Tim Zoph, Chair, KLAS Interoperability Measurement Advisory Team, Chapter 1

What a difference a year makes. In the fall of 2015, the organizers of the first KLAS Cornerstone Summit weren’t sure if industry leaders would even show up, much less stick around long enough to provide input. Luckily, their fears were unfounded. Not only did a dozen CEOs show up, but they gave enough feedback to get the ball rolling on a groundbreaking interoperability survey that KLAS hopes will provide some much-needed answers. We spoke with Tim Zoph, Independent Chair of KLAS’ Interoperability Measurement Advisory Team, about what they’ve learned from the preliminary survey results, the roadmap being developed to more effectively measure interoperability, and why he believes this is only the beginning.

Chapter 1

• The 2015 initial Cornerstone Summit
• True clinical impact of interoperability — “What really makes a difference in these transitions?”
• Goal to have measures in place by 2018
• “No one can solve interoperability alone.”
• Technical vs context challenges
• “We were just measuring interoperability of EMRs. There’s more to be done.”

Bold Statements

Our intent with the survey was not just to talk to CIOs or just folks within IT; it was important that we actually got out to clinicians so that we could understand how effective the use of this information was in the caregiving and decision-making process.

We believe it’s important when the day is done that there be measures related to the clinical impact of interoperability. In other words, it’s not just going to be sufficient when the day is done to say we delivered the information; we want to know what really makes a difference in these key transitions or handoffs of care.

No one can solve interoperability alone; it’s not simply a function feature in the product. It’s going to take commitment on the provider’s side. It’s going to take commitment on the federal side, and ultimately, the engagement of patients, which today are really the messengers of interoperability in the industry.

Bringing this forward into your own native systems, getting it into the workflow, and making sure it makes a difference in solving these multi-stakeholder needs — I think the context challenges of interoperability are certainly equal to the technical challenges.

Gamble:  In late August, KLAS held its second annual Cornerstone Summit in Utah. This being the second year, can you talk about what the expectations were and what you hoped would come out of it?

Zoph:  Sure. I think the best way to do that is to give some background. At the initial summit in October of 2015, we brought the industry together, which meant the 12 largest providers of electronic record solutions in the industry, along with providers, to create with support of KLAS an interoperability measurement framework. We had a group called the Fantastic Four with Micky Tripathi, John Halamka, Stan Huff, Daniel Nigrin, who came together to create a questionnaire that would create a way to determine the effectiveness of interoperability.

The framework was presented, and we had group discussions around it and made modifications to it, and then it was adopted jointly by both the provider and vendor communities as a reasonable first step in getting at measuring the state of interoperability for the industry. KLAS, being a neutral and honest third party, agreed to take on the challenge.

A couple of things happened. While this was going on, we recognized it was important to keep the momentum, and so with health IT being a collaborative industry by nature, we formed an interoperability measurement advisory team. I’ve been chair over that the past year. The purpose of that team was we knew going in that there would be challenges with administering the tool and maybe with clarity around the tool. So we wanted to have a mechanism for KLAS to have a group that they could go to say, here’s a challenge we’re facing in the survey process, how do we meet that challenge? And with interoperability, it’s not just a technical measure, but it’s also a use and a usability measure. Our intent with the survey was not just to talk to CIOs or just folks within IT; it was important that we actually got out to clinicians so that we could understand how effective the use of this information was in the caregiving and decision-making process.

KLAS then set out to administer the survey. They’ve had more than 503 conversations with more than 440 unique organizations — half of those conversations were either directly or attributed to clinicians, so we were able to get a lot of clinical feedback about interoperability. And then with the conference this year, we brought the group back together to do a few things. First, to share preliminary results and feedback from the survey (which will be officially published October 4) and then, to hear from KLAS about what were the challenges with the survey, what do we need to modify, and how do we use this conference — which really was a full-day working session — to address some of the challenges we saw in the first year. They could have been question challenges, for example, how do we get to the right people, do we have the right sampling, is this feedback going to help you improve your products and services so that interoperability improves.

We wanted to make sure we provided KLAS with feedback, and so we spent the second session of the day looking at performance benchmarking — how we aggregate all of this into a scoring system so that the industry can understand its overall performance and ultimately turn that into a scoring tool for the vendors so that if it becomes an important buying decision as to how well a solution is interoperable beyond just the function feature, there will be a scoring mechanism that could be used by the industry to rate vendor solutions on their ability to interoperate.

And then finally, the third working session was spent creating what we call a long-term roadmap. We started this last year with a fairly narrow focus. We looked primarily at interoperability of electronic medical records. We did that with the 12 leading industry vendors looking at what we called critical exchange partners; that is where information was most needed in the health delivery ecosystem. By and large, it was acute care and ambulatory that were included in the questionnaire.

We recognized that that was a good start, but we needed to go further and measure more technology, like medical devices, for example. We needed to look at a broader expanse of the delivery system that goes beyond simply acute care and goes out to secondary care and post-acute care, and ultimately goes to the patient. And in looking at the measures themselves, we really want to get at, does this make a difference? We also believe it’s important when the day is done that there be measures related to the clinical impact of interoperability. In other words, it’s not just going to be sufficient when the day is done to say we delivered the information; we want to know what really makes a difference in these key transitions or handoffs of care.

