Mike Davis Looks at the Past, the Present, and the Future of EHR Adoption—and Customization

Mike Davis, a well-known healthcare IT leader, is perhaps best known as one of the developers of the HIMSS Analytics EMRAM [electronic medical record adoption model] schematic, which he and other members of the HIMSS Analytics team developed in 2005. HIMSS Analytics is a division of the Chicago-based HIMSS (Healthcare Information & Management Systems Society). The EMRAM, according to the website devoted to the subject, “is an eight stage (0-7) model that measures the adoption and utilization of EMR functions required to achieve a near paperless environment that harnesses technology to support optimized patient care.” Since the development of the original EMRAM, HIMSS Analytics introduced its an Outpatient Electronic Medical Record Adoption Model, or O-EMRAM, in 2010. The EMRAM model has also been modified for use in various regions of the world.

Meanwhile, in June 2017, Davis joined the Orem, Utah-based KLAS Research, as lead analyst. He continues to assess and analyze the state of electronic health record (EHR)/electronic medical record (EMR) development across the inpatient and outpatient healthcare sectors in the U.S., and internationally. The Colorado-based Davis spoke recently with Healthcare Informatics Editor-in-Chief Mark Hagland, regarding the current state of EHR development, and where things are headed. Below are excerpts from that interview.

When you take a 40,000-feet-up view of the past ten years in terms of EHR development, and consider the trajectory for the next few years, what to do you see? Do you think perhaps that we’re at an inflection point in terms of EHR adoption and optimization right now?

I think you’re right that we’re at an inflection point in all this. As always, I think the government is trying to do the right thing, but sometimes, the execution isn’t right. Meaningful use pushed hospitals to implement EHRs as quickly as possible, but it wasn’t always with the engagement of clinicians. So we’ve got the EHRs going now, but a lot of times, the clinicians are up in arms, because some of them see them purely as billing systems. So where we are today is that, now that we’ve got them implemented, and clinicians have to use EHRs, to report on quality measures—how do we go back now and really work with the clinicians to make these EHRs usable? What are the best approaches to actually accomplishing that, so the systems are able to drive quality of care and patient safety? Those are the things you want to look at now.

In essence, given the legislation that created the meaningful use process, EHRs kind of had to be implemented by shotgun, correct? That was one of the key challenges beginning in 2009 [with the passage of the HITECH Act—the Health Information Technology for Electronic and Clinical Health Act, signed into law in Feb. 2009].

Yes. There were some people who knew what was coming in advance, and that’s what the EMRAM was all about.

Do you feel satisfied that the EMRAM has been used correctly in the industry?

You have to consider the broad history of the efforts to measure EHR adoption both quantitatively and qualitatively. EMRAM was the first objective measure of what was going on with the EMR [in patient care organizations]. Before we introduced EMRAM, there had been efforts, such as the Davies Awards, Most Wired, things like that. But we introduced EMRAM in order to look at the full spectrum of EMR capabilities. And now, at KLAS, we’ve created an initiative called the Arch Collaborative. The icon of KLAS is the Delicate Arch in Canyonlands [Utah—the state where KLAS Research is based]. So they incorporated that symbol. The thing about the organizations that did EMRAM early—the organizations that early on got to Stages 6 and 7—is that those organizations had cultures that were very focused on improving the EMR, and making it something that would be a benefit to their clinicians. And the groups involved in the Arch Collaborative, are comparable groups of people.

The thing is, it’s a bit discouraging every time I hear someone say, “Well, we just bought an EMR from one of the big vendors, and it cost $100 million or $200 million on our EMR, and now, we’re hoping to get something out of it”—that speaks to some of the challenges we still face [in terms of strategically implementing and using EHRs].

Totally agreed. The thing is, the way in which the meaningful use process came about, while it tended to straitjacket EHR implementation processes, without meaningful use under HITECH, we would still have relatively low EHR adoption even now, correct?

Yes, I absolutely agree; had the government not pushed this, we would have made very little progress. The challenge is that the academic people want outcomes data, without understanding the impact on clinician workflow; and the vendors want to quickly modify systems to get payments. And meaningful use has slowed down, and we’ll see if Stage 3 even gets implemented. But I think everyone understands that the data being collected is valuable; and that we’ll be able to get better insights into best practices, and how they’re impacting outcomes, and making sure we reward quality as we pay people; those are pretty important concepts.

When I was in Spain last autumn, meeting with healthcare IT leaders in hospitals and health systems there, they noted that the Stage 7 requirement that hospitals establish single-dose-based medication administration, has been an obstacle to achieving Stage 7 there; they simply don’t administer meds in that format. And they’ve been hoping that HIMSS will make certain customizations to allow more international hospitals to achieve Stage 7 EMRAM. Do you have any thoughts on that?

I can’t really comment on that. Before I left HIMSS Analytics—and we were working with Denmark at that time—and I said, look, Stage 5 is closed-loop meds administration, and so many things get implemented through that. And when I go to a hospital, I’d like to see the five rights of meds administration followed; that’s a patient safety issue.

But you can see where HIMSS International might consider modifying that requirement around single-dose?

