Re: [Gnumed-devel] Measurement groups

[Karsten Hilbert]

On Fri, Jul 05, 2013 at 07:44:01AM +0000, Jim Busser wrote:

> and the above then also brings me to a trickier question … if the test INR is
> 
> - normally < 1.2
> - target 2.0 - 3.0 for general anticoagulation
> - target 2.5 - 3.5 for mechanical valves and other high-risk conditions
> 
> is there a way that some condition suffered by the
> patient, or purpose of medication use, can guide "which"
> target range is referenced by GNUmed?

There is a way, sure. It requires quite a bit of fairly
complex code (ideally, an inferential engine, such as CLIPS)
- which needed to be written once. It also requires a
dedicated team of clinicians looking after the clinical
rules. That's the real crux, I suppose.

> Maybe this can only be practically managed at the level of
> 
> individual patient -- link table -- individual test
> 
> which each patient can have their own individualized
> target range, or alphanumeric "reference" information such
> as a comment that their result cannot be properly
> interpreted because of antibodies or other individual
> situation.

That is exactly what the fields you mentioned represeent. In
GNUmed one can define target and/or reference ranges *per
result* (not just per patient), and say why (comment).

Of course, the workflow is prone to be improved, say,
somehow display previous comments existing on the same
measurement type or some such.

Karsten
-- 
GPG key ID E4071346 @ gpg-keyserver.de
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346