CDA in the Wild: Conclusion (Installment #8)
Apologies—it has been almost a year since my last entry. For a long time, I was so traumatized that I thought I would never write about Wild CDAs again, but this last story must be told. You must be prepared. You must protect yourself.
Shortly after my last report, I found myself short on funds and long on spare time. I reluctantly accepted a job from a wealthy but eccentric benefactor. The task: research what I considered to be nothing more than a conspiracy theory, the Wild CDA equivalent of Big Foot or the Loch Ness Monster. I was to prove the existence of the mythical MeaninglessUseless Rex, the titan of the CDA genus, for which only the most fragmentary fossil evidence was ever found.
Easy paycheck then back to real work—or so I thought.
The beast lives!
I’m at a loss to describe the encounter. We all thought this genus of Wild CDA to be extinct, if it ever truly existed in the first place. After all, it served no useful purpose. But MeaninglessUseless Rex, has survived, unlike my research team.
We began our search in the East Haven Refuge (EHR), where the most convincing fossils of the beast were found. Imagine my surprise when we found not bones, but footprints so large that an intern could curl in the fetal position and fit inside them with room to spare.
Surely this must be a hoax! We followed the tracks, expecting to find a prank among the grad students on my team. We found horror instead.
Patient record after patient record messily devoured and excreted in a terribly summarized form. Practitioner directories ripped to shreds with only their most basic contact information left to identify them. Steaming piles of coded allergies, problems, and medications fermenting on the putrefied landscape.
Then the beast itself emerged from the EHR thicket.
My team did not survive the encounter. A mangled patient record is all that remains of them. But I did not consent to such a demise. I stood tall and roared back at the beast, hoping that a display of aggressive narrative that refused to be codified would be enough to allow my escape; but that proved to be wishful thinking, for MeaninglessUseless Rex devours all, leaving only an indigestible common clinical data set. I am now half the patient I used to be, reduced to codes and null flavors, everything else ruthlessly amputated.
My story ends here – your story can begin. Band together. Save the medical record from regulatorily-correct mish-mash as you cherish traditional values in a narrative-safe electronic record!
Obesity is a huge problem in the US; and it applies to CDA documents as well. You can think of CDA obesity as the kitchen sink problem: how do I fit everything I know about a patient into a single document? Meaningful Use decided for us that Allergies, Problems, Meds, Results were the key elements to be exchanged using the document paradigm. Key elements, certainly, but sufficient? Not necessarily and no one knows where to stop. “Valid” CDA documents became 60+ page “summaries”.
Some examples of CDA kitchen sink issues:
- Dump everything we know about a patient into one Continuity of Care Document (CCD): 10,000 lab tests, all medications administered ever, etc.
- Cram dozens, or even hundreds, of discrete documents into one CCD
- Enter one question/answer pair per section (i.e., Q is the section title, A is section text)
And this is only a smattering of the abuses I have seen.
Developers, clinicians, I implore you. When creating such a document, please bring it up in a web browser using a CDA rendering stylesheet. Ask yourself what your recipients will say when they try to read it. The answer, per the results of the Relevant and Pertinent project, is that most summaries are not clinically relevant. In my opinion, they are little more than EHR data dumps. Their only excuse for living is the absence of an API for querying EHRs for the data you are looking for (i.e., since I can’t query the EHR today for what I want, send me everything in a big blob).
We can do better.
Today, we have FHIR. I’m not going to describe FHIR—if you are curious a google search will suffice. I have a strong opinion about what FHIR means for these bloated 60+ page “summary” documents we see exchanged under Meaningful Use.
Clinical Documents were never meant to hold 10 pounds of coded data. CDA is a 5-pound bag, containing the actual words, either written directly by a clinician or spoken and transcribed, with the minimal coding necessary for the use case. CDA could have been a way to exchange simple, narrative notes with key data as coded elements. Instead, they have been a stand-in for an API, loosening a deluge of coded data from system to system.
I hope the rollout of FHIR spells the death of CCDs. And from the ashes of that death, I hope for the following:
- A reduction in the number of summary documents exchanged over the wire replaced by queries for the needed data elements when a selected set of data elements will suffice.
- The documents exchanged should be clinically relevant, and largely narrative with only those codes that add value.
- A shift from HL7 V3 syntax (CDA) to the simple, easily implementable syntax of FHIR documents (e.g., C-CDA on FHIR).
CCDs are not going away today. Implementers must comply with existing regulations; however, one path to the future allows compliance with existing regulations…FHIR to CDA conversion. I will explore this option in my next blog series.
In related news, Politico ventured to “the wilds of Virginia” [sic], and reports ONC head Don Rucker repeated, “the administration’s now-familiar promise to focus on interoperability and usability” at the Capital Health Tech Summit. We’ll see where that goes…
So, farewell CDA in the Wild, it’s been fun—but the future is on FHIR. Let’s take the capital “M” out of Meaningful and get back to plain meaning. That’s where my next blog series will go. Look for more here on the value of narrative in an electronic record. Can we say narrative is so hot today, it is on FHIR?
Click here to read previous installments in the CDA in the Wild series.