So we had, I think, a really solid day. We worked on all three of those areas. We affirmed that the baseline survey makes sense with modifications, we came up with a scoring system that we’re going to publish, and we developed an interoperability roadmap that will seek to measure interoperability more completely and have all those measures in place no later than 2018.

That’s what we used the conference for, and so we’re going to publish in October, and go through another round of measurement with a modified survey. I think the big takeaway from the conference was that collaboration is essential, interoperability is hard, and actually having a forum where we could bring the provider side, the vendor side, and ONC is invaluable. We had the ONC team there and we had [National Coordinator for Health IT] Vindell Washington join us and speak that evening. We think collaboration is going to be important, and having all those perspectives in the room is going to be important.

Another thing we realized is that no one can solve interoperability alone; it’s not simply a function feature in the product. It’s going to take commitment on the provider’s side. It’s going to take commitment on the federal side, and ultimately, the engagement of patients, which today are really the messengers of interoperability in the industry and these key transitions of care. The patients need to be engaged because it’s their information; it’s these handoffs that are so critical to patients and families, and so they need to be a part of the interoperability solution as well. And so keeping the industry together on a set of agreed-upon measures may be one of the more important things that we’ve accomplished over the last year.

Gamble:  Right. Now, the interoperability survey hasn’t yet been released, but what can you tell us about the preliminary results?

Zoph:  I think we have a good baseline survey. If anything, people want more clinical perspective, and even more than the 500-plus conversations we had. One of the things you realize is that there are a lot of users of interoperable information. If you go into the care setting today and you look at the information that’s being transferred with the patient or being used in a care handoff, you see that physicians are involved, care coordinators are involved, nurses are involved, social workers are involved, and the patient is involved, so you have a lot of stakeholders. What we want to do is make sure we’re bringing in those perspectives.

You’ll see in the scorecard itself that we’re not just measuring the technical interoperability. I think if we’ve learned anything, it’s that although we’re getting more information out in the industry — a lot of it through fairly traditional HIEs, which are the prominent method of communicating information — what’s happening now is even though it’s being communicated, a lot of that information is not readily available through the electronic record. It’s not easily found, it’s not necessarily within the workflow, and people are having a hard time really saying that it definitively makes a difference. So if we’ve learned anything, it’s that context and workflow matter in interoperability. In that regard, I’ll share some takeaways we’ve learned.

One, we need it more integrated into workflow; we need it more intelligent so that systems can prompt and say, something’s different here, or here’s how this information ties into clinical decision support. We want the information to be more accurate and more trusted, and we want to make sure people have confidence in it — the source of it, its language, and its terminology. And as it’s being sent around, people trust that tests don’t get reordered and questions don’t get asked and answered several times.

I would say it still feels like we’re in the very early stage of this interoperability; there’s information flowing, but we need to do a better job of getting it to the right place, at the right time, and in the right context for the users of that information.

Gamble:  Does it become difficult when there are so many constituents involved and so many opinions as far as how you’re going to get this right?

Zoph:  Yes. And looking back to the conference and the 70-plus people we had there, it’s really important to be collaborative and have lots of different perspectives. I think over time, it would be great to have patients at this conference; we had clinicians, but it would be beneficial to have even more diverse clinician views. The stakeholder needs in interoperability are very challenging and you want to make sure you have all those perspectives, because the users of the information ultimately are the ones to judge the impact. And so I would say one of the more challenging things about measuring interoperability is making sure we’re getting to the right stakeholders to get the right perspective.

Gamble:  Which, of course, is no easy task.

Zoph:  And it’s not just talking about function and features. Sometimes in a product rating system, the function either exists or doesn’t; the feature either exists or doesn’t. This one is far more challenging because you’re talking about how you’re bringing forward information from lots of moments of care for patients and bringing that into a context of institutions on record, it’s challenging.

Gamble:  What do you think will be the biggest takeaways from the survey findings?

Zoph:  I think one of the most important things we’ve learned is that we still feel like we’re at a very early stage of interoperability, and that the challenges are not technical alone. We’ve realized too that it’s important that providers, patients, and vendors all work together to make sure that we’ve got the information in the right place. In terms of the technical measures, I think it’s going to get easier to move information around; people are more optimistic about emerging standards like FHIR. But bringing this forward into your own native systems, getting it into the workflow, and making sure it makes a difference in solving these multi-stakeholder needs — I think the context challenges of interoperability are certainly equal to the technical challenges.

What I’ll tell you with the report, and you’ll see this going forward, is to keep in mind that we were, by and large, just measuring interoperability of electronic records — there’s more to be done. We’ve got to get post-acute care, we’ve got to medical devices, and we’ve got to get mobility. So I think there’s a lot of work to be done to have a broader statement of interoperability other than simply how well do electronic medical records exchange information.

Chapter 2