This wiki has undergone a migration to Confluence found Here
<meta name="googlebot" content="noindex">

Consolidated CDA July 2012 Errata

From HL7Wiki
Jump to navigation Jump to search

Return to SDWG page. (Items on this list must first have been approved for inclusion through a SDWG vote).


Approved Errata

1. Published IG states "HL7 Data Types Release 1 requires the codeSystem attribute unless the underlying data type is Coded Simple or CS, in which case it is prohibited. The displayName and the codeSystemName are optional, but recommended, in all cases".

  • Because displayName and codeSystemName are prohibited in CS data type, the recommendation only applies where the data type is NOT CS (e.g. CE or CD).


2. Section 1.8.1, Table "Constraints format example", shows that template constraints are presented in various formats in order to meet different user needs. We have found a handful of discrepancies between the constraints enumerated in the tables and the corresponding constraints enumerated in the list of constraints following each table. We have identified the source of discrepancies, which will resolve this issue in future publications. For now, we declare that the constraints enumerated in the list are the source of truth, and take precedence over the constraints enumerated in the table. We will include this declaration in future implementation guides. Specific discrepancies identified include:

  • Table "Comment Activity Constraints Overview" incorrectly lists and links to CONF 9430 and 9431, which do not exist.
  • Table "Family History Organizer Constraints Overview" incorrectly lists and links to CONF 8610, 8611, 8613, 8614, and 8615, which do not exist.
  • Table "Referenced Frames Observation Constraints Overview" incorrectly lists and links to CONF 15923, which does not exist.


3. Table "Template Ids Alphabetically by Template Type"

  • Lists an Implants Section template, which does not exist (see Procedure Implants Section instead).
  • Lists a Surgery Description Section template, which does not exist (see Procedure Description Section instead).


4.The implementation is full of constraints like the following: (reviewed and approved on the Sept 20, 2012 SDWG call)

  • SHALL contain exactly one [1..1] statusCode="completed" Completed (CodeSystem: ActStatus 2.16.840.1.113883.5.14 STATIC) (CONF:7321).

Which imply that statusCode should directly contain the string "completed" (i.e. <statusCode>completed</statusCode>) when what is wanted is statusCode/@code="completed" (i.e. <statusCode code="completed"/>) )


5.The medications section contains the following constraint: (reviewed and approved on the Sept 20, 2012 SDWG call)

  • SHALL contain exactly one [1..1] title="Medications" (CONF:7793).

other sections do not constrain their titles to a fixed string. This is an error and the correct conformance statement should be this:

  • SHALL contain exactly one [1..1] title (CONF:7793).


6.There are many constraints on code elements (observation/code, etc.) requiring @xsi:type to be present when that was not the intent. For example: (reviewed and approved on the Sept 20, 2012 SDWG call)

  • SHALL contain exactly one [1..1] code="11323-3" Health status with @xsi:type="CE" (CodeSystem: LOINC 2.16.840.1.113883.6.1 STATIC) (CONF:9073).

These were data entry errors. There was no intent to require @xsi:type to be present and constrain it to anything other than the default datatype for applicable code element.

7.Inconsistent use of serviceEvent/performer/@typeCode. There are several single value bindings for typeCode throughout consolidation. The general header constraints require "PRF" (Participation physical performer, see conf 14840), but other document type specific header constraints require other codes like "PPRF" (Primary Performer, see conf 8495). SDWG agreed to remove the single value binding from CONF 14840 in the general header constraints, allowing all document types to use whatever codes are appropriate. (Reviewed and approved on the Nov 1, 2012 SDWG call).

8. The H&P constains the following constraint: SHALL contain exactly one [1..1] Reason for Visit Section (templateId:2.16.840.1.113883.10.20.22.2.12) (CONF:9627). However, this conflicts with CONF 10057 if a "Chief Complaint and Reason for Visit" section is present, which states: SHALL NOT include a Chief Complaint and Reason for Visit Section with either a Chief Complaint Section or a Reason for Visit Section. (CONF:10057) . SDWG agreed that CONF 9627 will be removed, and a new constraint will be added that states SHALL either have a "Chief Complaint and Reason for Visit" or separate "Chief Complaint" and "Reason for Visit" sections. (Reviewed and approved on the Nov 1, 2012 SDWG call).

9. Update the heading in the Consultation Note LOINC Document Codes from "Root Level Document Type Code" to "Preferred Code" to align with the other tables. (Reviewed and approved on the Nov 1, 2012 SDWG call).

10. Conf 8903 makes act/code optional, which violates the base CDA spec, which requires 1..1 act/code. "SHOULD contain zero or one [0..1] code with @xsi:type="CE" (CONF:8903)". should be changed to "SHALL contain exactly [1..1] code (CONF:8903)". (Reviewed and approved on the Nov 1, 2012 SDWG call).

11. There are several places in the IG where the "sdwg" namespace prefix is referenced (Conf 2763 and conf 8933). These are typos, and the correct namespace prefix is "sdtc". (Reviewed and approved on the Nov 1, 2012 SDWG call).

Errata to be reviewed by SDWG

1. Smoking Status Observation templateId and value set OIDs are incorrect.

  • Update template from 2.16.840.1.113883.10.22.4.78 to 2.16.840.1.113883.10.20.22.4.78
  • Update value set reference in CONF:14817 to Smoking Status 2.16.840.1.113883.11.20.9.38 STATIC 2012-07-01