Interoperability: How close are we?

October 25, 2015

If interoperability is to become a reality, the government and private sector must overcome existing obstacles. But how should the industry proceed? Our experts explain.

To date, the concept of exchanging patient health data-interoperability-is more vision than reality. While existing in pockets across the healthcare landscape, there is still no true electronic information highway for enabling physicians to share patient data efficiently.

Recently, Medical Economics convened a panel of experts to discuss the current state of interoperability and how to move closer to a goal of truly connected patient data. The panel consisted of:

  • Robert Rowley, MD, a practicing family physician, health IT consultant, and entrepreneur.

Is interoperability a feasible goal?

Michael J. McCoy: Well, I think first of all the question has to be ‘what is the definition of interoperability?’ I think that, as in many other industries, there are pockets of what would be considered interoperability now. But if you put the focus [on a definition as] the movement of data that’s relevant and important for patients, so that it’s usable by clinicians wherever they may reside and whatever the setting is, that to me is the most important definition of interoperability. And you get beyond the general, technical framework of what is interoperability, to the “what does it really mean?” in use. Yes, I think interoperability is a reasonable goal when you focus it on the movement of data that’s required for taking care of patients.

If you broaden that to research, to population health, again, I think that there are still some priority data elements that ONC and our federal advisory committees have pointed out as the crux of the most important information that needs to be moved around. We’re making great progress in that area as well.

John D. Halamka: I think Mike is exactly right, the definition of interoperability is so key. I was editing an article last night where an author was lamenting that if they in their practice referred to another practice, a patient for say, cardiac care, but they in the future, had no idea if that patient ever went or was treated, that that is an interoperability problem. So of course, if we describe this as a closed loop referral workflow, is that interoperability?

Related:The debate over healthcare interoperability

Or I was testifying to a group and I said exactly what Mike said, ‘Interoperability is the exchange of defined elements, such as problems, meds, allergies, notes, labs, diet, care plan, [etc.].’ They said, ‘No it is everything, for every purpose, for everyone!’ And I said, “Well, if you define interoperability that way, you are never going to achieve your goal.” Because you want hair color and toenail length? … So let us cleanly define interoperability.

Leigh C. Burchell: It’s true that healthcare is arguably the most complex industry in which to create an interoperable data exchange environment. But other industries have done it and they’ve done it well. I think there are some parties who sometimes let their impatience for the end goal get the best of them and have some urgency to figure this out. But we’re well on our way.

There are a number of metrics that provide encouraging news in this area. From the percent of healthcare organizations that now use robust electronic health records and I think that matters in the context of interoperability, because it’s a fundamental requirement for electronic data exchange.

The number of hospitals exchanging data with providers outside of their walls has increased rapidly. There has been exponential growth in the utilization of Direct protocols for example, [and] the rapidly increasing volume of primary care physicians who are submitting electronic immunization to local public health agencies. All of those have increased very rapidly in the last several years.


NEXT: ONC as the coordinator and collaborator


Robert Rowley: Let me take a different tack on this. Because I think we’ve all been talking about how we can get the current systems that exist in the marketplace to exchange data, and I like the point of defining what kinds of data need to be exchanged. To me that is a fairly organization- and physician-focused way of looking at interoperability.

I’m going to throw in a somewhat different, somewhat radical perspective, which is that if we are trying to achieve interoperability between data silos, but we leave the silos intact, then we are asking the wrong question. I think the goal is universal longitudinal patient center data; something that follows a patient through their lifetime, as they move through different health plans, different providers-in and out of different systems.

Related:ONC aims for improved interoperability by 2017

I don’t believe this can be achieved, so long as patient data remains locked within the electronic health records (EHRs) that we see on the landscape or on the horizon today. I think that the next generation of health IT will have freed the data into external, universal locations. And EHRs will interact with those universal sources. And they can move away from a one-size-fits-all approach, which is I think what we have now, especially with the major vendors. We can use smaller user interfaces that are tailored to specific use cases. And it won’t really matter because they are all interacting with the same data source.

John D. Halamka: If we look at our millennials and their involvement in say multi-player online role playing games, here we have a cohort of 20 team members interacting in real time virtual reality from their iPhones and laptops. It puts medicine to shame. And one asks “Does Wikipedia and Facebook do a better job at creating shared care plans, and the idea of teams working together [better] than an EHR?” [It’s] probably true.

