Interoperability is incomplete without context and delivery – The fix and the future
As an information system, the EHR has a longer history supporting administrative tasks than supporting clinical care. Seamlessly delivering information into the proper clinical context during the work of a healthcare provider is important for quality, safety and efficiency. Interoperability in healthcare should include the concept of how information is delivered to clinical providers with the right clinical context. CORES is an example of this kind of interoperability and demonstrates the potential of the EHR as a platform upon which innovation can drive improvements in care delivery.
The disconnect between workflows and EHR design
EHR systems were not originally engineered to be patient care systems. Instead, they grew organically out of computerization of administrative systems such as accounting software, laboratory reporting systems, and hospital hotelkeeping databases. Consequently, many EHR systems are primarily designed to accomplish administrative tasks, rather than focus on supporting clinical staff and delivering patient care.
The impact of unsupported workflows
Without adequate delivery of information and its proper clinical context during patient care, quality, safety, efficiency, and value suffer. The promise of the EHR to improve health care delivery, or to maintain care quality while reducing costs cannot be met when information systems add complexity and fragment information delivery. When compared to interfaces designed for specific clinical workflow, standard EHR interface designs are rated worse for patient care by their users. Clinical workflows represent the actual delivery of care. Decisions about diagnoses, which directly affect care quality targets, and about treatments, which directly affect cost, are made during clinical work. Unless the right information is provided to the right provider at the right time in the workflow, both suffer.
Interoperability really means information that supports workflows
Traditionally, interoperability means that systems and devices can communicate, exchange information, and use the exchanged information. This definition encourages EHR systems to conform to standards for information interchange. What this definition lacks is the concept of contextual presentation of the information. To permit timely delivery of EHR information into the user’s workflow, we propose true interoperability include not only information interchange with retention of semantic meaning, but also the clinical or research context of that information. The unpredictable ways in which patients can experience changes in health status do not allow for a framework where information is simply available across different systems. For true interoperability to exist, information must present itself in context at the right time, to the right provider, and be actionable.
For inpatient clinical care, today’s systems primarily rely on the interaction design of desktop computers, which use a video screen, keyboard, and mouse. Some EHR vendors have created smartphone apps that allow limited interaction with EHR data. However, most EHR vendors largely disregard other interaction modalities, such as printed reports, pagers, and electronic whiteboards. Because care delivery happens with patients, largely away from desktop computers, interoperability should not be limited to a session at a desktop computer. Healthcare interoperability requires information to move seamlessly among the EHR system, including other interaction modalities that are integral to inpatient clinical care workflows.
Examples of our interoperable apps, with impact
CORES (TransformativeMed Inc. Seattle, WA) is one example of a truly interoperable application by the definition above. It can obtain clinical data directly from the EHR, regardless of the source that delivered it to the EHR. It is embedded in the standard desktop EHR interaction modality, yet also delivers contextualized information to pagers, printed paper reports, and bidirectionally to smartphones. CORES can deliver bundled information in the form of widgets to external applications, such as Picture Archiving and Communication Systems.
CORES demonstrates the cyclical information link among inpatient providers that has been poorly understood and unsupported by EHR vendors: the compilation of a list of inpatients to be seen on rounds, including those that were new from pending ED admissions and were not properly assigned yet by the EHR system; the review of overnight patient events and new objective data, including summarization of those data onto a portable format to carry on rounds; the formation on rounds of a shared understanding of the patient’s current state, problem list, likely trajectory, and the creation of a list of tasks that are needed; the conversion of that shared understanding and task list into a medical record document; and the recompilation of the list of inpatients with a summary of tasks and contingency instructions to guide overnight providers that will cover these patients but who don’t know them well. During this entire cycle, new information arrives without regard to whether the intended recipient is in front of a desktop computer, walking between patients, or in the middle of a procedure. In addition, specific questions arise that require interaction with other providers or roles in a system where those others are not readily available.
CORES permits users to reorganize EHR information to mirror their clinical needs and how clinicians work. For example, the CORES smartphone app can convert a laboratory value alert into a task with a follow-up reminder; use a desktop notification about patient census imbalance to trigger a request for patient reassignment for census load-balancing; forward a secure text-message conversation into the body of a clinical note; reorganize a subset of structured problems and annotate them with contingency plans for cross-covering providers.
Future needs and our direction
CORES re-imagines today’s vendor-managed EHR system as a platform upon which innovation can create creative and intuitive features that meet caregivers needs as they work. For example, specific clinical use-cases can act as templates to create interoperable, workflow-optimized, clinical care tools that interconnect the EHR, mobile devices, and API-driven embedded apps. Using a secure context, full read and write privileges, and clinical input into design, these tools can decrease cognitive overhead, deliver contextualized information at the right time to the right provider, and have the potential to bring the full quality and efficiency promise of computerized health care to fruition.
 Eason K, Waterson P. Fitness for purpose when there are many different purposes: Who are electronic patient records for? Health Informatics J. 2014 Sep;20(3):189-98. doi: 10.1177/1460458213501096.
 Ahmed A, Chandra S, Herasevich V, Gajic O, Pickering BW. The effect of two different electronic health record user interfaces on intensive care provider task load, errors of cognition, and performance. Crit Care Med. 2011 Jul;39(7):1626-34. doi: 10.1097/CCM.0b013e31821858a0.
 HIMSS Dictionary of Healthcare Information Technology Terms, Acronyms and Organizations, 2nd Edition, 2010, Appendix B, p190, original source: Wikipedia.
 Van Eaton EG, Lober WB, Pellegrini CA, Horvath KD. User-driven design of a computerized rounding and sign-out application. AMIA Annu Symp Proc. 2005:1145. PMID: 16779431.