In February, Black Book Research’s annual health information exchange (HIE) stakeholder survey concluded that the current state of operative HIEs in the U.S. can best be described as “persistent unpredictability,” and the industry appears a ways away from achieving meaningful interoperability.
The survey revealed that while some simple healthcare information is being exchanged among parallel electronic health record (EHR) systems in pockets of communities, 94 percent of America’s providers, healthcare agencies, patients and payers persist as meaningfully unconnected in Q1 2015. Outside of their garden walled EHR networks, providers are dropping HIE as a priority, as evidenced by a 5 percent drop in regional connectivity from last year, the data showed.
What’s more, the federal government and healthcare stakeholders seem to be at odds when it comes to addressing interoperability issues. To this end, Donald M. Voltz, M.D., department of anesthesiology and medical director of the main operating room at Aultman Hospital in Canton, Ohio, says there is an answer that solves healthcare’s interoperability problems that other vertical markets such as retail, banking, and transportation have shown to work—middleware. Middleware is software that is used to connect one or more different software applications; it has been simplified as the glue or plumbing used to pass data between applications. It is currently being used to connect completely unrelated software into a single user-friendly interface, and also to connect legacy and emerging technology that have been developed using different designs, data models, or architecture, Voltz says.
A board-certified anesthesiologist, researcher, and medical educator with more than 15 years of experience in healthcare, Voltz has been involved with many facets of medicine. He has performed basic science and clinical research and has experience in the translation of ideas into viable medical systems and devices. “I wanted to try to solve these interoperability problems, but implementing processes that work both on the administrative hospital and health system side, and the clinical side, is almost impossible,” he says. “We’re constrained by whatever platform it is, be it a single EHR or multiple EHRs, based on what they bring to the table. You need to build systems that operate but also bring action to the data on the clinical side.”
Donald M. Voltz, M.D.
Voltz says that EHRs are not smart systems, but are more about collecting data right now. They are still at the database level, he says. “We talk about putting business logic on top of them, but we’re not there yet in the sense that we haven’t addressed the needs of how we interact with the system or what kind of information can be collected from an ambient type interaction,” he says. “In anesthesia, there are issues with what am I documenting, when should I be documenting, and what’s being tracked in the EHR Open loops are putting us at risk from a medical legal standpoint but they are also blocking my ability to communicate with other providers that are taking care of the same patient.”
As such, middleware technology can connect to various pieces of information and develop on top of those connections without having to move or duplicate all of the data around it, Voltz says. Specifically, for the last several years, Voltz has been using software from the San Jose, Calif.-based Zoeticx, whose platform’s architectural design has been successfully used to link data from multiple databases, irrespective to the database platform or where the database is located, Voltz says. Essentially, it is a padding layer between EHR systems that provides interoperability, he notes. Voltz says there aren’t many “true” middleware solutions on the market today— the ones that are out there are more attempts at middleware but fall short as full platforms, and are thus more like messaging systems, he says. “They’re worried about interconnecting data but not addressing the problems in healthcare,” Voltz says. Nonetheless, earlier this year, the Plymouth Meeting, Pa.-based consulting company ECRI Institute dubbed middleware as one of 10 key technologies that healthcare CIOs need to be watching.
One of the problems that actually isn’t complained about by physicians, because they don’t know the term of it, is the concept of data provenance, Voltz continues. “I don’t know care where the data resides, but I need to know that it’s accurate or I’ll end up duplicating it,” he says. Another problem, he adds, is not knowing who on the care team has addressed an issue—or if anyone even did. “As a surgeon for instance, I don’t even have access into the flow of an EHR system. So I don’t know who has looked at the information or what new information has bubbled to the top of an EHR system, whether that’s a consultant, a nurse putting in a concern about a patient, or a lab value that just came back. So as an anesthesiologist, I’m sitting in the middle of this trying to orchestrate and coordinate not only the patient but how I allocate resources in the OR and how I respond to issues in the ER. There’s no way to do that well with an EHR system,” Voltz says
Middleware, on the other hand, allows Voltz to connect these systems and connect the people, and have a more efficient way to communicate with other providers or nurses that are on the care team, he says. “We can now message information even though it’s coming from disparate systems. So if I have Allscripts in one of the offices and Cerner in the hospital, I can connect those two in a secure and stable fashion, and get the data I need,” he says. For instance, currently says Voltz, he might need to get an EKG for comparison purposes at midnight in his office in order to make a decision on the fly. “Do I proceed with this patient without comparing an EKG or another lab value or do I not? These are real issues right now and they are not solvable with our current platforms,” he says. Another example he gives is connecting an Epic platform to the McKesson drug dispensing system. “I shouldn’t have to be the connector where I have to manually re-enter all of the patient’s information and go through multiple allergy screens again on two different systems,” he says.
MIddleware versus HIEs
While some might argue that HIEs take care of these issues, Voltz is adamant that’s not the case, noting that they are modeled conceptually just like EHRs are. “Granted they will have standardization at a certain point, but they are collecting a subset of the data, and when you do that, you’re taking a lot of the meaning out of that data,” he says. “Every one of us on the clinical realm is adding pieces to that data based on how we interact, and that’s all lost unless you can synthesize everything from multiple systems. The systems are not going to fix this problem, because like technology, medicine is evolving as well,” he says.
“The next big iteration to Cerner or the next big money dump into Epic is not going to fix our needs because by the time we fix what we think we need to fix, we will have different workflows and processes that we need to address. So with the evolution and emergence of technology, which I think will take off, we will have a huge influx of more data and more technology that will come into our healthcare system to better address the needs down the road. You will always need some way to develop, innovate, and connect on the technology that’s currently in existence,” he says.
While Voltz is a strong proponent of middleware technology to sit on top of other systems, he doesn’t think that they are a replacement for EHRs—or even HIEs. “I won’t say that middleware would ever replace EHRs, but it could extend and enhance the processes that we develop in healthcare. When I look at EHRs, I have to not only search for the information that I’m looking for, but I have no idea if that piece of information is even in the system or not,” he says. “If we can build systems so we know how a team collaborates on a patient, and I can say, for instance, my nurse practitioner addressed that point, then that allows me to make better logical leaps.”
Similarly, Voltz says that HIEs and data warehouses are valuable to aggregate large volumes of data and process that data for reporting and things that don’t have to happen in real time. “If you look at both EHRs and HIEs in health IT, they’re presented in this abstract way to solve all the problems in healthcare. They’re not going to do that, though; an HIE is not going to design for me in the trenches to get my day-to-day work done. But yet it’s a great place to take subsets of data and process those, and say let’s look at pneumonia admissions across the country in January, for example. That’s a way to use those systems,” he says.
Even if HIEs link together EHRs and they do that seamlessly and in real time where a patient’s information can be accessed, and it doesn’t matter where that data is coming from, Voltz still says there needs to be another platform that can be developed on top of the HIE to make the data actionable and bring meaning to it. “Look at the iPhone. There are limits with its sensors, but the interfaces that can be developed and the way I interact with that platform can be all kinds of things,” he says. “You can do anything from a blood glucose test to applying for a mortgage to track your kids at the playground. Comparatively, the HIE mindset is just trying to connect the data, and it’s much more beyond that that data that we need to work with,” he says.