A CDA document shall have at least one author. Authors could be either human (ClinicalDocument/author/assignedAuthor/assignedPerson) either devices (ClinicalDocument/author/assignedAuthor/assignedAuthoringDevice).

For definition “The author element represents the creator of the clinical document. If the role of the actor is the entry of information from his or her own knowledge or application of skills, that actor is the author. If one actor provides information to another actor who filters, reasons, or algorithmically creates new information, then that second actor is also an author, having created information from his or her own knowledge or skills.” [From Implementation Guide for CDA Release 2: Imaging Integration – UV Realm, March 2009].

According to this definition, not any device that generates the electronic document has to be considered as an author:

• a spider collecting and filtering information from different repositories, according to defined rules and policies, for the scope of creating a Patient Summary is definitely a document author (and maybe the only one for this document);
• an application that transforms a prescription record into a epSOS eP CDA may not be an author instead;
• The NCP that modifies the concepts conveyed, should appear as one of the authors.
• The author of the parent document is often also an author of the document generated through a transformation process (in that case also the author time should be the same).

Further to this, is therefore possible to determine the Nature of Patient Summary as follows [Excluding the NCPs as authors]:

• if there is a person author only, there the Patient Summary is the result of a practitioner clinical act;
• if there is a device author only, the summary was automatically generated according to well defined rules defined by the responsible organization;
• if there are both a person and a device as authors, the summary was created via a mixed approach.

The CDA provides a mechanism to better specify who authored what within the document, allowing the specification of authorship at the whole document level, at the section level and finally at the entry level. In any case is not required to repeat this information for each level, taking advantage of the context conduction propriety. Infact “context that is specified on an outer tag holds true for all nested tags, unless overridden on a nested tag. Context specified on a tag within the CDA body always overrides context propagated from an outer tag. For instance, the specification of authorship at a document section level overrides all authorship propagated from outer tags.” (HL7 CDA R2 Standard).

/ClinicalDocument/author/assignedAuthor
/ClinicalDocument/documentationOf/serviceEvent/performer (this applies to the prescriber and the dispenser).

The template ID referenced here refers to HP information in the /ClinicalDocument/documentationOf/serviceEvent/performer structure. In this document, the same requriements apply to the person author of the document (if there is one), and to the prescrober and dispenser (see body).

When there is no HP but we have the method of assembly of data by a device, such as the “spider” method, we have the following expression; ClinicalDocument/author/assignedAuthor/assignedAuthoringDevice.

When the data is collected from different sources & pre-existing documents that are part of a bigger system. In that case the organization responsible of that collection “signed” the PS as responsible.

ClinicalDocument/author/assignedAuthor/representedOrganization