

We have various committees that represent different factions of the organization — the clinical, the education, the research, the infrastructure — and we structure that and have it well-documented and then get buy-in from the leadership. I think the key to that is constituting those committees and councils with the right people, people who have an interest and a passion and have the time and who are going to participate in the settings. That really forces stakeholders to take ownership and responsibility. It doesn’t put all the responsibility on IT.

Often issues will come up and people will automatically say, “That’s an IT issue”. Because it has some computing faction to it, people just want to default to IT. The governance really forces users to look at workflow and process and management issues. More often than not, it’s not a technology issue. It comes down to poor workflow, poor process, poor management. So it really pushes back. It doesn’t let people dump on IT. It really forces people to take ownership. It brings people’s common interest and needs together, and looks at technologies.

I think it avoids duplication and re-duplication of functionality because it forces people to look at the inventory of stuff that you already have. Often, people just automatically say, “We need to go out and get a system for this.” More often than not, that functionality exists, so if it meets 85 or 90% of our needs, then we’re not going to go out and buy another system.

What would you say your most important or most pressing projects are right now?

These clinicals that I mentioned, which are nursing documentation, barcode administration, meds administration, and CPOE. We have a lot of research systems that are going on. We’re trying to really streamline and organize the research process. We do about $200 million in funded research and we want to try to scale that to $300 million over the next four or five years. So we’re working with those provosts of research, reorganizing that whole research support sector. Once they’ve identified their strategic imperatives, goals, and objectives, we will look at how we can use technology to assist them.

Going back to the governance issue, it forces people to not just look for technology solutions, but to identify what their strategic imperatives are and their goals and objectives. Then, from an IT perspective, we try to align our initiatives in support of those goals and objectives so we don’t get the cart before the horse. Often this happens — people tend to throw technology at it. It’s been my experience that if you’ve got a bad process and you automate it, you just it do it bad faster. I think that governance helps people to realize that technology is not the panacea or the magic elixir.

So if you’re with McKesson that means you’ve got Meds Manager, HEO, AdminRX, and HED, probably. Tell me where you are with those and what you’ve learned along the way.

Meds Manager is fully deployed. Expert Documentation, we’ve got half of those beds that I mentioned a moment ago. CPOE, we started the initial roll-out in April. AdminRX is probably, like Expert Documentation, about half rolled out.

I guess the vendor is not always right. They still have problems. You would think that as many implementations that they’ve had with these products that most of the issues have been worked out, but I guess that’s not always the case. The people on the ground, the McKesson people who are here on site, are often not the most knowledgeable or the most skilled.

What I’ve learned is that you try to identify within the organization, say McKesson — and it’s huge — it’s a difficult challenge to find the right person to get the right answer, rather than trying to work through the people who are here. I know McKesson doesn’t like that. They like for you to work through the team and the protocol, but we don’t do that.

We just had a team go to McKesson’s clinical brain trust outside Boulder. We send a team out there twice a year. They talk with the right people. So you establish names and contacts. These are the people that we go to when we have issues. That’s been my experience and it’s proven to be successful. That’s just the way I work. McKesson can like it or lump it.

If they’re not performing, we just don’t pay them. Of course, we’re a big McKesson client, so it doesn’t take long for the AR to build up. When it gets to be a million or two million dollars, they pick up the telephone and they call you. I say, “Well, look, when you start performing, I’ll start paying.” I think that’s one of the advantages of the single vendor. You’re doing so much business with them that you really show up on their radar screen. If you stop that revenue stream, it doesn’t take long to get their attention.

You’re hitting some big change management projects that involve a lot of clinical users. What kind of structure are you putting in place, on the informatics side, to get this done?

Going back to the governance model, big projects like this call for an overall steering committee. They’re at a high level, the 50,000 foot steering committee that makes the very big, broad decisions. For each of those products that you mentioned, there’s an implementation oversight committee or council that has people at the operational level. They look at workflow and change management. How is Expert Documentation going to change the way nurses deliver care?

We spent a lot of time before the implementation in thinking through that and talking with other organizations like Vanderbilt and Duke, organizations in our state — Spartanburg Regional and Anderson. These are community-based McKesson clients. Talking with counterparts — “How has that changed the way you deliver care and the workflow?” We’ve really paid a lot of attention on the front end to those kinds of issues.

What are you looking at in terms of success metrics?

Being an academic medical center, we have access to people who are very helpful, skillful, knowledgeable, and experienced in this area. A lady who is in the College of Health Professionals — we bought part of her time. She has worked with the nurses to identify metrics. She did a baseline on these metrics. Like how long does it take to do assessments and how long does it take to get vital signs into the chart?’ Now she’s gone back, once we got a sample size large enough, and done the post-measures.

