Re: [Gnumed-devel] Medication lists - Corrections vs Changes

[Busser, Jim]

On 2013-07-23, at 12:14 AM, Karsten Hilbert <address@hidden> wrote:

> On Mon, Jul 22, 2013 at 09:44:46PM +0000, Jim Busser wrote:
> 
>> … why can't we remove the constraint which prevents for
>> a single patient to have, in clin.substance_intake, more
>> than one row with the same clin.substance,
> 
> Because it requires thinking through, and handling, the
> consequences.

First of all, we should agree that among the substances that are inactive, 
there should be no reason why we cannot allow duplicates of the same substance, 
to reflect

1) different dates started and discontinued
2) optionally, differences in the schedule or instructions

I would argue a difference between

- editing something to "make it right" (with the prior version going into the 
audit table)

and

- capturing, in the record information that -- for any period of time -- 
represents a new, or different state

and where the second purpose is what provides value to any *clinical* history 
in the record. Not the audit tables.

If this is disagreed with, then GNUmed should not bother with progress note*s* 
and should instead support a maximum of only one progress not*e* per patient, 
editing of which should always cause the *only* copy of the original to be 
relocated to the audit tables.

Presently, I can input three times that a patient has diabetes, with variant 
spellings so they won't even be able to be sorted together.

What exactly is the concern?

With medications, duplicates among the inactive are not even visible, unless 
one chooses to see them.

If the argument is to be that some clinician might "mistakenly prescribe" from 
among the inactive substances, then they can as easily mistakenly prescribe 
from among unique-but-inactive substances.

In fact the latter is an even bigger risk for non-duplicated substances, 
because at least any duplication in any prescription between an inactive and 
active substance would draw attention to the duplication.

-- Jim