[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Demographics revisited-with PIDS (fwd)
Jinny
Apologies for delay in replying.
You ask about similarities between Pids and Demographics.
My feeling is that demographics and PIDS are very similar, and are,
indeed both very similar to clinical observations. All three are
concerned with properties of the person or subject. The difference
is in business use. Pids is about identification and the trait is a
relatively straigtforward thing that (if I recall correctly) has few
or no qualifiers except for its domain and trait-type. Demographics
is more "real-world" in that you may, for example, want to know about
relaionships between subjects and the nature of those relationship,
you may want to qualify with start and end dates etc.. Clinical
observations is the most qualified, with things like laterality,
clinical site, severity, urgency etc.. I think that it
may be theoretically possible for a clinical observations service
could provide the information needed for person identification and
demographics, as well as clinobs themselves, although it wouldn't
provide the kinds of matching algorithms that Pids can provide.
The distinction is a functional one. Once the COAS service and model
have been fully specified, we may want to look at it and decide
whether we provide demographic info via the COAS interface.
Does this help?
Peter
> --------------------------------------------------------------
> Hello all,
> Here are more issues for demographic service.
>
> As of now, many systems store some demographic information. I was
> wondering what is the vision for a demographic service, is it one server
> with an interface for demographic queries, or is it just a dmeographic
> service specification.
>
> In the Person Demographics paper, the traits listed are name,
> addresses etc.. How about traits like, height, weight, color of eyes,
> images of fingerprints. How would you classify them. Somebody called
> these as medical demographic properties. So are they covered under the
> same service as names etc..?
>
> How is PIDS going to be different from Demographics? Is demographic an
> extension of PIDS? Or a different service, that will use PIDS like all
> other services (COAS etc.) do?
>
> The following is from one of the previous messages from Peter:
> (response at the end)
>
> > [view diagram in 16 point courier]
> > >
> > > > ---------- ------------- ---------
> > > > | Person | 0..1 m | Role | |Time |
> > > > | |-----------------------| |-----| Period|
> > > > |________| |___________| |_______|
> > > > | |m
> > > > | |
> > > > ---------------- |
> > > > | Person | m |
> > > > | property |----------------------
> > > > |______________| <<context for person property
> > >
> > > Perfecto! This really makes sense to me!
> > > At the same time, isnt one person property going to be provided by
> > > one role. Do you have an example where it is really a :many
> > > relationship?
> >
> > You are right. I remember thinking about this for quite a while, but
> > I can't, for the life of me, remember why i put the "m" at the role
> > end of the association. If I remember, I'll let you know.
>
> I thought of the following example that explains an m:m relatinship:
> Consider a person who is a nurse and hence has a nurse-role. She has a
> name Jane. When she becomes a patient she still has the same name and
> will share the same name object with her other role.However, it is
> important to note that the same name with the same time period will be
> shared between two roles. I dont know how often this will happen in
> practice!
> Jinny.
>
> For details regarding list subscriptions and the list archive see:
> http://cadse.cs.fiu.edu/omg/halfem-rfi/
>
>
_____________________________________________________________________
Peter Nicklin, NHS IMC,
c/o CHSR, 21 Claremont Place, Newcastle Upon Tyne, NE2 4AA, UK
Tel: +44 191 230 3614 Fax: +44 191 230 4563 Mobile: +44 831 198319
X400: imc/G=Peter/S=Nicklin/O=nhs_imc/OU=cbs@mhs.attmail.com
For details regarding list subscriptions and the list archive see:
http://cadse.cs.fiu.edu/omg/halfem-rfi/