On 28 February 2016, the Journal of the American Medical Association and IEEE Pulse hosted a one-day symposium in Las Vegas, Nevada, concurrent with the Healthcare Information and Management Systems Society’s 2016 annual meeting. The event, which was the first in a series of technical talks dubbed IEEE Pulse On Stage, focused on the struggle of the U.S. health care system to have a scalable adoption of efficient and interoperable electronic health record (EHR) systems. Keynote speakers from both the medical and technical communities shared their insights on usability roadblocks as well as possibilities for reimagining the future of the EHR. The event was attended by a mixed audience of clinicians, informaticians, and health IT researchers.
The major takeaway from the event was no surprise: while it is not hard to reach agreement about how we got to this point of major struggle with nationwide EHR implementation, it is a lot harder to agree on a vision that can move us beyond this challenging situation. The importance of solving the EHR problem is even more critical because any viable vision of the future of health care delivery requires a sound vision of the future of EHR. But before we can formulate that vision, it is important not just to agree about the nature of the problem, but also about its main symptoms and major causes.
The list of symptoms is long. Much of the time that should be dedicated to clinician– patient interaction is in reality committed to clinician–computer interaction, mostly for data entry. Even worse, clinicians are not able to find data when they need it in a reasonable amount of time, if at all! Tests are repeated unnecessarily because clinicians are unaware of the existence of their results or just unable to find them. Patients often do not have access to their own records.
Yet the most debilitating symptom of all is the lack of interoperability across solutions from different vendors. As a result, both cost-effectiveness and quality of care are fatally wounded and continue to bleed. More tragically, continuity of care is compromised to the extent that patient data cannot follow patients’ geographic movement to other providers or health care facilities. Lack of interoperability is, therefore, the main roadblock between us and a brighter EHR future.
But why has interoperability proven so elusive? The answer to this question must include a compounded list of causes, topped by the lack of any encompassing standard governing the design of EHR systems (what to capture, how to capture it, and how it should be stored and shared). This lack of an encompassing standard has been coupled with (or rather led to) an evolution of disparate EHR solutions over several decades resulting in what is called the “baggage” challenge: hundreds of noninteroperable vendors’ solutions and thousands of already existing implementations.
Interventional efforts by government agencies and the health IT community to fix the EHR problem in light of this challenge have proven inadequate, if not totally misguided. The fundamental question now is not how to solve the EHR interoperability problem. Rather, it is how to solve the EHR interoperability problem despite the baggage challenge.
These are two fundamentally different questions that require profoundly different answers. Having accumulated and invested in all this baggage means a greater resistance to change backed by entrenched interests. It also means dismantling the old (at least partly) before building the new. Moreover, it means that it is absolutely necessary to agree on a unifying vision of how we want future generations to have a different, and certainly better, national-level EHR system. In my view, such a unifying vision can be founded on the concept of vendor-neutral EHRs.
Having vendor-neutral EHRs would not be a real novelty. It comes straight out of the radiology experience, which is the area having the most successful interoperability story in the history of health IT. At the center of this success story is the Digital Imaging and Communications in Medicine (DICOM) standard, one of the most forward-looking standards in health IT. The success of interoperability, as enabled by DICOM, has had a substantial and positive impact on the modern practice of medicine.
It allows, for example, an X-ray taken in Dallas, Texas, in the middle of the night to be read by a physician in Sydney, Australia—and a diagnosis to be provided quickly and accurately. Despite this success, the radiology community realized that better interoperability can still be achieved if the community embraces the concept of vendor-neutral archives (VNA). Not only does VNA provide larger scale and more flexible interoperability, it also protects health care facilities (and, subsequently, patients and health care consumers) against dependence on a single vendor’s proprietary solution (or, “vendor lockup”).
As a result, VNA has become the main part of the future vision for radiology. It may make sense, then, for the general EHR community to extract this learned lesson and adopt the moral of this success story.
So, the ingredients of the future national EHR system may not be difficult to envision: vendor-neutral EHRs enabled by an encompassing EHR standard and enforced by proper regulations. Before we can assess the feasibility of this proposition, more information is needed, particularly regarding the nature of the critically needed EHR standard. Doing this will necessitate deconstructing the main steps required to obtain any EHR data: capture (what and how), storage (how and where), processing (what, how, and who), retrieval (how and who), and visualization (how and who).
In the status quo, the vendor (or the system builder) has most of the control over how these steps are performed. There have been minimal standardization efforts that govern the various steps and the different questions (between brackets). The solution to the problem might begin by having the health IT community standardize—and the government regulate—what needs to be captured in various contexts, how exactly to store it (data format), and where to store it.
Assuring the standardization of these parts is presumed to be the fundamental prerequisite to achieve the vendor-neutral EHR vision. Although this prerequisite actually necessitates the painful dismantling of much of the status quo, it would still be necessary to assume that each vendor will maintain the ability to retrieve, process, and present the data in a way that can show the competitive and differentiating merits of that particular vendor. It is also necessary to emphasize that data retrieval must be heavily regulated to ensure patients’ and community rights of privacy and security. Finally, vendor neutrality is as important to personal health records as it is to EHR.
More details may be needed for this proposed vision to make more sense and to reveal its potential cost. Nevertheless, given a viable vision to solve the problem, filling in the details and orchestrating the different roles will require the coming together of all of the stakeholders. These stakeholders understand very well how we got to this point and can understand why getting out of it is almost guaranteed to be difficult and even painful. However, rallying their efforts to move things forward, it is important for them to see that the proposed unifying vision is, indeed, feasible.
For more information about the IEEE Pulse On Stage event, visit onstage.embs.org.