Mark Hulse, CIO, Moffitt Cancer Center, Chapter 2

If you think most hospital CIOs have a lot on their plates, you’d be absolutely right. And if you also think cancer center CIOs have even more, you’d win another prize. That’s because not only must they sate the clinical side of the house, but a voracious need for data on the research side constitutes a thirst that never gets quenched. As such, Mark Hulse, CIO at Moffitt Cancer Center, is a busy man. In this interview, he talks about the center’s march toward ever greater degrees of automation, and how every step marks an advantage in research that means hope for cancer patients.

Chapter 1

Chapter 2

• Taking decision support to the next level
• Using Cerner’s platform to analyze data in real-time
• Creating a research information exchange
• Partnering with Oracle, Deloitte and TransMed
• Rolling out Cerner’s core and ancillary clinical systems
• Taking patient portals to the next level

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 (13.1MB)

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

Bold Statements

Today, decision support is really based more on evidence-based guidelines, and that’s great. That’s a major step forward from where we’ve been in the past. But going forward, how does the patient that’s in front of me right now as a clinician compare to all of the other patients that are very similar to that patient — how they’ve done on this therapy versus a different therapy?

Ideally, while the first stage of what we’re doing relative to rolling out electronic health records from a Meaningful Use standpoint is critically important for quality of care, patient safety and all the good things, they’re still very much isolated systems in many ways.

It was really going to take a partnership across several parties to be able to pull this off. So what we were looking for were potential partners that really shared that vision — not only in terms of, ‘sure, we have a product that will do what you want to be able to do,’ but more importantly, ‘we understand the future that Moffitt wants to step into and it’s very much aligned with our product strategy.’

Sometimes we were thinking pretty broadly and had to push some of the vendors that think differently, and in other cases they pushed back on us as well and very appropriately. So that’s also I think oftentimes what can form that type of true partnership — keeping each other honest to that ultimate vision.

We created a disease status field result and we’re starting to roll that out across the clinics now. This will give us the ability to be able to analyze across hundreds or thousands or even tens of thousands of patients with similar profiles to know at any given point of time how they will respond to a specific therapy.

Gamble:  When you talk about the platform that you put together that can be used to run queries to identify candidates, is this the same platform for which data can be used for the creation of evidence-based guidelines? Is it the same bank of information?

Hulse:  It is. So what we started with for phase 1, which as I mentioned went live last year, we knew we wanted to get out of the gate focused on the research piece of it first. So what we started with was cancer registry data, because that already has a lot of outcomes data collected as a part of the cancer registry, which is a nationwide requirement that anyone who is treating cancer patients capture a well-defined data set in a conformed way. So there wasn’t a tremendous amount of work to do at Moffitt or at the other consortium hospitals to bring that data into the data warehouse. That gave us a good first set of clinical data. We also brought in biobanking data— obviously the consent information, that was critical. At Moffitt we have patients complete an electronic patient questionnaire so we brought that data in as well.

Having said that, that gives us a good data set to begin with, but to your point, to really begin to look at the clinical level — how are patients responding in probably not quite real time but certainly more near time, because cancer registry data can be quite dated by the time it’s abstracted and brought in. So now as part of our next phase we’ll be bringing in at Moffitt, we’re on Cerner platform for our clinical application, so we’d be bringing in that core set of Cerner data as well.

And that’s where we’ll really begin to analyze data closer to real-time and be able to provide diagnostic or treatment decisions for point-of-care and really take what we think of as decision support to the next level. Today, decision support is really based more on evidence-based guidelines, and that’s great. That’s a major step forward from where we’ve been in the past. But going forward, how does the patient that’s in front of me right now as a clinician compare to all of the other patients that are very similar to that patient — how they’ve done on this therapy versus a different therapy? So that will be able to provide some guidance.

Gamble: And I guess that’s where it comes into play trying to move other hospitals in the consortium off of paper, just because obviously things happen quicker if you do have that data exchange.

Hulse:  Exactly, research information exchange. It really is, as we said, both for point-of-care decision support as well as for research, so there’s a vision that’s been talked about even at the national level — the idea of real-time learning healthcare system. It’s been talked about by a number of folks and it’s still very much conceptual at this point, but ideally, while the first stage of what we’re doing relative to rolling out electronic health records from a Meaningful Use standpoint is critically important for quality of care, patient safety and all the good things, they’re still very much isolated systems in many ways.

They still support the individual point-of-care transactions, and ultimately where we’d like to be able to get to, again having standardized data by which we can just automatically analyze data from all of these various EMRs across the country and really begin to look at patterns — that’s going to be important for research. It’s also going to be important, for example, at a system level for organizations that are looking at more of an accountable care model. So the whole idea obviously is being able to keep patients healthy — provider organizations are more at risk — and being able to take interventions early. You have to be able to understand certain patterns in terms of how patient populations are being treated. So for all of those various reasons, we believe this type of a platform will be critical going forward.

Gamble:  Right. And just let’s touch a little bit on the approach that you used in going to different vendors and saying, ‘This is what we envision. How can you help us with that?’ That’s an interesting approach and I think that it seems like that’s what you had to do and maybe that isn’t the case most of the time.

