Data Presentation for Effective Care Transitions
President and Chief Strategy Officer
I have the honor of working with dozens of health information exchanges (HIE) nationally to improve interoperability. During the past decade, a lot of that time has focused on technical aspects of interoperability. How do we connect to the EHR? How do we convert data to an interoperable format? How do we coordinate a set of API requests and responses to move information to a desired endpoint?
Slowly but surely, every HIE breaks down these barriers to technical interoperability. Then, they are confronted with a stark reality. How SHOULD they display data to support higher quality care and better care transitions? How do they share data in a well-organized and consistent format (i.e. semantic interoperability)? These are particularly important considerations when sharing care summaries, like Consolidated Clinical Document Architecture (C-CDA) documents and Continuity of Care Documents (CCD).
Standards bodies have historically “punted” on this question, saying that data presentation should be use case dependent. That’s fair, but a vacuum of guidance creates tremendous variability. This vacuum existed for C-CDA and persists into Fast Health Interoperability Resources (FHIR). Specifically, there are only a few narrative (human-display) rules in FHIR:
- “The narrative SHALL have some non-whitespace content”
- “The narrative SHALL contain only the basic html formatting elements and attributes”
- “The narrative SHALL reflect all content needed for a human to understand the essential clinical and business information. It SHALL be safe to render only the narrative of the resource without displaying any of the resource’s discrete/encoded information”
While the above cements the importance of human readable content in interoperable standards, it grants tremendous discretion to data senders. What is essential to understanding clinical and business information? Should medications be presented in a list, table or paragraph format? Are medication names and dates enough, or do we need codes, routes and strength information?
Valuable perspective on this can be gleaned by what EHRs are already doing. In 2018, I participated with the Veterans Health Administration on a research study examining many different certified EHR technologies. One thing we did was to look at how the same three medications as shown from 38 different technologies were displayed. Here, we found some commonalities:
- 38 of 38 technologies showed the start date of the medication
- 37 of 38 technologies presented each medication as a row in a table
- 35 of 38 technologies had a dedicated column for the medication name
But there were also significant differences:
- 12 of 38 technologies showed the code of the medication (i.e. using RxNorm) in the human readable portion
- Only 7 technologies used TALLman lettering, a medication safety practice, on an eligible medication (e.g. cefTRIAXone); 4 technologies displayed all medication names as uppercase (e.g. CEFTRIAXONE) while others used lower or mixed casing
- Only 6 medications showed a generic name (e.g. acetaminophen) in addition to the ordered brand name (e.g. Tylenol)
- Only 2 of 38 technologies showed a strength field (e.g. 500mg) separate from dose-form (e.g. Tylenol Extra Strength 500 mg Oral Tablet)
The result was massive variation. Here are three examples, first a simpler presentation:
Another representing a typical midpoint of data complexity:
And finally, a more complex set of fields:
The variation shown above is stark, as even dates are shown in three very different formats. As we progress toward standards that facilitate rapid data sharing, like C-CDA and FHIR, it’s time to examine which human readable displays support effective care transitions. While Diameter Health strongly supports configurable displays for different use cases, there are opportunities to reduce the noise and variation through default behavior. Answering “how SHOULD display data support higher quality care and better care transitions?” is a question worth answering. Both clinicians and patients will thank us for getting this right.
What do you think? Use the contact us form on this page to keep the conversation going.