should this simplification be allowed?

Implementers (archive)
1

Hi,

I'm testing my implementation about the openEHR specification, with
some archetypes examples, and I'm finding things like this:

CLUSTER[at0018] occurrences matches {0 ..*} matches { name matches {
    CODED_TEXT matches {
          code matches {[ac0002]} }
}

I'd like to follow the specification as much as possible, and I think
that the correct way of expressing it, should have been:

CLUSTER[at0018] occurrences matches {0 ..*} matches { name matches {
    DV_CODED_TEXT matches {
          defining_code matches {
                               DV_CODE_PHRASE matches{
                                          code_string matches {[ac0002]}
                 }
                        }
                   }
                }
}

Am I wrong?
Thank you very much

2

Cati Martínez schreef:

Hi,

I'm testing my implementation about the openEHR specification, with
some archetypes examples, and I'm finding things like this:

CLUSTER[at0018] occurrences matches {0 ..*} matches { name matches {
    CODED_TEXT matches {
          code matches {[ac0002]} }
}

I'd like to follow the specification as much as possible, and I think
that the correct way of expressing it, should have been:

CLUSTER[at0018] occurrences matches {0 ..*} matches { name matches {
    DV_CODED_TEXT matches {
          defining_code matches {
                               DV_CODE_PHRASE matches{
                                          code_string matches {[ac0002]}
                 }
                        }
                   }
                }
}

Am I wrong?
Thank you very much
  

I thought, a cluster should have elements? (required)

Bert

3

Yes,

it should have, I explain it better:

CLUSTER[at0018] occurrences matches {0 ..*} matches {
           name matches {
               CODED_TEXT matches {
                   code matches {[ac0002]}
                }
            }

items cardinality matches {0..*; unordered} matches {
......
}
}

I was talking about the property "name" of the cluster, I've seen this
simplification in some archetypes...
I've been reading about it, and now I think that it is allowed by the
openehr specification, It's a way of constraining the CODE_PHRASE. It
avoids implying anything about the formal model of the type of the
code attribute.

Thank you.

4

The correct structure is shown in teh following example:

                                    ELEMENT[at0004] matches { -- Height
                                        name matches {
                                            DV_CODED_TEXT matches {
                                                defining_code matches {
                                                    [local::
                                                    at0005, -- Height
                                                    at0006] -- Length
                                                }
                                            }
                                        }
                                        value matches {
                                            C_DV_QUANTITY <
                                                property = <[openehr::122]>
                                                list = <
                                                    ["1"] = <
                                                        units = <"cm">
                                                        magnitude =
<|0.0..1000.0|>
                                                    >
                                                >
                                            >
                                        }
                                    }

You could use an [ac000n] instead as well. Your example is structurally
correct; the attribute names just need to be fixed. Did you see this in
an archetype online?

- thomas beale

Cati Martínez wrote:

5

Ok, thank you,

Yes, this example is online at:
http://oceaninformatics.biz/archetypes/en/openEHR-EHR-OBSERVATION.microbiology.v1.html

6

Cati - these are very old and ADL 1.2. We have not been able to find any current link to these files. Please use the archetypes on openEHR.org
Cheers, Sam

Cati Martínez wrote:

(attachments)

OceanCsmall.png