What is the Problem with Interoperability?

In a recent article by Dave Levin, MD, the astute and experienced physician points out the "reasonable, but incorrect assumption that two installations of the same EMR can easily share data...The hard truth is that every implementation of an EMR is different and even same-brand EMRs do not seamlessly connect." Why not? It seems almost shameless in an industry that has had a directive to achieve interoperability, at least since George W. Bush's executive order 13335 in April of 2004: "Incentives for the Use of Health Information Technology and Establishing the Position of the National Health Information Technology Coordinator".

That order was made over 13 years ago, the same period in which Facebook became the foremost social network in the world (with extreme interoperability), and about the same time that Amazon expanded into jewelry, shoes, and electronics, rapidly becoming the largest retailer in the world and making its founder, Jeff Bezos, nearly the richest man in the world. Why are EHRs not able to obtain the same level of interoperability?

The missing piece is an absence of standardization. Even within a single vendor, achieving interoperable standards is apparently harder than climbing Everest, and less survivable too. A number of organizations have attempted to create EHR standards, notably HL7. Their valiant efforts have been largely ignored. Again one must ask: why? For the most part, one must point a finger at the government's inability to establish and mandate simple interoperability standards. The efforts that have been made have been an abysmal failure.  Meaningful use? Seriously! Many of us say: meaningless use. It is not a standard, it is a hand-tying imposition on progress. There would be no Facebook, no Amazon, if such impediments were operative in the general internet commerce and social media space.

Think of it this way. GE, Samsung, Amana, Westinghouse, all make refrigerators and other appliances. They all seem to work fine on 120v current. Why? because that is the standard, and the companies know that is how they have to be built. What if (as with EHRs), companies were allowed to construct their innards any which way? We would not have cold food, or ice, dishwashers or air conditioning. None of the devices would be able to talk to the electrical grid. But now, EHR vendors can build what they want, can tell potential client hospitals (as well as the VA and the DOD) that they alone have it together and you better stick with them exclusively. Would we not rather choose an EHR based on how easy it worked? Knowing that it spoke the needed language as a baseline. The culprit in this is VA's and DOD's boss. In 2004, George Bush saw it, but the ability to stand up to the big companies that deliver health care tools is absent.

Until the government gets out of the business of trying to tell doctors how to practice and instead tells vendors they have to make EHRs that meet an interoperable standard, chaos will continue to reign.

Interoperability via Apps- A guest blog by Dr. Donald Kamens

I came across the article What Doe Epic's App Store Mean for EHR Interoperability?  I had the same conclusion as the author, "This will cement their long-term legacy."

In 2004, then President GW Bush inaugurated the US push toward electronic medical records.  Even at that time, it was recognized that if the effort to have a functional system were to be successful, systems would have to talk with one another, that is, be interoperable.

Fundamental to the concept of interoperability is the principle (or hope) of document exchangeability.  That is, a patient who enters hospital B in city Y should be able to have information from his visit to hospital A in city X sent electronically and incorporated into the care now being delivered.  Interoperability remains a pipe dream.  Why?

There are two fundamental forces at play; one is the complexity of the task such that different parts of the scheme do not inherently understand one another.  The structure of an order set, for example, requires integration to be metabolized and seen within a narrative of care.  Hence narratives are either foregone or synthetic nonsense when left to the cognitive devices of the machine itself.  The other force is the economic benefit that large scale system vendors garner by preventing interoperability with other systems.  MedX can sell more systems if its clients are "stuck" with buying theirs to maintain harmony within the sum of parts.  If a hospital complex as a number of care-settings on campus, and some off-campus, why would they want to risk purchasing an EHR that would not allow conversation between the sites?  Of course, they wouldn't.  And yet, enterprise vendors are rarely able to deliver on the promise of interoperability even between settings that the vendor has supplied, on or off campus.  And that is just the history of a typical single large scale vendor trying to talk to itself.  In one language, in one city, in one country.  You can image the babel when the conversation "should include" machines created by other vendors, at other locations.  It is a mess!  The fall back becomes the fail-safe dependable device for transmitting information: paper.  No one who works in medicine has been immune to the exasperation of creating coherent and reliable, non-fragmented, data accessible, helpful and supportive, medical records.  Pipe dream...

 

In fact, it has been enough of a distant dream, and enough of a perceived necessity, that the US government has incorporated a requirement for interoperability with an "Interoperability Roadmap" within its plans for increased adoption of electronic records and increased abandonment of paper.  Get interoperable, it is saying, or get off the playing field.  This command makes some of the big vendors sweat.

