Re: [Gnumed-devel] Matching was Re: Mirth (2 of 3): HL7 import for the GNUmed project - test source message

[Jim Busser]

Luke asked on the lab wiki page (under todo) about what datetime stamps to use.

There exists:

1. the datetime when the blood, urine etc was obtained from the patient (we 
have this as clin_when)

2. the internal-to-the-lab-datetime when they generated the result AND any 
subsequent internal-to-the-lab-datetime when they altered / corrected / updated 
the result

3. the datetime when the outgoing HL7 message was created (which not happen 
until requestor connects to "pull" it)

4. the datetime when the result was picked up, and processed, by the receiving 
facility (praxis)

5. the datetime when the result was acknowledged by a end-user qualified to 
take responsibility for it ("signing")

6. the "on update" stamp for when any portion of a record is changed (we have 
this already as "modified_when")


Over on the wiki, Luke wondered whether we should establish a new field "last 
updated" and set it to #3 above:

        "MSH 007 - date/timestamp of message sound good enough?"
        
http://wiki.gnumed.de/bin/view/Gnumed/LabImporters#TODO_which_also_involves_reviewi

However I think that it would be better to use #2 which is

        OBX 014 Date-Time of Observation (Test resulted by laboratory timestamp)

because it disambiguates (together with OBR 025 Result Status) whether an 
actual result had been changed, and at what time that became known in the lab, 
and can help us to not have to resign anything when all the lab did was to 
resend us a redundant copy of unchanged information.

-- Jim