Relationship between a section and its entries
This page identifies some open issues related to the relationship between an entry and its narrative text. (this is not about relationships between entries, nor between entries and "objects external to the document"). See 188.8.131.52 and 184.108.40.206 in the CDA R2 specification for details. The resolution of these issues should be part of CDA R3.
The fact that specific pieces of narrative text can be tagged/highlighted and linked to entries suggests a possibility to render the narrative text ("Asthma") in some special highlighted manner, where e.g. the code (e.g. ICD-10) from the related entry is shown when the mouse hovers over the text, or where the details entry will be displayed if the tagget snippet of original text is selected. The use of <content> tags is entirely optional however - which raises the question when and how it should be used.
Narrative to entry
- content: The CDA <content> element is used to wrap a string of text so that it can be explicitly referenced, or so that it can suggest rendering characteristics. The <content> element can nest recursively, which enables wrapping a string of plain text down to as small a chunk as desired.
- renderMultiMedia: What, if anything, do the renderMultimedia XML tags in a narrative block encapsulate?
The renderMultiMedia tag in a narrative block encapsulates the caption of the multimedia content. The caption is expected to be text that may also have super and subscripts, footnotes and footnote references, and linkHTML elements. Kboone 01:09, 10 January 2007 (CST)
- linkHTML: What, if anything, do the linkHTML XML tags in a narrative block encapuslate?
The linkHTML tag encapsulates the hyperlinked text. This text may contain footnotes and footnote references. In an HTML rendering, the contents of the linkHTML tag might be rendered between <a></a> tags. Kboone 01:09, 10 January 2007 (CST)
Entry to narrative
- The originalText component of a RIM attribute present in any CDA entry can make explicit reference to the identifier (contained in a content element in the narrative block), thereby indicating the original text associated with the attribute in the CDA entry.
- In the case of DRIV, is there an expectation (should it be required) that content tags be used in the generated narrative at the same level of granularity as the data contained in the Entries?
In the case of DRIV, my personal preference is that the generated narrative be richly linked to the data contained in the entries using references to ID attributes in the narrative tags (e.g, <content></content>). Kboone 01:25, 10 January 2007 (CST)
- coded_datatype.orignalText is "The text or phrase used as the basis for the coding", "The original text exists in a scenario where an originator of the information does not assign a code, but where the code is assigned later by a coder (post-coding.) NOTE: The original text property is not meant to be a link into the entire source document. The link between different artifacts of medical information (e.g., document and coded result) is outside the scope of this specification and is maintained elsewhere in the HL7 standards. The original text is an excerpt of the relevant information in the original sources, rather than a pointer or exact reproduction. Thus the original text is to be represented in plain text form"
- which suggests a partcular relationship between the text encapsulated between content tags,
- in the case of DRIV, originalText doesn't seem applicable. If we want to link from en entry to the fragment of text that was generated based on that entry, we'd either need a new mechanism or the definition of originalText will need to be changed.
I personally do not see a need to change the definition. The original text can appear in the narrative with links using the <reference> tag in the entry, even when the narrative is derived from the entry. Kboone 01:25, 10 January 2007 (CST)
- If one switches language or locale, in a document with DRIV text sections, and <content> links between the entry and the text, is it OK to replace the text with a translated/transformed version? (example: ...<content>200 [miles/gallon]</content>.. with an entry Act which has a value of "200 miles/gallon": if the physician has sleceted a metric locale, can the TEXT be changed to "90 km/l"?)
I think the answer is no in this case, since the actually document content is changed, and so you would have a new document. Kboone 01:25, 10 January 2007 (CST)
- It should have stated "displayed as 90 km/l". The document itself isn't changed (nor a new one created), the exact same semantic concept (and we know it's the same, because it's derived from a coded entry) is merely displayed using different units. Rene spronk 06:10, 10 January 2007 (CST)
- How does one enforce or express any relationship requirements one may have in a template or in a constraint language? (example: in medication sections, where text has been DRIV from an entry, each and every entry SHALL have a corresponding <content> enclosed part in the narrative)
- (related, though different) How does one enforce or express requirements related to the structure of narrative text? (e.g. Lab results SHALL be sent in the form of a table with 5 columns)
There are numerous mechanism to validate content. One very easy one to implement uses Schematron (now an ISO standard) to enforce these constraints. I know of at least 7 implementation guides on CDA that use Schematron to validate implementation constraints, many from IHE, the HL7 Care Record Summary, the British Columbia ebMS project, and a few others. Kboone 01:30, 10 January 2007 (CST)