On Thu, 3 Jul 2003, Tim Cook wrote: > On Thu, 2003-07-03 at 18:08, Andrew Ho wrote: > > > For example, we can see the TORCH social history "data model" in > > > > <socialhistory> > > <updated_time>2000/01/01</updated_time> > > No, that is not a data model. It is an exported record.
Tim, Thanks for pointing out that "social history" and "updated_time" are elements of TORCH's exported record. However, are they not also elements of TORCH's data model? My impression was that TORCH's exported record has a direct one-to-one mapping to TORCH's data model. If this understanding is incorrect, please let us know where we can find a true description of TORCH's data model. > If it were a data model you could tell if this was a "date" type, a > "time string" or a "character string". ... I understand that an exported record is but an example (=instance) of the data maintained by TORCH. Thus, while it is possible to ascertain something about the data model by looking at a part of an exported record, what I quoted was certainly not a full description of the much larger TORCH data model! For the purpose of illustrating the embedding of terminology within an EMR (which was my purpose in mentioning this fragment of TORCH exported record), I thought it sufficient to show just a small example. --> In this example, "social history" is one of the "terminology" used by TORCH. Will TORCH reconcile this "social history" term with similar terms (if any) in SNOMED CT ? If so, how do you plan to carry out the mapping / translation between the two? (As you can see, answering these questions may or may not require a full review of the entire TORCH data model. :-) Just so that you won't mistakenly conclude that I am singling out TORCH, GnuMed, OSCAR, TkFP, and other EMR's will also have to address the same challenge. Your willingness to share your thoughts on this can help many of us plan a better strategy. Best regards, Andrew --- Andrew P. Ho, M.D. OIO: Open Infrastructure for Outcomes www.TxOutcome.Org
