[Category: Section Narrative Text]The article
Section narrative text, introduces the "section narrative text" that is part of every
CDA document
section, and overviews its syntax.
As noted in that article, everything that is clinically relevant and contained in the structured clinical statements of a section must also be captured in the narrative text of that section.
Intended for Human ReadersSection narrative text is intended for human readers. It is the part of the CDA document that is typically formatted by the
XML stylesheet to display the CDA document in a web browser for a human reader.
Thus, CDA elements that are intended for "machine readers" (software programs) of a CDA document do not need to be included in the section narrative text. For example, many
identifiers contained in CDA
XML elements such as
id,
typeId,
templateId,
setId, etc. are intended to connect entities in the CDA document to system databases and objects - in which case they are not useful to human readers.
Effective for Human ReadersKeeping in mind that the target audience of the section narrative text are human readers provides a lot of guidance as to what should or should not be included, and how it should be formatted.
For example, it's often necessary to use quite a bit of XML syntax in a CDA document to convey information contained in a short human-readable statement. There's no need to present redundant information for a human reader (just the opposite) so a formal/rote "translation" of every XML element/attribute is not effective.
Challenging to Provide a Detailed "Cookbook"In an article on his
Health Intersections website titled
CDA: What information from the Entries has to go in the narrative?, Grahame Grieve walks through many of the core structures in the CDA structured body (
clinical act statements,
participation elements,
act relationship elements, etc.) and provides specific guidance for some of the XML elements and attributes and how they should appear (or not) in the section narrative text. Like most things on Grahame's website, it's a helpful and recommended article.
In many cases, however, what Grahame is able to provide is qualified by "usually", "depending on...", and similar - and all of it is qualified as being "suggestions". It's not possible to provide formal automated translation rules that can fully account for "what matters clinically to a human reader".
For implementers, as Grahame notes in his article, the key takeaway is:
"The decision has to be made by some clinically aware person who knows what the data means – don’t let programmers do it (and I see a lot of documents where the programmers have done it, and what they did doesn’t make sense to a clinician)."Meeting the ObjectiveThe underlying principle that motivates the existence of "section narrative text" is that CDA documents are intended for both human and machine readers. It's part of the CDA design assumptions that not everything communicated to the human reader will always be fully captured for machine processing. But in the other direction, it's a requirement that a human reader of a CDA document know everything that a that a machine interpretation of a CDA "knows".
Although the structured body of a CDA document can be presented to a human reader by the systems that interpret and processes it - the "human readable" portion of a CDA is intended to be the section narrative text (and some formatting of the
CDA Header) - not an automated formatting of the structured clinical act statements. Thus, it's required that the section narrative text for each section be an
effective and complete human-readable version of what is communicated in the structured clinical act statements.
Keeping this higher-level objective in mind should help guide what should and shouldn't be included in the section narrative text.
Other CDA PRO Know Articles Referenced In This Article
