[EMAIL PROTECTED] wrote: > Hi Horst, in writing my answer to you last night I have been thinking about > some of these problems through the night so hopefully this answer helps. > May I shift the metaphorical description or modelling of what we are trying > to do away from your representation first. > We think of the the "patient record" as being a "patient story" because > "records" are for structured (typically relational) storage. For us each > episode in the story is a form that is completed by someone (anyone, even > the patient). The repository then has to collate the story from the > episodes and re-present it to any enquirer when asked- that is the > retreival aspect and we call those retrievals of "passive" episodes/forms. > However there is another type of episode which we call "active" which is a > derived or computed or analytical form and these can be either > automatically generated (as needed in an ED) or manually created as needed > by an analytical enquiry ( which importantly is not a retreival enquiry). > What you are now proposing is that we need a variant of the analytical > page which I will call a "compilated" page, that is a page that has > defined in it information from passive pages that are a compilation of > content from those pages. So the compilation page needs a function to > refer to content on other passive pages but also to accept conditions to > that content such as "most recent value" and even a floating cursor that > would show you a list of the last x number of values. Thanks for pressing > us on that issue. > > In terms of entering data for the wrong patient I'm not sure what > referential integrity you want to preserve. Clearly the data has to be > taken out of one episode and possibly the whole episode removed as it was > allocated to the wrong patient. Then the data placed in the correct > episode if it exists otherwise a new episode has to be created. > Conceptually that requires a function "transfer episode to another > patient" - not a problem computationally. So what happens if there is a > system crash in the hiatus between requesting the transfer and executing > the correct storage - that has to managed by the rollback functionality in > the repository store, and is the same issue for any management of data in a > persistent store. > cheers > jon > Yes. It's a transfer of the passive page and a deletion of the associated data in the compilation page and addition of that "longitudinal" data to the correct patient's compilation page. Does that stress your rollback function?
David
smime.p7s
Description: S/MIME Cryptographic Signature
_______________________________________________ Gpcg_talk mailing list [email protected] http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