Sure. And I thought they’d already made the modification myself, but yes, I think there’s a very good chance they might be making that change. And we’re surveying organizations in the UK and Australia, and we look at how they’re implementing the electronic patient record, or EPR, as they call it. And what we’re finding is that the issues that everyone is having, across all countries, are pretty much the same. We’re talking with other countries about participating in the collaborative. And when you work internationally, you have to be pretty sensitive to how other countries work, and we are.

What are your aspirations for the new Arch Collaborative?

We’d like to continue to take this and blow this out. We want to figure out the best practices for education, for personalizing EMRs, and for EMR governance.

What is the scope of the collaborative, or what will it be? Do you see entire countries joining and participating?

That’s the hope, with the UK; we’re working with one of the leading NHS [National Health System] hospitals over there, so yes, that’s the hope. In the US, the process is that we have a standard survey that we distribute; it involves 25 questions. We do allow some modification of that, including for demographics: for example, some versions of the survey include nurses, some don’t; some include specialists, some don’t. And maybe they use different EMRs for inpatient and outpatient. So we make modifications. Once we establish a complete survey for an organization, we help them work with their clinicians. We leave that open for 30-45 days. And we get about 20-percent participation, and somewhere between 85 and 90 percent complete the survey.

What have the results been so far?

We’re just doing this on Survey Monkey right now. We may look at using a standard tool that KLAS uses. The key thing is to keep this simple. When you look at the type of data that we get out of those 25 questions, it will blow you away. That generates about 125 PowerPoints. We create a net EMR experience score. There’s a total score for the entire organization; we can care inpatient and outpatient scores. So we can look at inpatient satisfaction, or ambulatory satisfaction, or any point.

You mentioned just now that physicians and nurses are involved?

It’s much bigger than that: it encompasses physicians, residents, fellows, advanced practice providers (PAs and nurse practitioners), and nurses.

When did this process begin?

They got their first survey done in the first quarter of 2017. We’ve been modifying it. Not only do they get the raw data and PowerPoints, but we get a formal report. We’ve done about 30 final reports. In process, another 40-plus organizations are at some point in process. If they signed up to say yes, we want to be a collaborative member, we take them through the process. And if an organization just wants to do the survey, we do that with them. So, here’s a question for you: when you buy a smartphone, how much time do you spend personalizing it?

Not much…!

So you would be one of those “bad” physicians! I just bought one recently, and spent about two hours personalizing it. My point is that, when it comes to EMRs, you want physicians to personalize their EMR, to meet their individual needs. Some organizations have limited that, saying, geez, if I give them that capability, it will be a nightmare to support. Other organizations will customize by specialty; and a third group says, we’ll let them do whatever they want to.

Is there a best approach, in that context?

There has to be some allowance for personalization. It depends on their EMR governance and education. With good education, they don’t have much trouble supporting a lot of customization.

Will customization will lead to greater usability and success?

Governance and culture go together, and are key. If organizations are really behind this and want to make it work, that’s number one. Number two is that the second-biggest impact is initial education or training. When you do the initial training—when you go in and work with a cardiologist, and help them set up their workflow, they find they’ve learned the EMR well enough to be successful. And the third thing is the personalization; the more they personalize that EMR and the more personalized, they more successful they’ll be.

What about data collection?

If you personalize workflow, that should make it much easier to collect and report on data. Now, we train the physicians on how to put data into those systems. Most people don’t do a great job of that. Our big example is customized report views. There are certain types of information as a cardiologist that I want to be able to get easily, and I don’t want to have to go to a lot of places. If they can go in very quickly, they can get it. That’s the stuff we don’t do a really good job right now on setting up for the clinicians, or even on training them to do it. We understand that you have to put the data in, to get paid. But getting the data out, is really important!

Will we get to a point where physicians will be in a better place with this? One of the complexities in all of this now has to do with physicians feeling overwhelmed, and turning to scribes. But the use of scribes itself introduces additional complexities.

It’s interesting that you brought up scribes. We’ve found that most people aren’t satisfied with their scribe services, or people entering their orders for them. So that has to be looked at. And what about voice recognition? What we’re finding is that a lot of people aren’t getting the expected benefits out of voice recognition. I think that they’re not doing the appropriate integration into the EMR. I think we will see it—I think at some point in time, the voice recognition will start to provide us with an approach that works. The other element is the templates and order sets—if those are done correctly, they should be able to very quickly document what they need to document, and as a byproduct of that, the ICD-9 and CPT-4 codes, should be encoded, to drive data collection. So I think you’ll see better designs of workflows that will codify information, and then you’ll see improvements in how that information is captured, and that will be a byproduct.

I think that’s what physicians are looking for. They’re complaining now about the number of clicks it takes them to get through an order set. And I don’t blame the vendors; they’re trying to stay afloat, and have been putting everything into R&D so that people could get paid. I think the vendors will do a much better job of making the EMRs more intuitive, and that will drive efficiency. I think that’s where I think this will go.

Do you think we’ll get to a point where EHRs will be user-friendly, physicians will no longer be deeply dissatisfied, and the workflow and data flow will easily facilitate the data collection and measurement we need in order to support population health management and value-based healthcare? How optimistic or pessimistic are you on all that, overall?

I am more optimistic now than I’ve ever been, actually. I think we’re getting closer to that reality. I think the clinicians recognize that the practice of medicine is changing, and EMRs are a part of that. So they’re getting engaged. So I think we’re getting there. And we’ll be successful.