Re: A question of Demographics

[jinny m uppal <juppal01@cs.fiu.edu>]

Mr Nicklin,
> I think Patient Demographics is just as much a part of the patient 
> record as it is a part of the management of the encounter.  However, 
> *Person* Demographics may be thought of as having a broader role, 
> where some persons have no patient role at all, but are nontheless 
> described in some detail (this would especially apply to clinicians).
there is a model that i have inmind for demographiccs, i dont know how to
explain in words, but I'll try.....

I believe we would like one model that captures all demographic
information, whether it is of a patient or a non-patient.
Tell me, are we looking to model 'persons' (read non-patients) as one
entity and patients as a different peer entity? Ofcourse, given a patient
you should know who the person is. 
Lets go back to the roles that I mentioned before. Supose we treat patient
as one of many roles a person can have, then "Patient demographics" is
nothing but a combination of 1."Person demographics" for that patient and
2. some demographic information that is specific to patients. do I make
sense? Maybe this figure will help:


    ----------                       -------------     ---------
    | Person | 1                  m  | PersonRole|     |Time   |
    |        |-----------------------|           |-----| Period|
    |________|                       |___________|     |_______|
       |                                 |
       |                                 |
    ----------------                 -------------
    | Person       |                 | Role-based|
    | demographics |                 |properties |
    |______________|                 |___________|
     

PersonRole may have many subtypes: Patient, Guarantor, Nurse, Physician
etc.. Time period is basically to capture the time during which this role
was active. For that matter all demographic properties will also have some
time period since names, addresses etc change all the time.

> A Person has some properties which are not dependent upon role.  
> Examples would be: DoB, Gender, Primary language, ethnicity, 
> Primary name, home address, next of kin.  We might also include such 
> things as professional qualifications, experience, demonstrable 
> skills, although we might argue that these have some broad role or, 
> at least, domain.
yes, i agree that there are some role-based and some person based
properties. At the same time, we are mostly interested in a peron only if
they have a role in our organization. In other words, we collect
name from a person when he is, say, a patient. So, shouldnt we somehow
keep track of which role provided "name" (which according to the above
paragraph is a property of a person)?

> corbamed/98-03-10: Person Demographics
I have read this article several times and find it most informative!

> On the other hand we have very role dependent properties for a 
> person.  A nurse may be authorised to perform certain types of 
> professional activities in certain organisational contexts, for 

Or: a patient has a relationship with a guarantor (another role), and this
is a role-based property. it would be a good idea to identify the roles in
a healh care organization and find their corresponding role-based
properties. Any work that has been done here...?

> It's a seriously heavy topic that kept us very busy at the CORBAmed
> meeting in Helsinki.  The relationships between healthcare providers
> and patients can be various.  Examples: