Can EHRs achieve the interconnectedness we need?
Robert Rowley, MD
Twitter: @RRowleyMD
Electronic Health Records (EHRs) have come a long way in moving U.S. healthcare onto an electronic platform in the past few years. As of 2013, the majority of practicing community physicians, as well as most hospitals, have adopted an EHR system in their practices.
Like was the case with paper beforehand, EHRs have maintained the siloed nature of health care data, keeping records within the institutions (hospitals, medical practices) that created them. This is no real surprise – EHRs, after all, are provider-focused tools. More than being physician tools, they are really better thought of as practice tools. Many of the workflows addressed by EHRs, both in hospitals and in medical practices, involve the whole in-house care team, and only a subset of an EHR’s capabilities are about physician documentation of encounters.
What is also clear, though, is that we need to move beyond the fragmentation, and connect a patient’s health data across all the places where care is sought. We need to build what has sometimes been referred to as a “community chart” – one that exists centered around the patient, not the provider, and draws seamlessly from all the provider’s EHR records wherever they exist.
Can EHRs achieve the interconnectedness we need? Or do we need to see the rise of new, EHR-independent technologies in order to reach this needed next stage?
EHRs and Enterprise Charts
One way that interconnected health data has evolved is by creating institution-wide Enterprise Charts – a single chart, accessible by everyone within a delivery system. This has become more widespread as medicine changes away from small, independent practices, and toward larger groups, both physician-owned and hospital-owned. In such larger settings, a single large enterprise EHR is common (e.g. Epic), and is deployed across a hospital and all the affiliated practices. Within the walls of such an integrated delivery network (e.g. Kaiser), a single chart available everywhere is the realistic goal of their EHR.
Whether in tightly coordinated networks, or in looser arrangements, achievement of an Enterprise Chart is no trivial matter. It involves uniquely identifying patients and merging duplicate records. This is generally done using EMPI systems (Enterprise Master Patient Index) – a whole area of software that needs to be leveraged (often at considerable cost) in order to achieve a workable Enterprise Chart.
However, even if done well, and even if there is an actual or virtual Enterprise Chart in a healthcare delivery network, this still falls short of the goal. A patient-centered, portable record that accompanies an individual through one’s life journey into and out of these integrated delivery system simply cannot be achieved within the domain of a given EHR system, regardless of how well integrated it is within a network.
EHRs as patient data repositories
Much has been said about the value of EHRs in collecting patient data, and using that data for various purposes. And, to date, EHR installations have done that – accumulated data within the confines of their installation. Large installations with Enterprise Charts that cover millions of patients (e.g. Kaiser) have used this data to gain insights into disease trends, clinical quality and best practices.
Some companies, in fact, base much of their business value on the unique nature and scale of their internal patient data (e.g. Practice Fusion). Again, to date this has been true.
However, the need for interoperability is critical. Patient data needs to be shared across enterprises. Meaningful Use Stage 2 sets some requirements for vendors to be able to export and import summary patient data (via creation / consumption / transport of C-CDA standard documents).
At the same time, we are seeing the emergence of the next level of aggregation of patient data. New “next generation” platforms, though still quite nascent, are being built and deployed that allow patients to request data from all their providers, regardless of what EHR they use, and collect them in a central place that is outside the data silos of any installation or vendor. It is these new, global patient data repositories that will, over time, become the aggregated center of patient data. EHR vendors and installations will see their data leach out, as patients request copies of their records from their doctors, and put them into new, aggregated places outside the EHR.
EHR’s role in this next era
So if not the place for global patient data, what, then, is the role that EHRs will play down the road? The data displayed by an EHR may come from local or external sources. What will EHRs be used for, then?
EHRs are practice tools, and need to enable efficient workflows in a clinical setting – a doctor’s office or a hospital. Many of the current complaints about EHRs are about how clunky they are and how they slow down workflows, reducing efficiency, and turning physicians into educated data-entry staff. These frustrations have led to much of the discussion about EHR usability, and the need to make these tools more modern and efficient.
Here are some elements that modern EHRs should achieve. I would suggest these be considered overarching goals – if a vendor can accomplish this, they will be successful. If not, then a rip-and-replace decision will need to be addressed.
- Physician documentation of a patient encounter must be very quick. Much of the volume of documentation currently done (and EHRs are good at turning small notes into big ones) is done for billing support (a by-product of a fee-for-service system) and forensics. Natural Language Processing is a technology that is emerging, and may play an important role in helping a physician document the encounter. The goal of “two clicks and you’re done” should be targeted. Swift documentation capability, customizable macros, voice or click creation of notes – these are what we need to achieve. An EHR should not make it necessary for physicians to hire scribes to follow them around and do data entry.
- Streamlined workflows need to be handled by the EHR. In ambulatory settings, this means good handling of external documents, messages, and communication with patients. Some of the new aggregated technologies may have unified patient portals, so that a patient only needs to log in to one place to access all one’s physicians and communicate with them – a modern EHR should be able to facilitate this.
- Open access to internal data is another thing EHRs will need to do. There is no way that a vendor will be able to anticipate all the reporting that a given practice will want to do – clinical quality measures, disease trends, demand that will affect staffing levels are all examples of reports or dashboards that a practice may want to achieve. Rather than expecting a vendor to do this themselves (it will take a long time, and will likely be expensive), an open API access to one’s data can allow a whole ecosystem of “satellite” companies that can create the reporting, analysis and visualizations that a practice (or hospital) might need.
Conclusions
The role of EHRs will change. Now that we are past the “implementation phase” of EHR use, and are moving more to the “optimization phase”, it is clear what we need our tools to be. EHRs need to facilitate practice (or hospital) workflows. They need to move away from documentation time-sinks and embrace technologies that will allow documentation to be finished by the time the clinician leaves the exam room.
In the era of connecting patient’s data across different care settings, the role of EHRs as primary data repositories will diminish. EHRs will collect local data, but will need to share (in a two-way fashion) internal data with external sources. Good medical practice will be the result of this.
The next level of health IT can be seen from here. My entrepreneurial side says, “now, let’s build it!”
This article was original published on Dr. Rowley’s blog. It is republished here with permission.