Russell P. Branzell: The reality is we have interoperability today. It’s just small amount and it’s only occurring usually in fairly small silos. So the concept of whether it’s realistic, well, it’s already here. Is the end state, as John stated earlier, everything, everywhere for all time? Well, that’s going to take a really, really long time. Well beyond any of our tenures in our jobs, or even in the workplace.

But the realistic view of this is, what needs to be exchanged and what time frame? And if we need to get it done in a short period of time, 18 to 24 months, that’s a much different prospect than say a 10-year road map. So the question is: “How fast can we go at a reasonable pace of adoption to accelerate the exchange?”

Michael J. McCoy: But I think overall, the federal government has done a good job in helping guide to the point where we are. I think we still have a lot of work to be done. ONC is not the doer in this regard. It is again, the coordinator and the collaborator with a variety of other people represented on this call that can help further the goals of what we need for better healthcare for the country.


NEXT: "Information blocking" in healthcare today


Let’s discuss the process of “information blocking” when it comes to sharing data via EHR. ONC, in its report to Congress, said it had received approximately 60 unsolicited reports of the practice, mostly directed at IT developers. What is the current level of information blocking in healthcare today?

Michael J. McCoy: As we did indicate to Congress, we think that there are some vendors blocking, which may be either intentional or unintentional. I think some are indeed from vendors and their business practices; some are from health systems and their business practices.

I find that in my experience in dealing with individual physicians, most physicians are less problematic in intentionally having data blocking though they may be unintentionally doing so, by not participating in portals, not participating in other kinds of exchange that are possible. It may be financial reasons that they are not participating in some of those areas.

Related:An interoperability report from the field: It's not pretty

So when we look at a large landscape of things that can lead to the data not moving, again, my personal view is that not necessarily data blocking, but the data moves very, very slowly, and perhaps in an intentional fashion to move slowly. Because as been discussed many times, how many hospital CEOs wake up saying, “Oh, how can I make it easier for my patients to go to another hospital?”

Leigh C. Burchell: There are a lot of anecdotal stories. And that could be about software developers, it could be about providers, it could be about anybody. So there’s a lot of anecdotal information flying around, but not a great deal of quantifiable data or real research. So the acknowledgement of that [from ONC] was something we really appreciated …

The business model for information exchange right now isn’t fully baked, shall we say. In fact, the market drivers, the payment models that we have in place right now, don’t really encourage providers to exchange information for fear of losing a patient to somebody else, and losing that revenue source.

So I think, from that perspective, it raises questions about information blocking. For example, if a hospital receives two requests for connectivity from two physicians, one is in a local area, where there’s a strong referral pattern, and the other is 200 miles away in a frequent retirement area for their patients, but clearly the information volume would be less. The question would be, if the hospital prioritizes connectivity with the local person and the work on the more distant one takes longer, is that information blocking? I don’t think so. I think that you have to look at the business model environment and be realistic and pragmatic about the work that is going into this.


NEXT: In the middle of EHR adoption


John D. Halamka: Do our vendors have chief information blocking officers, who wake up every morning, figuring out how to reduce interoperability? Of course they don’t. The challenge, of course, is exactly as was described by the others, if interoperability is exchanging every byte of data with every person for every purpose. Oh, well the lack of that could be information blocking.

I will tell you in Massachusetts, we have a very strong business case for sharing data because we have so much global capitated risk. We have been very early adopters of this idea of pay for quality, pay for outcomes and not fee-for-service. It is a business imperative throughout the state to share data. So where there is a business driver, I will tell you I have never seen in the state of Massachusetts, any issue with sharing data from any vendor or any provider. I just have not seen it. And it’s because there’s a business driver.

Russell P. Branzell: I think what’s occurred here is we are in the middle of a journey as we’ve headed down this acceleration of adoption of electronic records and we’re in the middle of the journey.