We’ll do the same thing on trying to measure the reduction of adverse drug incidents. On CPOE, we’ll look at the reduction in lab orders like Chem7’s and portable x-rays. So, we have some well-established metrics that we are measuring. We will report those out. The same thing is true in the perioperatives, OR and anesthesiology. Measuring throughput. We’ve been able to move more people through the OR. Also, the anesthesia coding and charting. We’ve got a project to measure that as well.

How are you going to implement the Oacis repository?

Oacis has been in here since 1994 or 95 and its predecessor before that. Don Simborg started that company many, many years ago. So they’d been in here a very long time. We have a number of very knowledgeable Oacis users as well as IT people. We know Oacis probably as well as anybody, collecting data and the ODR, the clinical data repository, since ’95. That’s thirteen years of clinical data.

It’s rather elegant. The commitment I made when we signed the McKesson contract was that if the McKesson Physician Portal, which is their physician viewer, was not superior to the existing product, we wouldn’t change it out. So about a year ago, we began to look at installing the portal and we got our physicians to look at it. They said, “No, this won’t work.” It won’t work in an academic setting because most of the work is done by the house staff. A resident might have to cover 80 patients, so its not like a community-based physician who comes into the hospital and has one or two patients. The portal is great for them. I had installed the portal in a community-based setting and it worked well.

I thought it had matured and evolved over time, but when we looked at it a year ago, in my judgment, it hadn’t evolved much. So we came to the conclusion that it would not work. It just so happened that Emergis Oacis had a new release, a Java-based release of the viewer and the repository, so we stuck with it. We’ve rolled it out and it will be our enterprise-wide clinical viewer. We import Expert Documentation information into it. Physicians can launch CPOE out of it. So, it’s kind of a single sign-on type environment.

So you are responsible for doing the back-end integration with the Horizon database back into Oacis?

Yes. We’ve worked with Vanderbilt. Vanderbilt did it. It’s not a strategy that’s strikingly different from what is being done at Vanderbilt and Duke, which are both McKesson clients. In fact, Vanderbilt developed the McKesson CPOE. It was called WizOrders and McKesson licensed it. Vanderbilt has installed most of these products that you alluded to a moment ago — Meds Manager, Expert Documentation, barcode administration. They have a homegrown product that’s similar to Oacis called StarCharts/Star Panel. It’s something they developed there. It’s very similar. So they had cracked that nut as far as importing Horizon stuff into it, so we worked closely with them and emulated what they’ve done.

Do you think this will get people’s attention to look at Oacis as an alternative?

Yes, I think it will. In fact, my counterpart in Greenville, Doran Dunaway of Greenville Hospital System, which is one of the largest hospital systems in the upper state — he’s very keen on it. I don’t know whether he’s signed the contract. He looked at the Vanderbilt StarChart/StarPanel, which is being marketed. He looked at a number of different products. He came to the conclusion that it was as good as anything around.

Who’s commercializing the Vanderbilt product?

It’s called ICA, Informatics Corporation of America.

That’s right, I know those fellows.

They’ve got an install in Bassett Health in Cooperstown and it’s being used over in Memphis and it’s a nice product. We looked at it long and hard, but when our caregivers looked at it they said, “This is good, but what we have is equally good, so why would we change it out? If it was vastly superior then we could do it, but what we have is good.”

You’re right, I think the Oacis product is one of those jewels in the rough. It’s widely used in Canada and Australia. Texas Southwestern Medical Center uses it. A lot of people jettisoned it in the nineties –- Atlantic Health, University of Chicago pushed it out. I guess we were on the threshold of doing it until we looked at the McKesson portal and it wasn’t a good fit for academic medicine in our judgment.

As a matter of fact, we signed the contact with Emergis to put in their data warehouse. That’s a Sybase product. We’re in the process of bringing it up. It will be a true research warehouse. In the past, we’d gone against our transaction systems to extract data, but we’ll pull stuff out of our production systems and put it into this warehouse. The schema is optimized for research.

You’ll be able to take your data that’s historic, since you’ve got all that longitudinal data, and move that over to the warehouse?

Absolutely. We’ll pull out all thirteen years, extract it out, put it into Star schema, and optimize it for research.

That’s interesting. What kind of projects do you think will come out of that?

