Re: [Gnumed-devel] Gnumed early usability? (was: Lab data support in OSCAR)

[Karsten Hilbert]

Jim,

this is fairly exciting. My current *immediate* goal is to
make GnuMed clinically usable for retrieval/storage/reuse of
lab data. This is what my parents need today and we are almost
done.

Can you be more specific on your *exact* requirements ? Eg.
what format data do you download, exactly how do you plan to
use that data, how would you want to enter patients into
GnuMed, how do you associate samples to the lab with the
results received from the lab. Give me someone who knows your
technical details and we can work out the issues easily.

> Circumstances have required that I see patients at more than one 
> location and matters have gotten to the point that for me, paper 
> results handling will no longer work safely / reliably without an 
> excessive time drain.
Do you need to access the data remotely or will you carry
around and installation of GnuMed/OSCAR and PostgreSQL/MySQL
on a laptop ?

> I have contacted some local IT people (the same couple of fellows who 
> installed and support OSCAR for a local clinic) to determine some 
> server and support requirements which, at this stage, are partly 
> generic. If people are interested, I can post some of their suggestions 
> about hardware, accessibility and other planning considerations.
Surely so.

> I will prefer to use GNUMed as I think it's been designed more 
> carefully, and I have been unable to penetrate "OSCAR-think"... they 
> seem to not offer / support the same type of dialog as we have at 
> GNUMed. My conundrum remains the uncertain timeline for GNUMed.
The timeline is mainly defined by "I need *this* *then*". Some
basic functionality must be there of course but beyond that it
needs people with a vision overcoming the intertia who say
"GnuMed is the way I want to go so I will make *this* particular
thing happen." This doesn't mean they must write it themselves
but they must organize contacts and data and interface with
the people in the know. Like you telling me: "Look, this and
this is the requirements, this is the data format, here are
some data files for testing, I want to achieve that and that.
Here's a contact for questions."

> Option one would be for me to set up OSCAR, and to use it purely for 
> the receipt of electronic lab results, until such time as GNUMed is 
> more ready for use. I could then freeze and lookup in OSCAR historical 
> data, or I suppose I could archive the data downloads and re-import 
> them into GNUMed once it is ready.
Both would work. Why not queue up results for import into
both? Then keep *using* OSCAR until GnuMed fills your needs.

> Option two would be to use GNUMed, but ONLY if:
> - its table design,
Quite stable re patient identity and results handling.

> and widgets by which to
> create
- afaict usable but (?) lacking

> and find patients / 
works, stable

> match results
works, fairly stable, brand-new, however

> / view / 
works, roughly stable

> print patient data,
no code exists

> are quite doable in the next 
> 3-6 months
I *will* install the German lab handling within the next 2
months in our practice *and use it*.

> - its table design is stable,
again, re the lab/identity stuff it is quite stable

> and/or we create and maintain a reliable 
> path (update scripts) by which to update any real clinical data into 
> any table modification(s)
we will have to do that anyways, it won't be click-a-button
first, though

> The trick would be to identify people who, 
> with their code, would be  'compatible' with GNUMed.
They can always submit patches/code to me. This has worked
exceptionally well with Carlos, his code getting more
"compatible" steadily.

> I am not sure how 
> it would work... I could serve as the intermediary,
That would of course be the greatest guarantor for success.

> but of course they 
> will have technical questions beyond my ability,
no problem, post them or write us

> and of course Karsten, 
> Ian, Carlo and other key (but recently lower profile) GNUMedders would 
> need to be able to bless, perhaps help integrate, whatever is being 
> added, Do we think a positive scenario can be fashioned out of this?
It depends on whether you can spend (or find someone willing
to spend) enough effort into working with us. I for one am
certainly willing to work with you on my end.

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346