Related:Disconnected: How stalled interoperability hurts patient care

The way I try to compare it, is we all start in Seattle, we’re all heading to Miami, but none of us took the same path to get there. So we are in the middle of a period of pretty serious dysfunctional adoption, which I think accelerates the perception of people not willing to exchange data or in some cases even the concept of the information blocking. And I agree with everything that’s been said relative to, it’s very difficult to find examples of true information blocking and I’ve spent quite a bit of time with our vendor partners, as part of CHIME and just like John, have never found the office of information blocking in any of their headquarters, maybe it’s hidden in a dark secret place but by their words and their actions they’re working very hard.

Robert Rowley: I think [information blocking is] something that sort of happens, more tthan he matter of corporate priority. I think that the problems, though they are a challenge … are primarily technical. I think they are mainly about organizations wanting to share information.

The barrier is a political barrier. I think so long as there are multiple organizations competing for the same market, there is going to be a resistance to share data with your competitors. And that drives more reluctance. It’s not refusal to share data because the patient can ask for their data. They just make it more difficult...

I don’t think it’s willful but I think it’s just a reality that a lot of organizations are really tapped out, they’re trying to deal with things like ICD-10, they’re trying to deal with software upgrades, and things that they’ve purchased in the 90s that are no longer functional and they just can’t get around to dealing with new ways of looking at interoperability.

Michael J. McCoy: I would echo and one of the things that the Medicare Access and CHIP Reauthorization Act legislation is trying to do is and what HHS is trying to do is align so that we are making those drivers a reality. It is about data sharing and overall health with payments being tied to outcomes. It’s going to require physicians and hospitals and others in the community to understand the need for sharing data so that it’s relevant and timely and effective in reducing costs and improving outcomes.


NEXT: Who takes the lead on interoperability?


Who should take the lead on interoperability, the government or the private sector?

Robert Rowley: I think the government will set standards. They’ll create or specify standard vocabularies, standard message types, standard [human-computer interaction] functionality needed for certification of products. But I think the marketplace will determine the particulars.

An analogy I like use is the auto industry. There are thousands of different kinds of cars, but the government has set specific safety and functionality standards that everybody must meet. I think the EHR industry will evolve in that way. Everything that comes to market and is certified needs to have certain capabilities to be able to exchange data easily. And I think the previous state of EHRs was to facilitate workflows within an organization, but wasn’t really thought of as exchanging data with others.

Related:The right kind of interoperability must become reality

Now the rules of the road are that exchanging data with others is the most important thing. And I think that will come from the government. But I think the marketplace will determine the particulars.

Michael J. McCoy: Well, I think as stated, the role of the government is to help facilitate, not so much to regulate. I think when you look at our opportunity, it is to help convene those stakeholders–vendors, developers, large hospital systems, those representing small physicians and clinicians, and again, don’t forget the consumer or patient involved in all of this. And looking at balancing the requirements, making it happen in an expedient fashion, is a challenge that we all have to take on in order to achieve the ultimate best outcome that we can.

Leigh C. Burchell: We hear from our clients regularly that they just want some breathing room to work all of this into the new way that they do business. But I do think that there are some areas that we think government support is probably still necessary. Privacy and security being one; that’s certainly something we all need to stay on top of.

We need to be thinking through cybersecurity threats and also looking at variation in interstate laws and regulations around opt-in and opt-out-those types of topics is certainly something that still needs some attention, promotion in an appropriate way, and mapping out the best opportunities for telemedicine. I think this obviously still requires some effort because there’s such variation from state to state.

Russell P. Branzell: I think this is definitely an area where we can take a more aggressive approach. There are so many equivalent solutions where we are able to cross this country in different ways, whether that be the transportation system or financial system. We’ve figured out fairly quickly how to cross this country in an electronic format and/or even manual format.

We need to learn from those other industries and come up with a way to do this. If our Federal Reserve system worked like what we are trying to exchange [health data] on, we’d have different currencies for every state and we’d all be having to exchange currency when we travel every single day. There’s got to be a better solution to this in which we can look on fairly quickly.

John D. Halamka: I think it is unlikely we’re going to end up with a giant database in the basement of the White House, run by President Trump. I think in fact, we are going to have government working on standards, frameworks for governance, and to ensure that as a country we have regional solutions that solve the business problems of each locality.

I really think we are on the cusp of success. The private sector is taking a new role, with such things as the Argonaut Project and new standards are getting balloted every day. So that combination of some government, some private sector and the business cases will push us over the finish line.