But now there arrives on the scene a large vendor with an idea.  It is a "now why didn't I think of that" type of ah ha moment when one hears the idea.  Epic, perhaps the largest and fastest growing EHR company in the world has said it will solve the interoperability issue by permitting apps t be created, and function within their systems as semi-autonomous modules.  We all know apps, they are on our phones, n our iPads, on our computers.  Short for "application" an app is the ideal platform with to push interoperability.  Apps do not necessarily produce or accept interoperable data, but the vendor that allows an app to function within its technical borders can require that the application is able to function in that respect.  If you live in Rome, the vendor might say, do as the Romans.

Of course, this does not solve the overall nightmare of inter-system interoperability, but it puts it into the hands of module vendors who are historically more capable of solving the issue than the big players.  The larger nightmare will only be awakened from when a cogent interoperability standard, and standard maintenance format, is adopted and required.  Some organizations, for example, HL7 Standards group have been working on this for decades.  There are good schemata that can be applied, therefore, to the issue.

The requirement for standards to achieve interoperability is not a new concept.  Apps to achieve interoperability is entirely new.  Just like the multitude of app vendors one can explore on an Apple or Android device, so the potential for functioning modules within the enterprise EHR systems opens a door that will make clinician choice the center of future EHR purchases.  One considers, the app availability and quality, usually when deciding on an Apple versus Android phone.  Either that, or one is confident the choices will be sufficient to meet needs.

Can't you see yourself reading app reviews to choose the EHR system for your ER or clinic?  Trying out a demo for a few weeks, etc.  It is about to be a new world.  How the economics, security, reliability of this will work, is currently unknown.

One thing is for sure, Epic, not the darling of many clinicians because of its inherently cumbersome interfaces, content building requirement, and progressively ubiquitous presence, has come up with a terrific idea.  Let's see if it comes into play.  If it does, small boutique vendors that have fine-tuned the ways to make ERs and clinics hum, will now be able to offer their products to those who have wanted them.  It will be like putting a fine stereo system in a car.  Hopefully, the music will be good.

Where Has All the Interoperability Gone?

The entire meaningful use project intended to encourage provider adoption of Electronic Health Records (EHR), with the promise of easy access to patient records with two underlying goals: 1) improve the quality of health-care and 2) control rising costs.

Interoperability represents the ability of systems and organizations to work together (inter-operate).  Due to technical constraints systems often impose, the essence of interoperability in health-care has become, in essence, the need for easy, reliable exchange of information between these systems.

The EHR was supposed to interact with other EHRs to synchronize individual's medical history including tests and treatments.  Such synchronization in the US is important, as many patients wind up in different care settings.  This synchronization would give a care taker permission to all of the patient's clinical information.  This clinical data would be stored in a Health Information Exchange (HIE).  The HIE is defined as a mobilization of health-care information electronically across organizations within a region, community or hospital system. 

Although well-intended, the meaningful use program with its billions of spent dollars has further spurred the development of multiple competing health records that are by nature proprietary.  Enterprise systems, that provide the backbone of huge hospital systems are reluctant to interface with practice specific specialty "boutique" programs.  Why should these enterprise systems enable integration with boutique programs, when monopolizing their implantation over a hospital system is immensely lucrative for a vendor?  However, the reality is the boutique systems are often focused on end-user acceptance and walk-up usability, while enterprise systems--by their very nature, and especially by their hybrid, unfocused character--are unable to satisfy all end-users, if any at all.

How hard is it to interface through the HL7 standards developed for meaningful use and HIE?  According to most experts, accepting these standards takes the willingness of the participants involved.  For business reasons, many large organizations do not support standardization, and so interoperability--a functionality fully dependent on standardization-- is not an easy task.  Even if a hospital may be willing to accept the interoperability challenge, the cost of implementation is often artificially elevated by the proprietary vendors, making it prohibitive.

A related factor, referred to above with the term "hybrid", is the fantasy of a one size fits all product.  Hospital IT departments love hybrids because they theoretically means less hassle.  Their focus is less on the end-user than on avoiding potential problems for the IT department.

Hospitals, CEOs, private practitioners, politicians, and all health-care advocates should encourage their EHR vendors to allow the concept of interoperability to be enacted upon and enhanced.  You can find more information on the subject at the Center for Medical Interoperability.