Any of the principal investigators of clinical trials, research … as I said, we do about $200 million now and we hope that the warehouse is going enable us to grow that in scale. Most of it is NIH-funded research, but we do a lot of clinical trials. I think it will make it easier for the researchers to get access to patient data and financial data. We’ll use it for outcomes, accreditation reporting, and CMS.

It will be the gold standard, the system of record. We hope that we can terminate the existence of a lot of these pop-up databases. We’ve got a myriad of them and hopefully we can consolidate it all into the warehouse so we’ll know that any information that leaves this organization came out of that repository, that warehouse, and hopefully it’s accurate and consistent.

Other than Oacis, are there any other applications or vendors that you’ve run across that you think, “Wow, the average hospital has probably never heard of this product or this company, but it’s really cool and it’s doing a lot of good for us.”

There’s a product called Novo out of Georgia.

Novo Innovations.

Yes, Robert Connely. I think Robert is a smart guy. He used to be with McKesson. I think that product seems to be on a very strong trajectory. They seem to be really winning business.

Any kind of tools or anything you’ve found made a big difference or fixed a major problem?

No. I think it’s difficult for these niche players to break in because of the really big players like McKesson, Cerner, Eclipsys, Epic, and GE-IDX. I  think more and more organizations are going to move towards preferred vendors because most of these big players now have a fairly robust suite of products, both clinical and financial.

It’s going be difficult for these small players to continue to exist in the major product areas. In the small niche areas, they’ll continue, but for basic HIS kind of stuff, I think its going to be difficult and for standalone labs systems or standalone PACS. We see the integration of radiology and PACS. All these big players have that product now, so I think it’s going to be difficult for some of the small players to continue.

The HIMSS leadership survey seemed to indicate pessimism about funding, capital, and IT resources. Are you seeing any effects?

The housing market is in a rut, but people will continue to get sick and continue to need care. All the predictions show that by 2017 we’re going be spending $4.3 trillion. I don’t see that dissipating.

We seem to be doing OK. Our margins are very respectable. As I said, we just opened the new adult hospital. So I don’t see that healthcare sector being impacted by this so-called bad economy or recession. I think the demand for healthcare will continue to grow. People will continue to get sick and need the service. I don’t see the pessimism, unless it’s a spill-over from the general mood of the country.

Do any vendors stand out as either very well positioned or struggling?

I think Epic is in a good position because of their work with Kaiser. As you know, with the relaxation of Stark, hospital systems are going to do more with community-based physicians. Those organizations that have a suite of products which allow them to do that, I think, are going to be in an attractive position.

All the big players are scrambling to integrate the outpatient and the inpatient. I think that world is going to change. In the community-based setting, it’s always been bifurcated, but the model is going to be more like us. More like the academic medical centers, where you have a closed staff model; and more like what we’re trying to do with Oacis and the viewer is to have an enterprise-wide clinical environment, where a caregiver can access a patient’s information and it’s transparent to them as to where this information was gleaned, whether it was captured in the clinic or whether it was captured in an acute care facility.

With the relaxation of those laws, I think hospitals are going to be able to woo physicians and say, “OK, if you will bond with us, if you will only admit to us through this ASP model, we’ll provide an electronic medical record. We’ll house your data. It will be your data." I think that’s an issue that will have to be resolved — who owns this data. Hospitals will be able to say, “We’ll host this and you won’t have to outlay any cost”. So it think those vendors that are positioned to do that, and I think Epic is because if all that work with Kaiser, I think they’re going to be in an enviable position. I know McKesson is scrambling to try to close that gap. That’s true of Cerner, GE IDX, and others.

Anything else important going on in your world that we can talk about?

We are just trying to finish out this clinical implementation and start the revenue cycle because that’s where the money is. We’re trying to capture more of the money and collect more of the money. It all goes to the bottom line and provide the margins to fund other kinds of things.

One of the big buzzwords was PHR. A lot of the big players are moving into that, players like Microsoft and Google. McKesson has RelayHealth and, I’m sure Cerner and others. Medem, I don’t know if you’re familiar with them …

Yes. Ed Fotsch.

They look very attractive and I think they’re well positioned to do that. My understanding they have partnerships with Google and Microsoft. I think they are going to begin to gain a lot of this market share. That will be a big initiative to us — driven by marketing — trying to have more of what people want and that is online services: read the bill, pay the bill, do some pre-admission/pre-registration scheduling, online consultation with physicians.

We don’t have a lot of referring physicians, but we do have some physicians who have a special case, a transplant or whatever, and they need to move their patient into the center here. How do we push information or pull information back to those referring physicians? Also, as consumers take more control of their healthcare, they will want these health records stored somewhere. So I think that’s going to be a big push over the next few years.