Hulse:  I think oftentimes as CIOs we’re just faced with okay, there’s a problem, you develop a solution set, some options around that, and typically it ends up being either the development or purchase and implementation of a particular system that addresses whatever that’s needed. So it’s a replacement for an older system or whatever it is. So I think the standard approach of a request for information or response to functional requirement makes sense in that case. But when you’re really looking much more broadly, and particularly when you’re looking at something — and we truly believed at the time, and I think it’s been validated by the work we’ve done — that where we wanted to go really didn’t exist in the marketplace, that’s why we were very clear we wanted this to be about partnership. And we knew that as big and well-known and competent as many of these companies are, it was really going to take a partnership across several parties to be able to pull this off. So what we were looking for were potential partners that really shared that vision — not only in terms of, ‘sure, we have a product that will do what you want to be able to do,’ but more importantly, ‘we understand the future that Moffitt wants to step into and it’s very much aligned with our product strategy that we want to be able to create.’

And so both in the case of absolutely Oracle, Deloitte, and TransMed, we found that partnership. We spoke with a lot of folks in the industry who very bright and some great vendors out there, but at the end of the day, sitting across the table with people from those companies, we knew they were all part of the same team and vision. So that works well for us in this case, and again, I think as we look to really doing true transformation in the healthcare industry, it really is going to take a combination of these public and private partnerships to be able to think very differently and to be able to look at this in a very coherent and sort of neutrally beneficial way.

Gamble:  Absolutely, you’re trying to find the best solution and that means not necessarily doing what’s traditionally done, but like you said, really finding the best combination and the best partnership to make that work.

Hulse:  Exactly, and also being able to challenge each other. Sometimes we were thinking pretty broadly and had to push some of the vendors that think differently, and in other cases they pushed back on us as well and very appropriately. So that’s also I think oftentimes what can form that type of true partnership — keeping each other honest to that ultimate vision.

Gamble:  That’s how it’s supposed to work, right?

Hulse:  Yeah, at least ideally. If only it was that case all the time.

Gamble:  Sure. Okay, obviously that’s a huge part of what you’re doing, but I would imagine as a CIO you’ve got a few other projects on your plate right now. So what else are you looking at now and in the near future?

Hulse:  So like most of my colleagues across the country, we’re focused on Meaningful Use. We’re focused on  ICD10 remediation — grateful in some ways that we got the timeline pushed out on that, but that probably just gives us the time we’re going to need to complete that work and to do it the right way.

But I’ll touch on a couple of them. From the Meaningful Use standpoint, like many colleagues across the country, we’re implementing an electronic medical record. I mentioned that we use Cerner for both our core and ancillary clinical systems. We’re also a development partner with Cerner for their oncology-specific electronic medical record. So I think maybe they have two or three other partners in this space, but we’ve been working very closely with them. I touched on this a little bit, but what we found is that again, while most EMRs are creating great improvements in terms of efficiency and quality of care, safety, and all of that, they’re still insufficient in terms of capturing the kind of structured data that really we feel is needed to understand how are patients doing relative to other patients, so the whole idea of how does this patient compare.

So we’re working with Cerner to embed some of the structured data elements. One example of that is every time a patient comes to one of our clinics, the physician assess the patient and determines whether that patient is responding well to therapy or not, but how they’re responding is typically captured somewhere in an unstructured note that’s free text or is dictated using speech recognition to EMRs. So we can’t compare from a structured data standpoint how all those patients are doing.

So we created a disease status field result and we’re starting to roll that out across the clinics now. So again, this will give us the ability to be able to analyze across hundreds or thousands or even tens of thousands of patients with similar profiles to know at any given point of time how they will respond to a specific therapy. It’s a simple thing in some ways and yet again, we’re still very much focused on the EMRs as a replacement for the paper record and less about the whole value of the secondary use of data.

The other piece that we’re very proud of is we’ve done a lot of the internal development for our patient portal. It does all the kinds of things that you think a patient portal should be able to do — it gives patients the ability to view their schedule, request appointments, prescription renewals, review their own clinical data, obtain evidence-based information on their own disease from high quality content sites like, for example, the Lance Armstrong Foundation, which does a great job focused on cancer survivorship. But I had mentioned briefly we also have an electronic patient questionnaire, and this is a fairly sophisticated research instrument. It was actually developed by our population scientist and epidemiologist, and it takes patients about literally 45 minutes to complete it. It takes quite a while to do; you can imagine the amount of data that’s captured. We store all of that in the health and research informatics platform and that’s linked in to clinical data. We also created a clinical summary of that information and we store that in the electronic medical record so it’s visible for the physician.

And interestingly, it continues to surprise me that because we have generally an 85 to 90 percent completion rate on that and about 75 percent of it is actually done online before the patients come. So to me, that really speaks to not only the engagement of patients in the disease but really sort of the value of that they see in being able to have all of this data information online.

The next step we want to take is again, being a longitudinal research study, we want to be able to do follow-up questions with patients to understand from their standpoint how they’re doing.

Gamble:  Those patient engagement numbers seem really high. What’s your secret? How are you able to get that kind of adoption? That’s pretty impressive.

Hulse:  We’ll take a little bit of credit for that, because I think it’s the individual that we hired as a product manager for the patient portal, Jenn Camps, has just done a phenomenal job in really listening to the voice of the patient in terms of what’s of value to them, and has helped us develop a tool that really fits not only the needs of the patients but our researchers and clinicians as well.

But I think the reality of the situation is when you have a diagnosis of cancer — and while this is true certainly of any diseases, it’s particularly true in cancer — you become very invested and very interested in understanding everything that you can about the disease and its treatment options. So as unfortunate as the diagnosis still is — and obviously we’re making great progress, what’s exciting about working in a place like Moffitt is that the patients are extremely engaged and it’s a very inspiring place to work.

Chapter 3