gnumed-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnumed-devel] gmdrugs update (fwd)


From: Hilmar Berger
Subject: Re: [Gnumed-devel] gmdrugs update (fwd)
Date: Tue, 8 Oct 2002 00:08:35 +0200 (CEST)

On Thu, 26 Sep 2002, Horst Herb wrote:

> I have considerably updated / modified gmdrugs.sql

Well done, I really like to see how different ideas are merging to one
common implementation. 
Still, I see some problems (I'm sorry for stressing anyones patience so
much):
1. Right now, table *product* has one single link to *generic_drug*, which
seems to indicate that there can only be one generic drug in one product.
What about packages having more than one generic drug (e.g. tablet +
cream of different substances in one package) ?

2. In AMIS, there is additional information on adjuvants contained in a
drug, and in drugs containing more than one active substance ('generic
drug'), there is information on the amount of each substance in the drug
unit (e.g. 1mg Trimethoprim + 5mg Sulfamethoxazol + x mg adjuvant in one
tablet). I consider these data a useful piece of information and would
like to represent it in some table. 

3. 'is_compound' in *generic_drug*: I feel that this is not really
necessary. From my point of view Cotrimoxazole is a rare exception. I see
it this way: drugs are made of many substances, which have at least one
name (usually the generic name) and possibly more alias names (drug X,
drug X-base, drug X-hydroxide) which might have different weight per
drug, too. I would suggest the following (it is the way AMIS handles this 
problem):

single active substance:
product Y -> substance A (has : amount, ...) -> substance_name A_01
                                             -> substance_name A_02
          -> adjuvant a

where A_02 is an alias of A_01

more than one active substance that together have a single name:
product X -> substance C (has amount, ...) -> substance_name C_01
                 -> extended_substance T (has amount) -> substance_name T_01
                                                       -> substance_name T_02
                 -> extended_substance S (has amount)  -> substance_name S_01

I know that this is not UML, but I hope it gives at least an idea of what

I mean. I guess we will have to fine-tune everything once we get the
frontend up and running, which is what I should try to do right now :)
instead of constant criticism.

Hilmar







reply via email to

[Prev in Thread] Current Thread [Next in Thread]